Plasma Adrenomedullin Concentrations in Essential Hypertension
Abstract We designed the present study to assess any changes in plasma concentrations of the novel vasorelaxant peptide adrenomedullin in patients with essential hypertension. Plasma adrenomedullin concentrations were measured in 45 patients with untreated essential hypertension, 15 patients with borderline hypertension, and 30 normotensive control subjects. After 4 weeks of effective calcium channel blocker–based antihypertensive therapy, adrenomedullin concentrations were measured again. The concentrations were higher in hypertensive patients with increased serum creatinine levels or decreased glomerular filtration rates compared with borderline hypertensive patients and normotensive subjects, although values in normotensive and hypertensive individuals overlapped. Plasma adrenomedullin concentrations were positively correlated with serum creatinine levels and inversely correlated with glomerular filtration rates in the hypertensive patients, whereas adrenomedullin values were not correlated with blood pressure level, left ventricular mass index, or left ventricular ejection fraction. Despite blood pressure control with antihypertensive therapy, plasma adrenomedullin concentrations were not changed. Reversed-phase high-performance liquid chromatographic analysis showed that a major component of immunoreactive adrenomedullin in the plasma of normotensive subjects and hypertensive patients is human adrenomedullin-(1-52). These results indicate that plasma adrenomedullin concentrations are elevated in many hypertensive patients with renal dysfunction and its major component is human adrenomedullin-(1-52).
A novel vasorelaxant peptide, adrenomedullin, has been isolated from the acid extract of human pheochromocytoma.1 This 52–amino acid peptide has one intracellular disulfide bond and shows homology with CGRP.1 It has been demonstrated that intravenous injection of adrenomedullin causes a potent and long-lasting hypotensive effect in anesthetized rats and that this peptide binds to specific receptors on platelet membranes to increase intracellular cAMP.1 Recently, Ishizaka et al2 and Eguchi et al3 have demonstrated that adrenomedullin potently stimulates cAMP formation in rat vascular smooth muscle cells via its specific receptors. Subsequently, we4 have shown that rat and human adrenomedullin stimulates cAMP formation in cultured rat glomerular mesangial cells. More recently, Hirata et al5 have demonstrated that adrenomedullin increases GFR and urinary sodium excretion and markedly decreases renal vascular resistance in anesthetized rats. These observations may raise the hypothesis that adrenomedullin modulates vascular tone or glomerular function in certain pathological states.
To assess any changes in plasma adrenomedullin concentrations in essential hypertension, we measured plasma immunoreactive adrenomedullin concentrations in patients with untreated essential hypertension. We compared the results with those of patients with borderline hypertension and normotensive control subjects. We also measured adrenomedullin concentrations after 4 weeks of effective calcium channel blocker–based antihypertensive therapy.
Between January and December 1994 we recruited 45 patients with established essential hypertension and 15 patients with borderline hypertension for this study from a population of approximately 200 patients seen in our department. Routine laboratory studies of all patients included assays of serum electrolytes, serum creatinine, blood urea nitrogen, and fasting blood glucose level; liver function tests; urinalysis; a chest roentgenogram; and an electrocardiogram. On the basis of the results of the laboratory tests and WHO,6 subjects were placed into one of the following three diagnostic categories: Normal BP was defined as a systolic pressure equal to or less than 140 mm Hg and a diastolic pressure equal to or less than 90 mm Hg. Hypertension was defined as a systolic pressure equal to or greater than 160 mm Hg or a diastolic pressure equal to or greater than 95 mm Hg, or both. The term borderline hypertension was used to denote BP values between the normal and hypertensive ranges as described above. Secondary hypertension was excluded by clinical history; physical examination; and routine laboratory tests, including measurements of plasma renin activity, aldosterone, catecholamine, cortisol, and thyroid hormones; and an excretory urogram or renal arteriogram. None of these patients had signs or symptoms of cardiac or hepatic failure or of diabetes. In addition, none had clinical evidence of pulmonary disease, angina pectoris, or myocardial infarction. Only hypertensive patients who had no previous antihypertensive drug treatment or whose antihypertensive drug treatment had been washed out for at least 2 weeks were included in this study.
After the initial evaluation 34 hypertensive patients, who agreed with the aim of this study, were started on antihypertensive therapy with a calcium channel blocker (slow-release nicardipine, slow-release nifedipine, barnidipine, or amlodipine). This therapy was continued in the majority of patients for 4 weeks. Fifteen of the 34 hypertensive patients required a second or third drug to adequately lower BP. Plasma adrenomedullin concentrations were determined before the initiation of therapy and after 4 weeks of effective therapy with antihypertensive drugs.
Arterial BP was measured by a mercury sphygmomanometer after the patients had rested sitting for 30 minutes in a quiet, warm room. The mean of three BP measurements was used to classify the subjects.
A blood sample (5 mL) was drawn immediately into ice-chilled tubes containing Trasylol (5×105 kallikrein inactivator units/L) and EDTA (1 g/L). Plasma was separated by centrifugation for 10 minutes at 4°C and immediately frozen and stored at −80°C until radioimmunoassay.
Immunoreactive adrenomedullin was extracted from plasma by a modified method previously described in our laboratory.7 Briefly, 2 mL of plasma was diluted with 3 mL of 4% acetic acid. After centrifugation the solution was pumped at the rate of 1 mL/min through a Sep-Pak C18 cartridge (Waters Associates). After the cartridge had been washed with distilled water, the adsorbed peptides were eluted with 4 mL of 50% acetonitrile containing 0.1% trifluoroacetic acid. After evaporation with a centrifugal evaporator (model RD-31, Yamato Scientific Co), the dry residue was dissolved in an assay buffer. The recovery rate was calculated by the addition of three amounts of cold human adrenomedullin-(1-52) (5, 20, and 100 pg/mL [0.8, 3.3, and 16.6 pmol/L]) to plasma pretreated with the Sep-Pak C18 cartridge. The recovery rate was 75±4%.
The plasma immunoreactive adrenomedullin concentration was measured with an antibody against synthetic human adrenomedullin-(1-52) and 125I–human adrenomedullin-(1-52) (Peninsula Laboratories Inc). This antibody does not show any cross-reactivity with human adrenomedullin-(13-52), rat adrenomedullin-(1-50), human CGRP, endothelin-1, α-human atrial natriuretic peptide-(1-28), brain natriuretic peptide-32, or C-type natriuretic peptide-22.
Radioimmunoassay was done in the assay buffer of 0.01 mol/L sodium phosphate, pH 7.4, containing 0.05 mol/L sodium chloride, 0.1% bovine serum albumin, 0.1% Nonidet P-40, and 0.01% sodium azide. Sample (100 μL) or standard human adrenomedullin (100 μL) was dissolved in the assay buffer and then incubated for 24 hours at 4°C. Approximately 15 000 cpm of 125I–human adrenomedullin was added to each reaction and incubated for an additional 24 hours. After the second 24-hour incubation, 100 μL of diluted normal rabbit serum (Organon Teknika Corp) and 100 μL of diluted goat anti-rabbit IgG (Peninsula Laboratories Inc) were added, and the mixture was again incubated for 24 hours. After the third incubation the precipitate was collected by centrifugation at 1700g for 30 minutes. The supernatant was removed by aspiration and the pellet counted for 125I with a gamma counter. The effective range of the standard curve was between 2 and 200 pg of human adrenomedullin per assay tube. The 50% intercept was 19 pg of human adrenomedullin (Fig 1⇓). The interassay variation was 12%, and the intra-assay variation was 5%.
Reversed-phase HPLC was performed with an octadecylsilica column (0.46×25 cm, TSK-gel, ODS-80, TOSOH Co) eluted with a linear gradient of acetonitrile from 10% to 80% (10% to 60%, 35 minutes; 60% to 80%, 10 minutes; 80%, 5 minutes) in 0.1 mol/L sodium chloride with a flow rate of 1 mL/min by a method essentially the same as described previously.8 One-milliliter fractions were collected and assayed by radioimmunoassay. For chromatographic analysis of immunoreactive adrenomedullin, 30 mL of pooled plasma of hypertensive patients and normotensive control subjects was separated and treated by reversed-phase HPLC.
Reversed-phase HPLC profiles of immunoreactive adrenomedullin in extracts of pooled plasma from normotensive subjects and hypertensive patients are shown in Fig 2⇓. Adrenomedullin immunoreactivity in the plasma consisted of one major component, which was eluted in the position of standard human adrenomedullin-(1-52). The elution profile of the hypertensive patients was not different from that of the normotensive control subjects.
Serum sodium and potassium, serum and urinary creatinine, and blood urea nitrogen were measured by routine methods in the Department of Clinical Chemistry, Osaka City University Medical School. GFR was calculated by endogenous creatinine clearance as previously described.8
LV mass was evaluated by M-mode echocardiography (Sonolayer αSSA-270A, Toshiba) as previously described.7 Measurements were made according to the recommendations of the American Society of Echocardiography with the leading-edge to leading-edge convention.9 The LV internal dimension, ventriculum septum, and LV posterior wall were measured at end diastole as defined by the onset of the QRS complex. Measurements made in accordance with American Society of Echocardiography criteria were used in the formula of Troy et al10 : LV Mass (g)=1.05×[(LV Internal Diameter+LV Septal Wall Thickness+Posterior Wall Thickness)3]−(LV Internal Diameter).3 LV mass was normalized for body surface area (values are reported here as LV mass indexes). LV ejection fraction was calculated by standard techniques.7
Statistical analysis was done by linear regression analysis or Scheffé’s test for multiple comparisons preceded by ANOVA where appropriate.11 Comparisons between the values before and after therapy were analyzed by paired ANOVA and reexamined by the method of Greenhouse and Geisser.12
Table 1⇓ shows various parameters of the normotensive subjects and borderline hypertensive and hypertensive patients. There were no significant differences in age, sex distribution, serum sodium, and serum potassium among the three groups. As expected, the mean arterial pressure of hypertensive patients was significantly higher than those of normotensive subjects and borderline hypertensive patients. Hypertensive patients had slightly but significantly higher serum creatinine levels and blood urea nitrogen than borderline hypertensive patients and normotensive control subjects. The mean GFR of hypertensive subjects was 76±19 mL/min (range, 42 to 118 mL/min). This value was slightly but significantly lower than the mean value for GFR in age-matched healthy subjects (P<.05, n=20, 98±9 mL/min). Hypertensive patients in this study had almost normal LV ejection fraction (range, 65% to 86%). The mean LV ejection fraction of age-matched healthy subjects was 81±5% (n=15). Most hypertensive patients also had almost normal LV mass index, but some hypertensive patients had moderate LV hypertrophy (range of LV mass index, 115 to 162 g/m2). The mean LV mass index of age-matched healthy subjects was 117±4 g/m2 (n=15). The mean LV posterior and septal wall thicknesses of hypertensive subjects in this study were 11.4±2.4 and 10.5±1.8 mm, respectively. These values were slightly greater than each normal value (n=15; posterior wall thickness, 9.9±0.4 mm; septal wall thickness, 9.4±0.5 mm).
The plasma adrenomedullin concentrations of the three groups are shown in Fig 3⇓. The mean concentrations in the normotensive subjects and borderline hypertensive and hypertensive patients were 18±2, 20±4, and 34±14 pg/mL, respectively (3.0±0.3, 3.3±0.7, and 5.6±2.0 pmol/L, respectively). The mean concentrations of the hypertensive patients were significantly higher than those of borderline hypertensive patients (P<.05) and normotensive control subjects (P<.05). There was no significant difference in plasma adrenomedullin concentrations between the borderline hypertensive patients and normotensive subjects.
In the hypertensive group the plasma adrenomedullin concentrations were strongly correlated with serum creatinine levels (Fig 4A⇓). Adrenomedullin concentrations were also correlated with blood urea nitrogen levels (Fig 4B⇓), but this correlation was weaker compared with the correlation of adrenomedullin concentrations and serum creatinine levels. A strong inverse correlation between GFR and plasma adrenomedullin concentrations was found in the hypertensive group (Fig 4C⇓).
In the hypertensive group adrenomedullin concentrations were not correlated with mean arterial pressure, LV mass index, or LV ejection fraction (Fig 5⇓).
Table 2⇓ shows the plasma adrenomedullin concentrations of hypertensive patients with or without renal dysfunction. The 28 hypertensive patients with increased serum creatinine levels (≥1.2 mg/dL [106.1 μmol/L]) or 30 patients with decreased GFR (≤80 mL/min) had clearly higher concentrations of plasma adrenomedullin than those of the normotensive subjects or borderline hypertensive patients. In contrast, the hypertensive patients with normal serum creatinine level and normal GFR did not have higher concentrations of plasma adrenomedullin than the normotensive subjects or borderline hypertensive patients.
Fig 6⇓ shows mean arterial pressure and plasma adrenomedullin concentrations initially and 4 weeks after effective antihypertensive therapy in 34 hypertensive patients. Mean arterial pressure fell significantly from an initial value of 129±7 to 111±50 mm Hg (P<.01). In contrast, plasma adrenomedullin concentrations were not changed (35±14 to 34±14 pg/mL [5.8±2.3 to 5.6±2.3 pmol/L]). Serum creatinine levels were also not changed (1.5±0.6 to 1.5±0.5 mg/dL [132.6±53.0 to 132.6±44.2 μmol/L]).
Fig 7⇓ shows the correlations of plasma adrenomedullin concentrations and serum creatinine and BP levels 4 weeks after effective antihypertensive therapy. Adrenomedullin concentrations were strongly correlated with serum creatinine levels, whereas they were not correlated with mean arterial pressure.
Adrenomedullin is a novel vasorelaxant peptide recently isolated from pheochromocytoma.1 Ichiki et al13 and Nishikimi et al14 have demonstrated that this peptide is present not only in human adrenal medulla but also in human plasma. The plasma adrenomedullin concentrations reported here were not very different from values reported by these investigators.13 14 Furthermore, we showed that adrenomedullin immunoreactivity in the plasma of hypertensive patients and normotensive subjects consisted of one major component with the same retention time as that of authentic human adrenomedullin-(1-52). It is therefore likely that human adrenomedullin-(1-52) is the major circulating form of adrenomedullin in both normotensive and hypertensive individuals.
In the current study we found that plasma immunoreactive adrenomedullin concentrations were elevated in many patients in whom renal function was impaired. In fact, adrenomedullin concentrations were positively correlated with serum creatinine levels and inversely correlated with GFR in the hypertensive group. Ishimitsu et al15 have recently demonstrated that the plasma adrenomedullin concentrations in hypertensive subjects with signs of organ damage (WHO stage II) were significantly higher than values in normotensive control subjects and hypertensive subjects without organ damage (WHO stage I). Our data seem to be compatible with their report.15 However, the present study indicates that plasma adrenomedullin concentrations are not elevated in hypertensive patients with normal renal function compared with normotensive control subjects. Indeed, plasma adrenomedullin concentrations were not correlated with BP level in both the untreated and treated hypertensive groups. Furthermore, elevated plasma adrenomedullin concentrations in hypertensive patients were not changed despite BP control by 4 weeks of antihypertensive therapy. It seems, therefore, that high BP itself is not the main factor involved in the observed elevation of plasma adrenomedullin concentrations. In addition, plasma adrenomedullin concentrations were not correlated with LV mass index or LV ejection fraction in the hypertensive group. Taken together, these observations suggest that elevation of plasma adrenomedullin concentrations in hypertensive patients is related to renal dysfunction and not to high BP itself, cardiac dysfunction, or cardiac hypertrophy, at least in our study population. In this respect, our results are different from the data reported by Ishimitsu et al.15 In their study plasma adrenomedullin concentrations were slightly but significantly elevated in uncomplicated essential hypertension (WHO stage I). The precise reasons for this difference are not clear at present. The most likely explanation is that in their study some elderly hypertensive patients classified as WHO stage I might have had modestly reduced GFR. Actually, the GFR or renal blood flow in elderly subjects with high BP is often modestly reduced16 17 18 despite normal levels of serum creatinine and blood urea nitrogen and no proteinuria.
Animal experiments suggest that adrenomedullin elicits a potent and long-lasting hypotensive activity comparable to that of CGRP.1 In addition, adrenomedullin potently stimulates cAMP formation in cultured rat vascular smooth muscle cells.2 3 Recently, adrenomedullin was shown to be synthesized in and secreted from vascular endothelial cells.19 If the elevation of plasma adrenomedullin concentration reflects increased synthesis and if the local concentration of adrenomedullin at the site of production is greater than the plasma concentration, this peptide may modulate vascular tone via paracrine effects on adjacent vascular smooth muscle. Furthermore, we have recently shown that both rat and human adrenomedullin potently stimulate cAMP formation in cultured rat glomerular mesangial cells.4 Very recently, Hirata et al5 demonstrated that adrenomedullin increases GFR and urinary sodium excretion and markedly decreases renal vascular resistance in rats. In the same study their videomicroscopic analysis revealed that adrenomedullin increases the diameters of both afferent and efferent arterioles of the glomeruli. Adrenomedullin also inhibits the production of the vasoconstrictive peptide endothelin-1 in vascular smooth muscle cells, probably through a cAMP-dependent process.20 Although the exact mechanism of the elevation of plasma adrenomedullin concentrations in essential hypertension is not clear, these observations suggest that the observed elevation of plasma adrenomedullin in hypertensive patients may be part of a compensatory mechanism to offset further development of renal dysfunction. However, further studies will be necessary to clarify the physiological significance of adrenomedullin in essential hypertension. It is also necessary to clarify the exact pharmacokinetics of adrenomedullin, including degradation in hypertensive patients with renal functional damage.
Selected Abbreviations and Acronyms
|CGRP||=||calcitonin gene–related peptide|
|GFR||=||glomerular filtration rate|
|HPLC||=||high-performance liquid chromatography|
|WHO||=||World Health Organization|
This work was supported by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science, and Culture, Japan (572-690-231-646) and by a research grant of Osaka City University. The authors thank Atsumi Ohnishi and Yuka Inoshita for their technical assistance.
- Received June 5, 1995.
- Revision received August 8, 1995.
- Accepted September 11, 1995.
Hirata Y, Hayakawa H, Suzuki Y, Suzuki E, Ikenouchi H, Kohmoto O, Kimura K, Kitamura K, Eto T, Kangawa K, Matsuo H, Omata M. Mechanism of adrenomedullin-induced vasodilation in the rat kidney. Hypertension. 1995;25:790-795.
World Health Organization. Arterial Hypertension: Report of a WHO Expert Committee. Geneva, Switzerland: World Health Organization, 1978. WHO Technical Report Series 628.
Kohno M, Yasunari K, Murakawa K, Yokokawa K, Horio T, Fukui T, Takeda T. Plasma immunoreactive endothelin in essential hypertension. Am J Med. 1990;80:614-618.
Sahn DJ, De Maria A, Kisslo J, Weymon A. Recommendations regarding quantitation in M-mode echocardiography: results of a survey of echocardiographic measurements. Circulation. 1978;58:1072-1083.
Troy BL, Pombo J, Rackley CE. Measurement of left ventricular wall thickness and mass by echocardiography. Circulation. 1972;45:602-611.
Wallenstein S, Zucker CL, Fleiss JL. Some statistical methods useful in circulation research. Circ Res. 1980;47:1-9.
Greenhouse SW, Geisser S. On methods in the analysis of profile data. Psychometrika. 1954;24:95-102.
Nishikimi T, Kitamura K, Saito Y, Shimada K, Ishimitsu T, Takamiya M, Kangawa K, Matsuo H, Eto T, Omae T, Matsuoka H. Clinical studies on the sites of production and clearance of circulating adrenomedullin in human subjects. Hypertension. 1994;24:600-604.
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.
Hollenberg NK, Adams DF, Solomon HS, Rashid A, Abrams HL, Merrill JP. Senescence and the renal vasculature in normal man. Circ Res. 1974;34:309-316.
Messerli FH, Frohlich ED, Suarez DH, Reisin E, Dreslinski GR, Dunn FG, Cole FE. Borderline hypertension: relationship between age, hemodynamics and circulating catecholamines. Circulation. 1981;64:760-764.
Kohno M, Kano H, Horio T, Yokokawa K, Yasunari K, Takeda T. Inhibition of endothelin production by adrenomedullin in the vascular smooth muscle cells. Hypertension. 1995;25:1185-1190.