AT1 and AT2 Angiotensin Receptor Gene Expression in Human Heart Failure
Background The availability of selective antagonists for angiotensin II receptors has focused interest on the gene expression of angiotensin II–receptor subtypes in the human heart.
Methods and Results We analyzed expression of the AT1 and AT2 subtypes of the angiotensin II receptor in ventricular myocardium taken from 9 donor hearts before implantation and from 12 patients with heart failure (6 with dilated cardiomyopathy and 6 with ischemic heart disease). Competitive reverse transcription–polymerase chain reaction with synthetic RNA internal standards was used to detect mRNA for both subtypes and to quantify relative differences in levels between failing and nonfailing ventricular myocardium. AT1- and AT2-receptor mRNA could be detected in all samples. AT1-receptor gene expression was 2.5-fold greater in nonfailing hearts than in patients with failing hearts (P=.015). There was no significant difference in AT2-receptor mRNA expression in failing and nonfailing hearts.
Conclusions The level of expression of the angiotensin AT1 receptor appears to decrease in the failing human ventricle whereas the level of AT2 expression is unaffected. These changes parallel the changes found in human ventricular myocardium at the receptor level, suggesting that the changes in receptor level may result from changes in gene expression or mRNA stability.
The survival benefit conferred by blockade of the renin-angiotensin system in patients with heart failure has led to intense interest in the mechanisms underlying the action of angiotensin II. Recently, a new class of drugs that antagonize angiotensin binding to angiotensin II receptors has been developed1 ; these drugs are undergoing clinical trials in patients with heart failure. Although beneficial actions have been reported from initial research with these agents, details of their mechanisms of action at the receptor level are limited. Angiotensin II has been shown to bind to receptors present in human myocardium, and the first reports analyzing failing and nonfailing ventricular myocardium showed that there was a nonsignificant trend toward decreased angiotensin II receptor binding in hearts from patients with dilated cardiomyopathy.2 Further comparisons of atrial myocardium from failing and nonfailing human hearts found a significant decrease in total angiotensin II–receptor subtype binding.3
In the last 3 years, the angiotensin II receptor population has been subdivided. The human genes encoding the angiotensin II receptor subtypes (AT1 and AT2) were cloned in 19924 and 1994,5 respectively. Subsequently, the AT1 subtype has been further subdivided into the AT1a and AT1b genes that have 98% amino acid homology. The AT1b gene seems restricted in its tissue expression to placenta, lung, and liver and is not expressed in human heart.6 The AT1a receptor-subtype gene is a single-copy gene located on chromosome 3.7 The human AT2 receptor-subtype gene is located on the X chromosome8 and shares 92.6% amino acid homology with the rat AT2 subtype gene.5 In the rat, there is only 32% amino acid–sequence homology between the AT1a and AT2 receptor subtype genes.9
The present study investigated the level of gene expression at the mRNA level for the AT1a and AT2 receptor subtype genes in failing and nonfailing human ventricular myocardium.
Right ventricular myocardium from the endocardial surface of the intraventricular septum was obtained at the time of cardiac transplantation from patients with heart failure. Control myocardium was obtained from the endocardial surface of the right side of the intraventricular septum from donor hearts before implantation.
Reverse Transcription–Polymerase Chain Reaction
In all cases, the myocardial tissue obtained was immediately frozen in liquid nitrogen and stored at −80°C. Total RNA was isolated from tissue samples by use of the method of Chomozynski and Sacchi10 using Trizol reagent (Gibco BRL Life Technologies Inc). RNA was extracted with the use of chloroform and precipitated by isopropanol. The RNA was washed in 75% ethanol and then treated with DNAse 1 (Gibco BRL) for 15 minutes at room temperature before reextraction using the same method. After reextraction, the RNA was dried and stored at −70°C until required. RNA was quantified by spectrophotometry using the A260/280 method. Before use, samples were centrifuged at 4°C and resuspended in RNase free water.
Synthesis of RNA Internal Standards
Synthetic RNA internal standards were manufactured by use of a polymerase chain reaction (PCR)–based technique. For each target mRNA, an oligonucleotide primer pair was synthesized that under the conditions specified in Table 1 ⇓gave amplification of a PCR product of the predicted size visible as a single band on ethidium gel. The PCR product was purified and cut with a restriction enzyme to confirm amplification of the correct sequence. A third oligonucleotide primer was then synthesized corresponding to a position in the amplified target mRNA sequence that was inset from either the 5′ or 3′ end by ≈200 base pairs (bp). This third primer was then used in combination with the appropriate forward or reverse primer from the original pair and conditions were optimized to result in amplification of a PCR product of the predicted size visible as a single band on ethidium gel (≈200 bp shorter than the original PCR product) (Fig 1⇓). If the third primer was a forward primer, a new forward primer was then synthesized that linked a T7 recognition sequence to the original forward-primer sequence, which was in turn linked to the third primer (forward-primer) sequence. A new reverse-primer sequence was also synthesized linking the reverse primer to an 18 poly-T tail. These primers were used to synthesize a PCR product that formed a DNA template for transcription of RNA by use of a T7 RNA polymerase (Promega Co). The RNA formed was then extracted, treated with DNAse 1, and reextracted. The quantity of RNA present was determined by A260/280 spectrophotometry, and reverse transcription–PCR (RT-PCR) amplification from RNA was confirmed with the original primer pair with an RNAse-treated sample used as a negative control to ensure no amplification arose from any template DNA still contaminating the RNA.
The PCR product resulting from amplification of the AT1 target cDNA was 255 bp in length, and the AT1 internal standard PCR product was 137 bp. The AT2 target cDNA product was 293 bp long, and the AT2 internal standard product was 254 bp. To verify that the AT1 PCR product arose from the target cDNA, a restriction enzyme cut was performed with the use of Ssp I. This gave a single cut at position 83 on the product sequence. The PCR product amplified by the AT2 primer pair was subcloned into a vector with the use of the TA cloning kit (Invitrogen Inc) and sequenced to confirm it matched the cDNA sequence.
Serial dilutions of the synthetic RNA mimics were made and added to equal quantities of total RNA extracted from failing or nonfailing hearts in a series of RT-PCR reaction mixes. Each tube therefore contained 500 ng of total RNA and a specific concentration of both the AT1 and AT2 mimic RNA strands. These were then reverse transcribed and amplified by one of the original primer pairs over 45 cycles. First-strand cDNA was synthesized from total RNA with the use of monkey Moloney leukemia virus reverse transcriptase (Perkin-Elmer Cetus) and random hexamers. Aliquots (5 μL) of the cDNA from each combination of total RNA and mimic RNA were amplified against a single pair of primers. Amplification by the PCR was performed in a total reaction volume of 50 μL. The sequences of the primers and the conditions used are shown in Table 1⇑. Thermal cycling was performed on a PTC 100 thermal cycler (MJ Research). Electrophoresis of the amplified products was performed on 1.5% agarose gel containing Tris acetate/EDTA and ethidium bromide. A Hae III digest of φ174 DNA (Gibco BRL) was used as a molecular size standard. Gels were visualized with UV irradiation and photographed. When the synthetic mimic RNA strand exceeded the patient's target mRNA, competition for the primers favored the mimic, and a PCR product resulted that was the size of the synthetic internal standard. When the patient's target mRNA exceeded the mimic RNA, the product was the longer target product, but when the quantity of synthetic mimic RNA in the tube equaled the quantity of the target mRNA present, amplification of both sized bands was equal (Fig 2⇓). This allowed comparison between the failing and nonfailing heart series in relation to the dilution sequence so that relative changes in concentration of the target mRNA between samples could be derived. Ethidium gels were analyzed by blinded observers to determine the point of equivalence between the intensity of the mimic and target product bands. The point of equivalence determined from ethidium gel analysis for an individual total RNA sample was not found to differ between consecutive RT-PCR analyses. To quantify the level of reproducibility, we ran five separate RT-PCR analyses on serial runs of the thermocycler against 1 pg of the synthetic target RNA. The coefficient of variance was .053.
Noncompetitive amplification of β-actin was used to demonstrate the presence of intact mRNA in each total RNA sample and to help to demonstrate approximate equivalence of mRNA loading in each patient's RT-PCR reaction series.
RT-PCR Control Experiments
The absence of genomic contamination in the cDNA samples was confirmed by the use of a PCR reaction using primers that amplify the promoter region of apolipoprotein(a)-related gene C, a nontranscribed region of genomic DNA.11 These primers were designed to sensitively and specifically amplify genomic DNA only, and the technique was validated against serial dilutions of genomic DNA and total RNA with or without RNAse pretreatment before RT-PCR (G.A.H., MD, MRCP, and C. Byrne, MD, MRCP, unpublished data, 1994). This method allows screening of cDNA for genomic contamination without the need to use reverse-transcriptase–negative RNA controls, thus saving on the consumption of total RNA. A subgroup of samples were also put through the reverse-transcription stage without the addition of reverse transcriptase to act as supplementary negative controls. The AT2 primers spanned intron sequences (T.K., MD, and M.H., MD, unpublished data, 1995) to enable identification of amplification of cDNA from any genomic DNA amplification by the size of the product. The AT1 primers were located within the single intron that contains the entire coding sequence.12
Validation of Competitive RT-PCR Measurements
A synthetic RNA standard containing terminal sequences for the forward and reverse primers of both the AT1-receptor and AT2-receptor subtypes was synthesized by PCR, purified on an electrophoretic gel, and quantified by spectrophotometry. One picogram was then amplified against a serial dilution of the mimic RNAs for AT1 and AT2 using each of the primer pairs in turn to test whether differences in amplification efficiency resulted in different measurements of the quantity of the synthetic RNA standard present.
Variability in absolute quantities of mimic RNA in successive dilutive series was controlled for by comparison with the synthetic RNA standard, and appropriate corrections were applied.
In addition, total RNA from a patient with heart failure was diluted 10-fold and run against a 10-fold serial dilution of the mimic RNA to check that the point of equivalence showed a 10-fold difference between the two concentrations.
All results are expressed as mean±SD. Statistical analysis was performed with the use of a statistical software package (Statview, Abacus) on an Apple Macintosh computer. The two-tailed Mann-Whitney U test was used to compare relative concentrations of the AT1 and AT2 mimics in failing and nonfailing hearts and relative concentrations of AT1 versus AT2 in each group. Differences between patients with ischemic heart disease, dilated cardiomyopathy, and donor hearts were analyzed by use of ANOVA with appropriate post hoc tests.
Subjects (n=9) from whom donor control myocardium was obtained were significantly younger (33.3±12.6 years; range, 19 to 52 years) than patients with heart failure (50.2±8.9 years; range, 33 to 65 years; P<.01). However, there was overlap in the ages of patients from the two groups, and there was no significant correlation between age of the donor hearts and either AT1 (r=−.51, P=NS) or AT2 (r=.014, P=NS) mRNA level. None of the subjects from whom donor control myocardium was obtained showed any evidence of cardiac disease. Clinical characteristics of the patients with heart failure are shown in Table 2⇓. None of the patients with ischemic heart disease had clinical evidence to suggest they had suffered acute myocardial infarction within 6 months of cardiac transplantation (range, 6.5 to 30 months).
Representative gels for the AT1- and AT2-receptor subtypes analyzed by competitive RT-PCR amplification to 45 cycles are shown in Fig 3⇓.
Amplification of β-actin was confirmed in all samples. In all cases, screening for genomic contamination after DNAse treatment of the RNA was negative. The AT1 mRNA levels were higher in nonfailing than in failing hearts (200±104 versus 79±41 fg; P=.015). There was no significant difference in the levels of AT2 mRNA between the two groups (113±65 versus 137±150 fg; P=NS) The differences between levels of mRNA for AT1 and AT2 in the two groups are shown in Fig 4⇓. AT1 mRNA levels were similar in patients with ischemic heart disease versus dilated cardiomyopathy (81±45 versus 77±41 fg; P=NS). However, the numbers in each group were small (n=6 in each group) and thus the statistical power for the detection of any significant difference between the two etiologies was low.
Dilution of a sample of total RNA by 10-fold resulted in a shift in the point of equivalence against the mimic serial dilution of 10-fold (Fig 5⇓). Amplification of AT1 and AT2 mimic serial dilution ranging from 20 pg to 10 fg against 1 pg of the synthetic target RNA confirmed that the point of equivalence was at the 1-pg dilution of each mimic (Fig 5⇓). The ratio of mRNA for AT1 compared with mRNA for AT2 in nonfailing hearts was 1.78:1 (P=NS); the ratio of mRNA for AT1 compared with mRNA for AT2 in failing hearts was 0.58:1 (P=NS).
The decrease in the concentration of AT1 mRNA in failing compared with nonfailing myocardium agrees with the findings of a previous report that analyzed atrial myocardium3 and with findings in failing human ventricular myocardium at the receptor level.13 The lack of significant change in the level of AT2 mRNA between failing and nonfailing myocardium also reflects findings at the receptor level in human ventricular myocardium13 but differs from the findings reported at the receptor level in human atrial myocardium, in which it appears that the level of AT2 receptors is also decreased under the conditions of chronic heart failure.3
The similarity of relative changes in receptor density and mRNA level suggest that the principal mechanism regulating the level of angiotensin II receptors may be the concentration of mRNA for the two receptor subtypes present in the myocardium. This is not necessarily the case, because the mRNA concentrations could be just an epiphenomenon associated with another regulatory mechanism, but regulation at the mRNA expression level is thought to be the primary means by which cells regulate the availability of proteins.14 The cause of the decrease in the level of mRNA for the AT1 subtype cannot be determined from the present study. The mechanism may be factors resulting in decreased transcription, decreased message stability, or both.
Although the results obtained in the present study give a definitive answer to the question being investigated, it is important to avoid concluding too much. Because others have performed quantification at the receptor level in detailed studies in subjects with heart failure, we did not attempt to obtain receptor density data on the patients we studied. Paired comparisons of mRNA and receptor concentrations in individual patients therefore were not possible.
Because of the low level of message present and the very limited quantities of donor myocardium available, we were unable to localize mRNA expression to individual cell types and could only provide data for ventricular myocardium as a whole. The observations were confined to myocardium taken from the right side of the interventricular septum because this is the area from which the donor biopsy samples were taken, and thus regional differences in the level of RNA expression could not be assessed. Finally, the functional significance of the observation that the level of AT1 gene expression decreases in failing hearts cannot be determined from the present study. Even if the numbers of subjects studied were greatly increased, it would be difficult to draw associations between the level of gene expression and clinical features because the number of uncontrolled variables present in such patients is necessarily high.
The findings in the present study have interesting implications for the predicted actions of angiotensin II receptor inhibitors. The preservation of gene expression for the AT2 receptor in patients with heart failure indicates that a situation will exist in the presence of AT1-receptor blockade in which angiotensin II will act unopposed on the AT2 receptors. The removal of feedback inhibition by the actions of angiotensin II via the AT1 receptor may also result in an increase in circulating and tissue levels of angiotensin II,15 thus increasing the occupation of the AT2-receptor population. Although the functional effects mediated by the AT2 receptor remain unclear, there is increasing evidence that in some tissues the AT2 receptor may activate pathways that result in inhibition of cell growth16 or even stimulation of apoptotic pathways.17 18 Although proapoptotic effects acting via the AT2 receptor have not been investigated in myocardium, there is evidence in PC12W (rat pheochromocytoma) cells and R3T3 (mouse fibroblast) cells that apoptosis may be induced by dephosphorylation of mitogen-activated protein kinase secondary to AT2-receptor stimulation.18 Thus, it is possible that a change in the balance of the effects mediated by angiotensin II on the two receptor subtypes may not only decrease the stimulation of hypertrophy through the AT1 receptor but also potentially increase antihypertrophic, proapoptotic effects on the myocardium.
The decrease in AT1-receptor subtype expression seen in human heart failure is in marked contrast to the increase observed in the ventricle of animal models of acute myocardial infarction.19 In the rat infarct model, there appears to be upregulation of AT1–receptor-subtype gene expression, both acutely and persisting up to 8 months after infarction.20 Despite the fact that patients with ischemic heart disease were included in the heart failure group in our study, the effects of myocardial infarction and ischemic cardiomyopathy appear to have been opposite. This raises intriguing questions about what the major regulators of AT1-receptor gene expression in the myocardium are. One possibility is that the cell types expressing AT1 in these two situations differ. There is some evidence that the upregulation of AT1 after infarction localizes to scar tissue and may be related to fibroblasts20 ; the downregulation observed in failing hearts, however, may be related to cardiac myocytes. It is also possible that species differences are important. In the rat, there is a transcription factor AP1 recognition site present on the AT1-promoter region,9 which might act to stimulate AT1 transcription under hypoxic conditions. This site is absent on the human AT1 promoter and may result in oxidative mechanisms predominating and causing inhibition of the binding of oxidation-sensitive transcription factors such as stimulator protein 1 (SP1).21 SP1 recognition sites are present just upstream of the human AT1 transcription start site, and inhibition of the binding of SP1 would be predicted to decrease transcription of the gene.
In the interventricular septum, myocardial levels of mRNA for the angiotensin II AT1-receptor subtype are 2.5-fold greater in nonfailing hearts than in patients with heart failure. In contrast, levels of AT2-receptor subtype mRNA are similar. These results are very similar to the changes observed in the receptor density of the two subtypes in failing and nonfailing hearts, suggesting that regulation of the alteration in expression is at the mRNA level. The alteration in the balance of gene transcription between the two subtypes may alter the actions of angiotensin II on failing hearts and may influence the effects of specific AT1-receptor antagonists in patients with heart failure.
This work was supported by a medical school grant from Merck and Co, West Point, Pa. Dr Gullestad was supported by The Research Council of Norway. We are most grateful to Gail Yee for expert technical assistance with the extraction and preparation of RNA. We would also like to thank the surgical staff at Stanford, and in particular Hermann Reichenspurner and John Stevens, who assisted us in the collection of the myocardial samples.
- Received September 21, 1995.
- Revision received March 20, 1996.
- Accepted October 28, 1996.
- Copyright © 1997 by American Heart Association
Regitz-Zagrosek V, Friedel N, Heymann A, Bauer P, Neuss M, Rolfs A, Steffen C, Hildebrandt A, Hetzer R, Fleck E. Regulation, chamber localization, and subtype distribution of angiotensin II receptors in human hearts. Circulation. 1995;91:1461-1471.
Inagami T, Guo D, Kitami Y. Molecular biology of angiotensin II receptors: an overview. J Hypertens. 1994;12(suppl 10):S83-S94.
Byrne C, Schwartz K, Meer K, Cheng J-F, Lawn R. The human apolipoprotein (a)/plasminogen gene cluster contains a novel homologue transcribed in liver. Arterioscler Thromb. 1994;14:534-541.
Asano K, Minobe W, Mitchusson K, Dutcher D, Roden RL, Port JD, Bristow MR. Selective down-regulation of angiotensin II AT1 receptors in failing human heart: relationship to β1-receptor down regulation. J Am Coll Cardiol. 1995;25:291A. Abstract.
Goldberg M, Tanaka W, Barchowsky A, Bradstreet TE, McCrea J, Lo M-W, McWilliams EJ, Bjornsson TD. Effects of losartan on blood pressure, plasma renin activity, and angiotensin II in volunteers. Hypertension. 1993;21:704-713.
Stoll M, Steckelings U, Paul M, Bottari S, Metzger R, Unger T. The angiotensin AT2-receptor mediates inhibition of cell proliferation in coronary endothelial cells. J Clin Invest. 1995;95:651-657.
Yamada T, Horiuchi M, Dzau V. The novel angiotensin II type 2 receptor mediates programmed cell death. Proc Natl Acad Sci U S A. In press.
Meggs L, 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.
Lefroy D, Wharton J, Crake T, Knock GA, Rutherford RAD, Polak JM, Poole-Wilson PA. Angiotensin II receptor expression after myocardial infarction. Br Heart J. 1995;73:P32. Abstract.