Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

The renin-angiotensin system (RAS) and the sympathetic nervous system (SNS) are important regulators of cardiovascular function. Angiotensin II (Ang II), the effector peptide of RAS, elicits potent vasoconstrictor effects on interacting with specific Ang II receptors in vascular smooth muscle (Mendelsohn, 1985). Experimental data indicate that it also modulates peripheral sympathetic neurotransmission in vitro and in vivo by enhancing the release of the adrenergic transmitter in several tissues (Rand et al., 1990) and augmenting the effects of the transmitter at the postjunctional sites (Nicholas, 1970), thereby exerting a facilitatory effect at the adrenergic neuroeffector junction. These observations have been substantiated in studies with pithed rats wherein endogenous Ang II was shown to facilitate sympathetic neurotransmission after spinal cord stimulation (Wong et al., 1992). Ang II-induced facilitation of peripheral adrenergic transmission has also been demonstrated in hand veins and resistance vessels of humans (Benjamin et al., 1988; Seidelin et al., 1991). Consistent with these findings, treatment with angiotensin-converting enzyme inhibitors has been reported to decrease circulating norepinephrine (NE) concentrations (Wenting et al., 1983).

The nexus between the SNS and the RAS could have serious implications in the pathogenesis of various cardiovascular disorders. An increase in sympathetic neural activity is believed to be important in the pathogenesis of hypertension in spontaneously hypertensive rats (Judy et al., 1976). Consistent with this view, an increased transmitter turnover was detected in some vascular beds and in the heart during the development of hypertension (Adams et al., 1989). The activity of the SNS was also found to be augmented in congestive heart failure (CHF) (Rector et al., 1987; Francis, 1989). The ensuing increase in cardiac NE spillover has been associated with malignant ventricular arrhythmia (Meredith et al., 1991), presumably accounting for the positive correlation noted between plasma NE concentrations and mortality rates in this condition (Rector et al., 1987). Furthermore, Ang II levels are also reportedly elevated in CHF (Francis, 1989), suggesting that the augmentation of the activity of the SNS in this condition is secondary to an upsurge in the levels of the peptide.

Ang II elicits its vast array of pharmacological actions by binding to specific receptors located on the membranes of its target cells. Based on the differential binding affinities of selective ligands, losartan, CGP42112, and PD 123319, two receptor subtypes were identified and subsequently categorized as Ang II type 1 (AT₁) and type 2 (AT₂), respectively (Chiu et al., 1989; Whitebread et al., 1989; Bumpus et al., 1991). The AT₁ subtype appears to be the principal mediator of all the known physiological actions of Ang II, whereas the function of the AT₂ subtype is poorly defined at present. The receptor mediating the prejunctional facilitatory effects of Ang II on sympathetic neurotransmission was proposed to be of the AT₁ subtype based on the antagonistic effect of losartan, the prototypical AT₁ receptor blocker (Tofovic et al., 1991; Wong et al., 1992; Foucart et al., 1996). Given the prognostic and pathophysiological significance of the interaction between the RAS and the SNS regarding cardiovascular disorders, it was considered desirable to ascertain and quantify the inhibitory effects of valsartan, a potent Ang II receptor blocker (Criscione et al., 1995), on the prejunctional actions of Ang II. An ancillary goal of the present investigation was to compare the potency of valsartan with those of three other AT₁ receptor blockers (losartan, eprosartan, and irbesartan) for effecting this inhibitory action. The isolated rat left atrial preparation was used as a model system in this investigation because it is amenable to direct measurements of the parameters of interest without being encumbered by confounding physiological mechanisms operative in more complex in vivo systems.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animal care and experimental procedures were in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and were approved by the Animal Care Committee of Novartis Institute for Biomedical Research. Male Sprague-Dawley rats (300-400 g) were anesthetized with sodium pentobarbital (65 mg/kg i.p.). After opening the chest, the hearts were removed, cannulated at the aorta, and immediately suspended in a Langendorf apparatus for retrograde perfusion (4 ml/min) of the coronary system with warm (37.5°C) oxygenated (95% O₂, 5% CO₂) physiological salt solution (PSS). The PSS contained 118.0 mM NaCl, 4.7 mM KCl, 1.03 mM KH₂PO₄, 0.45 mM MgSO₄, 2.5 mM CaCl₂, 25.0 mM NaHCO₃, 11.1 mM dextrose, 0.14 mM ascorbic acid, and 0.067 mM disodium EDTA. The procedure described by Foucart et al. (1996) was adopted with some modifications for the assessment of agents affecting the overflow of the radiolabeled neurotransmitter. Briefly, the left atrial walls were dissected from the suspended and perfused hearts and incubated for 25 min in 5 ml of PSS to which [³H]NE (5.7 µCi/ml) was added. The radiolabeled incubation solution was maintained at 37.5°C and continually oxygenated as before. At the end of the incubation period, the tissues were lightly blotted on a filter paper to remove superficially bound [³H]NE and transferred to 0.5-ml perfusion chambers (one atrium per chamber). The perfusion system used was a Brandel suprafusion system (SF-06; Brandel Inc., Gaithersburg, MD). The perfusion rate was set at 0.4 ml/min, and the temperature of the perfusion solutions was kept constant at 37.5°C with the help of a water bath and an environment cage. Electrical stimulation of the preparations was performed by a multichannel electrical stimulator (ES2-069-55; Brandel Inc.) and platinum screened electrical probes. Effluents were collected into 8-ml vials placed in vial trays. From reagent to effluent, each channel was completely isolated from the others.

The atria were washed for 65 min with PSS during which a priming stimulus (PS; 3 Hz, 50 mA, 1 ms, 60 s) was given at 50 min to eliminate the superficial or loosely bound [³H]NE. The effluent was subsequently collected once every 5 min for a total of 70 min (14 sampling periods). During this period, the atria were field stimulated twice (S₁ and S₂; 3 Hz, 50 mA, 1 ms, 60 s) at 20 and 55 min as described by Foucart et al. (1996). Thus, initiations of PS, S₁, and S₂ were each separated by 35 min. Ang II was included in the perfusing solution 20 min before the second stimulation (S₂) in select experiments. Test compounds (or vehicle) were typically included in the perfusing solution 20 min before S₁. Thus, the tissues were initially treated with each of the test compounds (or vehicle) for 35 min and then exposed to Ang II for 20 min in the continued presence of the agents before being subjected to S₂. The effects of test compounds on the control (or basal) stimulation-induced (SI) efflux were also ascertained by including the compounds in the perfusion solution 20 min before either S₁ or S₂.

At the end of the experiment, the atria were lightly blotted, weighed, and placed in 7-ml vials, each containing 1 ml of Soluene-350 (Packard Instrument Co., Meriden, CT). The vials were shaken at 50°C for 2 h to solubilize the tissues. The radioactivity present in the solutions (effluents, solubilized tissues) was determined by liquid scintillation counting (Beckman LS6500; Beckman Instruments, Irvine, CA) after the solutions were mixed with 5 ml of Pico-Fluor 40 (Packard Instrument Co., Meriden, CT). The spontaneous (resting) radioactive outflow during the 5-min period before the stimulation was measured, and the SI component of the outflow of radioactivity was determined by subtracting the resting radioactivity from the total radioactivity content of the 5-min sample collected during the stimulation period. The SI outflow of radioactivity measured during the second period of stimulation (S₂) was expressed as the percentage of the first period of stimulation (S₁). The values were initially standardized for the total radioactivity in the tissue at that time point by expressing them as fractional releases (FR) of radioactivity, and the ratio % FR₂/FR₁ was then used to indicate the effects of pharmacological interventions.

Drugs. Desipramine hydrochloride, oxymetazoline hydrochloride, fenoterol hydrobromide, and Ang II (synthetic, human sequence) were obtained from Sigma Chemical Co. (St. Louis, MO). Valsartan was synthesized in-house at Novartis (Summit, NJ). Losartan was a gift from DuPont Merck Pharmaceuticals (Wilmington, DE). Eprosartan and irbesartan were synthesized in Novartis (Basel, Switzerland). Stock solutions (10² M) of desipramine, oxymetazoline, and fenoterol were prepared fresh each day in PSS. Stock solutions (10³ M) of Ang II prepared in deionized water were stored in 100-µl aliquots at 80°C. All other compounds were prepared fresh each day in DMSO to a concentration of 10² M. Further dilutions were made in PSS. Tritiated NE (NE, levo-[ring-2,5,6-³H]) was purchased from NEN Life Science Products, Inc. (Boston, MA) with a specific activity of 62.3 Ci/mmol and a radioactive concentration of 1 mCi/ml.

Statistical Analysis. All data are expressed as mean ± S.E. Student's t test (two-tailed, unpaired) was used to determine statistical significance of differences between means of control and treatment groups. An ANOVA followed by Dunnett's test was used for multiple comparisons with a control group. Differences with P < .05 were considered significant.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Rat left atrial preparations loaded with [³H]NE and subjected to electrical field stimulation produced an increased efflux of the radioisotope. The efflux of radioactivity from the tissue into the superfusion solution in control experiments is shown in Fig. 1. The SI fractional release of radioactivity during S₂ (FR₂, 0.532) when expressed as a percentage of that released during S₁ (FR₁, 0.524) yielded a value of 101.5. The ability of the in vitro assay system to detect alterations in the overflow of the released NE was ascertained by exposing the tissues to agents known to modulate the reuptake or release of the neurotransmitter. Exposure of the rat atrial preparation preloaded with the [³H]NE to desipramine (10⁶ M), an inhibitor of the neuronal uptake of NE, 20 min before S₂ resulted in a significant augmentation of the radioactivity efflux on electrical stimulation (Table 1). Desipramine, however, did not cause any discernible augmentation of the resting efflux of radioactivity. Similar applications of the ₂ agonist oxymetazoline (10⁶ M) or the ₂ agonist fenoterol (10⁶ M) to the rat atrial preparation resulted in corresponding inhibitory or facilitatory effects on the SI overflow of [³H]NE (Table 1).

View larger version (25K): [in this window] [in a new window]   Fig. 1.   Mean effluxes of radioactivity into 5-min collections of the PSS bathing the atria in control experiments (n = 15). Electrical stimulations (3 Hz, 50 mA, 1 ms, 60 s) are indicated by S1 and S2. In these experiments, the mean content of radioactivity remaining in the atria at the end of the experiment was 2.36 × 106 dpm.

View this table: [in this window] [in a new window]   TABLE 1 Effect of a 20-min treatment with the indicated agent (106 M) on stimulation-induced outflow of radioactivity from rat atria FR1 and FR2 are the fractional releases of radioactivity evoked by the two electrical stimuli and are calculated by subtracting the resting radioactivity from the total radioactivity content of the 5-min samples collected during each of the two stimulation periods (S1, S2) and expressed as a percentage of the total tissue radioactivity at that time. The exposure of the tissue to the indicated agent (dissolved in PSS) commenced 20 min before S2 and continued for the remainder of the experiment. Results are expressed as mean ± S.E.

Exposure to Ang II (10⁹ to 10⁶ M) 20 min before S₂ did not significantly alter the resting efflux of [³H]NE but produced a significant increase in the SI efflux (Fig. 2). A maximal augmentation of about 60% was observed with 10⁸ M Ang II. No diminution of the response was evident on increasing the concentration of the peptide to 10⁷ or 10⁶ M. Pretreatment of tissues with each of the four AT₁ receptor blockers (10⁶ M) for 20 min before S₁ inhibited the subsequent Ang II (10⁷ M)-induced augmentation of the SI release of [³H]NE (Fig. 3A). The percent inhibition of the Ang II response ranged between 73.2% (losartan) and 92.7% (valsartan). Inclusion of lower concentrations of the AT₁ receptor blockers (10⁷ and 10⁸ M) in the perfusing solution 20 min before S₁ also inhibited the Ang II response. Although all four compounds attenuated the Ang II-mediated facilitatory effects, the inhibition produced by eprosartan and losartan did not attain statistical significance at these lower concentrations (Fig. 3, B and C). Furthermore, the percent inhibition of the Ang II response by the four compounds at each of the three concentrations tested (10⁶, 10⁷, and 10⁸ M) retained the same rank order of activity: valsartan > irbesartan > eprosartan > losartan. Valsartan, the principal focus of this study, was further tested in tissues challenged with a 10-fold higher concentration of the agonist. Under these conditions, valsartan (10⁶ M) effected a 40.3% inhibition (68.5 versus 40.9%; augmentation of the SI outflow of radioactivity by Ang II in the absence and presence of valsartan, respectively) of the response elicited by Ang II (10⁶ M, n = 6).

View larger version (76K): [in this window] [in a new window]   Fig. 2.   Effect of Ang II (109 to 106 M) on the stimulation-induced efflux of radioactive NE from [3H]NE-labeled isolated left atria of rat. The tissues were electrically stimulated twice at 35-min intervals. Ang II was introduced into the perfusing solution 20 min before the second period of stimulation. Columns represent the mean ± S.E. from 6 to 12 determinations. *P < .05, compared with control response (ANOVA followed by Dunnett's test).

View larger version (46K): [in this window] [in a new window]   Fig. 3.   Effects of losartan, eprosartan, irbesartan, and valsartan (106 M, A; 107 M, B; 108 M, C) on Ang II (107 M)-evoked augmentation of the stimulation-induced efflux of radioactive NE from [3H]NE-labeled isolated rat left atrial preparation. The atria were electrically stimulated twice at 35-min intervals. The AT1 receptor blockers or vehicle (DMSO) was included in the perfusing solution 20 min before the first period of stimulation. Ang II was included in the perfusing solution 20 min before the second period of stimulation in the continued presence of the drugs or DMSO. Data represent the mean ± S.E. from 5 to 11 determinations. *P < .05, compared with control Ang II response (ANOVA followed by Dunnett's test).

The effects of the Ang II receptor blockers on the control responses were ascertained by including the agents (10⁶ M each) in the perfusion solution before S₁ as before while omitting the subsequent application of Ang II before S₂. The fractional release of radioactivity during S₂ relative to that during S₁ (% FR₂/FR₁) remained unaltered (104.4 ± 12.1, DMSO, n = 5; 100.4 ± 3.3, losartan, n = 6; 104.6 ± 2.9, eprosartan, n = 6; 102.1 ± 2.4, irbesartan, n = 6; 104.6 ± 1.4, valsartan, n = 6) despite exposure of the tissues to each of the four agents for an additional 35 min. The fractional releases of radioactivity (FR₁) during S₁ were also unaffected by the 20-min pretreatment with the drugs (Table 2). The effects of the AT₁ receptor blockers on SI overflow of radioactivity were further explored by including a 10-fold higher concentration (10⁵ M) of the agents in the perfusion solution 20 min before S₂. This protocol enabled comparison of the SI efflux in the absence and presence of the agent in the same tissue. There was again no statistically significant difference between compound-treated and vehicle-treated tissues regarding the ratio of the fractional releases of radioactivity (Table 3), indicating that the Ang II receptor blockers do not alter either the resting or the basal SI efflux (in the absence of added Ang II) of radioactivity.

View this table: [in this window] [in a new window]   TABLE 2 Fractional release of radioactivity on electrical stimulation of tissues pretreated for 20 min with DMSO (0.01%, v/v) or the indicated AT1 angiotensin II receptor blocker (106 M) FR1 is the fractional release of radioactivity evoked by the electrical stimulus and is calculated by subtracting the resting radioactivity from the total radioactivity content of the 5-min sample collected during the stimulation period and expressed as a percentage of the total tissue radioactivity at that time. The exposure of the tissues to the vehicle (DMSO) or the indicated agent commenced 20 min before S1 and continued for the rest of the experiment. Results are expressed as mean ± S.E. The differences in the values obtained with tissues treated with DMSO and losartan, eprosartan, irbesartan, or valsartan were not statistically significant (Dunnett's test).

View this table: [in this window] [in a new window]   TABLE 3 Stimulation-induced outflow of radioactivity from tissues treated for 20 min with DMSO (0.1%, v/v) or the indicated AT1 angiotensin II receptor blocker (105 M) FR1 and FR2 are the fractional releases of radioactivity evoked by the two electrical stimuli and are calculated by subtracting the resting radioactivity from the total radioactivity content of the 5-min samples collected during each of the two stimulation periods (S1, S2) and expressed as a percentage of the total tissue radioactivity at that time. The exposure of the tissues to DMSO (vehicle) or the indicated agent commenced 20 min before S2 and continued for the rest of the experiment. Results (mean ± S.E.) obtained with two sets of experiments (DMSO/valsartan and DMSO/losartan/eprosartan/irbesartan) conducted separately are shown. The differences in the values obtained with tissues treated with DMSO or valsartan were not statistically significant (Student's t test). Similarly, the differences in the values obtained with tissues treated with DMSO and losartan, eprosartan, or irbesartan were also not statistically significant (Dunnett's test).

The likelihood ratio test for testing regression model 2 versus model 1 (i.e., testing for equality of the slopes of the four concentration-response lines as indicated in Statistical Analysis) yielded an F value of 0.0882. Comparison with an F distribution with 3 and 91 df yielded a P value of .97, indicating that there is no evidence that the slopes are different, thereby necessitating estimation of only one slope. The concentration-response lines thus obtained for each of the four agents is shown in Fig. 4. The log IC₅₀ values (log M, with the 95% CIs in parentheses) accordingly computed for the drugs were as follows: valsartan, 7.78 (8.19, 7.51); irbesartan, 7.65 (8.02, 7.40); eprosartan, 7.12 (7.37, 6.86); and losartan, 6.75 (7.00, 6.40). Thus, the log IC₅₀ values obtained with valsartan and irbesartan were significantly lower than those obtained with eprosartan and losartan. Furthermore, on translation into IC₅₀ values (16.6 nM, valsartan; 22.4 nM, irbesartan; 75.9 nM, eprosartan; 177.8 nM, losartan) these values reveal that valsartan is 4.6 and 10.7 times as potent as eprosartan and losartan, respectively, in inhibiting the prejunctional actions of Ang II (10⁷ M).

View larger version (22K): [in this window] [in a new window]   Fig. 4.   Regression analysis of the data presented in Fig. 3. The proportion inhibition of the Ang II response in each individual tissue exposed to the receptor blocker was determined by subtracting the individual response from the average response seen in the presence of Ang II alone and by dividing that value by the average increase effected by Ang II relative to the average basal response. Figure represents the proportion inhibition of the Ang II-mediated response by each of the four AT1 receptor blockers across the indicated concentration range. The regression lines were plotted and rendered parallel as described in Results. The dotted line indicates a proportion inhibition level of 0.5. The symbols for valsartan, eprosartan, and losartan at each of the three concentrations tested have been shifted slightly from their precise x coordinates to prevent their overlap and to help discern all data points.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Previous studies with atrial and other sympathetically innervated tissues incubated with [³H]NE have shown that [³H]metabolites mainly constitute the spontaneous (resting) outflow of radioactivity, whereas intact [³H]NE accounts almost entirely for the SI outflow of radioactivity (Angus et al., 1984; Rump et al., 1994). Thus, SI outflow of radioactivity from tissues preincubated with [³H]NE is frequently used as an index of NE release from sympathetic nerves (Fuder and Muscholl, 1995). The observation in this study that the fractional release of radioactivity during the second period of stimulation (FR₂) is essentially equivalent to that detected during the initial stimulus (FR₁) is consistent with the observations of Chulak et al. (1995) with atrial preparations obtained from Wistar rats and subjected to similar treatment. Thus, the fractional release of radioactivity from the tissue remains basically unchanged during two periods of stimulations spaced 35 min apart even though the resting outflow of radioactivity from the tissue declines somewhat, thereby emphasizing the stability of the preparation. The expression of the SI efflux as a fraction of the radioactivity in the tissue clearly adds to the precision of the index and facilitates comparison of releases during consecutive periods of stimulation.

The observation that desipramine, a tricyclic antidepressant known to inhibit the neuronal uptake of NE (Franco et al., 1976), caused a significant increase in the radioactive content of the effluent from electrically stimulated tissues demonstrates the ability of the in vitro assay system to detect alterations in the overflow of the released neurotransmitter after pharmacological interventions. The use of the preparation for assessing the effects of agents on sympathetic neurotransmission was further affirmed by using substances known to modulate release of NE. There is compelling evidence indicating that the release of NE from sympathetic nerve terminals is modulated by endogenous or exogenous substances acting at receptor sites associated with the nerve terminals (Westfall, 1977; Fuder and Muscholl, 1995). The prejunctional receptors at peripheral neuroeffector sites, which have been most thoroughly studied, are the inhibitory ₂- and the facilitatory ₂-adrenoceptors. Application of the ₂ agonist oxymetazoline or the ₂ agonist fenoterol to rat atria loaded with [³H]NE was found in this study to decrease or increase, respectively, the release of NE on electrical stimulation. These results are consistent with those reported by Abadie et al. (1996) with human atrial appendages subjected to similar treatment. Thus superfused rat atrial preparations, when used as indicated in this report, provide a stable and reliable in vitro model system for studying modulations of sympathetic neurotransmission.

The maximal augmentation (60%) in the SI efflux observed in this study with Ang II is consistent with the values reported with the peptide in other sympathetically innervated tissues (Brasch et al., 1993; Cox et al., 1995). The observed lack of any attenuation of the response on increasing the concentration of the peptide to 10⁷ or 10⁶ M, however, is in contrast to studies with other preparations wherein a decreased augmentation of the response was seen with supramaximal concentrations of Ang II (Cox et al., 1996; Guimaraes et al., 1998). The apparent resistance of the isolated rat left atrial preparation to any tachyphylaxis or desensitization on exposure to Ang II under these experimental conditions makes the preparation especially suitable for delineation of the effects of AT₁ receptor blockers on the prejunctional actions of Ang II.

The observed inhibition of the Ang II responses by all four Ang II receptor blockers tested suggests that the enhancement of sympathetic neuroeffector transmission in the rat heart by the peptide entails activation of AT₁ receptors. Similar inferences have also been drawn from studies using human atrial tissues (Munch et al., 1996; Rump et al., 1998). Thus, facilitation of neuronal NE release by Ang II acting via prejunctional AT₁ receptors apparently is a phenomenon evident across diverse species. The observation that the fractional releases of radioactivity during the two consecutive periods of stimulation spaced 35 min apart remained constant despite continued presence of the drugs indicates that the observed inhibition of the Ang II response by the drugs is not a consequence of any time-dependent attenuation of the basal SI efflux by the agents masking the Ang II response. This inference was reinforced by observations that inclusion of high concentrations of the agents (10⁵ M) between the two periods of stimulation (S₁, S₂) also does not alter the basal SI efflux. Furthermore, the observation that fractional releases (FR₁) during S₁ are similar across all groups treated with vehicle or drug for 20 min reiterates that the agents per se do not alter the basal SI efflux but that they selectively inhibit facilitation of the response by Ang II. These conclusions are in consonance with those drawn by Foucart et al. (1996) after their examination of the effects of losartan in the isolated rat atria.

Although all four Ang II receptor blockers that we tested inhibited the prejunctional actions of Ang II in the rat atria, significant differences were noted in their relative potencies to effect this action. The log IC₅₀ values computed from the concentration-response relationships of the individual agents indicated that valsartan and irbesartan are significantly more potent than eprosartan and losartan in inhibiting the prejunctional facilitatory actions of Ang II. The potency difference between valsartan and irbesartan, however, did not attain statistical significance. The ability of valsartan to significantly attenuate the actions of a very high concentration of Ang II (10⁶ M) further underscores the effectiveness of this agent in inhibiting the prejunctional facilitatory actions of the peptide. This is in contrast to the hypothesis advanced by Ohlstein et al. (1997) that eprosartan might be a more effective antagonist of prejunctional Ang II receptors relative to losartan, valsartan, and irbesartan based on an evaluation of the effects of the agents on activation of sympathetic outflow in the pithed rat. The results reported by Ohlstein et al. (1997) are also somewhat at variance with those reported by Wong et al. (1992), who have demonstrated in a similar experimental model that the prejunctional Ang II receptors modulating sympathetic nerve function are of the AT₁ type, sensitive to losartan. Although the reasons for the divergent results obtained by Ohlstein et al. (1997) are not readily apparent, they do not appear to be related to any meaningful qualitative difference in the abilities of the AT₁ receptor blockers to inhibit the prejunctional neuromodulatory effects of Ang II.

The inhibition of the prejunctional facilitatory effects of Ang II on peripheral sympathetic neurotransmission by the angiotensin receptor blockers has significant therapeutic implications. For instance, the inhibitory effects of valsartan are observed at concentrations that are clinically relevant. The peak and 24-h postdose plasma concentrations of valsartan exceed 1 and 0.1 µM, respectively, after oral administrations of the recommended therapeutic daily dose (80 mg) of the drug for 7 days to healthy subjects (Morgan et al., 1997). At these concentrations, valsartan inhibited the prejunctional facilitatory effects of Ang II (10⁷ M) in rat atria by 92.7 and 71.7%, respectively. Thus, it is conceivable that valsartan exerts significant antagonistic effects prejunctionally and inhibits neuronal NE release when used clinically to treat hypertension. This action may not only contribute to its net antihypertensive efficacy but may also help ameliorate other pathophysiological cardiovascular conditions that are exacerbated by elevations in interstitial or circulating NE levels. Based on the known pharmacokinetic properties of irbesartan, eprosartan, and losartan (Ohtawa et al., 1993; Marino et al., 1998; Martin et al., 1998), it is equally feasible that each of these agents, when used at their recommended therapeutic doses in humans, also exerts significant inhibitory effects on the prejunctional actions of Ang II with consequences qualitatively similar to those envisaged with valsartan. Although the potency of losartan was found to be lower than those of valsartan and irbesartan in this study, it may not truly foretell the overall activity of the compound in vivo because of the additional contribution anticipated from the active metabolite of losartan (EXP3174) generated in vivo. It is well documented that a modest fraction (14%) of an orally administered therapeutic dose of losartan is converted to EXP3174 (Lo et al., 1995), a metabolite that is reportedly about 15 times more potent than losartan in inhibiting the pressor responses to Ang II in the conscious normotensive rat (Wong et al., 1990).

The -adrenergic component of the peripheral sympathetic nervous system is known to play a major role in the pathophysiology, clinical manifestations, and natural history of human CHF. Chronic stimulation of myocardial -adrenergic receptors is believed to induce hypertrophy of cardiomyocytes and contribute to the development of catecholamine-induced cardiomyopathy (Leier et al., 1990). By virtue of their prejunctional inhibitory actions, Ang II receptor blockers can therefore be anticipated to redress the autonomic imbalance characteristic of patients with CHF and to exert salutary effects in this condition. Whether the observed high potency of valsartan for inhibition of the prejunctional actions of Ang II translates into a significant clinical advantage in the treatment of CHF vis-à-vis the other drugs tested has yet to be ascertained.