Evidence for a Physiological Role of Angiotensin-(1-7) in the Control of Hydroelectrolyte Balance
Abstract In this study we evaluated the possibility that angiotensin-(1-7) [Ang-(1-7)] acts as an endogenous osmoregulatory peptide by determining the effect of acute administration of its selective antagonist [d-Ala7]Ang-(1-7) (A-779) on renal function parameters in rats. In addition, we investigated the physiological mechanisms involved in the antidiuretic effect of Ang-(1-7). The antidiuretic effect of Ang-(1-7) (40 pmol/0.05 mL per 100 g BW) in water-loaded rats was completely blocked by A-779 (vehicle-treated, 3.34±0.43 mL/h; Ang-(1-7), 1.48±0.23; A-779, 2.72±0.35; Ang-(1-7) plus A-779, 3.26±0.49). In contrast, the antidiuretic effect of Ang-(1-7) was not significantly changed by a vasopressin V2 receptor antagonist in a dose that completely blocked the antidiuresis produced by an equipotent dose of vasopressin. In addition, Ang-(1-7) administration did not significantly change vasopressin plasma levels in water-loaded rats. The antidiuretic effect of Ang-(1-7) in water-loaded rats was associated with a reduction of creatinine clearance (0.68±0.04 versus 1.38±0.32 mL/min in vehicle-treated rats, P<.05) and an increase in urine osmolality (266.8±32.7 versus 182.8±14 mOsm/kg in vehicle-treated rats, P<.05). An effect of Ang-(1-7) in tubular water transport was demonstrated in vitro by a fourfold increase in the hydraulic conductivity of inner medullary collecting ducts in the presence of 1 nmol/L Ang-(1-7). Subcutaneous administration of A-779 (2.3 to 9.2 nmol/100 g) produced a significant increase in urine volume (4.6 nmol/100 g, 0.45±0.12 mL/h; vehicle-treated rats, 0.16±0.03 mL/h; P<.05) comparable to that of acute administration of a vasopressin V2 receptor antagonist. The diuretic effect of A-779 was associated with an increase in creatinine clearance and decrease in urine osmolality. In contrast, no significant effects on urine volume were observed after systemic administration of angiotensin subtype 1 or 2 receptor antagonists (DuP 753 and CGP 42112A, respectively). These findings suggest that endogenous Ang-(1-7), acting on specific receptors, participates in the control of hydroelectrolyte balance by influencing especially water excretion.
It has been assumed that the actions of the renin-angiotensin system in the brain and periphery arise from the interaction of Ang II with its receptors.1 However, increasing evidence indicates that the actions of the renin-angiotensin system are mediated by multiple angiotensin peptides [Ang II, Ang-(1-7), Ang III, Ang IV] acting on multiple angiotensin receptors.2 3
The heptapeptide Ang-(1-7) is a recently identified active component of the renin-angiotensin system that can be formed by a route independent of angiotensin-converting enzymes.4 5 It can be formed either from Ang I by cleavage of the Pro7-Phe8 bound by prolyl-endopeptidase or neutral endopeptidase 24.11 or from Ang II by removal of the carboxy-terminal phenylalanine with prolyl-endopeptidase or carboxypeptidases (see Reference 3 for review). Ang-(1-7) has been demonstrated in the plasma and tissues of a variety of species, including humans, dogs, sheep, and rats.3 6 Ang-(1-7) present in plasma may be derived from tissues or hydrolysis of its circulating precursors by enzymes present on the surface of endothelial cells.5
In addition to the different enzymatic routes for its generation, Ang-(1-7) differs importantly from Ang II by its selectivity. Although both peptides elicit some similar actions in the brain, such as AVP release,7 changes in blood pressure,8 9 10 and increases in neuronal activity,11 Ang-(1-7) is devoid of significant dipsogenic, vasoconstrictor, or aldosterone secretagogue actions (see References 2 and 3 for review). Ang-(1-7) can even present effects opposite those of Ang II, such as on baroreceptor reflex sensitivity, which is attenuated by Ang II and increased by Ang-(1-7).12
A growing body of evidence suggests that one of the major physiological actions of Ang-(1-7) is related to the control of hydroelectrolyte balance. Dense immunostaining for Ang-(1-7) immunoreactivity has been demonstrated in the supraoptic and paraventricular nuclei of the hypothalamus and neurohypophysis,13 and Ang-(1-7) is as potent as Ang II in releasing AVP from hypothalamus-neurohypophysial explants.7 In addition to its AVP-releasing activity in vitro, Ang-(1-7) possesses a potent peripheral antidiuretic activity in water-loaded rats.14 In vitro, Ang-(1-7) has been reported to increase fluid and bicarbonate reabsorption in the proximal straight tubule at physiological concentrations (10−12 mol/L) and to have an opposite effect at a supraphysiological concentration (10−8 mol/L).15 In addition, a natriuretic effect of supraphysiological doses of Ang-(1-7) has been reported in denervated kidneys16 in vivo and in in situ perfused rat kidneys.17 The natriuretic effect of Ang-(1-7) in these preparations is in accordance with the observation that 10−9 mol/L Ang-(1-7) inhibits (20%) Na+ flux in cultured renal tubular epithelial cells.18 A physiological role for Ang-(1-7) in the control of hydromineral balance is also suggested by the selective increase in its circulating levels in chronically salt-loaded rats.6 Moreover, the enzymes necessary to generate Ang-(1-7) from its precursors (Ang I or Ang II) are abundant in the kidney.19
Despite the several lines of evidence for a role of Ang-(1-7) in the control of hydromineral balance, the evaluation of the physiological relevance of this angiotensin has been delayed by the absence of a selective antagonist. It has been recently shown that the Ang-(1-7) analogue A-779 is a potent and selective Ang-(1-7) antagonist.20 21 This compound antagonizes several actions of Ang-(1-7), including its antidiuretic effect and the changes in MAP produced by Ang-(1-7) microinjection into the dorsomedial or ventrolateral medulla.10 20 In contrast, A-779 did not modify the antidiuretic effect of AVP and the myotropic, dipsogenic, or pressor effects of Ang II.20 Furthermore, this compound has a very low affinity for AT1 and AT2 receptors, with Ki values greater than 1 μmol/L.20 A-779 also produces selective blockade of the Ang-(1-7) stimulatory effect on the neuronal activity in the paraventricular nucleus.21
The availability of a selective Ang-(1-7) antagonist prompted us to evaluate the role of this heptapeptide in hydroelectrolyte balance. In the present study, we determined the effect of acute administration of A-779 on renal function parameters. In addition, we investigated physiological mechanisms involved in the antidiuretic effect of Ang-(1-7).
Male Wistar rats weighing 280 to 320 or 120 to 125 g were used for the in vivo and in vitro experiments, respectively. The rats were housed in plastic cages with free access to ordinary chow and water at an ambient room temperature of 25°C and a 12-hour light/dark cycle (lights on 7 am to 7 pm).
Characterization of the Antidiuretic Effect of Ang-(1-7) in Water-Loaded Rats
Protocol 1: Antidiuretic Effect of Ang-(1-7)
Water diuresis was induced by water load (5 mL/100 g BW, gavage). Immediately after water load, the rats (n=30) were treated subcutaneously with vehicle (throughout, vehicle is 0.9% NaCl, 0.05 mL/100 g BW) or graded doses of Ang-(1-7) (20 [n=9], 40 [n=32], or 80 [n=6] pmol/0.05 mL per 100 g BW). Urine was collected for 60 minutes after water loading.
Protocol 2: Blockade of the Antidiuretic Effect of Ang-(1-7)
Immediately after water load, the rats were treated subcutaneously with vehicle (n=30), A-779 (1 μg [1.1 nmol]/100 g BW, n=14), or A-779 (1.1 nmol/100 g BW) plus Ang-(1-7) (40 pmol/0.05 mL per 100 g BW, n=9). Urine was collected for 60 minutes after water loading.
Protocol 3: Evaluation of the Role of AVP on the Antidiuretic Effect of Ang-(1-7)
Immediately after water load, the rats were treated subcutaneously with vehicle (n=30), AVP (20 pmol/0.05 mL per 100 g BW, n=13), the AVP V2 receptor antagonist AVPa (1 μg [0.8 nmol]/100 g BW, n=26), AVP (20 pmol/0.05 mL per 100 g BW) plus AVPa (0.8 nmol/100 g BW, n=18), or Ang-(1-7) (40 pmol/0.05 mL per 100 g BW) plus AVPa (0.8 nmol/100 g BW, n=27). Urine was collected and urine volume measured 60 minutes after water loading. An additional group of water-loaded rats was used for determination of plasma concentration of AVP and Ang-(1-7). Twenty-four hours before blood sampling, an arterial catheter was implanted into the abdominal aorta through the femoral artery under ether anesthesia. Water-loaded rats receiving vehicle (n=6) or Ang-(1-7) (40 pmol/0.05 mL per 100 g BW, n=6) were used. Blood samples (0.75 mL/100 g BW) were collected 30 minutes after water loading with chilled syringes and transferred to tubes containing 1 mmol/L p-hydroxymercuribenzoate, 9.1 mmol/L 8-hydroxyquinoline, 1 mmol/L phenylmethylsulfonyl fluoride, 1 mmol/L pepstatin A (10 μL of each per milliliter of blood), and 7.5% EDTA (50 μL/mL blood). Blood samples were centrifuged at 2000g for 20 minutes at 4°C, and the plasma was stored at −20°C.6
Protocol 4: Evaluation of the Role of Prostaglandins on the Antidiuretic Effect of Ang-(1-7)
A group of rats was pretreated with indomethacin (5 mg/kg IM) 30 minutes before water loading, and immediately afterward the rats were injected subcutaneously with vehicle (n=10) or Ang-(1-7) (40 pmol/0.05 mL per 100 g BW, n=10). Urine was collected and urine volume measured 60 minutes after water loading.
Protocol 5: Effect of Ang-(1-7) on Renal Function Variables
Immediately after water loading, another group of rats was treated subcutaneously with vehicle (n=13) or Ang-(1-7) (40 pmol/0.05 mL per 100 g BW, n=16) and transferred to metabolic cages. Blood and urine samples for measurement of Ccr, osmolality, and sodium and potassium concentrations were collected 60 minutes after water loading.
Protocol 6: Effect of Ang-(1-7) on MAP
In an additional group of rats, a polyethylene catheter (PE-10 soldered to a PE-50) was inserted into the abdominal aorta through the femoral artery with rats under ether anesthesia. Twenty-four hours later, MAP was recorded with a pressure transducer (Statham, P23XL) connected to a data acquisition system (CODAS, Dataq Instruments Inc). MAP was continuously recorded for 1 hour after water loading in rats treated with Ang-(1-7) (40 pmol/0.05 mL per 100 g BW SC, n=5) or vehicle (n=5).
Effect of Ang-(1-7) in Isolated IMCDs
Isolated IMCDs were perfused by previously described techniques.22 23 Tubules were isolated from a small kidney slice immersed in a dish of chilled Ringer-HCO3− buffer, oxygenated, and kept at pH 7.4 by bubbling the solution with 5% CO2 and 95% O2. The IMCDs were dissected without the use of collagenase or other enzymes. After isolation, the segments were transferred to a temperature-regulated chamber (37°C) that was mounted on the stage of an inverted microscope. Unless indicated, Ringer-HCO3− buffer (mmol/L: NaCl 115.0, NaHCO3 25.0, sodium acetate 10.0, KCl 5.0, CaCl2 1.0, MgSO4 1.2, NaH2PO4 1.2, d-glucose 5.5) was used as bath and perfusion fluid. For the hypotonic solution (70.0 mOsm), NaCl (115.0 mmol/L) was withdrawn. FD&C green dye was added to perfusate as a visual marker.
Net water absorption (Jv) was measured with [14C]inulin dialyzed immediately before the experiments. The dialyzed isotope was added to the perfusion solution at a final concentration of 25 to 100 cpm/nL. Jv was evaluated as Vi−Vo/L, where Vi is the perfusion rate, Vo the collecting rate, and L the length of the tubule studied (tubule length and inner diameter can be measured to within 0.05 mm with a precalibrated micrometer in one eyepiece of the inverted microscope used to observe perfusion). Vo was directly measured on the basis of collection time, and Vi was calculated by the appearance rate of the impermeant marker [14C]inulin in the collection pipette according to the equation Vi=(Ino/Ini), where Ino is the counts per minute of the collected fluid and Ini is the counts per minute of the perfusate. Timed tubular samples were collected for analysis under mineral oil by aspiration into a calibrated pipette. All measurements were obtained 60 to 90 minutes after perfusion of a given nephron segment was started. The inner diameter of the perfused segments ranged from 25 to 35 μm and the tubular length from 1.5 to 2.0 mm.
Hydraulic conductivity (Lp) was determined by measurement of net fluid movement in response to an imposed gradient. Net fluid reabsorption was induced by perfusion with hypotonic perfusion solution (80 to 90 mOsm/kg H2O, 40 mmol/L NaCl) and 300 mOsm/kg H2O bath solution. Osmolalities of perfusate and bath fluid were measured, and Lp (in centimeters per second per atmosphere) was calculated in each experiment by the following equation23 :
where Cb and Ci are the osmolalities of the bath and perfusate, respectively, Vi is the perfusate rate, Vo the collection rate, R the gas constant, T the absolute temperature, and A the luminal surface area. This equation rests on the assumption that the reflection coefficient of NaCl is 1.0.23
The passive lumen-to-bath urea permeability (Pu, in centimeters per second) was determined from the rate of [14C]urea disappearance from the perfusion solution with the following expression22 : Pu=(Vi/A)ln(Ci/Co), where Ci and Co are the activities of [14C]urea in the initial perfusion fluid and collected fluid, respectively, and Vi and A are as defined above.
The baths were checked for osmolality and pH with an osmometer (Advanced Instruments, Inc) and pH meter (Iris 7 Tecnow), respectively. The bath fluid was changed every 10 minutes to reduce the effect of evaporation and consequently the increase in osmotic gradient. The bath osmolality did not change significantly during the 10-minute period (from 296±1 at 0 minute to 298±2 mOsm/kg H2O at 10 minutes).
Timed fluid collections were made with a constant-volume constriction pipette that was rinsed four times with 0.5 mL water; 5 mL scintillation liquid was then pipetted into the vial (Aquasol Universal Cocktail, New England Nuclear). The isotopic concentration was determined with a liquid scintillation spectrometer (Packard Tricarb 1600 TR).
Hydraulic conductivity and urea permeability were measured before, during, and after removal of a Ringer-HCO3− bath solution containing 10−9 mol/L Ang-(1-7) (five tubules for each measurement).
Effects of A-779 on Basal Diuresis
Protocol 1: Diuretic Effect of A-779
Male Wistar rats received subcutaneous injection of vehicle (n=36) or graded doses of A-779 (2.3 [n=10], 4.6 [n=12], or 9.2 [n=10] nmol/100 g BW). Immediately after treatments, the rats were transferred to metabolic cages and urine was collected during 4 hours.
Protocol 2: Diuretic Effect of AVP V2 Receptor Blockade
Male Wistar rats received subcutaneous injection of vehicle (n=36) or AVPa (1.6 [n=13], 3.2 [n=12], or 6.4 [n=10] nmol/100 g BW). Immediately after treatment, the rats were transferred to metabolic cages and urine was collected during 4 hours.
Protocol 3: Blockade of AT1 or AT2 Receptors
Male Wistar rats were injected subcutaneously with vehicle (n=36), the AT1 receptor antagonist DuP 753 (100 μg [217 nmol]/100 g BW, n=10, or 500 μg [1085 nmol]/100 g BW, n=8), or the AT2 receptor antagonist CGP 42112A (50 μg [47.5 nmol]/100 g BW, n=6, or 500 μg [475 nmol]/100 g BW, n=6). Immediately after treatment, the rats were transferred to metabolic cages and urine was collected during 4 hours.
Protocol 4: Effect of A-779 on Renal Variables
An additional group of rats received 0.5 mL H2O/100 g BW (gavage). Immediately afterward, the rats were treated subcutaneously with vehicle (n=9) or A-779 (4.6 nmol/100 g BW, n=13) and transferred to metabolic cages. Urine and blood were collected 6 hours afterward as described above for measurement of Ccr, osmolality, and sodium and potassium concentrations.
Immediately after urine collection, rats were anesthetized with ether and blood samples withdrawn by heart puncture. Blood samples were kept for 30 minutes at room temperature and then centrifuged at 2000 rpm for 10 minutes. Serum was used for measurement of serum osmolality, Na+, K+, and creatinine concentration.
After collection, urine samples were centrifuged at 3000 rpm for 5 minutes (room temperature) for measurement of urine osmolality, Na+, K+, and creatinine concentration.
Sodium and potassium concentrations were measured by flame photometry (Corning 400). Serum and urine osmolality were measured with a freezing-point osmometer (Fiske Osmometer). Ccr measurements24 25 were performed with a kit that minimizes the interference of endogenous chromogens (Labtest, catalog No. 35E). Plasma samples were extracted6 with C18 Bond-Elut cartridges (Analytichem International). Radioimmunoassay for Ang-(1-7) was performed as described previously.6 Plasma AVP was determined with a commercially available antibody (Arnel Products Co Inc) and AVP as the standard.
AVPa (MW 1239.5) and indomethacin were from Sigma Chemical Co and AVP from Peninsula Laboratories, Inc. The nonpeptide AT1 receptor antagonist DuP 753 (MW 461) was a gift from DuPont de Nemours & Co, and the AT2 receptor antagonist CGP 42112A (MW 1053) was provided by Ciba-Geigy Ltd. The Ang-(1-7) analogue A-779 (MW 873.1) was synthesized by Dr M.C. Khosla (Cleveland Clinic Foundation). Ang-(1-7) (MW 890) was synthesized by M.C. Khosla or obtained from Bachem (batch No. 25691). All other chemicals used were of the highest purity available.
All results are reported as mean±SE. For all experiments with IMCDs, statistical analysis was performed with the paired t test. For all other protocols, data were analyzed with one-way ANOVA followed by the Newman-Keuls test or Student’s t test for nonpaired data when appropriate. The level of significance was set at a value of P<.05.
Characterization of the Antidiuretic Effect of Ang-(1-7) in Water-Loaded Rats
Antidiuretic Effect of Ang-(1-7) in Water-Loaded Rats
As shown in Fig 1A⇓, Ang-(1-7) induced a dose-dependent decrease in water diuresis. A reduction of approximately 80% in the volume of spontaneously voided urine was observed with the high dose used (80 pmol/0.05 mL per 100 g BW).
Blockade of the Antidiuretic Effect of Ang-(1-7)
The antidiuretic effect of Ang-(1-7) was completely blocked by the selective Ang-(1-7) antagonist A-779 (Fig 1B⇑). On the other hand, at the dose used, A-779 did not significantly change water diuresis or the antidiuretic effect of AVP in water-loaded rats (Fig 1C⇑).
Role of AVP on the Antidiuretic Effect of Ang-(1-7)
The decrease of water diuresis produced by Ang-(1-7) in water-loaded rats was not modified by administration of the AVP V2 receptor antagonist AVPa (Fig 1B⇑). Ang-(1-7) administration (40 pmol/0.05 mL per 100 g BW) reduced urine volume of the first hour to 1.48±0.23 mL compared with 3.46±0.4 mL in vehicle-treated rats (P<.05). In AVPa-treated rats, the antidiuresis produced by Ang-(1-7) administration was similar to that observed in untreated rats (1.83±0.29 mL, P<.05 compared with the control group). The efficacy of V2 receptor blockade was demonstrated in experiments showing that the antagonist completely blocked the antidiuresis produced by AVP administration in water-loaded rats (Fig 1C⇑).
Thirty minutes after Ang-(1-7) administration to water-loaded rats, the plasma concentration of Ang-(1-7) was 134.6±14.7 pg/mL, compared with 37.6±2.2 pg/mL in vehicle-treated rats. The AVP plasma concentration in Ang-(1-7)–treated rats did not differ from that of control rats (2.6±1.3 versus 1.8±1.0 pg/mL).
Role of Prostaglandins on the Antidiuretic Effect of Ang-(1-7)
To evaluate the influence of prostaglandins in the antidiuretic action of Ang-(1-7), we performed additional experiments in indomethacin-treated rats. Indomethacin potentiated the antidiuretic effect of Ang-(1-7). In indomethacin-treated rats, Ang-(1-7) reduced urine volume in the first hour by approximately 10-fold (0.38±0.17 versus 3.46±0.40 mL in the control group, P<.05). Administration of indomethacin per se reduced water diuresis by approximately 60% (1.42±0.54 mL/h).
Effect of Ang-(1-7) on Renal Function Parameters
The effect of Ang-(1-7) on renal function parameters in water-loaded rats is shown in Table 1⇓. The reduction in urine volume produced by Ang-(1-7) was associated with an increase in urine osmolality and urinary sodium concentration and a decrease in Ccr compared with vehicle-treated rats (P<.05).
Effect of Ang-(1-7) on MAP
As expected, subcutaneous administration of Ang-(1-7) in water-loaded rats did not significantly change MAP (106±3 versus 103±1 mm Hg in vehicle-treated rats, P>.05).
Effect of Ang-(1-7) in Isolated IMCDs
The evidence obtained in water-loaded rats that Ang-(1-7) acts independently of circulating AVP was substantiated in vitro. As shown in Fig 2⇓, 10−9 mol/L Ang-(1-7) added to the bath in the absence of AVP was able to increase hydraulic conductivity from 0.35±0.06 to 1.38±0.24×10−6 cm·s−1·atm−1 (P<.05). Hydraulic conductivity returned to 0.30±0.12 10−6 cm·s−1·atm−1 after withdrawal of Ang-(1-7). Urea permeability did not change in the presence of Ang-(1-7) [control, 26.26±0.98 10−5 cm·s−1; Ang-(1-7), 25.87±1.02; recovery, 23.20±1.73].
Effect of A-779 on Basal Diuresis
Diuretic Effect of A-779
Subcutaneous administration of A-779 produced a significant dose-dependent increase in urine volume (Fig 3⇓). A threefold increase in basal urine output was observed in the groups injected with the higher doses of A-779 (4.6 nmol/100 g BW, 0.45±0.12 mL/h; 9.2 nmol/100 g BW, 0.43±0.10, compared with 0.16±0.03 in vehicle-treated rats, P<.05). Administration of the smallest dose of A-779 (2.3 nmol/100 g BW) produced only a slight and nonsignificant increase in urine volume (0.24±0.05 mL/h).
Diuretic Effect of AVP V2 Receptor Blockade
Subcutaneous administration of the AVP V2 receptor antagonist AVPa also produced a significant dose-related increase in urine output (Fig 3⇑). The maximal effect on urine volume was detected with the higher doses of AVPa (3.2 nmol/100 g BW, 0.55±0.17 mL/h; 6.4 nmol/100 g BW, 0.56±0.10, compared with 0.16±0.03 in vehicle-treated rats, P<.05). There was no statistical difference in the maximal urine output obtained with A-779 and AVPa (0.45±0.12 and 0.56±0.10 mL/h, respectively, P>.05). As observed for A-779, administration of the smallest dose of AVPa (1.6 nmol/100 g BW) did not change urine output compared with the vehicle-treated group (0.19±0.09 mL/h). However, the combination of the smallest doses of A-779 and AVPa (2.3 and 1.6 nmol/100 g BW, respectively) produced a significant increase in urine output (0.42±0.05 mL/h, n=10). Only a slight additive effect was observed when a combination of higher doses (A-779, 4.6 nmol; AVPa, 3.2 nmol) of both antagonists was used (0.66±0.17 mL/h).
Blockade of AT1 or AT2 Receptors
The effect of the angiotensin antagonists DuP 753 and CGP 42112A on basal urine output is shown in Fig 4⇓. Subcutaneous injection of the AT1 receptor antagonist DuP 753 (100 or 500 μg/100 g BW) did not produce a significant change in urine output compared with vehicle-treated rats (100 μg, 0.19±0.08 mL/h; 500 μg, 0.19±0.10 versus 0.16±0.03 in vehicle-treated rats, P>.05). A similar finding was obtained with the AT2 receptor antagonist CGP 42112A (50 μg, 0.05±0.04 mL/h; 500 μg, 0.10±0.05 versus 0.16±0.03 in vehicle-treated rats). Although DuP 753 did not change urine flow rate, it produced an increase in Na+ excretion from 0.38±0.03 mmol in vehicle-treated rats to 0.65±0.11 mmol at 100 μg/100 g BW (n=4, P<.05) and to 0.61±0.02 mmol at 500 μg/100 g BW (n=4, P<.05).
Effects of A-779 on Renal Parameters
The effects of the selective Ang-(1-7) antagonist A-779 on renal function parameters are shown in Table 2⇓. Subcutaneous administration of A-779 produced a significant increase in basal diuresis that was associated with an increase in Ccr and Na+ excretion. A decrease in urine osmolality compared with vehicle-treated rats (P<.05) was also observed.
In this study, we have obtained evidence that the heptapeptide Ang-(1-7) acts as an endogenous antidiuretic peptide by showing that (1) the in vivo Ang-(1-7) antidiuretic action in water-loaded rats was not blocked by an AVP V2 receptor antagonist; (2) in vitro, Ang-(1-7) increased water transport in IMCDs; and (3) in conscious rats, acute systemic administration of the recently described Ang-(1-7) antagonist A-77920 21 but not of classic AT1 or AT2 antagonists produced diuresis and natriuresis.
We have confirmed and extended our previous observation that peripheral administration of Ang-(1-7) reduces water diuresis induced by water loading in rats in a dose-dependent manner.14 The antidiuretic effect of Ang-(1-7) was associated with increases in urinary sodium concentration and urine osmolality, indicating that this effect is the result of an increase in water reabsorption.
The antidiuresis produced by Ang-(1-7) was not significantly changed by an AVP V2 receptor antagonist in a dose that completely blocked the antidiuresis produced by AVP administration (Fig 1B⇑). This observation argues against the participation of circulating AVP in the antidiuretic effect of Ang-(1-7). Although we have previously observed that in vitro Ang-(1-7) can release AVP from hypothalamic-neurohypophysial explants,7 in vivo experiments indicate that the AVP-releasing activity of Ang-(1-7) may be restricted to its central actions and is probably not involved in its peripheral biological effects (see Reference 14). Our finding that Ang-(1-7) administration in water-loaded rats did not significantly change plasma AVP levels is consistent with this view. Additional evidence against the dependence of the antidiuretic effect of Ang-(1-7) on AVP release was obtained by Pawloski-Dahm and Fink.26 These investigators observed that chronic infusion of Ang-(1-7) can produce antidiuresis without changing plasma AVP levels. Finally, our observation that the antidiuretic effect of Ang-(1-7) appears to involve important changes in GFR (see below), contrasting with the absence of a gross effect of AVP on this parameter,27 indicates that Ang-(1-7) is capable of reducing water excretion independent of AVP release. However, we cannot exclude the possibility that in situations of osmotic imbalance, the central AVP-releasing activity of Ang-(1-7), demonstrated in vitro,7 may contribute to increased plasma AVP levels.
In vitro studies also support a direct effect of Ang-(1-7) on water metabolism. We have observed that 10−9 mol/L Ang-(1-7) increases water reabsorption in the distal segment of rat IMCDs. To our knowledge, this is the first demonstration of an action of Ang-(1-7) in this nephron segment. However, this effect was not accompanied by changes in the passive lumen-to-bath urea transport under our experimental conditions. The results obtained with Ang-(1-7) differ from those described for AVP, which concomitantly increases water and urea transport.28 29 It has been described that AVP activates urea transport and water channels by different processes.30 Our data further support the evidence for separated routes for water and urea transport in the terminal IMCDs. The influence of Ang-(1-7) on water transport, however, is not restricted to the distal nephron. Recently, Garcia and Garvin15 demonstrated that in the rat proximal straight tubule, Ang-(1-7) in a concentration as low as 10−12 mol/L increases fluid and bicarbonate reabsorption. In a pharmacological concentration, however, an opposite effect was observed. A similar finding was also recently obtained by us in a toad skin preparation.31
Using in situ perfused17 or denervated16 rat kidneys, other investigators have observed a natriuretic effect of Ang-(1-7) associated with an increase in urine flow rate. An inhibitory effect of Ang-(1-7) on Na+ transport in cultured renal epithelial cells was also observed by Andreatta–Van Leyen et al.18 These apparently conflicting data may be related to the peptide concentration used or differences in experimental conditions. As demonstrated by Garcia and Garvin15 in proximal tubules, at a physiological concentration, Ang-(1-7) acts by increasing tubular transport, whereas at supraphysiological levels, an opposite action was observed. This biphasic effect of Ang-(1-7) on tubular reabsorption resembles the well-established effect of Ang II.32 33 At low concentrations, Ang II stimulates Na+ and bicarbonate reabsorption, whereas at concentrations higher than 10−9 mol/L, an inhibitory effect is observed.32 33 34 Regarding differences in experimental conditions, in the studies in which a diuretic effect of Ang-(1-7) was observed, anesthesia and denervated kidneys were used in addition to a relatively high peptide infusion rate (picomoles per minute). In these conditions, a substantial natriuresis was observed.16 17 Thus, diuresis observed in these experiments was probably secondary to the peptide effect on proximal tubular transport as discussed above. Conversely, in conscious, water-loaded rats, we have observed the opposite effect. The low Ang-(1-7) plasma concentration (approximately 1.5×10−10 mol/L) achieved 30 minutes after its administration, which is comparable to that found in chronically salt-loaded rats,6 may partly explain these differences. Furthermore, we have observed a significant reduction of Ccr after Ang-(1-7) administration, contrasting with a slight tendency for an increase in GFR observed in in situ perfused kidneys. These differences suggest that at least part of the Ang-(1-7) actions at the glomerular level are importantly blunted by anesthesia and/or kidney denervation. It has been recently described that infusion of large amounts of Ang-(1-7) (2.4 μg/100 g BW per hour) in Wistar-Kyoto rats did not change urine volume or sodium excretion.35 Infusion of the same amount of this peptide in spontaneously hypertensive rats elicited a transient diuresis and natriuresis associated with a significant decrease of plasma AVP concentration. No change in plasma AVP was observed in Wistar-Kyoto or Sprague-Dawley rats.35 Although more experiments should be done with different (lower) doses of Ang-(1-7), these results further illustrate the complexity of the renal actions of this heptapeptide.
We have obtained evidence using endogenous Ccr that Ang-(1-7) modified GFR. This methodology is widely used to permit the use of conscious, unrestrained rats.24 25 36 37 However, one may argue that the use of Ccr to estimate GFR is not appropriate in rats. Early reports have described a substantial tubular secretion of creatinine in these animals.25 However, more recent studies have shown that the handling of creatinine in the rat kidney depends on the rat strain. In male Fisher rats38 and BioBreeding diabetic-prone or diabetic-resistant rats,24 creatinine is actually reabsorbed (30% to 50%) by kidney tubules. A similar observation was made in our rat strain, in which a ratio of Ccr to the clearance of inulin (Cinulin) of 0.7 was obtained in anesthetized animals (R.A.S.S., K.T.P., N.C.V.B., unpublished results, 1995). The low Ccr value in our awake rats suggests that the difference between Ccr and Cinulin in this condition may be even higher. Thus, the results obtained with Ang-(1-7) in the water-diuresis experiments and with the use A-779 in the basal diuresis experiments, suggesting that Ang-(1-7) acts at the glomerular level reducing GFR, needs to be interpreted with caution in view of the low Ccr and low Ccr-Cinulin ratio observed in our rats. The decrease in Ccr produced by Ang-(1-7) probably results from an effect of Ang-(1-7) on mesangial cells decreasing Kf, because this peptide is essentially devoid of vascular effects.2 3 This possibility is strengthened by the recent identification of an angiotensin receptor subtype in rat mesangial cells that shows a high affinity (Kd approximately 30 nmol/L) for Ang-(1-7).39 However, we cannot exclude the possibility that the change in Ccr produced by Ang-(1-7) could be due to a differential effect of the heptapeptide on afferent and efferent arterioles, decreasing the filtration rate without substantially changing total renal blood flow, and/or to an interference of Ang-(1-7) with creatinine transport. Our data also indicate that the antidiuretic effect of Ang-(1-7) in water-loaded rats is not related to changes in MAP or sodium reabsorption.
Pretreatment with the cyclooxygenase inhibitor indomethacin potentiated the antidiuretic effect of Ang-(1-7). This observation is consistent with the prostaglandin-releasing activity of Ang-(1-7).40 41 Prostaglandins, released by Ang-(1-7) from endothelial cells41 or through activation of phospholipase A2 at the tubular level,18 could counteract the antidiuretic action of Ang-(1-7) by increasing urine output through their well-documented glomerular and tubular effects.42
Subcutaneous administration of A-779 in conscious rats produced a dose-dependent increase in urine volume and Na+ excretion. The diuretic effect of A-779 was associated with a significant increase in Ccr, suggesting that this effect was partially due to an increase in GFR, a finding that is consistent with the decrease in Ccr produced by Ang-(1-7) administration in water-loaded rats.
The possibility that the renal effects of A-779 could be due to interference with other peptides cannot be ruled out. However, we have shown that this compound does not influence the biological activity of several peptides, including Ang II, Ang III, substance P, and bradykinin, even at a molar ratio of 2500:1.20 The effects of A-779 are also apparently unrelated to systemic hemodynamic effects because only a minor increase in MAP was observed with A-779 infusion in normal rats.43
In contrast to the change in urine output observed after A-779 administration, subcutaneous injection of the AT1 receptor antagonist DuP 753 or the AT2 receptor antagonist CGP 42112A did not significantly modify urine output within 4 hours of administration. Interestingly, a tendency toward a decrease in urine output was observed after CGP 42112A administration. Our observations and those of others44 in conscious rats contrast with the finding in anesthetized rats that maximal doses of DuP 753 caused natriuresis, diuresis, and chlorouresis similar in magnitude to that elicited by maximal doses of the AT2 receptor antagonist PD 123177 (60 to 120 mg/kg IV).45 A significant effect of DuP 753 on urine flow rate was also observed in anesthetized rats by Xie et al.46 The absence of an acute effect of DuP 753 on urine flow rate in conscious compared with anesthetized animals should be interpreted by taking into account the fact that the renin-angiotensin system is activated in anesthetized animals.47 Thus, DuP 753 or other Ang II antagonists would be expected to have more pronounced renal effects in anesthetized than conscious rats. However, even in anesthetized animals, the effects obtained with DuP 753 on renal function are not always the same. For example, Clark et al48 reported a decrease in GFR, urine flow, and sodium excretion in anesthetized dogs (also see Reference 49 for review).
In micropuncture studies, Xie et al46 recently demonstrated that intravenous administration of DuP 753 in rats produces a significant decrease in electrolyte (Na+, HCO3−, Cl−) and fluid reabsorption in proximal convoluted tubules (S1 subsegment). Our observation that DuP 753 increased Na+ excretion is consistent with their findings. The effects of DuP 753 in the proximal tubular transport were previously interpreted as the result of blockade of endogenous Ang II.46 However, recent observations have raised concerns about the specificity of DuP 753.15 50 51 52 53 This compound can bind to non–angiotensin II binding sites50 or interfere with non–angiotensin-mediated responses.52 In addition, differential regulation of Ang II and DuP 753 binding sites was observed in rat glomeruli and human mesangial cells.54 DuP 753 can also block the effects of Ang-(1-7) in some sites, including the heart51 and kidney.15 53 We have observed that the antidiuretic effect of Ang-(1-7) in water-loaded rats is completely blocked by DuP 753.53 A similar observation was made by Garcia and Garvin15 in the proximal straight tubules and by Gironacci et al51 in the heart. Since Ang-(1-7) does not exert most of the Ang II actions mediated through AT1 receptors,3 these findings suggest the existence of at least two Ang-(1-7) receptor subtypes: one that is expressed, for example, in blood vessels55 and the brain (rostral ventrolateral medulla and caudal pressor area)10 20 and is not blocked by AT1 or AT2 antagonists10 20 and another, expressed in sites such as the hypothalamus,56 heart,51 and kidney,15 53 that can be blocked by DuP 753 and to a variable extent by AT2 antagonists. Nevertheless, the observation that DuP 753 can block at least some of the Ang-(1-7) effects on the kidney raises the possibility that blockade of endogenous Ang-(1-7) can contribute to its pharmacological effects.
The observation that acute systemic administration of an Ang-(1-7) antagonist significantly increased urinary water excretion, an effect not observed after acute treatment with AT1 or AT2 receptor antagonists, has important physiological implications. First, this finding strongly suggests that Ang-(1-7) acts endogenously as an antidiuretic hormone. Other data support this hypothesis: Ang-(1-7) immunoreactivity is present in brain areas related to hydromineral balance, the paraventricular and supraoptic nuclei of the hypothalamus and neurohypophysis13 ; Ang-(1-7) possesses a potent antidiuretic activity in vivo,14 substantiated by our findings in IMCDs and those of Garcia and Garvin15 in rat proximal straight tubules; this heptapeptide has a significant AVP-releasing activity in neurohypophysial explants7 ; and Ang-(1-7) plasma concentration increases approximately fourfold in salt-loaded rats, a condition in which levels of the other circulating components of the renin-angiotensin system decrease or do not change.6 A second important implication relates to the inability of the other angiotensin antagonists to modify urine output under the experimental conditions used in the present study. As discussed above, this finding strongly suggests that Ang-(1-7) exerts its actions through specific angiotensin receptors.
One important question regarding the possible role of Ang-(1-7) in the control of hydromineral balance is related to its participation as a short- and/or long-term factor in this regulatory mechanism. We have observed that Ang-(1-7) plasma levels increased with acute volume contraction (hemorrhage), with 24 hours of water deprivation, or with chronic increases in serum osmolality produced by salt loading.6 However, in recent studies we have found that acute (1 to 2 hours) increases in plasma osmolality did not increase plasma Ang-(1-7) levels (A.C. Carvalho and R.A.S.S., unpublished results, 1995). In addition, Ang-(1-7) plasma levels measured in water-loaded rats (37.62±2.19 pg/mL) did not differ from the values previously found in euvolemic rats in our laboratory (32.75±6.70 pg/mL).6 Although these observations do not exclude the possibility of local changes in Ang-(1-7) production in the kidney, it appears that Ang-(1-7) is likely to be more involved in the long-term control of osmoregulation. However, further studies are necessary to clarify the precise response of tissue and plasma Ang-(1-7) levels to acute and chronic imbalances of plasma osmolality.
In summary, we have found that the potent antidiuretic effect of Ang-(1-7) in water-loaded rats is probably mediated by glomerular and tubular effects, mostly related to water transport. More importantly, using a selective Ang-(1-7) antagonist, we obtained evidence that pointed toward an important physiological role for this peptide in the control of hydromineral balance. Finally, it should be pointed out that in patients treated with angiotensin-converting enzyme inhibitors, Ang-(1-7) plasma concentration has been reported to increase ninefold.57 This finding illustrates the importance of knowing the physiological significance of this heptapeptide, which may contribute to the pharmacological effects of angiotensin-converting enzyme inhibitors.
Selected Abbreviations and Acronyms
|AT1, AT2||=||angiotensin subtype 1, subtype 2|
|AVPa||=||[adamantaneacetyl1, O-Et-d-Tyr2,Val4, aminobutyryl6, Arg8,9-vasopressin|
|GFR||=||glomerular filtration rate|
|IMCD||=||inner medullary collecting duct|
|MAP||=||mean arterial pressure|
This work was partially supported by Fundação de Amparo a Pesquisa do Estado de Minas Gerais (FAPEMIG), Pro-reitoria de Pesquisa (PRPq-UFMG), Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), and Financiadora de Estudos e Projetos (FINEP) and by a National Institutes of Health grant (to M.C.K.). A.C.S.S. and K.T.P. were recipients of fellowships from Coordenadoria de Aperfeiçoamento de Pessoal (CAPES). N.C.V.B. was the recipient of a fellowship from CNPq. We are thankful to EI DuPont de Nemours & Co (Wilmington, Del) for the supply of DuP 753 and to Ciba-Geigy Ltd (Basel, Switzerland) for providing CGP 42112A. We also thank Dr Maria J. Campagnole-Santos for discussion.
- Received December 18, 1995.
- Revision received January 8, 1996.
- Accepted January 8, 1996.
Dzau VJ. Circulating versus local renin-angiotensin system in cardiovascular homeostasis. Circulation. 1988;77(suppl I):I-4-I-13.
Ferrario CM, Barnes KL, Block CH, Brosnihan KB, Diz DI, Khosla MC, Santos RAS. Pathways of angiotensin formation and function in the brain. Hypertension. 1990;15(suppl I):I-13-I-19.
Santos RAS, Brosnihan KB, Chappell MC, Pesquero JL, Chernicky CL, Greene LJ, Ferrario CM. Converting enzyme activity and angiotensin metabolism in the dog brainstem. Hypertension. 1988;11:153-157.
Santos RAS, Brosnihan KB, Jacobsen DW, DiCorleto PE, Ferrario CM. Production of angiotensin-(1-7) by human vascular endothelium. Hypertension. 1992;19(suppl II):II-56-II-61.
Schiavone MT, Santos RAS, Brosnihan KB, Khosla MC, Ferrario CM. Release of vasopressin from the rat hypothalamo-neurohypophysial system by angiotensin-(1-7) heptapeptide. Proc Natl Acad Sci U S A. 1988;85:4095-4098.
Campagnole-Santos MJ, Diz DI, Santos RAS, Khosla MC, Brosnihan KB, Ferrario CM. Cardiovascular effects of angiotensin (1-7) microinjected into the dorsal medulla of rats. Am J Physiol. 1989;257:H324-H329.
Felix D, Khosla MC, Imbodem H, Montani B, Ferrario CM. Neurophysiological responses to angiotensin-(1-7). Hypertension. 1991;17:1111-1114.
Campagnole Santos MJ, Heringer SB, Batista EN, Khosla MC, Santos RAS. Differential baroreceptor reflex modulation by centrally infused angiotensin peptides. Am J Physiol. 1992;263:R89-R94.
Garcia NH, Garvin JL. Angiotensin-(1-7) has a biphasic effect on fluid absorption in the proximal straight tubule. J Am Soc Nephrol. 1994;5:1133-1138.
Handa RK, Handa SE, Ferrario CM, Strandhoy JW. In vivo actions of angiotensin-(1-7) in the rat kidney. FASEB J. 1992;7:A221. Abstract.
Erdös EG, Skiedgel RA. Renal metabolism of angiotensin I and II. Kidney Int. 1990;38:24-27.
Santos RAS, Campagnole-Santos MJ, Baracho NCV, Fontes MAP, Silva LCS, Neves LAA, Oliveira DR, Caligiorne SM, Rodrigues ARV, Gropen C Jr, Carvalho WS, Simões e Silva AC, Khosla MC. Characterization of a new angiotensin antagonist selective for angiotensin-(1-7): evidence that the actions of angiotensin-(1-7) are mediated by specific angiotensin receptors. Brain Res Bull. 1994;35:293-298.
Magaldi AJ, César KR, Yano Y. Effect of insulin on water and urea transport in the inner medullary collecting duct. Am J Physiol. 1994;266:F394-F399.
Levinsky NG, Lieberthal W. Clearance techniques. In: Windhager EE, ed. Handbook of Physiology: Section 8: Renal Physiology. New York, NY: Oxford University Press; 1992; 1:227-247.
Pawloski-Dahm C, Fink GD. Chronic intravenous infusion of angiotensin-(1-7) produces increases in circulating vasopressin but not arterial pressure. Hypertension. 1992;20:69. Abstract.
Gellai M. Vasopressin antagonists in studies on the role of vasopressin in renal hemodynamics. In: Schrier RW, ed. Vasopressin. New York, NY: Raven Press Publishers; 1985:167-170.
Magaldi AJ, Rocha AS, Kudo LH. Effect of quinidine, prostaglandin E2, furosemide and calmodulin antagonist on water and urea permeabilities of rat papillary collecting duct. In: Proceedings of the IXth International Congress of Nephrology; June 11-16, 1984; Los Angeles, Calif. 424A. Abstract.
Nielsen S, Knepper M. Vasopressin activates collecting duct urea transporters and water channels by distinct physical processes. Am J Physiol. 1993;265:F204-F213.
Farina MDD, Santos RAS, Lopes MJ. Angiotensin-(1-7) has a biphasic effect on frog (Rana catesbeiana) skin osmotic permeability. Hypertension. 1995;25:1409. Abstract.
Douglas JG, Romero M, Hopfer U. Signaling mechanisms coupled to the angiotensin receptor of proximal tubular epithelium. Kidney Int. 1990;30(suppl 30):43-47.
Harris PJ, Skinner SL. Intra-renal interactions between angiotensin II and atrial natriuretic factor. Kidney Int. 1990;38(suppl 30):87-91.
Harris PJ, Navar LG. Tubular transport responses to angiotensin. Am J Physiol. 1985;248:F621-F630.
Benter IF, Ferrario CM, Morris M, Diz DI. Antihypertensive actions of angiotensin-(1-7) in spontaneously hypertensive rats. Am J Physiol. 1995;269:H313-H319.
Stoff JS, Rosa RM, Silva P, Epstein FH. Indomethacin impairs water diuresis in the DI rat: role of prostaglandins independent of ADH. Am J Physiol. 1981;241:F231-F237.
Namnum P, Insogna K, Boggish D, Hayslett JP. Evidence for bidirectional net movement of creatinine in the rat kidney. Am J Physiol. 1983;244:F719-F723.
Ernsberger P, Zhou J, Damon TH, Douglas JG. Angiotensin II receptor subtypes in cultured rat renal mesangial cells. Am J Physiol. 1992;263:F411-F416.
Jaiswal N, Diz DI, Chappell MC, Khosla MC, Ferrario CM. Stimulation of endothelial cell prostaglandin production by angiotensin peptides: characterization of receptors. Hypertension. 1992;19(suppl II):II-49-II-55.
Bello APC, Moura AM, Oliveira LA, Silva JR, Santos RAS. Efeitos cardiovasculares da infusão periférica de um antagonista da Angiotensina-(1-7) sobre a pressão arterial de ratos normotensos e hipertensos. Proc Fed Biol Exp. 1994:101. Abstract.
Camargo MJF, Lutterotti NV, Campbell WG, Pecker MS, James GD, Timmermans PBMWM, Laragh JH. Control of blood pressure and end-organ damage in maturing salt-loaded stroke-prone spontaneously hypertensive rats by oral angiotensin II blockade. Hypertension. 1993;11:31-40.
Cogan MG, Liu FY, Wong PC, Timmermans PBMWM. Comparison of inhibitory potency of nonpeptide angiotensin II receptor antagonists PD 123177 and DuP 753 on proximal nephron and renal transport. J Pharmacol Exp Ther. 1991;259:687-691.
Keeton TK, Campbell WB. The pharmacologic alteration of renin release. Pharmacol Rev. 1981;31:81-157.
Clark KL, Robertson MJ, Drew GM. Effects of the non-peptide angiotensin receptor antagonist, DuP 753, on basal renal function and on the renal effects of angiotensin II in the anesthetized dog. Br J Pharmacol. 1991;104:78-82.
Gironacci MM, Adler-Graschinsky E, Peña C, Enero MA. Effects of angiotensin II and angiotensin-(1-7) on the release of [3H]norepinephrine from rat atria. Hypertension. 1994;24:457-460.
Simões e Silva AC, Baracho NCV, Santos RAS. Contribuição dos receptores angiotensinérgicos AT1 e AT2 para o efeito antidiurético da Angiotensina-(1-7) em ratos. Proc Soc Bras Hipertensão. 1993:42. Abstract.
Chansel D, Bizet T, Vandermeersch S, Pham P, Levy B, Ardaillou R. Differential regulation of angiotensin II and losartan binding sites in glomeruli and mesangial cells. Am J Physiol. 1994;266:F384-F393.