Inhibition of Aldosterone Production by Adrenomedullin, a Hypotensive Peptide, in the Rat
Recently, we conducted in vitro studies and reported that adrenomedullin, a novel hypotensive peptide, inhibits aldosterone secretion by dispersed rat adrenal zona glomerulosa cells. To assess the physiological role of this inhibitory effect, we investigated the effect of adrenomedullin on aldosterone production in vivo. Male Sprague-Dawley rats were fed a normal sodium diet before the experiments. To begin the experimental procedure, we stimulated aldosterone production with a sodium-deficient diet or bilateral nephrectomy. After 3 days of sodium depletion or immediately after nephrectomy, we injected synthetic human adrenomedullin (2.5 nmol/kg SC) and repeated the injection three times at 6-hour intervals. Two hours after the last injection, the rats were decapitated and adrenal capsular tissue was collected. Adrenomedullin had no effect on plasma and adrenal aldosterone concentrations in the rats fed a normal sodium diet. Rats fed a sodium-deficient diet had significantly increased aldosterone concentrations in both plasma (4770.1±364.3 pmol/L) and adrenal gland (57.34±3.27 pmol per adrenal). Subsequently, injection of adrenomedullin significantly inhibited increases in concentrations (plasma, 2648.9±313.2 pmol/L; adrenal, 44.28±4.94 pmol per adrenal). In nephrectomized rats, increased aldosterone concentrations in plasma and adrenal gland were also significantly inhibited by adrenomedullin. In the second part of the study, plasma renin concentration, adrenal renin activity, plasma corticosterone concentration, serum potassium concentration, and plasma immunoreactive adrenomedullin concentration were examined for adrenomedullin effects. The first four were unaffected, and the last, plasma immunoreactive adrenomedullin, was elevated 15% to 30%. These in vivo results, together with our in vitro data, suggest that adrenomedullin may indeed play a physiological role in the control of blood pressure and electrolyte balance.
- adrenal gland
- zona glomerulosa
- adrenal renin
- renin-angiotensin system
A novel hypotensive peptide, AM, was discovered in human pheochromocytoma.1 This peptide consists of 52 amino acids in the human and 50 amino acids in the rat, and it shows approximately a 20% homology with calcitonin gene–related peptide.2 AM is distributed not only in pheochromocytoma but also in normal adrenal medulla and a variety of tissues, as well as circulated in blood.1 2 AM mRNA has been found in adrenal, lung, kidney, and heart tissues in both rats and humans2 3 and in vascular endothelial cells in rats.4
Although a strong vasodilating effect has been clearly demonstrated,1 2 little information concerning the physiological aspects of this peptide has been available. Recently, our laboratory reported that AM inhibits aldosterone secretion by dispersed adrenal glomerulosa cells in the rat, apparently specific to Ca2+-dependent pathways.5 AM inhibits aldosterone secretion stimulated by high potassium, Ang II, and the Ca2+ ionophore A23187 in vitro, yet again, the physiological effects of this inhibitory action have not been clarified.
In this study, we evaluated the in vivo effect of AM on aldosterone production in the rat. We stimulated aldosterone production using two different mechanisms: a sodium-deficient diet and bilateral nephrectomy. We demonstrated that the increased aldosterone production caused by these two stimuli was attenuated by treatment with AM. Furthermore, we evaluated the degree to which the renin-angiotensin system affects the relationship between aldosterone production and AM. We examined renin activity in both plasma and adrenal capsular tissue (zona glomerulosa) because the adrenal tissue renin-angiotensin system has been shown to be a regulating factor in aldosterone production and because the regulating mechanisms of adrenal renin and kidney renin differ.6 7 8 9 10 11 12 13 14
Male 6-week-old Sprague-Dawley rats (Charles River Japan Inc, Atsugi, Japan) were used. All rats were housed in a room with constant temperature and a 12-hour light/dark cycle (light, 7 am to 7 pm) and given water and a diet containing normal sodium (CRF-1, 2.9 g sodium/kg, Oriental Yeast Co Ltd) ad libitum at least 1 week before the experiments. All animal protocols were approved by the Ethics Committee of Animal Care at our institute.
Sodium-Deficient Diet Experiment
At 7 to 8 weeks of age (270±4 g, n=39), the rats were randomly divided into four groups that received a normal sodium diet (NORMAL, n=8), a normal sodium diet plus AM (NORMAL+AM, n=9), a sodium-deficient diet (0.05 g sodium/kg, Oriental Yeast Co Ltd) (LOW, n=11), and a sodium-deficient diet plus AM (LOW+AM, n=11). The sodium-deficient diet in the LOW and LOW+AM groups was started 3 days before the study. Synthetic human AM was purchased from Peptide Institute Inc and dissolved in normal saline solution (3 μmol/L). In the NORMAL+AM and LOW+AM groups, subcutaneous injections of AM (2.5 nmol/kg) were given four times at 6-hour intervals beginning at 11:30 am. Intravenous bolus injections of this amount of AM significantly decrease blood pressure in anesthetized rats.1 2 Before starting this study, we evaluated the effect of a subcutaneous injection of AM (2.5 nmol/kg) on blood pressure and heart rate. AM was injected into the rats subcutaneously under pentobarbital anesthesia (40 mg pentobarbital/kg IP), and their systolic pressure and heart rate were measured with the tail-cuff method (PS-200, Riken-kaihatsu Co Ltd) and recorded every 5 minutes over a 30-minute period after the injection. This injection had no significant effect on systolic pressure, and heart rate increased only slightly (from 398.4±18.3 to 421.4±20.2 beats per minute, n=5, P=NS). The NORMAL and LOW groups were injected with normal saline solution as a vehicle. Twenty hours after the first injection (2 hours after the last injection), the rats were decapitated, and blood was collected from the neck vessels into chilled tubes containing Na2EDTA (final concentration, 4 mmol/L). The plasma was separated and stored at −80°C for later analysis of PAC, PCC, and PRC. After the aorta was flushed with ice-cold normal saline solution, the adrenals were removed and cleaned of fat. The capsular portion (zona glomerulosa) of the adrenal was separated from the decapsular (fasciculata-medullary) portion, and the capsular tissue was frozen in liquid nitrogen and stored at −80°C.
At 8 to 9 weeks of age (333±16 g, n=26), the rats were divided into three groups: a sham-operated group (SHAM, n=8), a group that underwent bilateral nephrectomy (NEPEX, n=9), and a group that underwent nephrectomy and received AM (NEPEX+AM, n=9). In the NEPEX and NEPEX+AM groups, bilateral nephrectomy was performed between 2:30 and 3:30 pm with rats under ether anesthesia. In the NEPEX+AM group, AM was injected subcutaneously (2.5 nmol/kg) immediately after the surgery, and the injection was repeated three times at 6-hour intervals. In NEPEX rats, normal saline solution was injected (at the same frequency) as a vehicle after the surgery. In SHAM rats, the normal saline solution vehicle was injected after sham operation, again at the same intervals. Two hours after the last injection, between 10:30 and 11:30 am, the rats were decapitated, and plasma and adrenal capsular tissue were collected by the same method as described in the sodium-deficient experiment.
Measurement of Adrenal Renin Activity and Adrenal Aldosterone Concentration
The portions of the frozen adrenal capsular tissues from each rat were combined in 1 mL ice-cold Tris-acetate buffer (0.1 mol/L, pH 7.4) and homogenized with a polytetrafluoroethylene-glass homogenizer. After centrifugation for 20 minutes at 5000g and 4°C, the supernatant was aspirated and renin activity was measured by a slightly modified form of the method reported by Doi et al.15 Briefly, 50 μL supernatant was incubated with 85 μL Tris-acetate buffer containing bovine serum albumin (1 g/L), 5 μL 8-hydroxyquinoline (0.34 mol/L), 5 μL dimercaprol (0.16 mol/L), 25 μL EDTA (0.11 mol/L), 5 μL phenylmethylsulfonyl fluoride (0.10 mol/L), and 75 μL nephrectomized rat plasma as renin substrate. After a 1-hour incubation at 37°C, a 50-μL aliquot was taken for radioimmunoassay of generated Ang I. Ang I was measured with a commercially available kit (Toray-Fuji Bionics Inc).
After extraction of methylene chloride, aldosterone concentration in the supernatant was measured with a radioimmunoassay kit (Dainabot Co Ltd). Five milliliters of methylene chloride was added to 0.2 mL of supernatant; after centrifugation, the aqueous phase was discarded. Next, 0.5 mL of 0.1 mol/L NaOH was added to the solution, and again, the aqueous phase was discarded after centrifugation; then, 0.5 mL of 0.1 mol/L acetic acid was added, and the aqueous phase was discarded after centrifugation. The extract was evaporated by a SpeedVac concentrator (Savant Instruments Inc) and reconstituted with the zero standard for the radioimmunoassay kit.
Measurements of PAC, PCC, and PRC
Frozen plasma was thawed on ice and used for measurements of PAC, PCC, and PRC. PAC was measured by direct radioimmunoassay in all but the NEPEX rats, in which PAC was measured after extraction of methylene chloride with the method described for determination of the adrenal extract. PCC was measured by direct radioimmunoassay (Amersham). PRC measurements were done with the same method as the one described above for measurement of adrenal renin activity.
Measurement of Plasma ir-AM Concentration
Several aliquots of the frozen plasma were selected at random from each experimental group, and plasma levels of ir-AM were examined. This measurement was performed by radioimmunoassay with antiserum against human AM(40-52)-NH2, as previously reported by Sakata et al.16 This antiserum recognizes the C-terminal amide structure and shows 100% cross-reactivity between rat and human AM.16
Serum Potassium Concentration
Serum potassium is an important regulator of aldosterone production in nephrectomized animals.17 We evaluated the effect of AM on serum potassium concentration in the NEPEX and SHAM rats. A new set of rats was prepared for this experiment because arterial blood was collected through a puncture in the abdominal aorta to avoid hemolysis. Twenty male Sprague-Dawley rats at 7 weeks of age (241±3 g) were divided into four groups (SHAM, n=4; SHAM+AM, n=4; NEPEX, n=6; NEPEX+AM, n=6); bilateral nephrectomy or sham surgery was performed; and AM or vehicle (normal saline) solution was injected as described in the nephrectomy study. Twenty hours after surgery, the rats were anesthetized by ether inhalation and blood was collected from the abdominal aorta for measurement of serum potassium concentration.
Results of measurements of adrenal renin activity were normalized to picomoles of Ang I per one adrenal capsular tissue per hour of renin assay incubation, and measurements of adrenal aldosterone were normalized to picomoles per one adrenal capsular tissue. Results are expressed as mean±SE. Data were analyzed statistically by one-factor ANOVA and Fisher's probability of the least significant difference test. Student's unpaired t test was used for comparison of the two groups. Significance was defined as a value of P<.05.
Sodium-Deficient Diet Experiment
Effects of AM on PAC and adrenal aldosterone concentration are shown in Fig 1⇓. In NORMAL rats, PAC and adrenal aldosterone concentration were 170.6±33.6 pmol/L and 11.76±1.39 pmol per adrenal, respectively. The addition of AM to a normal salt diet (NORMAL+AM rats) had no significant effects on PAC (105.7±16.4 pmol/L) or adrenal aldosterone concentration (11.1±3.08 pmol per adrenal). A sodium-deficient diet (LOW rats) significantly increased PAC (4770.1±364.3 pmol/L) and adrenal aldosterone concentration (57.34±3.27 pmol per adrenal) by 28-fold and 5-fold, respectively. These increases were significantly inhibited by AM injections (LOW+AM rats; PAC, 2648.9±313.2 pmol/L, P<.01; adrenal aldosterone, 44.28±4.94 pmol per adrenal, P<.05).
Although PCC tended to increase slightly after a sodium-deficient diet (307.0±107.2 nmol/L in NORMAL rats versus 386.2±108.3 in LOW rats, P=NS), AM had no significant effect on PCC in rats on either diet (Fig 2⇓).
To assess the contribution of plasma and tissue renin activities to the inhibition of aldosterone concentration, we evaluated PRC and adrenal renin activity. Effects of AM on PRC and adrenal renin activity are shown in Fig 3⇓. In rats fed a normal diet, AM had no significant effect on either PRC (5.99±0.76 nmol Ang I/L per hour in NORMAL rats versus 3.52±0.65 in NORMAL+AM rats, P=NS) or adrenal renin activity (1.47±0.33 pmol Ang I per adrenal per hour in NORMAL rats versus 1.93±0.61 in NORMAL+AM rats, P=NS). Although LOW rats showed significantly increased PRC and adrenal renin activity compared with NORMAL rats (PRC, 15.74±1.12 nmol Ang I/L per hour in LOW rats, P<.01; adrenal renin, 5.93±1.86 pmol Ang I per adrenal per hour in LOW rats, P<.05), AM had no significant effect on either PRC (13.29±1.19 nmol Ang I/L per hour) or adrenal renin activity (7.21±1.56 pmol Ang I per adrenal per hour).
Effects of AM on PAC and adrenal aldosterone concentration are shown in Fig 4⇓. In control (SHAM) rats, PAC and adrenal aldosterone concentration were 230.8±45.2 pmol/L and 12.6±1.39 pmol per adrenal, respectively. Nephrectomy significantly increased both PAC (4037.2±530.4 pmol/L) and adrenal aldosterone concentration (43.22±2.22 pmol per adrenal) by 18-fold and 3-fold, respectively. In the NEPEX+AM rats, the increases in PAC and adrenal aldosterone were significantly inhibited compared with values in NEPEX rats (PAC, 2544.0±375.6 pmol/L, P<.05; adrenal aldosterone, 32.13±3.94 pmol per adrenal, P<.01). In addition, PCC increased significantly in response to nephrectomy (201.0±61.5 nmol/L in SHAM rats versus 1112.6±92.1 in NEPEX rats, P<.01); however, this increase was not inhibited by AM (1214.6±169.4 nmol/L in NEPEX+AM rats, P=NS) (Fig 5⇓).
Effects of AM on PRC and adrenal renin activity are shown in Fig 6⇓. As expected, PRC decreased to a nearly undetectable level after nephrectomy (7.78±0.93 nmol Ang I/L per hour in SHAM rats versus 0.15±0.03 in NEPEX rats, P<.01), whereas adrenal renin significantly increased (3.96±0.58 pmol Ang I per adrenal per hour in SHAM rats versus 11.91±1.56 in NEPEX rats, P<.01). The AM injection after nephrectomy (NEPEX+AM rats) had no significant effect on either PRC (0.19±0.04 nmol Ang I/L per hour) or adrenal renin (11.37±1.55 pmol Ang I per adrenal per hour).
Plasma ir-AM Concentration
Because a sodium-deficient diet had no effect on plasma ir-AM concentrations (6.78±0.18 pmol/L, n=5, in NORMAL rats versus 6.30±0.59, n=6 in LOW rats), rats injected with vehicle solution on either diet were recognized as a control group and were compared with the AM-treated rats. When the AM-injected rats were killed (2 hours after the last AM injection), plasma ir-AM concentration was increased by approximately 30% compared with the control rats (Table 1⇓). In the nephrectomy experiment, SHAM rats showed a relatively high plasma ir-AM concentration. Plasma ir-AM concentration in the NEPEX group was significantly higher than in the SHAM group, probably because of the disappearance of renal clearance. AM injection (NEPEX+AM rats) further increased plasma ir-AM concentration by approximately 15% (Table 1⇓).
Serum Potassium Concentration
Serum potassium concentration was not significantly altered by the addition of AM to either SHAM or NEPEX groups. However, it was significantly elevated when the NEPEX and SHAM groups were compared (4.75±0.12 mmol/L in SHAM rats versus 8.10±0.27 in NEPEX rats, P<.01) (Table 2⇓).
In this study, we found that subcutaneous injection of AM significantly decreased PAC levels stimulated by both a sodium-deficient diet and bilateral nephrectomy. Furthermore, we evaluated adrenal aldosterone content and found that the decreases in PAC after AM injection were associated with significant decreases in adrenal aldosterone, establishing a strong positive correlation between PAC and adrenal aldosterone (r=.91, P<.01 in the sodium-deficient experiment; r=.89, P<.01 in the nephrectomy experiment). However, the inhibitory effect of AM did not extend to the plasma level of corticosterone, which is a major product of adrenal zona fasciculata cells in the rat. It is most likely that after injection, AM suppressed aldosterone production in zona glomerulosa cells and subsequently suppressed secretion and the plasma level. Our study also suggests that the inhibitory effect of AM on adrenal steroidogenesis might be specific to zona glomerulosa cells.
A sodium-deficient diet and nephrectomy stimulate aldosterone production by different mechanisms. Although aldosterone production is regulated by the balance of many factors, the principal factors are known to include the renin-angiotensin system, the potassium ion, and ACTH.17 During sodium depletion, the plasma renin-angiotensin system plays an important role in the stimulation of aldosterone biosynthesis.18 After nephrectomy, PRC falls to nearly zero, whereas pituitary ACTH release and plasma potassium concentration increase.19 20
In the present experiments, increased PAC and adrenal aldosterone in response to sodium depletion were significantly inhibited by AM, but PRC was not inhibited. This suggests that the inhibition of aldosterone production is not due to the suppression of PRC levels. Furthermore, PCC was not inhibited by the introduction of AM, suggesting that the plasma ACTH level might not be significantly affected by AM. This in vivo observation contradicts a recent in vitro report by Samson et al21 that AM significantly inhibited basal ACTH secretion from cultured rat anterior pituitary cells and ACTH secretion stimulated by ACTH-releasing hormone.
By further examination of serum potassium concentrations, we found that inhibition of aldosterone production by AM might not be due to the suppression of the potassium ion serum level. In the present study, serum potassium concentration significantly increased in NEPEX rats compared with SHAM rats, but it did not change after AM injection.
The adrenal tissue renin-angiotensin system has been shown to be a regulating factor of aldosterone production. Several experiments, both in vivo and in vitro, have shown a correlation between adrenal renin and aldosterone levels.6 7 It has been reported that the production of renin in cultured rat and bovine adrenal zona glomerulosa cells is regulated by ACTH, potassium concentration, and Ang II7 8 9 10 and that angiotensin-converting enzyme inhibitors and angiotensin type 1 receptor antagonists can reduce ACTH- and potassium-stimulated aldosterone production.7 11 12 13 14 These reports suggest that the stimulation of aldosterone production by ACTH and potassium are mediated at least in part by the adrenal renin-angiotensin system. In our present study, we showed significant increases in adrenal renin after a sodium-deficient diet and after nephrectomy, consistent with earlier reports.6 15 20 22 AM did not affect basal and stimulated levels of adrenal renin. Thus, adrenal renin and plasma renin might not be primary determinant factors in the suppression of aldosterone production.
Our laboratory previously reported that AM significantly inhibits aldosterone secretion stimulated by Ang II, potassium, and the Ca2+ ionophore A23187 from dispersed rat adrenal zona glomerulosa cells but that aldosterone stimulated by ACTH and dibutyryl cAMP was not inhibited.5 These data suggest that AM inhibits the Ca2+-dependent increase in aldosterone production. The results of the present in vivo study confirm and extend our in vitro data. Presumably, AM acts directly on the zona glomerulosa cells and inhibits aldosterone production primarily by blocking the action of Ang II during sodium depletion and high serum potassium concentration after nephrectomy.
Although the AM receptor has not been well characterized, several investigators have shown that AM cross-reacts at least in part with the calcitonin gene–related peptide receptor23 24 25 and that the vasodilating effect of AM is mediated by intracellular cAMP.24 26 Recently, Kapas et al27 identified a cDNA encoding AM receptor in the rat lung. They reported that the AM receptor gene is expressed in the lung, adrenal, heart, spleen, cerebellum, and cerebral cortex, and they noted that when the receptor cDNA is expressed in COS-7 cells, cAMP production by the cells increases in response to AM. Our previously reported observations concerning adrenal glomerulosa cells, both in vivo and in vitro, cannot be simply explained by the stimulation of an adenylate cyclase–dependent pathway because cAMP is widely accepted as a stimulant of aldosterone production.28 One possible explanation is the existence of receptor subtypes that activate different signaling mechanisms. In support of this possibility, Owji et al29 reported that AM binding sites in the rat lung can be distinguished from those in the heart, both pharmacologically and by molecular weight. Furthermore, Shimekake et al30 reported that in bovine vascular endothelial cells, AM causes not only cAMP but also [Ca2+]i increases, leading to the activation of nitric oxide synthesis and release. This suggests that the signal transduction mechanism must be different from that in vascular smooth muscle cells.
In a study using human subjects, Ishimitsu et al31 have reported that plasma AM concentration increases in patients with hypertension and renal failure by 26% to 45% and 78% to 214%, respectively, compared with normotensive subjects. The amount of AM used in our experiments did not significantly decrease systolic pressure, and the increase in plasma ir-AM measured by radioimmunoassay was less than 30% at the time the rats were killed. It is therefore possible that aldosterone production is modulated by physiological changes of plasma AM level.
In conclusion, AM significantly inhibited aldosterone production stimulated by a sodium-deficient diet and bilateral nephrectomy but did not affect PCC, PRC, adrenal renin activity, and serum potassium concentration. The present results confirm and extend our previous in vitro data. This in vivo data together with our in vitro data suggest the possibility that AM inhibits aldosterone production in the adrenal gland and plays a physiological role in the control of blood pressure and electrolyte balance via this inhibitory action. The precise mechanism of this inhibitory effect remains to be established.
Selected Abbreviations and Acronyms
|Ang I, II||=||angiotensin I, II|
|PAC||=||plasma aldosterone concentration|
|PCC||=||plasma corticosterone concentration|
|PRC||=||plasma renin concentration|
- Received December 11, 1995.
- Revision received January 19, 1996.
- Revision received April 2, 1996.
Kitamura K, Sakata J, Kangawa K, Kojima M, Matsuo H, Eto T. Cloning and characterization of cDNA encoding a precursor for human adrenomedullin. Biochem Biophys Res Commun. 1994;194:720-725.
Doi Y, Atarashi K, Franco-Saenz R, Mulrow PJ. Effect of changes in sodium or potassium balance, and nephrectomy, on adrenal renin and aldosterone concentrations. Hypertension. 1984;6(suppl I):I-124-I-129.
Yamaguchi T, Naito Z, Stoner GD, Franco-Saenz R, Mulrow PJ. Role of adrenal renin-angiotensin system on adrenocorticotropic hormone- and potassium-stimulated aldosterone production by rat adrenal glomerulosa cells in monolayer culture. Hypertension. 1990;16:635-641.
Yamaguchi T, Franco-Saenz R, Mulrow PJ. Effect of angiotensin II on renin production by rat adrenal glomerulosa cells in culture. Hypertension. 1992;19:263-269.
Wang Y, Yamaguchi T, Franco-Saenz R, Mulrow PJ. Regulation of renin gene expression in rat adrenal zona glomerulosa cells. Hypertension. 1992;20:776-781.
Oda H, Lotshaw DP, Franco-Saenz R, Mulrow PJ. Local generation of angiotensin II as a mechanism of aldosterone secretion in rat adrenal capsules. Proc Soc Exp Biol Med. 1991;196:175-177.
Gupta P, Franco-Saenz R, Mulrow PJ. Locally generated angiotensin II in the adrenal gland regulates basal, corticotropin-, and potassium-stimulated aldosterone secretion. Hypertension. 1995;25:443-448.
Tan SY. Control of adrenal secretion of mineralocorticoids. In: Mulrow PJ, ed. The Adrenal Gland. New York, NY: Elsevier Science Publishing Co Inc; 1986:153-167.
McCaa RE. Role of the renin-angiotensin system in the regulation of aldosterone biosynthesis and arterial pressure during sodium deficiency. Circ Res. 1977;40(suppl I):I-157-I-162.
Douglas JG. Mechanism of adrenal angiotensin II receptor changes after nephrectomy in rats. J Clin Invest. 1981;67:1171-1176.
Baba K, Doi Y, Franco-Saenz R, Mulrow PJ. Mechanisms by which nephrectomy stimulates adrenal renin. Hypertension. 1986;8:997-1002.
Naruse M, Inagami T. Markedly elevated specific renin levels in the adrenal in genetically hypertensive rats. Proc Natl Acad Sci U S A. 1982;79:3295-3299.
Kapas S, Catt KJ, Clark AJL. Cloning and expression of cDNA encoding a rat adrenomedullin receptor. J Biol Chem. 1995;270:25344-25347.
Shimekake Y, Nagata K, Ohta S, Kambayashi Y, Teraoka H, Kitamura K, Eto T, Kangawa K, Matsuo H. Adrenomedullin stimulates two signal transduction pathways, cAMP accumulation and Ca2+ mobilization, in bovine aortic endothelial cells. J Biol Chem. 1995;270:4412-4417.
Ishimitsu T, Nishikimi T, Saito Y, Kitamura K, Eto T, Kangawa K, Matsuo H, Omae T, Matsuoka H. Plasma levels of adrenomedullin, a newly identified hypotensive peptide, in patients with hypertension and renal failure. J Clin Invest. 1994;94:2158-2161.