INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

THERE IS EVIDENCE THAT atrial natriuretic peptide (ANP), in addition to its well-defined acute hypotensive effect, may also play a physiological role in chronic regulation of blood pressure. Long-term infusions of ANP into conscious animals lead to a sustained decrease in arterial blood pressure (ABP) (8, 18, 19, 43). Similarly, transgenic mice overexpressing a transthyretin-ANP fusion gene (TTR-ANP) are markedly hypotensive compared with nontransgenic (NT) control mice, in association with lifelong 8- to 10-fold elevation in plasma ANP (47, 49). In contrast, "knockout" mice in which synthesis of ANP (/) (23) or its guanylate cyclase-A (GC-A) (29) receptor is interrupted by targeted homozygous disruption of their native genes develop hypertension relative to their wild-type (+/+) siblings. The ANP knockout mice develop salt sensitivity of ABP after prolonged high dietary salt intake, in association with an apparent inability to regulate plasma renin activity (33), and genetic decrease in GC-A receptor expression also leads to salt-sensitive hypertension (39).

The chronic hypotensive effect of ANP is brought about by a reduction in total peripheral resistance, consequent to generalized vasodilation in regional vascular beds (3, 8, 18, 19, 43). The resistance vasculature has, however, been reported to be insensitive to direct relaxation by ANP (14, 25, 41), suggesting that the chronic ANP-dependent dilation of resistance vessels is mediated by intermediary vasoeffector mechanisms whose single or cumulative action(s) reduces peripheral vascular resistance. In this regard, the vascular endothelium (VE), which is known to play a major role in local regulation of vascular tone (22, 48), may act as a modulator of chronic ANP-dependent vascular effects, especially in light of recent evidence that ANP may affect synthesis and secretion of locally acting vasoactive substances from VE (21, 31, 35, 46, 50) as well as their effects on target tissues (5, 24, 40). ANP inhibits production of vasoconstrictor endothelin-1 (ET-1) (13, 21, 26, 50) and stimulates synthesis of the vasodilators C-type natriuretic peptide (CNP) (37) and possibly nitric oxide (NO) (31, 45) from vascular endothelial cells. Furthermore, ANP attenuates the pressor effect of ET-1 (40) and may potentiate the vasodilatory action of CNP by downregulating endothelial C-type receptors (5, 24) involved in clearance of natriuretic peptides.

These findings imply that such ANP-endothelium interactions, if operative in vivo, may represent a novel mechanism by which ANP exerts its chronic hypotensive effect. Thus ANP could reduce peripheral vascular resistance chronically by regulating the synthetic activity of VE in the resistance vasculature, so that the vasoconstrictor moiety associated with ET-1 is reduced and vasodilatory CNP and NO are potentiated, and/or by modulating target responses to these substances. In the present study, we measured steady-state concentrations of ET-1, CNP, and endothelial constitutive NO synthase (ecNOS) in several tissues of ANP-overexpressing transgenic mice (TTR-ANP) and pro-ANP gene knockout mice (/) and their respective wild-type controls (NT, +/+) to determine whether the chronic effect of ANP on blood pressure is attributable to differences in the synthesis of these paracrine vasoactive factors by resistance vessel endothelium. In addition, we measured changes in ABP and heart rate (HR) in / and +/+ mice after antagonism of ET_A/B receptors, inhibition of NOS activity, or immunoneutralization of guanylate cyclase-B (CG-B) receptors to test whether the responsiveness of these target cardiovascular parameters to endogenous ET, NO, and CNP is affected by the level of ANP activity.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Animals. Production and molecular analysis of TTR-ANP mice and ANP knockout mice have been described in detail previously (23, 47). Male TTR-ANP mice and their respective NT littermates (C3HeB/FeJ background) and / and +/+ mice (C57BL/6J background) of both sexes, 8-13 mo old and weighing 27-42 g, were used in this study. The TTR-ANP mice were kindly provided by Dr. L. J. Field, Krannert Institute of Cardiology (Indianapolis, IN). The / and +/+ animals were obtained from a resident colony. The animals were housed according to sex and genotype in groups of two to four per cage and kept at ambient 23°C and 40% humidity in a room with a 12:12-h light-dark cycle.

Materials. All materials for the ET-1 and CNP RIA, including porcine ET-1 and CNP-(122) standards, ¹²⁵I-CNP-(122), rabbit polyclonal anti-ET-1 and anti-CNP sera, goat-anti-rabbit IgG serum, and normal rabbit serum were from Peninsula Laboratories (Belmont, CA), with the exception of ¹²⁵I-ET-1 (NEN DuPont, Markham, ON, Canada). Acrylamide, bis-acrylamide, Dowex-AG50W-X8 (H⁺) resin, Poly-Prep chromatography columns and protein assay kit (Bradford) were from Bio-Rad (Mississauga, ON, Canada). The ecNOS monoclonal antibody was from Transduction Laboratories (Lexington, KY). L-[³H]arginine, Hybond-C nitrocellulose membranes, and an enhanced chemiluminescence (ECL) kit were purchased from Amersham (Oakville, ON, Canada). SB-209670 was a generous gift of SmithKline Beecham (King of Prussia, PA). 3G12 monoclonal antibodies were kindly donated by Genentech (San Francisco, CA). CNP-22 was from Phoenix Pharmaceuticals (Mountain View, CA). All other reagents were purchased from Sigma Chemical (St. Louis, MO).

Tissue preparation. ET-1, CNP, and ecNOS immunoreactivities were measured in whole organ homogenates of kidney, heart, lung, brain, and hindquarter skeletal muscle. The tissues were frozen in liquid nitrogen immediately after dissection from anesthetized mice (Inactin, 100 µg/g body wt) and stored intact at 70°C until assayed. The tissues were washed with two rinses of cold (~4°C) PBS (pH 7.4), cut into small pieces, and homogenized (~5:1, ml:g tissue) on ice in protein lysis buffer (50 mM Tris · HCl, pH 7.4, 0.1 mM EDTA, 0.1 mM EGTA, 1 mM PMSF, 2 µM leupeptin, 2 µM pepstatin A, and 0.1% -mercaptoethanol) with two to four successive bursts (11,000 rpm, 10 s) as required, from a Polytron (PT 3300, Brinkmann Instruments, Littau, Switzerland). The homogenates were mixed with glycerol (10% vol/vol) and centrifuged at 10,000 g for 15 min in a refrigerated (4°C) Microfuge to remove cellular debris.

Tissue concentration of ET-1 and CNP. ET-1 and CNP immunoreactivities were measured by RIA, according to the method previously described (32). For ET-1 and CNP RIA, 100 µl of the cleared supernatants were incubated (1:1:1) with the respective rabbit polyclonal anti-ET-1 or anti-CNP sera and ¹²⁵I-ET-1 or ¹²⁵I-CNP tracers. The sensitivity of the ET-1 and CNP assays was 5.8 and 3.8 pg/100 µl respectively. Cross-reactivity of the ET-1 antiserum with big ET-1, ET-2, ET-3, and CNP was 35, 7, 7, and 0%, respectively. CNP antiserum cross-reactivity with CNP-(153) and ET-1 was 100 and 0%, respectively. All values for ET-1 and CNP were normalized for total protein concentration in the sample, as determined by the Bradford method.

Tissue concentration of ecNOS. ecNOS immunoreactivity in the tissue homogenates was measured by Western blot (16). The homogenates were treated with 20 mM 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate for 20 min at 4°C for solubilization of membrane-bound protein. One hundred micrograms of total protein extract were electrophoresed, according to the method of Laemmli, in 6% SDS-polyacrylamide gels (Mini-Protean II, Bio-Rad) under reducing and denaturing conditions, and transferred to Hybond-C nitrocellulose membranes by electroblotting. The membranes were blocked overnight (12-15 h) at 4°C with 4% BSA (fraction V, Sigma Chemical) in Tris-buffered saline and 0.1% Tween 20 (TBS-T; pH 7.5) and incubated with 1:2,000 dilution of mouse anti-ecNOS monoclonal antibody for 1 h. After three washes (15 min each) with TBS-T, the membranes were incubated with horseradish peroxidase anti-mouse IgG (1:5,000) secondary antibody for 2 h. The ecNOS signal was detected by ECL and quantified with Image version 1.52 (National Institutes of Health, Bethesda, MD).

Total NOS enzyme activity. NOS enzyme activity in the tissue homogenates was determined by measuring the rate of formation of L-[³H]citrulline from L-[³H]arginine (6). The tissues were homogenized as described above in lysis buffer (25 mM Tris · HCl, pH 7.4, 1 mM EDTA, and 1 mM EGTA). Fifty microliters of supernatant were incubated at 37°C for 1 h with reaction buffer (50 mM Tris · HCl, pH 7.4, 2.5 mM NADPH, 10 µM tetrahydropterin, 10 µM flavin adenine dinucleotide, 10 µM flavin adenine mononucleotide, 50 U calmodulin, 2 µM L-[³H]arginine, 10 µM L-arginine, and 5 mM L-valine). Corresponding blank reactions were prepared by combining each sample with reaction buffer supplemented with 300 µM each of N^G-nitro-L-arginine methyl ester (L-NAME) and N^G-monomethyl-L-arginine (L-NMMA). The reaction was terminated by adding 400 µl of ice-cold stop buffer (50 mM HEPES, pH 5.5, and 5 mM EDTA). The total volume was applied to Poly-Prep chromatography columns containing 1 ml of Dowex AG 50W-X8 (Na⁺) preequilibrated with 2 ml of stop buffer. L-[³H]citrulline was eluted with 2 ml of distilled water. The radioactivity was quantified in 10 ml of scintillant in a -liquid scintillation counter. Each blank was subtracted from its corresponding noninhibited sample. The results are expressed as counts per minute (CPM) per minute per milligram of protein.

In vivo studies. Mice (/, +/+) were anesthetized with Inactin (150 µg/g body wt ip) and kept at a body temperature near 38°C with a heat lamp. After tracheostomy, a jugular vein and carotid artery were cannulated with catheters (300- to 400-µm diameter) fashioned from PE-50 tubing for intravenous infusion and measurement of mean blood pressure (ABP) and HR, respectively. On completion of surgery, 0.12 ml of isotonic saline containing 2.25% BSA and 1% glucose was infused over 15 min as a priming dose, followed by constant infusion of the same solution at 0.12 ml/h for the duration of the experiment. The experiment was begun after an additional 30-min equilibration period. BP and HR were monitored continuously and recorded at 10-min intervals with a small displacement pressure transducer (model RP 1500, Narco Systems) connected to a MacLab/4e data acquisition system. Each experiment consisted of a 30-min control period, followed by a 30-min infusion of L-NAME (0.12 mg · kg¹ · min¹), SB-209670 (100 µg · kg¹ · min¹), or 3G12 (20 µg · kg¹ · min¹) to inhibit NOS (17), ET_A/B (27), or GC-B (11) receptors, respectively. At the end of the experiment, the inhibitors were coinfused with L-arginine (1.2 mg · kg¹ · min¹), ET-1 (100 ng · kg¹ · min¹), or CNP (100 ng · kg¹ · min¹), respectively, to assess the effectiveness and selectivity of inhibition.

Statistical analysis. All results are expressed as means ± SE. Unpaired t-test was used to compare differences between mutant (TTR-ANP, /) mice and their respective wild-type controls (NT, +/+), with respect to ET-1, CNP, and ecNOS concentration and NOS activity. One-way ANOVA was used to compare ABP and HR responses to the different treatments within and between genotypes. P < 0.05 was considered to indicate statistically significant difference.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

The baseline hemodynamic and physical characteristics of the various genotypes are summarized in Table 1. ABP was significantly lower (P < 0.0001) in TTR-ANP mice and higher (P < 0.0001) in / mice than in the respective NT and +/+ control mice. TTR-ANP and / mice did not differ significantly from their controls with respect to HR or hematocrit. Heart weight-to-body weight ratios were significantly decreased (P < 0.0001) and increased (P < 0.0006) in TTR-ANP and / mice, respectively, compared with their wild-type controls, in consonance with previously described lower and higher total peripheral resistance in TTR-ANP (3) and / mice (L. G. Melo, A. T. Veress, U. Ackermann, S. C. Pang, T. G. Flynn, and H. Sonnenberg, unpublished observations), respectively. Kidney weight-to-body weight ratios did not differ between mutant and control mice.

View this table: [in this window] [in a new window]   Table 1.   Baseline physical and hemodynamic characteristics of TTR, NT, /, and +/+ mice

Table 2 shows the concentration of immunoreactive ET-1 in tissues of TTR-ANP, NT, /, and +/+ mice. Variable amounts of ET-1 were detected in the different tissues, but the distribution and abundance patterns were similar in all genotypes, with lung and skeletal muscle expressing the highest and lowest concentrations, respectively. No statistically significant differences in tissue ET-1 concentration were observed between mutant (TTR-ANP, /) mice and the corresponding control (NT, +/+) mice. However, there was a prominent strain-related difference in kidney ET-1 concentration between animals of the TTR-ANP, NT background (C3HeB/FeJ) and those of the /, +/+ background (C57BL/6J).

View this table: [in this window] [in a new window]   Table 2.   Tissue concentration of ET-1 in TTR, NT, /, and +/+ mice

The tissue concentrations of immunoreactive CNP are shown in Table 3. As for ET-1, the abundance patterns of CNP in the tissues were similar between the various genotypes (Table 3). As expected, the brain contained the highest concentration of CNP. The lowest concentration was detected in skeletal muscle. There were also strain-related differences in kidney CNP concentration, with animals of the /, +/+ strain having lower concentrations than those of the TTR-ANP, NT strain. No significant differences in tissue CNP concentrations were found, however, between mutant mice and their respective control mice.

View this table: [in this window] [in a new window]   Table 3.   Tissue concentration of CNP in TTR, NT, /, and +/+ mice

Table 4 shows the concentration of immunoreactive ecNOS in the different tissues from the various genotypes. As for ET-1 and CNP, the relative tissue abundance of ecNOS was similar in all genotypes and did not differ statistically between mutant and control mice. The heart was found to have the highest ecNOS concentration per unit of total homogenate protein, and comparable amounts were found in all other tissues. Likewise, total NOS enzyme activity in the tissue homogenates (Table 5) was similarly distributed in all genotypes. NOS activity was severalfold higher in brain and muscle than in the other tissues, reflecting the contribution of neuronal NOS, which was detected only in these two tissues (Melo and Sonnenberg, unpublished observations).

View this table: [in this window] [in a new window]   Table 4.   Tissue concentration of ecNOS in TTR, NT, /, and +/+ mice

View this table: [in this window] [in a new window]   Table 5.   Total NOS enzyme activity in tissues of TTR, NT, /, and +/+ mice

The cardiovascular effects of antagonism of ET_A/B receptors with SB-209670 in / and +/+ mice are shown Fig. 1. Basal ABP differed significantly between genotypes (Fig. 1A). Basal HR was lower in / mice (Fig. 1B), but this difference did not reach statistical significance. The antagonist reduced ABP significantly (Fig. 1A) and decreased HR slightly (Fig. 1B) in both genotypes. The effects of the antagonist on ABP and HR were not reversed by coinfusion (10 min) with a dose of ET-1 previously titrated to produce an increase of 30-35 mmHg in ABP. The relative changes (%) in ABP and HR after SB-209670 administration were quantitatively similar in both genotypes (Table 6).

View larger version (24K): [in this window] [in a new window]   Fig. 1.   Average arterial blood pressure (ABP) (A) and heart rate (HR) (B) during baseline conditions (solid bars) and after infusion (30 min) of endothelin receptor A/B (ETA/B) antagonist SB-209670 (hatched bars) in / (n = 8) and +/+ mice (n = 8). Baseline ABP differed significantly between genotypes (* P < 0.0002). Significant differences were also found between baseline ABP and SB-209670 in both genotypes (/ mice, # P < 0.005; +/+ mice, #P < 0.0001) and between baseline ABP and SB-209670 + ET-1 (+ P < 0.05).

View this table: [in this window] [in a new window]   Table 6.   Change in ABP and HR in response to infusion (30 min) of SB-209670, L-NAME, or 3G12 in / and +/+ mice

The effect of immunoneutralization of GC-B receptor activity with 3G12 monoclonal antibody on ABP and HR is shown in Fig. 2. Basal ABP and HR for the two genotypes were similar to those of the previous group (Fig. 1). The antibody slightly, but not significantly, increased ABP (Fig. 2A) and HR (Fig. 2B) in both genotypes by comparable relative (%) amounts (Table 6) and reduced the hypotensive effect of CNP (10 min) infused at a dose producing a 20- to 25-mmHg decrease in ABP (Fig. 2A).

View larger version (29K): [in this window] [in a new window]   Fig. 2.   Average ABP (A) and HR (B) during baseline conditions (solid bars) and after immunoneutralization of GC-B receptors (intravenous infusion, 30 min) with 3G12 (hatched bars) in / (n = 5) and +/+ mice (n = 5). Baseline ABP differed significantly between genotypes (* P < 0.05). No significant differences in ABP or HR were found between baseline conditions and 3G12.

Figure 3 shows the effect of inhibition of endogenous NOS activity with L-NAME on ABP and HR in / and +/+ mice. As in the previous two groups, basal ABP was significantly higher in / mice than in +/+ mice (Fig. 3A), and HR was slightly, but not significantly, lower in / mice than in +/+ mice (Fig. 3B). L-NAME increased ABP significantly in both genotypes (Fig. 3A). HR did not change in / mice but increased slightly in +/+ mice (Fig. 3B). The effects of L-NAME on ABP and HR were not prevented by coinfusion (10 min) with a 10-fold molar excess of L-arginine. Comparable changes (%) in ABP and HR were found between genotypes after L-NAME infusion (Table 6).

View larger version (32K): [in this window] [in a new window]   Fig. 3.   Average ABP (A) and HR (B) during baseline conditions (solid bars) and after inhibition of NOS (intravenous infusion, 30 min) with NG-nitro-L-arginine methyl ester L-NAME (hatched bars) in / (n = 6) and +/+ mice (n = 6). Baseline ABP differed significantly between genotypes (* P < 0.0001). Significant differences were found between baseline ABP and L-NAME and L-NAME + L-arginine (L-Arg) (/ mice, #, + P < 0.0001; +/+ mice, #, + P < 0.0001) in both genotypes as well as between genotypes after L-NAME (** P < 0.005).

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

A contributory role of ANP to chronic regulation of ABP is implied by observations that lifelong genetic alterations in the level of endogenous ANP activity uncover a tonic hypotensive effect of this hormone (23, 29, 47), which is associated with vasodilation of the resistance vasculature (3). The rationale for the present study is based on the premise that the chronic hypotensive effect of ANP may be mediated by an intermediary vasoeffector mechanism, inasmuch as previous observations (14, 25, 41) had shown that the resistance vasculature per se is, at least acutely, insensitive to ANP. Our aim was to examine whether the chronic vascular effects of ANP are attributable to differences in activity of the locally acting vasoactive substances ET-1, CNP, and NO. We measured concentrations of ET-1, CNP, and ecNOS in whole organ homogenates from TTR-ANP, NT, /, and +/+ mice as an index of local synthesis from resistance vessel endothelium (15, 30, 36, 44, 46) and tested whether the responsiveness of ABP and HR to endogenous ET-1, CNP, and NO was affected by the level of ANP activity. Our results show that neither the synthesis of ET-1, CNP, or NO from the resistance vasculature nor their effects on cardiovascular regulation were altered by the chronic level of ANP bioactivity, thus indicating that long-term vascular effects of ANP are not determined by differences in activity of these endothelial modulators.

The lack of a detectable effect of ANP on the synthetic activity of VE is surprising. Endothelial cells have an abundance of both C and GC-A ANP binding sites (27, 28). Activation of the preponderant C-receptor subtype leads to inhibition of basal ET-1 synthesis (21) and may potentiate basal ecNOS activity via inhibition of adenylate cyclase (31, 35), whereas GC-A-dependent elevation in cGMP production inhibits agonist-mediated ET-1 synthesis (13, 26) and stimulates ecNOS (20, 45) and CNP synthesis (37). We hypothesized that the chronic vascular effects of ANP could be affected by these receptor-mediated interactions with the endothelium. Thus tonic activation of resistance vessel endothelium by ANP in TTR-ANP would lead to hypotension, concomitant with potentiation of CNP and ecNOS and inhibition of ET-1, whereas the absence of such modulation in ANP knockout (/) mice would result in hypertension. This is clearly not supported by our findings. A possible cause for the lack of differences in synthetic activity of VE between TTR-ANP and NT is that endothelial cell responsiveness to ANP is attenuated in the TTR-ANP mice, due to homologous receptor downregulation (24). Conversely, absence of ANP may result in an increase in the number of its endothelial receptors, and the consequent increase in basal activity of these receptors may account for the lack of differences in synthetic activity of VE between / and +/+ mice.

Alternatively, ANP may exert its effects on peripheral vascular resistance by modulating the responsiveness of target cardiovascular effects of endothelial factors. A functional interaction between ET-1 and ANP, for example, is suggested by the almost identical distribution of receptors for both peptides in many target tissues (38). ANP attenuates vascular reactivity to ET-1 by reducing intracellular calcium concentration (4) and possibly, by decreasing the number of ET_A receptors in vascular smooth muscle (VSM) (1, 12). ANP may also lead to upregulation of soluble guanylate cyclase in VSM (42) via its inhibitory action on adenylate cyclase (1) and prolong the vascular effects of CNP by downregulating C receptors (24). Our in vivo findings indicate that endogenous ET-1 and NO, and to a lesser extent CNP, participate significantly in cardiovascular regulation in / and +/+ mice, but their relative contributions to maintenance of basal ABP and HR are not influenced by the chronic level of ANP activity. These observations suggest that ANP interactions with the target cardiovascular effects of ET-1, NO ,and CNP are ineffective in the chronic state or else that they may be overcome by counteracting influences on cardiovascular function.

A direct effect of ANP on the resistance vasculature, however unlikely, cannot be totally discounted. On the basis of the similarities in cardiovascular phenotype between the GC-A and the ANP knockout models, it is tempting to speculate that the chronic hypotensive effect of ANP may be due to direct GC-A-mediated relaxation of the resistance vasculature. However, the scarcity of GC-A receptors and relative insensitivity to ANP-mediated cGMP production in resistance vessel smooth muscle (2, 9) would preclude a role of this pathway in ANP-induced dilation of the resistance vasculature. Indeed, the physiological significance of cGMP-mediated ANP vasodilation is doubtful, as it appears to be a pharmacological characteristic of large arteries (7, 51). Nevertheless, the GC-A knockout model firmly establishes the role of this receptor in chronic regulation of blood pressure. To this extent, the ANP and CG-A models share similarities in cardiovascular function, at least with respect to actions of ANP that are mediated by the GC-A receptor. We have preliminary evidence that hypertension in / mice may be associated with an increase in neurogenic tone, since autonomic ganglionic blockade leads to a significantly greater decrease in ABP in these mice than in the wild-type controls (L. G. Melo, A. T. Veress, U. Ackermann, S. C. Pang, T. G. Flynn, and H. Sonnenberg, unpublished observations). Because ANP exerts its generalized sympatholytic effects via the GC-A receptor (2), this may be the common mechanism by which hypertension is expressed in both GC-A and ANP knockout mice.

In conclusion, the results of the present study show that over a wide range of chronic ANP activity, neither the synthesis of ET-1, CNP, and NO from the resistance vasculature, nor their actions on the cardiovascular system are affected, thus indicating that the chronic effect of ANP on vascular resistance is not mediated by the endothelium.