Author + information
- Received August 29, 1999
- Revision received January 26, 2000
- Accepted March 30, 2000
- Published online August 1, 2000.
- Stefan Dhein, MD∗,* (, )
- Peter Röhnert∗,
- Silke Markau, MD†,
- Emanuel Kotchi-Kotchi, PhD∗,
- Karin Becker, PhD∗,
- Ulrike Poller, MD∗,
- Bernd Osten, MD† and
- Otto-Erich Brodde, PhD∗
- ↵*Reprint requests and correspondence: Dr. Stefan Dhein, Institute of Pharmacology, University of Halle-Wittenberg, Magdeburger Str. 4, D-06097 Halle (Saale), Germany
The purpose of this study was to elucidate whether cardiac beta-adrenergic effects may be blunted in patients on maintenance hemodialysis (HD) and may help to explain autonomic dysfunction.
Patients on HD often suffer from autonomic dysfunction.
We investigated the cardiovascular response of five HD patients (age: 46.1 ± 7.9 years) and six healthy volunteers (age: 48.2 ± 7.5 years) to isoprenaline, pirenzepine and phenylephrine. For analysis of underlying mechanisms of beta-adrenoceptor hyporesponsiveness, six-week-old male Wistar rats were rendered uremic by 5/6-nephrectomy (n = 9; SNX) and were killed for removal of the heart after six to seven weeks. Sham-operated rats (n = 15) served as controls.
In the patient study, isoprenaline (3.5, 7, 17, 35 ng/kg/min, i.v.) led to an increase in heart rate, and shortening of the heart rate corrected duration of the electromechanical systole (QS2c), both of which were significantly reduced in HD patients. Baroreflex sensitivity was significantly reduced in HD patients. The response to low parasympathomimetic doses of pirenzepine was unchanged. In the rat study, left ventricular strips were placed in an organ bath, electrically driven and exposed to isoprenaline (10−11 to 10−6 mol/liter). While pD2 values were unchanged, maximum effect at the highest concentration was significantly reduced in SNX rats. The response to carbachol was not altered, nor was the M2-cholinoceptor density. There was no difference in beta-adrenoceptor density, or in immunodetectable amount of Gs and Gi protein. Activation of adenylyl cyclase evoked by isoprenaline was significantly reduced in left ventricular membranes of SNX rats, whereas effects of 10 μmol/liter GTP, 10 mmol/liter NaF, 10 μmol/liter forskolin and 10 mmol/liter Mn2+ were not altered.
Cardiac beta-adrenergic responses are blunted in chronic uremia due to reduced isoprenaline-dependent activation of adenylyl cyclase. This might be caused by an “uncoupling” of the receptor or by an inhibition of the receptor by uremic toxins.
A common problem in patients on maintenance hemodialysis (HD) is the development of an autonomic dysfunction (1–3) with impaired baroreflex control (4,5). Thus, acute hypotension occurs in up to 30% of the patients (6–8). Hyporesponsiveness to adrenergic stimuli, characterized by reduced increase in heart rate after isoprenaline administration, has been shown in uremic, predialysis patients (9) and in patients on HD (10,11), as well as diminished baroreceptor sensitivity (4,10). Moreover, Campese and colleagues (9) could demonstrate attenuated cardiovascular responses to norepinephrine in patients with chronic renal failure (predialysis state) and with elevated plasma norepinephrine levels (12,13). While previous investigations were focused on the role of alpha-adrenoceptors (10,14), only little is known on the role of beta-adrenoceptors and beta-adrenoceptor signaling in HD patients.
Regarding end-organ responsiveness to catecholamines, a reduced response in heart rate to isoproterenol in HD patients was found (15,16). In addition, reduced (-)[125I]-iodocyanopindolol [(-)[125I]ICYP] binding to rat lung beta-adrenoceptors if incubated with uremic plasma has been reported (17) due to interference of endogenous uremic substances on the beta-adrenergic receptors (18). In a 48-h rat model of acute uremia, the effect of isoprenaline on heart rate and adenylyl cyclase activity was reduced (19). However, it remains unclear whether this may also play a role in chronic uremia. In addition, the role of beta-adrenoceptor density and subtype distribution is unclear, because Mann and coworkers (19) found no change in beta-adrenoceptor density but reduced beta-adrenoceptor responsiveness, while others found enhanced beta2-adrenoceptor densities on mononuclear cells in patients (20). Furthermore, Gs and Gi proteins, as the link between beta-adrenoceptor and adenylyl cyclase, have not been investigated in chronic uremia.
Thus, the purpose of the present study was to find out whether cardiac beta-adrenoceptor responses were altered in patients on HD. We found a reduced heart rate and reduced positive inotropic response to isoprenaline and studied the possible underlying mechanism in a rat model assessing cardiac beta-adrenoceptor density, beta1/beta2-adrenoceptor-ratio, Gs and Gi proteins and adenylyl cyclase activity.
Six healthy volunteers (three men, three women, age: 48.2 ± 7.5 [22–67.5] years; at least six weeks drug-free) and five patients (two men, three women, age: 46.1 ± 7.9 [24–65] years; medication: see Appendix) with chronic renal failure on HD (dialysis frequency: 3–4 h/week over a period of at least 2 years, mean: 4.4 ± 1.4 years; uremia was caused by glomerulonephritis [two], pyelonephritis [one] polycystic kidney disease [one] and sclerotic kidney [one]) were studied. The study protocol was approved by the Ethical Committee of the University of Halle-Wittenberg, and written informed consent was obtained from all participants. Only patients that did not exhibit signs of acute infection, asthma, chronic pulmonary disease or clinical signs of acute decompensation were included in the study. The patients on HD had an ejection fraction of 62.2 ± 4.1%, a left ventricular end-diastolic diameter of 49.4 ± 5.5 mm and left ventricular end-systolic diameter of 35 ± 5.5 mm; plasma norepinephrine was 447 ± 176 pg/ml and epinephrine was 30 ± 4.3 pg/ml (“normal” values for this test : norepinephrine: 150–275, epinephrine: 25–35 pg/ml).
After 30 min of rest in supine position in an air-conditioned (21°C), quiet room, baseline hemodynamic parameters were assessed. Thereafter, either isoprenaline (3.5, 7, 17, 35 ng/kg/min) (n = 4 patients, n = 6 volunteers) or (on another examination day) phenylephrine (0.1, 0.2, 0.5, 1.0 μg/kg/min) (n = 5 patients, n = 6 volunteers) was infused intravenously for 10 (isoprenaline) or 15 (phenylephrine) min per dose. In another protocol, the patients and volunteers received pirenzepine in low doses, which are known to exert parasympathomimetic effects (21–23). After 30 min of rest in supine position, pirenzepine (sequential doses of 0.16, 0.32, 0.64, 1.25 mg) was injected intravenously over 5 min, each dose step requiring 20 min. The interval between each experimental protocol was at least one week. Experiments were always carried out at the end of the long dialysis-free interval.
Cardiovascular effects of drug infusion were analyzed by assessing systolic and diastolic blood pressure, heart rate and systolic time intervals. During the study, an electrocardiogram was continuously monitored, from which heart rate was determined over 20 cardiac actions (20 RR intervals). Blood pressure was measured five times in the last 5 min of each dose step using a common sphygmomanometer (Diplomat Presameter). If the increase in systolic blood pressure exceeded 50 mm Hg or the decrease in diastolic blood pressure was more than 30 mm Hg, the infusion of the drug was terminated for safety reasons. Systolic time intervals were measured in the last minute of each dose step by simultaneous recording of a electrocardiogram, a phonocardiogram and the carotid pulse tracing (Bioset 8000 multichannel recorder; Hörmann Medizintechnik, Zwönitz, Germany) at high paper speed (100 mm/s). Heart rate corrected duration of the electromechanical systole (QS2c) was defined as the time interval between Q-spike of the peripheral electrocardiogram and the first high-frequency component of the phonocardiogram and was corrected for heart rate revealing QS2c as previously described (24).
Baroreceptor sensitivity was assessed according to the method first described by Smyth et al. (25) by infusion of cumulative doses of phenylephrine and monitoring heart rate and blood pressure. Calculating the linear regression for the changes in heart rate over the changes in systolic blood pressure reveals a linear function of Y = A + B × X, the slope of which is an indicator of the baroreceptor-reflex sensitivity (26,27). This phenylephrine method has been recently confirmed to be appropriate in clinical tests (28).
All animal experiments were performed according to the German laws for animal welfare and were approved by the local committee for animal studies. Male Wistar rats (six weeks old) were submitted to 5/6 nephrectomy (SNX) that is, subtotal, as described by Amann et al. (29). Briefly, the right kidney was removed under light anesthesia (ketamine), and one week later, the left kidney was resected (resection of the lower and upper kidney pole, leaving an intact kidney segment in between) (SNX, n = 40). Concomitantly, a group of rats was sham operated (SOP, n = 42). After the first operation, the SOP rats received the same amount of food as the operated SNX rats had consumed the previous day (diet consisting of 22.5% protein, 0.2% Na+ and 1% K+; Altromin, Lage, Germany). The experiment was terminated six to seven weeks after the second operation by sacrificing the rats by cervical dislocation.
After sacrificing the rats, the rats’ hearts were removed, placed into oxygenated Krebs-Henseleit solution and trabecular strips of 1 mm to 2 mm width, 1 mm to 1.5 mm thickness and 6 mm to 8 mm length were prepared from the left ventricle. These strips were placed in 10-ml organ baths containing Tyrode solution of the following composition (NaCl 136.9, KCl 5.4, CaCl2 2.5, MgCl2 1.05, NaH2PO4 0.42, NaHCO3 25.0, glucose 9.7 mmol/liter, equilibrated with 95% O2 and 5% CO2, 37°C) and electrically driven (1 Hz) with rectangular pulses (5 ms) 50% above threshold (1–4 V, mean: 2.5 V) (Stimulator II; Hugo Sachs Elektronik, Hugstetten, Germany). After prestretching the preparation (10 mN), the developed force was recorded via a strain gauge on a Hellige recorder (Hellige, Freiburg, Germany). After 60 min of equilibration, cumulative concentration response curves for isoprenaline (10−10 to 10−6 mol/liter) and calcium (2.5 to 13 mmol/liter) were carried out as detailed elsewhere (30). To evaluate a possible functional role for beta2-adrenoceptors in these preparations, we assessed the effects of isoprenaline (10−10 to 10−3 mol/liter) in the absence or presence of either 3 × 10−7 mol/CGP 20712A or 3 × 10−7 mol/CGP 20712A plus 5 × 10−8 mol/liter ICI 118.551 in additional experiments (n = 5–7).
In additional experiments (n = 3 for each group), the reactivity to a parasympathomimetic stimulus was tested by application of cumulative concentrations of carbachol (10−9 to 10−5 mol/liter) after prestimulation of the strips with 3 μmol/liter forskolin.
Radioligand binding study
Beta-adrenoceptors were assessed by (-)[125I]-ICYP-binding assay detailed elsewhere (23). Briefly, tissues were homogenized in 10 volumes ice-cold 1 mmol/liter KHCO3 with an Ultra Turrax (Janke and Kunkel, Staufen, Germany), diluted to 20 ml with 1 mmol/liter KHCO3, centrifuged at 700 g for 15 min, passed through four layers of cheesecloth and centrifuged again at 50,000 g for 20 min. Pellets were washed once by resuspension and recentrifugation and finally resuspended in incubation buffer (Tris-HCl 10, NaCl 154, ascorbic acid 0.55 mmol/liter, pH 7.4, 25°C) at a protein concentration of 0.1–0.2 mg/ml. Protein content was determined by the method of Bradford (31) using bovine immunoglobulin G as a standard.
The density of beta-adrenoceptors in cardiac membranes was determined by (-)[125I]ICYP binding at six concentrations ranging from 5 to 200 pmol/liter. Nonspecific binding of ICYP was defined as binding to membranes that could not be displaced by a high concentration of the nonselective beta-adrenoceptor antagonist (±) CGP 12177 (1 μmol/liter). Specific binding was defined as total binding minus nonspecific binding and normally was about 70 % to 80% at 50 pmol/liter ICYP.
To determine the relative amounts of beta1- and beta2-adrenoceptors, membranes were incubated with ICYP (100 pmol/liter) in the absence or presence of six concentrations (from 10−9 to 10−4 mol/liter) of the selective beta1-adrenoceptor antagonist CGP 20712A, and the specific binding assessed as described above. CGP 20712A competition curves were analyzed using the iterative curve fitting program InPlot (GraphPad Software, San Diego, CA). Statistical analysis was performed by using the F-ratio test to measure the goodness-of-fit of the competition curves for either one or two binding sites.
M2 muscarinic receptors were determined by [3H]N-methyl-scopolamine ([3H]NMS) binding as recently described (32). Briefly, tissue was homogenized as described above. After filtering the supernatant of the first centrifugation step through cheesecloth, it was centrifuged at 21,000 g for 45 min; the pellets were resuspended in the incubation buffer (Na2HPO4 10, NaH2PO4 10, pH 7.4) to yield a protein concentration of 0.6 mg/ml to 0.7 mg/ml. The M2 muscarinic receptor density was assessed at six different concentrations of [3H]NMS ranging from 0.1 to 10 nmol/liter as detailed elsewhere (32). Nonspecific binding of [3H]NMS was determined as radioactivity bound to membranes that was not displaced by atropine (1 μmol/liter). Specific binding was defined as total minus nonspecific binding. Statistical analysis was performed as described above.
Immunodetectable Gs and Gi subunits were measured by quantitative immunoblotting as described previously (23). Briefly, aliquots of the membrane preparations (approximately 100 μg protein/sample) were separated on sodium dodecyl sulphate polyacrylamide gels (10% acrylamide in the running gel) and thereafter blotted overnight to nitrocellulose membranes (Hybond, ECL; Amersham) at a constant voltage of 55 V. The blots were washed and incubated overnight at 4°C with a 1:500 dilution of the indicated antisera. The blots were washed again twice and then incubated for 1 h at room temperature with [125I]protein A solutions (8.5 μCi/μg, 129 μCi/ml). After another four washes, the blots were exposed on imaging plates (Fuji Photo Film Co. Ltd, Japan) for 2 h to 24 h depending on the actual radioactivity. Detection and quantification of the G-proteins were carried out by using the Bio-Imaging Analyzer BAS 2000 (Fuji Photo Film Co. Ltd.) equipped with the data analyzing program Tina 2.10 fram raytest (Isotopenmessgeräte, Straubenhardt, Germany). Photostimulated luminescence (PSL) was determined, background luminescence was subtracted and the results expressed as densitometric units (d.u.). Using the autoradiograms, the molecular weights of the specific bands were identified. Protein dependency was established for each antibody. To allow the detection of possible increases or decreases, protein amounts corresponding to the middle of the linear part of the protein dependency curves were used.
Adenylyl cyclase activity
The procedure for determination of adenylyl cyclase activity has been previously described in detail (23). In brief, membranes (30 μg to 40 μg protein) were incubated for 10 min at 30°C in a final volume of 100 μl containing 40 mmol/liter Hepes buffer (pH 7.4), 5 mmol/liter MgCl2, 1 mmol/liter EDTA, 10 μmol/l GTP, 500 μmol/liter ATP, about 100,000 cpm [α-32P]ATP, 100 μmol/liter cAMP and an ATP regenerating system (5 mmol/liter phosphocreatine and 50 U/ml creatine phosphokinase) in the presence or absence of isoprenaline (10−8 to 10−5 mol/liter). In additional experiments, the effect of 10 μmol/liter forskolin, 10 mmol/liter NaF, 10 μmol/liter GTP or 10 mmol/liter Mn2+ was assessed. For GTP, NaF and forskolin activation membranes were incubated in the buffer as described above but without GTP; for Mn2+ activation, membranes were incubated in this buffer without GTP and Mg2+. In all experiments, the reaction was stopped by the addition of 100 μl buffer containing 50 mmol/liter Tris, 40 mmol/liter ATP, 1.4 mmol/liter cAMP, 2% SDS and [3H]cAMP (about 10,000 cpm) at pH 7.5; 800 μl H2O was then added. The mixture was poured into a Dowex AG 50W-X4 anion-exchange column (mesh size: 200–400; hydrogen form) and ATP was eluted twice with 1 ml water. The columns were then placed over neutral alumina columns, and cAMP was eluted from the Dowex columns with 4 ml water. The alumina columns were placed over scintillation vials, and the cAMP was eluted from the alumina columns with 5 ml 0.1 mol/liter imidazole (pH 7.3). Fifteen milliliters Lumasafe plus scintillator was added to the eluate and counted at 42% efficiency. The determined amount of [3H]cAMP in each vial was used to calculate the recovery of cAMP for each column, and the amount of [32P]cAMP collected from each column was corrected for the recovery rate, which usually was in the range of 70% to 80%.
All values are presented as means ± SEM of n experiments. Experimental data (rats) were fitted and analyzed by computer-supported iterative nonlinear regression analysis using the InPlot program (GraphPAD Software). Statistical significance of differences was analyzed by unpaired two-tailed Student’s t test or, if appropriate, by repeated measures analysis of variance followed by t test using Bonferroni corrections for multiple comparisons. A p < 0.05 was considered to indicate a significant difference. The statistical evaluation was performed using the InStat program (GraphPAD Software).
Chemicals were obtained from the following sources: isoprenaline (patients) (Aleudrina; Boehringer Ingelheim, Ingelheim, Germany), (-)isoprenaline bitartrate, atropine sulfate for the animal experiments and carbachol (Sigma, Deisenhofen, Germany), phenylephrine (Neo-Synephrine; Sanofi-Winthrop, New York, New York), pirenzepine (Gastrozepine; Thomae, Biberach, Germany), CGP 20712A (kindly donated by Ciba Geigy, Basel, Switzerland), ICI 118.551 (kindly donated by ICI Pharma, Planckstadt, Germany); [3H]-N-methylscopolamine (specific activity 85 Ci/mmol), (-)[125I]ICYP (specific activity: 2,200 Ci/mmol), [125I]-protein A (8.5 μCi/μg), [α-32P]ATP (specific activity 30 Ci/mmol) and [3H]cAMP (specific activity 44.5 Ci/mmol) (New England Nuclear, Dreieich, Germany). All other chemicals were of the purest commercially available grade. The antibodies AS/7 and RM/1, which specifically recognize GIα1/2 and Gsα, respectively (33), were obtained from New England Nuclear.
Systolic, mean and diastolic blood pressure, as well as QS2c at baseline, that is, assessed after 30 min of rest in supine position, were not significantly different between both groups, as can be seen in Table 1. Only the heart rate was elevated in patients on HD. Isoprenaline infusion caused a dose-dependent increase in heart rate (Fig. 1 ) and shortening QS2c (Fig. 2). Both responses were significantly attenuated in patients on HD (maximum change at the highest dose used: heart rate, 43.8 ± 3.4 vs. 33.8 ± 4.3 beats/min, p < 0.05; QS2c, −99.7 ± 7.6 vs. −73.1 ± 8.4 ms, p < 0.05).
Baroreflex sensitivity was assessed by monitoring changes in heart rate and in systolic blood pressure while infusing cumulative doses of phenylephrine, which resulted in a dose-dependent increase in systolic blood pressure and decrease in heart rate in both groups (Table 2).
Thereafter, changes in heart rate were plotted against changes in systolic blood pressure, and a linear regression of the form Y = A + B × X was fitted. We found no change of the Y-axis intercept A (healthy, −2.5 ± 0.14; patients on HD, −2.1 ± 0.58), while the slope B of the function was significantly lower in patients on HD (healthy, −0.41 ± 0.01; patients on HD, −0.32 ± 0.04 beats/min/mm Hg, p < 0.05; R was >0.9 for both fits), indicating a reduced baroreflex sensitivity in these patients (Fig. 3).
Injection of low doses of pirenzepine (0.16 mg to 1.25 mg) resulted in a significant (p < 0.05) decrease in heart rate but without any difference between both groups, as shown in Figure 4. Pirenzepine led to slight but not significant prolongation of QS2c from 514.3 ± 6.22 ms (healthy) to 522.2 ± 6.17 ms (healthy) and from 509.9 ± 8.73 ms to 511.9 ± 10.6 ms (patients on HD).
First, the rat model of chronic uremia was characterized. In rats rendered uremic by 5/6 nephrectomy, we found elevated plasma creatine (SOP, 50.2 ± 1.4, n = 34; SNX, 77.9 ± 1.6 μmol/liter, n = 42, p < 0.05) and urea values (SOP, 7.2 ± 0.3, n = 34; SNX, 11.5 ± 0.5 mmol/liter, n = 42, p < 0.05) six to seven weeks after operation. Both groups did not differ with regard to their body weight (SOP, 314 ± 6, n = 42; SNX, 309 ± 4, n = 40; NS). In contrast, in uremic rats, cardiac hypertrophy became obvious (cardiac index: SOP, 3.24 ± 0.05, n = 42; SNX, 3.78 ± 0.06 mg/g, n = 40, p < 0.05), which was due to hypertrophy of the left ventricle. Thus, left ventricular cardiac index was significantly enhanced (SOP, 1.89 ± 0.07, n = 15; SNX, 2.6 ± 0.13 mg/g, n = 23, p < 0.05), while right ventricular cardiac index remained unchanged (SOP, 0.52 ± 0.03, n = 15; SNX, 0.57 ± 0.02 mg/g, n = 23, NS). Systolic blood pressure, measured at the tail vein, was slightly elevated in uremic rats (SOP, 122.2 ± 1.6, n = 20; SNX, 130.8 ± 4.4 mm Hg, n = 28, NS, p = 0.12). Plasma epinephrine levels did not differ between both groups (SOP, 20 ± 18, n = 11; SNX, 36 ± 16 pg/ml, n = 11, NS), whereas plasma norepinephrine was significantly increased in uremic rats (SOP, 285 ± 28, n = 11; SNX, 400 ± 68 pg/ml, n = 11, p < 0.05).
Exposure of the left ventricular strips to cumulative concentrations of isoprenaline (10−11 to 10−6 mol/liter) led to a concentration-dependent increase in contractile force in both groups. However, while pD2 values were not altered (SOP, 7.9 ± 0.02, n = 11; SNX, 8.1 ± 0.02, n = 11), the maximum effect at the highest concentration was significantly reduced in uremic rats (SOP, 5.16 ± 0.11, n = 11; SNX, 3.22 ± 0.07 mN, n = 11, p < 0.05) (Fig. 5A). For control, the inotropic response to external CaCl2 was investigated. It became obvious that the calcium-induced increase in force was not significantly different between both groups: maximum increase was observed after application of 8 mmol/liter CaCl2, as can be seen in Figure 5B.
In additional experiments, we investigated the functional role of cardiac beta1- and beta2-adrenoceptors in SOP (n = 5) and SNX (n = 7) rats. The addition of 3 × 10−7 mol/liter CGP20712A led to a similar rightward shift of the isoprenaline concentration response curve in both groups. The addition of 5 × 10−8 ml/liter ICI 118.551 had no effect on this rightward shift in both groups (pD2 values: SNX: isoprenaline, 8.05 ± 0.05; CGP, 4.84 ± 0.06; CGP 20712A + ICI 118.551: 4.85 ± 0.05, n = 7; SOP: isoprenaline, 7.84 ± 0.05, CGP, 5.30 ± 0.11; CGP 20712A + ICI 118.551, 4.92 ± 0.1, n = 5). The KB for CGP 20712A was 0.8 ± 0.3 or 0.2 ± 0.1 nmol/liter in SOP and SNX rats, respectively (not significantly different).
To investigate the cardiac M2-receptor function, we applied carbachol (10−9 to 10−5 mol/liter) after stimulation with 3 μmol/liter forskolin in three additional experiments. Carbachol led to the expected decrease in contractile force without differences between both groups (pD2 values: SOP: 6.48 ± 0.28, n = 9; SNX, 6.16 ± 0.16, n = 9) (Fig. 5C). Regarding the density of M2 cholinoceptors (assessed by [3H]NMS binding), we found no significant difference between SNX and SOP rats (SNX: 65.9 ± 7 fmol/mg protein, n = 11, vs. SOP: 62.6 ± 8.9 fmol/mg protein, n = 12, NS). The KD values for [3H]NMS were not different between both groups (SNX: 0.7 ± 0.15 nmol/liter, n = 11, vs. SOP: 0.64 ± 0.1 nmol/liter, n = 12, NS).
To investigate the mechanism underlying the reduced contractile response to isoprenaline (mentioned above) but not to calcium, we assessed the components of the beta-adrenoceptor-G-proteins-adenylyl cyclase system in left ventricular tissue.
Left ventricular beta-adrenoceptor density as assessed by (-)[125I]ICYP binding was not different between both groups (SOP: 21 ± 2.9, n = 11; SNX: 23.1 ± 3.9 fmol/mg, n = 11). In addition, beta1:beta2-adrenoceptor ratio, as determined from competition curves of the highly selective beta1-adrenoceptor antagonist CGP 20712A (10−10 to 10−1 mol/liter) with the ICYP binding (100 pmol/liter), was not significantly altered in uremic rats (beta1-adrenoceptors: SOP, 65.7 ± 1.7%, n = 11; SNX: 68.6 ± 1.4%, n = 10, NS).
Next, we investigated the immunodetectable amount of the G proteins Gs and Gi in the left ventricles. The Gs-specific antiserum RM/1 detected two bands with molecular weights of 45 and 53 kDa corresponding to Gsα-short and Gsα-long. No significant differences could be detected between SOP and SNX rats (53 kDa Gs: SOP, 1,093 ± 102, n = 6; SNX, 855 ± 107, n = 6, NS; 45 kDa Gs: SOP, 1,517 ± 135, n = 6; SNX, 1,279 ± 131 densitometric units, n = 6, NS).
The antiserum AS/7, which recognizes the alpha subunits of Gi1, Gi2 and transducin, detected one band with an apparent molecular mass of 43 kDa; no significant difference in this band between SOP and SNX rats could be detected (Gi: SOP, 206 ± 28, n = 7; SNX, 181 ± 24 densitometric units, n = 5, NS).
Finally, we assessed adenylyl cyclase activity in left ventricular membranes from SNX and SOP rats. Basal activities did not differ between both groups (Table 3). However, the isoprenaline-induced activation of adenylyl cyclase was at each concentration significantly larger in SOP than it was in SNX rats (Fig. 6). Thus, at maximum stimulation (with 10−5 mol/liter isoprenaline), cAMP formation was significantly diminished in uremic rats (SOP: 42.9 ± 5, n = 9; SNX: 27.5 ± 4.7, n = 7, p < 0.05). pEC50 values of isoprenaline, however, were not different between SOP (8.18 ± 0.25, n = 9) and SNX (7.95 ± 0.24, n = 7) rats. On the other hand, activation of adenylyl cyclase by 10 mmol/liter NaF, 10 μM forskolin, 10 μmol/liter GTP or 10 mmol/liter Mn2+ did not differ between SNX and SOP rats (Table 3).
Human study: cardiac beta-adrenoceptor responses and baroreflex sensitivity are reduced in chronic HD patients
A common problem in patients on chronic hemodialysis is the development of an autonomic dysfunction (1), including decreased responsiveness of vascular α-adrenoceptors (34), increased plasma norepinephrine levels and a reduced baroreceptor sensitivity (4,5,10,11), as was also seen in our study. Increased plasma norepinephrine levels and decreased baroreflex sensitivity (35–38) are also found in patients with chronic heart failure, and it is well known that in these patients, cardiac beta-adrenoceptor function is decreased (for review see reference 39). However, relatively little was known on cardiac beta-adrenoceptors in patients on HD. Thus, we first investigated the effects of isoprenaline on heart rate and QS2c (as a measure of inotropism ) in patients on HD and controls. The results clearly demonstrate that both the increase in heart rate and the shortening of QS2c were blunted in patients on HD, which cannot be due to age-dependent changes, because both groups were age matched. Thus, beta-adrenoceptor responsiveness appears to be decreased.
While isoprenaline effects on heart rate are mediated through beta1- and beta2-adrenoceptors (for review see references 39,41), its effects on contractility appear to be predominantly mediated via beta1-adrenoceptor stimulation (24). Thus, it can be concluded that in these patients beta1-adrenoceptors in the heart are desensitized. Unfortunately, we could not test whether beta2-adrenoceptor responses are also changed in these patients.
Because both the sympathetic and parasympathetic pathways control heart rate, it was also necessary to investigate the responsiveness to parasympathomimetic stimuli. Recently, we demonstrated that low doses of the M1-receptor antagonist pirenzepine decreased heart rate in humans possibly via inhibition of presynaptic muscarinic autoreceptors, thus enhancing release of endogenous acetylcholine (32). Because we did not observe a difference in the pirenzepine-induced decrease in heart rate between patients on HD and normal subjects, it can be concluded that muscarinic receptor function is probably not altered, a conclusion shared with other investigators (10), although Zoccali and coworkers (42) did find reduced parasympathetic activity (however, unrelated to duration of hemodialysis).
Regarding the underlying mechanisms, only a few studies on beta-adrenoreceptors, mainly on blood cells such as lymphocytes, have been performed in chronic HD patients. Thus, reduced beta2-adrenoceptor responsiveness (in lymphocytes) was observed (43,44) with unchanged beta2-adrenoceptor number (43), while increased beta2-adrenoceptor density and increased epinephrine levels were found by others (17), which was interpreted as an indication for endogenous uremic substances blocking the beta-adrenoceptors. Many substances, including polyols, sugars, 3-D-hydroxybuyrate, dimethylglycine, sulfoconjugated catecholamines and others (45,46), have been found to be increased in uremic plasma. Ferchland and coworkers (18) showed that such endogenous uremic substances can inhibit beta-adrenoceptor binding, which may contribute to the reduced responsiveness to isoprenaline in vivo.
Regarding the impaired baroreflex, it should be noted that the patients included in our study were normotensive and on hemodialysis for 4.4 years. The differentiation seems to be necessary because Armengol et al. (11) showed that normotensive HD patients (mean duration of hemodialysis, 3.2 years) exhibited less pronounced disturbance of autonomic tests, while in hypotensive subjects (mean duration of hemodialysis, 10.8 years), the impairment of the autonomic function was more distinct. However, the overall integrity of the baroreflex was impaired in both normotensive and hypotensive hemodialysis patients (10,11). The reduced responsiveness to isoprenaline observed in our study indicates a possible impairment of the efferent sympathetic branch.
Rat study: cardiac beta-adrenoceptor responses are reduced in chronic uremic rats
In order to study the mechanisms underlying the beta-adrenoceptor hyporesponsiveness, we decided to investigate the components of the beta-adrenoceptor system, i.e., beta-adrenoceptor, G-proteins and adenylyl cyclase in cardiac tissue in a rat model of chronic uremia, the subtotally nephrectomized rat. This rat model exhibited a similar decrease in isoprenaline response as observed in humans (Fig. 5A) and elevated plasma norepinephrine levels but unaltered plasma epinephrine levels as seen in HD patients and as described in the literature (19). Furthermore, similar to HD patients, an unchanged cardiac M2-receptor function was found in these uremic rats. Thus, this rat model seems to be suitable to serve as a model, at least partly, of chronic hemodialysis patients as investigated in this study.
The beta-adrenergic hyporesponsiveness seen in our study may be caused by either reduced beta-adrenoceptor density (as common in heart failure ), by blockade of the receptors by endogenous substances or by a defect in the signaling cascade. In our study, beta-adrenoceptor density was found to be unchanged, indicating the absence of beta-adrenoceptor downregulation, which is in accordance with the other findings (19). In addition, the cardiac beta1/beta2 distribution was not altered. Accordingly, the rightward shift of the isoprenaline concentration response curve by CGP 20712A was not altered in uremic rats, and in both groups, additional ICI 118.551 had no effect on the curve. In sum, this means that the positive inotropic effect of isoprenaline is mediated via beta1-adrenoceptors in rats, which agrees with the literature (30,39), and that in uremic rats the beta1-adrenoceptor responsiveness is reduced.
Furthermore, the immunodetectable amount of Gsα and GIα was not different between SNX and SOP rats, although Gsα (both the short and the long form) showed a tendency to decrease. Moreover, activation of adenylyl cyclase by GTP (acting at Gs and Gi[23,47]) and NaF (acting under these experimental conditions at Gs) did not differ between both groups of rats, indicating unchanged functional activities of Gs and Gi.
Activation of adenylyl cyclase by forskolin (acting predominantly at the catalytic unit of the enzyme, but involving partly Gs[23,47,49]) and Mn2+ (acting at the catalytic unit ) was not different between SNX and SOP rats; this is in agreement with the idea of an unchanged adenylyl cyclase activity in chronic uremia. On the other hand, isoprenaline-induced activation of adenylyl cylase was at each concentration higher in SOP rats than in SNX rats. Thus, in chronic uremic rat hearts, beta-adrenoceptor number, amount and activity of G proteins and activity of adenylyl cyclase are not changed, while beta-adrenoceptor responsiveness (activation of adenylyl cyclase and positive inotropic effects) is reduced.
At present, the mechanism underlying this beta-adrenoceptor hyporesponsiveness (in the face of unchanged number and unchanged activity of G proteins and adenylyl cyclase) is not clear. It could be due to uncoupling of the receptor from the effector system. Taking into account the high amount of spare beta-adrenoceptors in the rat heart (30), “uncoupling” of the beta-adrenoceptors should result in a rightward shift of the isoprenaline concentration response curve but not in a decrease in maximum response with unchanged pD2 and pEC50 values. Alternatively, uremic toxins might inhibit beta-adrenoceptor binding (17,18). This could result in a decrease in maximum response without change in pD2 and pEC50 values as observed in our study. In sum, beta-adrenoceptor responsiveness (predominantly beta1-adrenoceptors, see preceding text) in the chronic uremic rat heart is decreased; this decrease is due to the uncoupling of the beta-adrenoceptor or to an inhibition of beta-adrenoceptors by uremic toxins.
In summary, in patients on HD as well as in chronic uremic rats, cardiac beta-adrenoceptor responsiveness is decreased; cardiac muscarinic receptor function, however, is unaltered. The mechanism underlying this reduced beta-adrenoceptor responsiveness remains unclear; it might be due to an “uncoupling” of the receptor or to an inhibition of the receptor by uremic toxins. In chronic hemodialysis patients, this blunted beta-adrenoceptor responsiveness may functionally contribute to the impaired baroreflex response.
The skillful technical assistance of Anja Struppert and Pia Matthes is gratefully acknowledged. We are thankful to Professor Dr. E. Ritz, Department of Nephrology, University of Heidelberg, for introduction into the technique of subtotal nephrectomy.
Medication of the patients
Patient number and medication (long-term):
1. Digitoxin 0.07 mg/day, ranitidine 300 mg/day, calcium-carbonate 3 g/day, alfacalcidol 1 μg/week, resonium A 15 g/day, etilefrin 30 mg/day, meloxicam 7.5 mg/day, Fe3+ 40 mg/month, vitamin B12 9 mg/week, folic acid 60 mg/week, L-carnitine 3 g/week.
2. Resonium A 15 g/day, magnesium 120 mg/day, losartan 100 mg/day, clonidin 0.475 mg/day, calcium-diactetate 2.85 g/day, medazepam 10 mg/day, amlodipine 10 mg/day, isosorbitdinitrate 80 mg/day, urapidil 180 mg/day, Fe3+ 40 mg/week, alfacalcidol 1 μg/week, epoetin beta 21,000 IU/week.
3. Ranitidin 300 mg/day, furosemide 120 mg/day, resonium A 30 g/day, folic acid 320 μg/day, biotin 60 μg/day, ascorbic acid 200 mg/day, vitamin B1 16 mg/day, vitamin B2 16 mg/day, vitamin B6 20 mg/day, nicotinamide 100 mg/day, pantothenic acid 20 mg/day, NaHCO3 1.5 g/day, Ca2+ 1.5 g/day, Fe3+ 40 mg/week, epoetin alpha 3,000 IU/week, L-carnitine 3 g/week.
4. Furosemide 80 mg/day, allopurinol 200 mg/day, calcium-diacetate 2.85 g/day, aluminiumhydroxide 1.2 g/day, indometacine 25 mg/day, etilefrine 10 mg/day, Fe3+ 40 mg/week, alfacalcidol 1 μg/week, epoetin beta 15,000 IU/week.
5. Furosemide (retarded) 250 mg/day, alfacalcidol 2 μg/week, omeprazol 20 mg/day, simvastatin 5 mg/day, calcium-diacetate 4.3 g/day, resonium A 15 g/day, aluminium-hydroxide 1.2 g/day, dalteparine-sodium 15 mg/day, Fe3+ 40 mg/month, epoetin alpha 6,000 IU/week, thiamine 1.5 mg/day, pyridoxine 10 mg/day, riboflavine 1.7 mg/day, vitamin B12 6 μg/day, ascorbic acid 60 mg/day, nicotinamide 20 mg/day, biotine 0.3 mg/day, pantothenic acid 10 mg/day, folic acid 0.8 mg/day.
☆ This work was supported by a grant (DFG OS 131/3-2 to B.O. and O.-E.B.) from the Deutsche Forschungsgemeinschaft (Bonn, Germany).
- diastolic blood pressure
- maintenance hemodialysis
- heart rate
- mean arterial pressure
- heart rate corrected duration of the electromechanical systole
- photostimulated luminescence
- systolic blood pressure
- 5/6-nephrectomized rat
- sham-operated rat
- Received August 29, 1999.
- Revision received January 26, 2000.
- Accepted March 30, 2000.
- American College of Cardiology
- Rostand S.G,
- Brunzell J.D,
- Cannon R.O III.,
- Victor R.G
- Armengol N.E,
- Amenos A.C,
- Illa M.B,
- Bertran J.G,
- Ginesta J.C,
- Fillat F.R
- Brecht H.M,
- Ernst W,
- Koch K.M
- McGrath B.P,
- Ledingham J.G.G,
- Benedict C.R
- Ulman A,
- Drüeke T,
- Zingraff J,
- Crosnier J
- Mann J.F.E,
- Jakobs K.H,
- Riedel J,
- Ritz E
- Poller U,
- Nedelka G,
- Radke J,
- Pönicke K,
- Brodde O.-E
- Schäfers R.F,
- Adler S,
- Daul A,
- et al.
- Smyth H.S,
- Sleight P,
- Pickering G.W
- Pitzalis M.V,
- Mastropasqua F,
- Passantino A,
- et al.
- Michel M.C,
- Farke W,
- Erdbrügger W,
- Philipp T,
- Brodde O.-E
- Brodde O.-E,
- Daul A
- Ellenbogen K.A,
- Mohanty P.K,
- Szentpetery S,
- Thames M.D
- Ferguson D.W,
- Abboud F.M,
- Mark A.L
- Ferguson D.W,
- Berg W.J,
- Roach P.J,
- Oren R.M,
- Mark A.L
- McDevitt D.G
- Katada T,
- Northup J.K,
- Bokoch G.M,
- Ui M,
- Gilman A.G
- Yan S.Z,
- Huang Z.H,
- Andrews R.K,
- Tang W.J
- Strittmatter S,
- Neer E.J