Regulation, Chamber Localization, and Subtype Distribution of Angiotensin II Receptors in Human Hearts
Background To assess the chamber localization, subtype distribution, and regulation of human myocardial angiotensin II receptors in heart failure, we determined the binding of angiotensin II, Sar1Ile8–angiotensin II, and the subtype-specific antagonists Dup 753 (AT1-specific) and PD 123319 (AT2-specific) in atria from patients with normal (left ventricular ejection fraction >55%) or moderately impaired (left ventricular ejection fraction 30% to 55%) cardiac function and in atria and ventricles from explanted end-stage failing hearts. Sarcolemmal and combined fractions, the latter including internalized receptors, were studied. In addition, AT1 mRNA content was analyzed by polymerase chain reaction after reverse transcription.
Methods and Results The number of angiotensin II binding sites (Bmax) in sarcolemmal fractions was significantly reduced in explanted end-stage failing hearts in comparison with control subjects and moderate heart failure (Bmax 3.9±0.8 versus 11.2±1.7 and 9.6±0.8 fmol/mg protein, respectively). A comparable 65% reduction in receptor numbers was found in combined fractions from end-stage failing hearts, indicating that the loss of binding sites was not due to their internalization. The dissociation constants were comparable in sarcolemmal and combined fractions and in nonfailing and failing hearts, ranging from 0.5±0.2 to 1.2±0.5 nmol/L. In nonfailing hearts, 69±4% of binding sites were blocked by the subtype-2–specific inhibitor PD 123319 and were therefore classified as AT2; 33±5% were blocked by the subtype-1–specific inhibitor DUP 753 and thus classified as subtype 1. In explanted hearts, comparable ratios of 66±5% AT2 sites and 34±5% AT1 sites were found. AT1 cDNA amplification signals by polymerase chain reaction were reduced to about one third of the level in control subjects in end-stage failing hearts.
Conclusions Angiotensin II receptors in human myocardium are present in relatively low numbers, and AT2 is the predominant subtype. A significant loss of angiotensin II receptors occurs in end stage but not in moderate heart failure. The loss of receptors affects both subtypes to a comparable degree. The data suggest that the decrease in receptor density is due to a decrease in steady-state mRNA abundance.
Angiotensin-converting enzyme (ACE) inhibitors improve survival in patients with heart failure and successfully prevent remodeling—a complex of changes in myocardial gene expression, protein pattern, and function—of the left ventricle after myocardial infarction.1 2 3 Direct effects of angiotensin II on the myocardium may contribute to these advantages.
Two angiotensin II receptor subtypes, AT1 and AT2, are present in mammalian hearts.4 5 6 7 AT1, a G-protein–coupled receptor, mediates positive inotropic and chronotropic effects in animal models and human atrial myocardium.8 9 10 It mediates protein synthesis and hypertrophy in isolated myocytes and stimulates protein and DNA synthesis in rat fibroblasts.11 12 13 By contrast, AT2 does not interact with guanine nucleotide-binding proteins.14 It has been claimed recently that it mediates collagen synthesis in human cardiac fibroblasts.15
AT1 is downregulated by agonist exposure in vascular smooth muscle cells, in rat mesangial cells and in pressure-overload–induced rat myocardial hypertrophy.16 17 18 Receptor upregulation occurs in hypertrophic hearts of spontaneously hypertensive and renovascular hypertensive rats and in postinfarction models in rats.19 20 In human heart failure, myocardial tissue ACE activity is increased, indicating the possibility of elevated tissue angiotensin II levels and receptor downregulation.21 As cardiac hypertrophy frequently precedes clinical heart failure, upregulation of angiotensin receptors (ATR) also seems possible in human heart failure, and we investigated both possibilities. Since internalization may contribute to the upregulation or downregulation of ATR-numbers,16 we measured ATR densities in two different subcellular fractions, one of which included internalized receptors.22 The number of binding sites was determined in atrial myocardium obtained from end-stage failing hearts at cardiac transplantation and in organs with normal left ventricular ejection fraction (LVEF) obtained at cardiac surgery. Analysis of mRNA content by polymerase chain reaction (PCR) supports the binding data.
Since only preliminary reports on receptor subtypes in human hearts and few data on regional distribution are available,4 23 we also investigated the relative proportion of AT1 and AT2 subtypes in normal and failing human hearts with subtype-specific ligands and examined the regional distribution of ATR between atria and ventricles.
Receptor Quantitation Studies
Myocardial tissue from both atria, right atrial appendage, and both ventricles was obtained at orthotopic heart transplantation from 25 male patients with end-stage heart failure, 16 with dilated cardiomyopathy, and 9 with ischemic heart disease. Clinical and hemodynamic data and preoperative cardiac therapy of all patient groups are summarized in Table 1⇓. Sarcolemmal fractions were prepared from 14 of 25 end-stage failing hearts and combined sarcolemmal/light vesicle fractions from 11 of 25 (see below).
Fifteen patients undergoing aortocoronary bypass surgery (n=14) or mitral valve replacement (n=1) without clinical signs of congestive heart failure and with normal cardiac function (left ventricular ejection fraction >55%, normal right and left ventricular filling pressure) served as control subjects (Table 1⇑). The right atrial appendage was excised at cannulation of the heart through the right atrial appendage. In 10 patients a sarcolemmal fraction and in 5 a combined fraction was studied. Three patients in the group of 10 with a normal LVEF used for the preparation of sarcolemmal fractions were pretreated with the ACE inhibitor captopril (25 to 50 mg/d). This low-dose treatment caused no significant difference in the number of binding sites (Bmax) or in the dissociation constant (Kd) in patients with and without ACE inhibitor therapy (Bmax: captopril, 10.2±1.4; no ACE inhibitors, 12.9±2.4 fmol/mg protein, NS; Kd, 0.5±0.1 and 0.6±0.2 nmol/L, NS). Therefore, the three captopril-treated patients were subsequently included in the control group.
The group with moderate heart failure consisted of 13 patients with LVEF 30% to 55% who were undergoing coronary or valvular surgery (Table 1⇑). Tissue was sampled as in the control group. In 6 of these 13 patients a sarcolemmal fraction was studied, and in 7 of 13 a combined fraction was studied (Table 1⇑). In a first series of experiments we prepared sarcolemmal fractions from all samples obtained, and in a second series we prepared combined fractions (see below).
Subtype analysis with selective antagonists was conducted in a separate series of 13 patients (Table 2⇓). As receptor quantitation yielded no significant differences between patients with LVEF >55% and those with LVEF 30% to 55%, 7 patients with normal or moderately impaired LVEF (mean, 54±3%) undergoing aortocoronary bypass surgery or valve replacement were used as control subjects and were compared with 6 patients with end-stage heart failure undergoing orthotopic heart transplantation (Table 2⇓). Informed consent was obtained from all patients before tissue sampling. The study was approved by the Ethical Committee of the Free University of Berlin.
Tissue Sampling/Regions Investigated
Myocardium obtained at surgery or at heart transplantation was frozen immediately in the operating room on dry ice. To investigate the effect of possible delays between excision and freezing on Bmax and Kd, tissue samples obtained at surgery from three nonfailing hearts were divided in two; one half was frozen immediately and the other half was frozen after a delay of 30 minutes. Bmax (7.1±1.3 and 10.2±0.9 fmol/mg protein) and Kd (0.8±0.4 and 0.8±0.4 nmol/L) were not different. Comparably, samples from four explanted hearts were frozen immediately or kept at room temperature for 15, 30, or 60 minutes before freezing. No significant differences in Bmax and Kd were found (data not shown).
Samples from the right atrium were obtained in all patient groups, whereas left atrial and ventricular myocardium was only studied in explanted hearts. To exclude a difference in Bmax and Kd between right atrial anterolateral wall obtained at transplantation and right atrial appendage obtained at bypass or valvular surgery, samples from both regions were excised and compared in four explanted hearts. The respective values for Bmax (combined fractions, 23.0±8.4 and 23.9±6.2 fmol/mg protein) and the dissociation constants (0.5±0.2 and 0.6±0.2 nmol/L were comparable.
Preparation of Sarcolemmal and Combined Membrane Fractions
Tissue (200 to 300 mg) was homogenized in a 10-fold volume (wt/vol) of ice-cold 25 mmol/L Tris buffer/300 mmol/L sucrose, pH 7.4, with proteinase inhibitors at 15 000 rpm with an Ultraturrax (Janke and Kunkel). For the preparation of sarcolemmal fractions, the homogenate was centrifuged for 10 minutes at 500g, the supernatant was centrifuged for 30 minutes at 48 000g, and the pellet was resuspended in 0.6 mol/L KCl, 30 mmol/L histidine, pH 7.0, and again centrifuged for 20 minutes at 48 000g. The final pellet was resuspended in Tris buffer (50 mmol/L Tris, 10 mmol/L MgCl2, 0.2% bovine serum albumin, pH 7.4). Protein was determined in triplicate. Homogenates were shock-frozen and stored at −80°C.
For combined fractions including sarcolemmal and internalized receptors, both centrifugations were carried out at higher speed,22 24 that is, the first spin at 5000g and the second spin at 110 000g. All other steps were performed as for the sarcolemmal fraction.
Yield of Protein
Sarcolemmal fractions from the right atrium in explanted hearts (10.1±1.7 μg protein/mg wet wt), in patients with LVEF 30% to 55% (10.5±0.7), and patients with LVEF >55% (10.5±0.8) yielded comparable amounts of protein. Yields in right and left ventricles of explanted hearts (right ventricle, 17.3±3.1 and left ventricle, 12.0±1.8 μg protein/mg wet wt) did not differ.
The total yield of protein in combined light vesicle/sarcolemmal fractions was lower than in sarcolemmal fractions because the first centrifugation was performed at higher speed. Within this fraction, protein yields from explanted hearts (4.3±0.5 μg protein/mg wet wt), from hearts with LVEF 30% to 55% (4.4±0.4), and from hearts with LVEF >55% (4.4±1.3) were comparable.
Receptor Binding Studies
In most experiments, 125I–angiotensin II (2200 Ci/mmol NEN) was used as a radioligand. Assays were run for 60 minutes at 18°C in a total volume of 100 μL (final concentrations: 50 mmol/L Tris, 10 mmol/L MgCl2, 0.2% bovine serum albumin, pH 7.4, and proteinase inhibitors) with 100 (sarcolemmal fractions) or 30 (combined fractions) μg protein/40 μL, 30 μL radioligand, and 30 μL competitor as determined by the actual protocol. All samples were run in triplicate. Incubation was stopped by adding 4 mL cold buffer. Samples were filtered through Millipore 0.45-μm filters. Filter-bound radioactivity was counted in a Packard gamma counter (with an efficiency of 80%). Saturation of specific binding sites was reached with angiotensin II concentrations of 1.2 nmol/L (Fig 1⇓).
To reduce the amount of protein and radioactivity needed, cold saturation studies were run in most experiments. A set amount of radioligand, resulting in final concentrations of 0.08 or 0.3 nmol/L (see below) in 30 μL, was incubated with 100 or 30 μg protein/40 μL (sarcolemmal or combined fraction), and the total concentration of ligand was increased by adding progressively larger amounts of unlabeled ligand to obtain final concentrations of 0.3×10−12 to 3×10−6 mol/L. Nonspecific binding was determined in the presence of 10−6 mol/L unlabeled ligand. The ebda/ligand program (P.I. Munson, D. Rodbard, modified by G.A. McPherson, biosoft) was used for the analysis of curves by the least squares fitting method for the determination of the Bmax, for the calculation of the Kd after transformation of the data according to Scatchard, and for the calculation of Hill coefficients.25 26 Replacement curves of normal steepness, linear Scatchard plots, and Hill coefficients close to unity in all three patient groups (LVEF >55%, 1.01±0.12; LVEF 30% to 55%, 0.98±0.09; explanted hearts, 0.95±0.09) support the presence of only one angiotensin II binding site. Because the possibility of detecting a second binding site depends on the concentrations of radioligand,27 we used two different concentrations in sarcolemmal (0.08 nmol/L) and combined fractions (0.3 nmol/L).
The effect of Gpp(NH)p (a nonhydrolyzable analogue of GTP) on agonist binding was studied in three separate patients with normal LVEF. Replacement curves in the presence of Gpp(NH)p (10−4 mol/L) did not differ significantly from the curves without Gpp(NH)p (10−4 mol/L) (Fig 2⇓). No effect of increasing Gpp(NH)p concentrations on angiotensin II binding was detected (data not shown).
For subtype analysis, 30 μg protein (in 40 μL) was incubated with 0.3 nmol/L 125I-Sar1Ile8–angiotensin II (30 μL) and 30 μL binding buffer (50 mmol/L Tris, 10 mmol/L MgCl2, 0.2% bovine serum albumin, pH 7.4, proteinase inhibitors). The radioligand was replaced by increasing concentrations of either unlabeled Sar1Ile8–angiotensin II or the subtype-specific competitors DUP 753 (AT1-specific) or PD 123319 (AT2-specific) (range, 10−10 to 10−6 mol/L for all three compounds) in separate experiments.6 7 Percent inhibition of 125I-Sar1Ile8–angiotensin II binding was calculated for each competitor, and mean values were calculated for patients and control subjects. In separate experiments in 3 patients, the combination of DUP 753 and PD 123319 (both 10−6 mol/L) was used to replace 125I-Sar1Ile8– angiotensin II binding. In 4 different patients, the AT1 site was blocked with 10−6 mol/L DUP 753 and the 125I-Sar1Ile8–angiotensin II bound to the remaining sites was replaced by increasing concentrations of PD 123319. The labeled antagonist 125I- Sar1Ile8–angiotensin II was used in these experiments to avoid interference with high- and low-affinity AT1 binding sites, which may occur with agonist binding.
Myocardial AT1 mRNA Content
Atrial samples from 4 patients with LVEF >30% and five explanted end-stage failing hearts were randomly selected from patients used for subtype analysis (Table 2⇑). Total RNA was extracted by the acid guanidinium thiocyanate/phenol/chloroform method.28 29 Yields are shown in Tables 3⇓ and 4⇓. Five hundred nanograms of total RNA was transcribed into cDNA with Superscript (Life Technology). The resulting cDNA was amplified by polymerase chain reaction with primers for AT1 and for pyruvic dehydrogenase (PDH), which was used as a standard. In a separate study, the PDH mRNA content was analyzed in atrial myocardium from 12 control subjects and 12 end-stage failing hearts by the method described below, and no significant differences were found between the two groups (Table 3⇓). Primer sequences were as follows: AT1-1: 5′CCTTCGACGCACAATGCTTG-3′, AT1-2: 5′-AGCCCTATCGGAAGGGTTGA-3′, PDH-1: 5′-GGTATGGATGAGGAGCTGGA-3′. PDH-2: 5′-CTTCCACAGCCCTCGACTAA-3′, resulting in amplification products of 173 bp for AT1 and 103 bp for PDH at the cDNA level.30 As the PDH primers span an intron, a 185-bp fragment results at the DNA level, and this can be used to estimate genomic DNA contamination of the RNA preparations, which was comparably low in all patients. PCR amplification with the AT1 primers without the addition of reverse transcriptase but including the reverse transcription reaction conditions yielded negligible amounts of amplification products that were comparable in both groups (high-performance liquid chromatography [HPLC] area: control subjects, 858±164, explanted hearts, 804±170).
PCR was run for 30 cycles for PDH and AT1 as described previously (denaturation: 95°C, 45 seconds, annealing 60°C for 60 seconds, extension 72°C for 45 seconds) in a Perkin Elmer 9600.31 Both reactions were still in the exponential phase at this number of cycles (data not shown). The amplification efficiency in patients and control subjects was compared for PDH and AT1 mRNA. The AT1 signal amplified in a collinear fashion in both groups, as did the PDH signal (Fig 3⇓). To determine the sensitivity of the method, we quantitated mRNA from a patient sample by comparison with a standard RNA of known concentrations and assayed a series of dilutions. About 100 copies could still be detected in the PCR/HPLC system, whereas 5 to 10 copies could be detected by Southern blot (data not shown).
The reaction mixture from the PCR amplification tubes was run through an HPLC for quantitation (Gynkotek; DEAE-column: 3.5×0.46 cm, Perkin Elmer), as described previously.30 Samples were applied in a volume of 80 μL. Running conditions were as follows: flow 1 mL/min, gradient from 75% buffer A (25 mmol/L Tris/HCl, pH 9.0), 25% buffer B (25 mmol/L Tris/HCl, pH 9.0, 1 mol/L NaCl) to 100% buffer B after 15 minutes. Peaks were detected using an SP 4 Gynkotek spectral photometer at a wavelength of 260 nm and integrated using the PC integrator software (Nelson Analytical). For quantitation, the area under the curve of the PDH and AT1 peaks and the respective quotients were calculated.
Angiotensin II Binding Sites on Nonmyocyte Cells in Human Myocardium
Human cardiac nonmyocyte cells were isolated by a modification of a method described for the isolation of rat cardiac fibroblasts.32 In brief, pieces of the left ventricular free wall of explanted end-stage failing human hearts were minced and digested with a collagenase (Sigma Chemical Co)/trypsin (Serva)/dispase (Boehringer Mannheim) solution. After pelleting, cells were seeded on cell culture dishes, rinsed once with phosphate-buffered saline (PBS) after an incubation period of 30 minutes at 37°C, and subsequently cultured in DMEM (Gibco BRL) supplemented with 10% fetal calf serum, 2 mmol/L glutamine, and antibiotics. In immunofluorescence assays, isolated cells stained negative for desmin (Boehringer Mannheim), human factor VIII (Behring), and smooth muscle cell actin (Immunotech) and positive for vimentin (Boehringer Mannheim); they are therefore likely to represent human cardiac fibroblasts. For binding studies, cells were seeded in multiwell dishes and cultured to subconfluence in the above-mentioned medium. After removal of the culture medium, cells were incubated overnight at 4°C with 125I–angiotensin II with or without competitor. Unbound radioactivity was removed by washing three times with PBS. Cells were lysed with 2N NaOH, and wells were washed once with water. Lysate and washing were combined and counted in an automated gamma counter.
DUP 753 (Losartan) and PD 123319 were generous gifts from Du Pont Merck and Parke-Davis Pharmaceutical Research. 125I–angiotensin II and 125I-Sar1Ile8–angiotensin II were from Anawa. All other reagents were of the highest purity available.
Data are represented as mean values and standard error of the mean. The Wilcoxon rank test was used to compare the two groups. For multiple comparisons, we used a closed test procedure according to Marcus et al.33 After rejection of the null hypothesis, individual F tests were conducted for the various steps. Significant differences were accepted for P<.05.
Number of Angiotensin II Binding Sites in Heart Failure and Control Subjects
Specific 125I–angiotensin II binding was significantly reduced in atrial sarcolemmal fractions from explanted end-stage failing hearts (n=14) in comparison with control subjects (n=10, P<.01) (Fig 4A⇓). The difference between patients with LVEF 30% to 55% (n=6) and control subjects was not significant (Fig 4A⇓). Accordingly, the number of binding sites calculated by Scatchard analysis was significantly reduced in explanted end-stage failing hearts (Fig 5A⇓) but not in moderate heart failure (LVEF 30% to 55%) in comparison with control subjects (end-stage failing hearts, 3.9±0.8; LVEF 30% to 55%, 9.6±0.8; LVEF >55%, 11.2±1.7 fmol/mg protein). Specific angiotensin II binding in percent of total binding was comparable in the three patient groups (Fig 4B⇓), indicating that the affinity of the receptor for its substrate was unchanged. Calculated dissociation constants were not different in patients and control subjects (0.7±0.2, 0.6±0.2, and 0.5±0.1 nmol/L, respectively) (Fig 5B⇓). The number of angiotensin II receptors and the dissociation constants in sarcolemmal fractions from explanted hearts of 8 patients with dilated cardiomyopathy and 6 patients with ischemic heart disease were not significantly different (Bmax, 4.0±1.0 and 3.9±1.0 fmol/mg protein; Kd, 0.8±0.15 and 0.7±0.10 nmol/L, respectively).
Regional Distribution and Cellular Localization of Angiotensin II Receptors
In a subgroup of seven explanted hearts, specific angiotensin II binding in the right atrium was significantly higher than binding in the left atrium and the right or left ventricle (Fig 6⇓). Specific binding in percent of total binding was unchanged (data not shown). Accordingly, the calculated receptor densities were significantly higher in the right atrium than in other regions of the heart (right atrium, 3.9±0.8; left atrium, 2.0±0.6; right ventricle, 1.7±0.6; left ventricle, 1.6±0.3 fmol/mg protein) (Fig 5A⇑), and dissociation constants in all four regions were comparable (range, 0.7±0.2 to 0.4±0.2 nmol/L) (Fig 5B⇑).
To investigate the cellular localization of angiotensin II receptors in human myocardium, nonmyocyte cultures were prepared from human hearts. Nonmyocytes did not correspond to endothelial or smooth muscle cells but, based on immunostaining, corresponded to fibroblasts (see “Methods”). Significant specific 125I–angiotensin II binding (Bmax, 1.5 fmol/mg protein; Kd, 0.6 nmol/L) was found on these cells (Fig 7⇓).
Angiotensin II Receptors in Combined Sarcolemmal/Light Vesicle Fractions
The number of binding sites in right atrial combined sarcolemmal/light vesicle fractions in 11 end-stage heart failure patients was significantly lower than in 7 patients with LVEF 30% to 55% or in 5 control subjects with LVEF >55% (Fig 8⇓) (Bmax: end-stage heart failure, 17±5; LVEF 30% to 55%, 54±12; LVEF >55%, 71±14 fmol/mg protein). Patients with LVEF 30% to 55% differed significantly from end-stage failing explanted hearts but not from patients with LVEF >55%. The dissociation constants did not differ significantly between the three groups (0.8±0.2, 1.0±0.2, and 1.4±0.4 nmol/L) and were comparable to those in the sarcolemmal fractions.
Subtype Analysis and Antagonist Binding
Replacement of the labeled antagonist 125I-Sar1Ile8– angiotensin II by Sar1Ile8–angiotensin II, PD 123319, and DUP 753 in 7 patients with LVEF >30% (mean, 54±3%) and in 6 explanted hearts is shown in Fig 9⇓, A and B. In the nonfailing hearts, 69±4% of the binding sites were blocked by 10−6 mol/L PD 123319 and are therefore classified as AT2. Conversely, 33±5% of specific binding sites were blocked by DUP 753 and are regarded as AT1. In explanted hearts, 66±5% of binding sites were occupied by PD 123319 and 34±5% by DUP (Fig 9⇓, A and B), resulting in comparable ratios of AT1 and AT2 subtypes in failing and nonfailing hearts. Antagonist binding confirmed significant differences in receptor density between control subjects (Bmax, 110±17) and explanted end-stage failing hearts (Bmax, 45±8 fmol/mg protein, P<.05), indicating a loss of about 60% of receptors in end-stage heart failure. The dissociation constants were not significantly different in the two groups (Kd, 1.7±0.4 and 1.3±0.16 nmol/L, respectively). To assess whether the AT1-specific and the AT2-specific ligand together occupy all angiotensin II binding sites, the labeled antagonist 125I-Sar1Ile8–angiotensin II was replaced with the combination of DUP 753 and PD 123319, both 10−6 mol/L. Together, the two substances replaced 100% of 125I-Sar1Ile8–angiotensin II binding (Fig 10A⇓). In 4 other patients, the AT1 site was blocked with DUP 753 (10−6 mol/L), and the 125I-Sar1Ile8– angiotensin II was replaced by PD 123319 (Fig 10B⇓). The bound 125I-Sar1Ile8–angiotensin II was completely displaced, indicating that the AT1 and the AT2 subtypes account for 100% of receptors in human atria.
Loss of AT1 mRNA in Human Heart Failure
Analysis of AT1 mRNA content by PCR showed significantly reduced amounts of the AT1 amplification product in end-stage failing hearts in comparison to hearts with LVEF >30% (P<.01, Fig 11⇓ and Table 4⇑). The content of PDH mRNA was used as a standard (see “Methods”). Quantitation of the PCR products by HPLC yielded a reduction in the ratio of the AT1/PDH signal to about one third of control subjects in the explanted end-stage failing hearts, indicating a significant reduction in specific AT1 mRNA content.
We report in detail for the first time the regulation, chamber localization, and subtype distribution of angiotensin II receptors in human myocardium. A low density of angiotensin II receptors, about 5 to 10 times less than β-adrenergic receptors, was found in normal human hearts, and a loss of about 60% of receptors was observed in right atrial tissue from explanted end-stage failing hearts. Since the loss of receptors is found in sarcolemmal and in combined sarcolemmal/light vesicle fractions, which include internalized receptors, it cannot be explained by internalization of receptors. PCR analysis suggests that the downregulation is due to a decrease in steady-state mRNA abundance. About 70% of receptors in control subjects and in end-stage failing hearts are of the AT2 subtype. Since this ratio does not change in heart failure, yet the total number of receptors is significantly decreased, it can be concluded that downregulation of both receptor subtypes occurs in end-stage heart failure.
Physiological Role and Localization of ATR in Human Myocardium
The ATR subtype AT1 mediates contraction of isolated myocytes as well as growth in myocytes and fibroblasts.8 9 11 12 Whereas the hypertrophic effects of angiotensin II have not yet been quantitated in human hearts, a positive inotropic effect of angiotensin II on atrial myocardium was demonstrated.10 The inotropic effect obtained by stimulation with angiotensin II in human atria was much weaker than the effect obtained with β-adrenergic agonists in the same preparations.10 This is in agreement with the lower number of angiotensin II receptors in comparison with β-adrenergic receptors in our study. No contractile effect of angiotensin II was found in ventricles of failing human hearts,10 and this is also in agreement with the small number of angiotensin II receptors and the low percentage of AT1 in human end-stage failing ventricles in our study. In addition, a considerable number of ventricular angiotensin II receptors in human hearts is expressed on nonmyocytes, most likely fibroblasts, and are therefore not expected to contribute to a contractile response. Stimulation of collagen production in human cardiac fibroblasts already has been suggested as a potential role for myocardial angiotensin II.15 Further studies are needed to determine the precise cellular localization of angiotensin II receptors. However, the downregulation observed corresponds to the physiological response of angiotensin II receptor pathway stimulation in failing and nonfailing preparations of human heart.10
The higher Bmax in the right atrium in comparison to the explanted end-stage failing ventricles parallels the localization of the angiotensin II–converting enzyme,34 suggesting either that components of the cardiac renin-angiotensin system are mainly expressed in the right atrium or that the loss of receptors in heart failure is more pronounced in failing ventricles than in the atria.
Characterization of ATR and ATR Subtypes in Human Myocardium
High-affinity angiotensin II binding sites have been demonstrated in the myocardium of mammals.5 6 21 35 36 37 38 Whereas some authors observed a single angiotensin II binding site in human myocardium and bovine heart,23 24 others describe two angiotensin II binding sites in calf and rabbit, which differ in agonist and angiotensin II but not in antagonist Sar1Ile8–angiotensin II binding and are modulated by the addition of GTP, GTP-γ-S, or the nonhydrolyzable GTP analogue Gpp(NH)p.36 37
Indeed, the myocardial angiotensin II binding site consists of two angiotensin II receptor subtypes that bind angiotensin II with very similar affinities; in the rat, for example, 1.5 and 1.2 nmol/L, and cannot be differentiated on the basis of angiotensin II binding.6 19 These subtypes differ, however, in their affinity to subtype-specific ligands such as DUP 753 (AT1-specific) and PD 123319 (AT2-specific).6 7 Subtype AT1 is coupled to a guanine nucleotide binding protein, whereas AT2 is not.12 Thus, shifting of binding curves by GTP-γ-S or Gpp(NH)p was found in species in which subsequent studies have shown that AT1 is the dominant subtype, for example, in the rat.19 20 About 70% of human ATR is of the AT2 subtype, which is not modulated by GTP binding proteins, whereas AT1 represents only about 30% of binding sites. Therefore, the effect of Gpp(NH)p may be below the detection threshold in human myocardium. Accordingly, we found no significant shift of the angiotensin II binding curve by Gpp(NH)p. The use of subtype-specific ligands—DUP 753 and PD 123319—allowed the detection and quantification of the receptor subtypes. As the use of PD 123319 and DUP 753 in combination replaced 100% of Sar1Ile8–angiotensin II binding, the two receptor subtypes account for all the angiotensin II receptors in human myocardium.
ATR Downregulation in Human Heart Failure and Potential Mechanisms
The decrease of ATR density in failing human myocardium is likely to be due to a true downregulation and not to a general loss of membrane protein in heart failure, since calculation of wet weight–related receptor densities results in a comparable loss of receptors in heart failure (Table 5⇓). The sarcolemmal fraction used by us and others is enriched in sarcolemmal marker enzymes such as Na/K-ATPase and contains about 90% of β-adrenergic receptors.39 In this fraction, the plasma membrane marker 5′-nucleotidase is unchanged in heart failure, whereas the number of β-receptors is significantly decreased.40 A second fraction was prepared by high-speed centrifugation, containing the sarcolemmal as well as the intracellular light vesicle fraction.22 24 Since we found a 76% decrease in receptors in this combined light vesicle/sarcolemmal fraction and a 65% loss in the sarcolemmal fraction, the loss of surface receptors cannot be explained by their internalization. A comparable loss of 60% of receptors is confirmed by antagonist binding.
Receptor downregulation in human heart failure was not found by Urata et al.23 This group used tissue from multi-organ donors, whose hearts were not transplanted. Based on the special pathophysiological conditions of these hearts before explanation, they cannot be considered completely normal. Our own studies showed highly pathological myocardial catecholamine concentrations in such hearts.41 Furthermore, the tissue was frozen up to 6 hours after explanation of the hearts, and the effect of such a delay on receptor numbers has not been excluded.
As we found no differences in receptor density between patients with dilated cardiomyopathy and coronary heart disease, downregulation of ATR in human heart failure appears not to represent a disease-specific mechanism. Hypertrophy and postinfarction models in the rat show ATR upregulation as well as downregulation.16 17 18 19 20 We propose that in some cases of human heart failure, increased myocardial angiotensin II levels may be present for long periods, since myocardial ACE is probably activated in human heart failure and may not be completely inhibited by all forms of ACE inhibitor therapy.21 42
Confirmation of Receptor Downregulation at mRNA Level
PCR analysis of the AT1 and PDH cDNA amplification signals indicates a reduced AT1 mRNA content in end-stage failing hearts in comparison to control subjects. A reduction in AT1 mRNA content supports the loss of myocardial AT1 established by binding studies and suggests regulation at the transcriptional level and/or by modification of mRNA stability. However, these studies are only based on small numbers, and no internal standard was available to assess the efficiency of reverse transcription and the accuracy of mRNA quantitation by PCR. The results need confirmation by a quantitative PCR technique based on an internal mRNA standard as well as by the investigation of larger patient groups.
In view of the observation that the number of angiotensin II receptors is significantly reduced in end-stage but not in moderate human heart failure, the myocardial effects of ACE inhibitors and angiotensin II receptor antagonists may be more pronounced in early than in late stages of heart failure. Cellular localization of angiotensin II receptors, potential effects of AT2 stimulation, and intracellular signaling pathways of AT2 in human myocardium will require further investigation.
This study was supported by grant DFG Re 662/1-3. We wish to thank Daniela Hauth and B. Krüdewagen for their valuable technical assistance in the receptor-binding studies and the HPLC quantitation of PCR fragments. We thank Carla Weber for her graphical work.
- Received August 19, 1994.
- Accepted September 7, 1994.
- Copyright © 1995 by American Heart Association
Rogg H, DeGasparo M, Graedel E, Stulz P, Erne P. Identifikation von Angiotensin II Rezeptor Subtypen am humanen Vorhofgewebe. Schweiz Med Wochenschr. 1991;121(suppl 36):23. Abstract.
Secchi LA, Chandi AG, Grady EF, Kalinyak JE, Schambelan M. Characterization of angiotensin II receptor subtypes in rat heart. Circ Res. 1992;71:1482-1489.
Allen IS, Cohen NM, Dhallan RS, Gaa ST, Lederer WJ, Rogers TB. Angiotensin II increases spontaneous contractile frequency and stimulates calcium current in cultured neonatal rat heart myocytes: insights into the underlying biochemical mechanisms. Circ Res. 1988;54:286-293.
Dempsey PJ, McCallum ZT, Kent KM, Cooper T. Direct myocardial effects of angiotensin II. Am J Physiol. 1971;220:477-481.
Holubarsch C, Hasenfuss G, Schmidt-Schweda S, Knorr A, Pieske B, Ruf T, Fasol R, Just H. Angiotensin I and II exert inotropic effects in atrial but not in ventricular human myocardium. Circulation. 1993;88:1228-1237.
Sadoshima J, Izumo S. Molecular characterization of angiotensin–induced hypertrophy of cardiac myocytes and hyperplasia of cardiac fibroblasts. Circ Res. 1993;73:413-423.
Baker KM, Aceto JF. Angiotensin II stimulation of protein synthesis and cell growth in chick heart cells. Am J Physiol. 1990;259:H610-H618.
Schorb W, Booz GW, Dostal DE, Conrad KM, Chang KC, Baker KM. Angiotensin II is mitogenic in neonatal rat cardiac fibroblasts. Circ Res. 1993;72:1245-1245.
Brilla CG. Angiotensin II type 2 receptor-mediated stimulation of collagen synthesis in human cardiac fibroblasts. Circulation. 1992;85:(suppl I):I-89. Abstract.
Ullian ME, Linas SL. Role of receptor cycling in the regulation of Angiotensin II surface receptor number and Angiotensin II uptake in rat vascular smooth muscle cells. J Clin Invest. 1989;84:840-846.
Tang SS, Diamant D, Rogg H, Schunkert H, Lorell BH, Ingelfinger JR. Rat hearts contain angiotensin II (ANG II) receptors that are downregulated and differentially expressed during hypertrophy. Hypertension. 1992;20:418. Abstract.
Nishimura J, Kobayashi S, Chen X, Shikasho T, Kanaide H. Angiotensin II receptor mRNA is regulated by angiotensin II: possible involvement of protein kinase C in receptor downregulation. Circulation. 1992;86:(suppl I):I-289. Abstract.
Suzuki J, Matsubara H, Urakami M, Inada M. Rat angiotensin II (type 1A) receptor mRNA regulation and subtype expression in myocardial growth and hypertrophy. Circ Res. 1993;73:439-447.
Meggs LG, Coupet J, Huang H, Cheng W, Li P, Capasso JM, Homcy CJ, Anversa P. Regulation of angiotensin II receptors on ventricular myocytes after myocardial infarction in rats. Circ Res. 1993;72:1149-1162.
Studer R, Müller B, Reinicke H, Just H, Holtz J, Drexler H. Quantified RNA–polymerase chain reaction demonstrates augmented gene expression of angiotensin-converting enzyme in ventricles of patients with heart failure. Circulation. 1992;86(suppl I):I-119. Abstract.
Maisel AS, Motulsky HJ, Insel PA. Externalization of β-adrenergic receptors promoted by myocardial ischemia. Science. 1985;230:183-186.
Feldmann AM, Ray PE, Silan CM, Mercer JA, Minobe W, Bristow MR. Selective gene expression in the failing human heart. Circulation. 1991;83:1866-1872.
Rolfs A, Schuller I, Finkh U, Weber-Rolfs I. PCR, Clinical Diagnostics and Research. Berlin/Heidelberg/New York: Springer Laboratory; 1992.
Rolfs A, Weber-Rolfs I, Regitz-Zagrosek V, Kallisch H, Riedel K, Fleck E. Genetic polymorphisms of the angiotensin II type 1 (AT1) receptor gene. Eur Heart J. 1994;15(suppl D):108-112.
Villarreal FJ, Kim NN, Ungab GD, Printz MP, Dillmann WH. Identification of functional angiotensin II receptors on rat cardiac fibroblasts. Circulation. 1993;88:2849-2861.
Marcus R, Peritz E, Gabriel KR. A closed testing procedure with special reference to ordered analysis of variance. Biometrika. 1976;63:660-665.
Yamada H, Fabris B, Allen AM, Jackson B, Johnston CI, Mendelsohn FAO. Localization of angiotensin-converting enzyme in rat heart. Circ Res. 1991;68:141-149.
Baker KH, Campanile CP, Trachte GJ, Peach MJ. Identification and characterization of the rabbit angiotensin II myocardial receptor. Circ Res. 1984;54:286-293.
Rogers T. High affinity angiotensin II receptors in myocardial sarcolemmal membranes. J Biol Chem. 1984;259:8106-8114.
Wright GB, Alexander RW, Ekstein LS, Gimbrone MA. Characterization of the rabbit ventricular myocardial receptor for angiotensin II. Mol Pharmacol. 1983;24:213-221.
Kent RS, DeLean A, Lefkowitz RJ. A quantitative analysis of β-adrenergic receptor interactions: resolution of high and low affinity rates of the receptor by computer modeling of ligand binding data. Mol Pharmacol. 1980;17:379-385.
Bristow MR, Ginsburg R, Umans V, Fowler M, Minode W, Rasmussen R, Zehra P, Menlove R, Shah P, Jamieson S, et al. β1- and β2-Adrenergic-receptor subpopulations in nonfailing and failing human ventricular myocardium: coupling of both receptor subtypes to muscle contraction and selective β1-receptor downregulation in heart failure. Circ Res. 1986;59:297-309.
Regitz V, Sasse S, Strasser R, Bossaller C, Fleck E. Noradrenaline in healthy and diseased human ventricular and atrial myocardium. Eur Heart J. 1989;10(suppl):247. Abstract.