Abstract
We recently reported that in the ischemic human heart, locally formed angiotensin II activates angiotensin II type 1 (AT1) receptors on sympathetic nerve terminals, promoting reversal of the norepinephrine transporter in an outward direction (i.e., carrier-mediated norepinephrine release). The purpose of this study was to assess whether cardiac sympathetic nerve endings contribute to local angiotensin II formation, in addition to being a target of angiotensin II. To this end, we isolated sympathetic nerve endings (cardiac synaptosomes) from surgical specimens of human right atrium and incubated them in ischemic conditions (95% N2, sodium dithionite, and no glucose for 70 min). These synaptosomes released large amounts of endogenous norepinephrine via a carrier-mediated mechanism, as evidenced by the inhibitory effect of desipramine on this process. Norepinephrine release was further enhanced by preincubation of synaptosomes with angiotensinogen and was prevented by two renin inhibitors, pepstatin-A and BILA 2157BS, as well as by the angiotensin-converting enzyme inhibitor enalaprilat and the AT1 receptor antagonist EXP 3174 [2-N-butyl-4-chloro-1-[2′-(1H-tetrazol-5-yl)biphenyl-4-yl] methyl]imidazole-5-carboxylic acid]. Western blot analysis revealed the presence of renin in cardiac sympathetic nerve terminals; renin abundance increased ∼3-fold during ischemia. Thus, renin is rapidly activated during ischemia in cardiac sympathetic nerve terminals, and this process eventually culminates in angiotensin II formation, stimulation of AT1 receptors, and carrier-mediated norepinephrine release. Our findings uncover a novel autocrine/paracrine mechanism whereby angiotensin II, formed at adrenergic nerve endings in myocardial ischemia, elicits carrier-mediated norepinephrine release by activating adjacent AT1 receptors.
Angiotensin II (Ang II) is a potent facilitator of norepinephrine (NE) release from peripheral (Zimmerman, 1962) and cardiac sympathetic nerve endings (Seyedi et al., 1997; Farrell et al., 2001). A renin-angiotensin system is present in the heart (Dostal and Baker, 1999; Bader et al., 2001), and local Ang II formation increases in myocardial ischemia (Jalowy et al., 1999). Locally formed Ang II could therefore play a role in the release of NE associated with myocardial ischemia. Indeed, in an isolated guinea pig heart model of ischemia/reperfusion, blockade of Ang II AT1 receptors (AT1R) reduced both NE release and the severity of associated arrhythmias (Maruyama et al., 1999).
We recently reported that in the ischemic human heart, ACE-independent formation of Ang II from angiotensin I (Ang I) promotes carrier-mediated release of NE by activating AT1R (Maruyama et al., 2000) located on sympathetic nerve terminals (Seyedi et al., 1997). Renin and prorenin are present in adrenal chromaffin cells (Berka et al., 1996), which are functionally comparable to postganglionic adrenergic nerves. Furthermore, a new form of renin, which is activated by ischemia, has been recently described in the heart (Clausmeyer et al., 2000).
The purpose of this study was to assess whether cardiac sympathetic nerve endings may not only be the target of Ang II (Seyedi et al., 1997), but also contribute to local Ang II formation. We report the presence of immunoreactive renin in sympathetic nerve terminals isolated from the human heart and its increased abundance during ischemia. When the activity of this renin was inhibited pharmacologically, the release of NE elicited by ischemia was markedly attenuated. This suggests a novel autocrine/paracrine mechanism by which Ang II, generated at adrenergic nerve endings in ischemic conditions, elicits the release of NE by activating adjacent AT1R.
Materials and Methods
Source of Human Cardiac Tissue.
Specimens of right atrium (i.e., surgical waste tissue) were obtained from 34 patients undergoing cardiopulmonary bypass (29 males and 5 females, age 64.9 ± 1.6 years; coronary artery bypass grafting in 30, valve replacement in 4), following a protocol approved by our institutional review board. Of the 30 coronary artery bypass grafting patients, 8 were chronically treated with beta adrenoceptor blocking agents. Preoperative treatment with beta blockers did not affect the ischemic release of NE. All patients chronically treated with ACE inhibitors were excluded from the study. At the time of surgery, a piece of atrial appendage measuring ∼1 cm3 was removed from the atriotomy site.
Incubation of Atrial Tissue.
Atrial specimens were immediately transported to the laboratory in ice-cold oxygenated Krebs-Henseleit solution (KHS) of the following composition: 118.2 mM NaCl, 4.83 mM KCl, 2.5 mM CaCl2, 2.37 mM MgSO4, 1.0 mM KH2PO4, 25 mM NaHCO3, and 11.1 mM glucose. After removal of fat and connective tissue, specimens were divided into several fragments (each weighing 24.5 ± 1.1 mg, wet weight, measured at the end of incubation). Each fragment was incubated for 15 min at 37°C in 2 ml of KHS gassed with 95% O2 and 5% CO2 (PO2 ∼550 mm Hg, pH ∼7.4) containing the monoamine oxidase inhibitor pargyline (1 mM). Following the 15-min stabilization period, fragments were incubated for an additional 20 min in oxygenated KHS in the absence or presence of pharmacological agents.
Preparation of Cardiac Synaptosomes.
Atrial specimens were freed from fat and connective tissue and minced in ice-cold 0.32 M sucrose containing 1 mM EGTA, pH 7.4, and 1 mM pargyline, to prevent enzymatic destruction of synaptosomal NE. Synaptosomes were isolated as previously described (Imamura et al., 1995). Briefly, minced tissue was digested with 120 mg of collagenase (Type II; Worthington Biochemicals, Freehold, NJ) per gram of wet heart weight for 1 h at 37°C. After low-speed centrifugation (10 min at 120g), the resulting pellet was suspended in 10 volumes of 0.32 M sucrose and homogenized with a Teflon/glass homogenizer. The homogenate was spun at 650g for 10 min, and the pellet was rehomogenized and respun. The pellet containing cellular debris was discarded, and the supernatants from the last two spins were combined and equally subdivided into four to eight tubes and recentrifuged for 20 min at 20,000g at 4°C. This pellet, which contained cardiac synaptosomes, was resuspended either in HEPES-buffered saline (HBS; 500 μl, normoxic conditions) or in glucose-free HBS, which contained the reducing agent sodium dithionite (500 μl, ischemic conditions), and incubated in the absence or presence of angiotensinogen (for 1 h) or other pharmacological agents for 20 min in a water bath at 37°C prior to ischemia (see below). HBS contained 50 mM HEPES, pH 7.4, 144 mM NaCl, 5 mM KCl, 1.2 mM CaCl2, 1.2 mM MgCl2, and 10 mM glucose. Each suspension functioned as an independent sample and was used only once. In every experiment, one sample was untreated (control, basal NE release).
Purity of the synaptosomal preparation was verified by Western blot analysis of myosin using MF20, an antibody against sarcomeric myosin heavy chain (Fig. 1) (Bader et al., 1982). Sarcomeric myosin was used as a marker of synaptosomal purity, since it is present in myocardial tissue but not in nerve endings. Equal amounts of protein (1.5 μg) for both atrial homogenate and synaptosomes were run in parallel and in triplicate on the same gel. The density of each band within a gel was analyzed using NIH Image Software (version 1.60). As shown in the immunoblot (see Fig. 1), detection of the 205-kDa band associated with myosin was barely detectable in the synaptosomal fraction compared with the atrial homogenate (OD, 768 ± 281 and 9744 ± 705 for synaptosomes and atrial homogenate, respectively; means ± S.E.M.).
Induction of Ischemia.
Ischemia was induced by incubating either atrial fragments or synaptosomes for 70 min in glucose-free KHS (atrial tissue) or HBS (synaptosomes) bubbled with 95% N2 and 5% CO2, containing the reducing agent sodium dithionite (3 mM; PO2∼0 mm Hg, pH ∼7.3; ischemic NE release) (Hatta et al., 1997). Matched control fragments and synaptosomes were incubated for an equivalent length of time with oxygenated KHS and HBS, respectively (normoxic NE release). Drugs, when used, were continued throughout the entire normoxic and ischemic periods.
Norepinephrine Assay.
Incubating media were assayed for NE by high-pressure liquid chromatography with electrochemical detection (Hatta et al., 1997). Perchloric acid and EDTA were added to samples to achieve final concentrations of 0.01 N and 0.025%, respectively. The NE present in the effluent was adsorbed on acid-washed alumina adjusted at pH 8.6 with Tris-2% EDTA buffer, and then extracted into 150 μl of 0.1 N perchloric acid. These final sample aliquots were injected onto a 3-μm ODS reverse-phase column (3.2 × 100 mm; Bioanalytical Systems Inc., West Lafayette, IN) with an applied potential of 0.65 V. The mobile phase consisted of monochloroacetic acid (75 mM), sodium EDTA (0.5 mM), sodium octylsulfate (0.5 mM), and acetonitrile (1.5%) at pH 3.0. The flow rate was 1.0 ml/min. No NE breakdown occurred during the 70-min ischemic period. Dihydroxybenzylamine was added to each sample as an internal standard prior to alumina extraction and used for calculation of the recovery during the extraction procedure. This recovery was 77% or better. The detection limit was approximately 0.2 pmol.
Western Blotting.
Human right atrial homogenate or synaptosomal preparations were mixed with 10 μl of 2× Novex Tris-glycine SDS sample buffer (Invitrogen, Carlsbad, CA) and boiled for 4 to 5 min. Samples were separated by electrophoresis on 4% and 10 to 20% gradient Tris-glycine SDS-polyacrylamide gels, for myosin and renin, respectively. Electrophoresis was carried out at 50 V/gel for 60 min. Gels were soaked in transfer buffer (25 mM Tris-base, 0.2 M glycine, and 20% methanol, pH 8.5) and electrotransferred onto 0.22-μm nitrocellulose membranes (Invitrogen) for 90 min at 200 V and 4°C. After transfer, the nitrocellulose was blocked for 1 h in blocking buffer [Tris-buffered saline (TBS), containing 0.1% Tween 20, 5% (w/v) nonfat dry milk]. Anti-sarcomeric myosin heavy chain antibody (MF20; Bader et al., 1982) or anti-renin antibody (BR1–5;Campbell et al., 1996), kindly donated by Drs. D. A. Fischman and D. F. Catanzaro, respectively, was incubated with the nitrocellulose overnight at 4°C, diluted 1:50,000 or 1:12,500 in primary antibody dilution buffer (TBS containing 0.1% Tween 20, 5% bovine serum albumin), respectively. The nitrocellulose was washed three times with TBS, then horseradish peroxidase-coupled secondary antibody was added at a 1:2000 dilution in blocking buffer for 1 h. After three further TBS washes, myosin or renin was detected using enhanced chemiluminescence (LumiGLO; Cell Signaling Technology Inc., Beverly, MA). Chicken pectoralis myosin and mouse kidney extracts were used on appropriate gels as positive controls. Prestained molecular weight standards (Invitrogen) were included in all gels.
Statistics.
Values are expressed as mean ± S.E.M. Analysis by one-way ANOVA was used, followed by Dunnett's multicomparison testing. A value of p < 0.05 was considered statistically significant.
Drugs and Chemicals.
Human plasma angiotensinogen, desipramine hydrochloride (DMI), pepstatin-A, and 5-(N-ethyl-N-isopropyl)-amiloride (EIPA) were purchased from Sigma-Aldrich (St. Louis, MO). Enalaprilat and EXP 3174 were gifts from Merck Research Laboratories (West Point, PA). BILA 2157BS was a gift from Dr. B. Simoneau, Boehringer Ingelheim (Canada) Ltd., Research and Development (Laval, QC, Canada). EXP 3174, DMI, pepstatin-A, and EIPA were dissolved in dimethyl sulfoxide. BILA 2157BS was dissolved in 0.02 M Na2HPO4 buffer. Further dilutions were made with distilled water; at the concentration used, dimethyl sulfoxide and Na2HPO4 buffer did not affect NE release.
Results
Carrier-Mediated NE Release from the Human Myocardium.
The incubation of human right atrial tissue for 70 min in glucose-free KHS in ischemic conditions (PO2 ∼0 mm Hg; pH ∼7.3) caused a 7-fold increase in the release of endogenous NE above basal level in normal oxygenated conditions (ischemic, 3.5 ± 0.21 versus basal, 0.5 ± 0.11 pmol/mg of protein; means ± S.E.M.; n = 12). As previously reported (Hatta et al., 1997), this release is carrier-mediated, since it is Ca2+-independent and inhibited by the NE transporter inhibitor DMI. As shown in Fig.2, inhibition of renin activity with either the aspartyl protease inhibitor pepstatin-A (panel A) or the more potent and selective renin inhibitor BILA 2157BS (panel B) caused a concentration-dependent decrease in NE release, which amounted to ∼70% with 30 μM pepstatin-A and ∼60% with 30 nM BILA 2157BS.
Carrier-Mediated NE Release from Sympathetic Nerve Terminals Isolated from the Human Myocardium.
Sympathetic nerve endings (cardiac synaptosomes) were isolated from human right atrial tissue. As shown in Figs. 3 and 4, incubation of human cardiac synaptosomes for 70 min in glucose-free HBS in ischemic conditions, elicited a significant release of endogenous NE (i.e., an ∼70% increase above basal level in normal oxygenated conditions). This release was inhibited by ∼50% by the NE transporter inhibitor DMI (300 nM) and by the inhibitor of the Na+/H+ exchanger (NHE) EIPA (30 μM), indicating that it was carrier-mediated (Fig. 3). The increase in NE release caused by ischemia was markedly reduced (∼80%) by the ACE inhibitor enalaprilat (3 μM) and by the Ang II AT1R antagonist EXP 3174 (300 nM), indicating the participation of endogenous Ang II in this process (Fig. 3). Furthermore, the renin inhibitors pepstatin-A (30 μM) and BILA 2157BS (100 nM) suppressed the enhancement in NE release elicited by ischemia by ∼70%, implying a role of renin in the neuronal formation of Ang II (Fig. 3).
A role of renin in the neuronal formation of Ang II was further supported by the findings depicted in Fig. 4. Incubation of human cardiac synaptosomes with angiotensinogen (400 nM) increased ischemic NE release by ∼70%. This increase was prevented by the renin inhibitor BILA 2157BS (100 nM), the ACE inhibitor enalaprilat (3 μM), and the Ang II AT1R antagonist EXP 3174 (300 nM)(Fig. 4).
Presence of Renin in Sympathetic Nerve Terminals Isolated from the Human Myocardium.
Sympathetic nerve terminals were screened by Western blot for the presence of renin with a specific antibody, BR1-5 (Campbell et al., 1996). Equal amounts of protein (1.0 μg) for both normoxic and ischemic synaptosomes were run in parallel and in triplicate on the same gel. Figure 5demonstrates that renin is present in sympathetic nerve endings isolated from the human heart and, more importantly, that renin increases ∼3-fold with a 70-min ischemia period (OD, 820 ± 135 and 2159 ± 339 for normoxic and ischemic synaptosomes, respectively).
Discussion
In protracted myocardial ischemia, NE is released in massive amounts by a nonexocytotic, “carrier-mediated” mechanism (i.e., the NE transporter carries NE in an outward direction) (Schömig et al., 1991; Dart and Du, 1993). This reversal of the normal NE reuptake process results from impeded NE storage into synaptic vesicles, due to failure of the H+-ATPase pump, leading to NE accumulation in the axoplasm, coupled with a decrease in intracellular pH due to ATP deficit. This intraneuronal acidosis activates the NHE, leading to an increase in intracellular Na+ that causes reversal of the NE reuptake process when combined with the increased axoplasmic NE, eliciting carrier-mediated NE release (Levi and Smith, 2000). That NHE plays a pivotal role in the initiation of carrier-mediated NE release in the ischemic human myocardium is supported by our findings that agents known to stimulate NHE (particularly, Ang II) potentiate this type of NE release (Maruyama et al., 2000), whereas NHE inhibitors (e.g., amiloride derivatives) prevent it (Hatta et al., 1997).
Since local Ang II production increases in the ischemic myocardium (Jalowy et al., 1999) and Ang II is a potent NHE activator (Gunasegaram et al., 1999), it is likely that Ang II formed in the heart plays a role in carrier-mediated NE release and ventricular arrhythmias during myocardial ischemia (Maruyama et al., 1999, 2000). There is considerable debate as to where in the heart, myocytes, and/or interstitium Ang II is generated (Dostal and Baker, 1999; Bader et al., 2001). Since renin is present in chromaffin cells (Berka et al., 1996), which are functionally comparable to postganglionic adrenergic nerves, we considered the possibility that sympathetic nerve terminals may be a source of Ang II, as well as a target (Seyedi et al., 1997), in the ischemic human heart.
Knowing that inhibition of Ang II formation from Ang I, via both ACE-dependent and -independent pathways, attenuates carrier-mediated NE release in the ischemic human heart (Maruyama et al., 2000), we first questioned whether Ang I may be produced in situ and eventually contribute to ischemic NE release after its conversion to Ang II. Indeed, we found that the aspartyl protease inhibitor pepstatin-A (Guyene et al., 1976) and the more potent and selective renin inhibitor BILA 2157BS (Duan et al., 1996; Simoneau et al., 1999) markedly diminished NE release in the ischemic human heart (see Fig. 2). Notably, this is the same human model in which we previously found NE release to be inhibited by blockade of Ang II AT1R (Maruyama et al., 2000). Thus, our present and past findings in the ischemic human heart strongly suggest a renin-dependent local generation of Ang I and its subsequent cleavage to Ang II. Ang II then further activates NHE, thus promoting carrier-mediated NE release.
A renin-angiotensin system has long been described in the heart (Dzau, 1987; Lindpaintner and Ganten, 1991), but the cardiac synthesis of renin has been controversial (Von Lutterotti et al., 1994). Although renin can be taken up from the coronary circulation and play a role in the local formation of Ang II (Muller et al., 1998), a new intracellular nonsecretory form of renin (exon 1A-renin mRNA) has been found in rat cardiac tissue (Clausmeyer et al., 2000). Notably, the expression of exon 1A-renin mRNA in the heart is greatly increased following infarction (Clausmeyer et al., 2000). Another distinct renin transcript is also present in neural tissue (Lee-Kirsch et al., 1999), and both renin and prorenin have been localized in adrenal chromaffin cells (Berka et al., 1996), which are analogous to postganglionic adrenergic nerves.
Therefore, we next investigated whether cardiac sympathetic nerve terminals specifically harbor a renin-angiotensin system capable of fostering NE release during ischemia. To this end, we established a new model of cardiac neuronal ischemia. Thus, we isolated sympathetic nerve endings from the human heart (Imamura et al., 1995) and incubated them in the same ischemic conditions as the atrial tissue. This resulted in a large release of NE via reversal of the NE transporter. Indeed, this release was attenuated by the NE transporter inhibitor DMI and by the NHE inhibitor EIPA (see Fig. 3).
Although the amount of NE released from sympathetic nerve endings during ischemia (∼70% increase over basal levels) was not as large as in atrial tissue (7-fold increase), this discrepancy is less relevant if one considers that the absolute values of NE released per milligram of protein at the end of the 70-min ischemia period were relatively similar (synaptosomes, 2.45 ± 0.09 pmol of NE per milligram of protein; n = 64; atrial tissue, 3.5 ± 0.26 pmol of NE per milligram of protein; n = 12), whereas basal NE release from synaptosomes (1.46 ± 0.03 pmol of NE per milligram of protein; n = 64) was far greater than basal release from atrial tissue (0.5 ± 0.18 pmol of NE per milligram of protein; n = 12). We had previously recognized that basal NE release from synaptosomes can be relatively high (Seyedi et al., 1997). Indeed, in addition to its high concentration in the nerve terminals, NE also exists in lower concentrations throughout the neuron. Thus, in the course of the homogenization-centrifugation process, when nerve terminals detach and reseal, variable amounts of NE may escape in the final high-speed supernatant (Whittaker, 1993).
The synaptosomal preparation was highly pure (see Fig. 1). Therefore, a major portion of the ischemic synaptosomal NE release results from the activation of the renin-angiotensin system in sympathetic nerve endings. Notably, due to the inherent leakiness of the synaptosomal preparation, and the consequent higher basal NE release, the contribution of the synaptosomal renin-angiotensin system to ischemic NE release is likely to be even greater in vivo.
That cardiac sympathetic nerve endings are endowed with a renin-angiotensin system, which is activated by ischemia and thus promotes NE release, is supported by several of our findings: 1) the renin inhibitors pepstatin and BILA 2157BS, as well as the ACE inhibitor enalaprilat and the Ang II AT1R antagonist EXP 3174, each abolished NE release in this model of cardiac neuronal ischemia (see Fig. 3); 2) incubation of sympathetic nerve endings with angiotensinogen greatly enhanced NE release during ischemia, and this effect was prevented by either renin inhibition, ACE inhibition, or Ang II AT1 receptor blockade (see Fig. 4); and 3) renin is present in sympathetic nerve terminals, and exposure to ischemia for 70 min markedly enhanced its abundance (see Fig. 5).
Although the essential components of the renin-angiotensin system had already been identified in the heart (Dostal and Baker, 1999; Bader et al., 2001), this is the first report of an active renin-angiotensin system in a cardiac synaptosomal preparation. It implies that angiotensin produced at the nerve endings could influence cardiac adrenergic transmission in an autocrine/paracrine fashion, particularly when adrenergic activity and NE release are increased, as occurs in myocardial ischemia (Schömig et al., 1991; Imamura et al., 1994,1996), a condition associated with increased local angiotensin production (Jalowy et al., 1999).
Most importantly, our data reveal the presence of renin in sympathetic nerve terminals of the human heart and its activation during ischemia. From our immunoblot studies (see Fig. 5), the approximate molecular weight of synaptosomal renin appears to be 38 kDa, which characterizes it as a truncated form of renin, compared with human renal renin (Do et al., 1987). Three renin isoforms have been identified in human brain, lung, and kidney (Sinn and Sigmund, 2000). It is not clear which of these isoforms is expressed in human heart and whether they are activated by cleavage of the rest of the propeptide, which would result in a lower molecular weight.
We cannot exclude the possibility that a transcript similar to exon 1A-renin (Clausmeyer et al., 2000) exists in the human heart, and that such a transcript may be related to the enzymatic form activated by ischemia in cardiac synaptosomes. Exon 1A-renin mRNA was found to be increased 5-fold in the left ventricle 4 to 5 days after ligation of the left descending coronary artery (Clausmeyer et al., 2000). In contrast, we found an ∼3-fold increase in renin protein after only 70 min of ischemia. Clausmeyer et al. (2000) did not measure exon 1A-renin mRNA prior to 4 days after infarction, based on earlier studies by other investigators (Passier et al., 1996). Possibly, exon 1A-renin mRNA levels might have been elevated at an earlier stage.
The local synthesis of renin in the heart has been controversial, and earlier studies argued that renin mRNA was present in the heart only in minuscule levels, insufficient to yield biologically significant amounts of renin (Von Lutterotti et al., 1994). The recent discovery byClausmeyer et al. (2000) that an alternative renin transcript is stimulated in the infarcted heart, together with our evidence in ischemic sympathetic nerve terminals, strengthens the notion that, independently of its isoform, cardiac renin may have a relevant role in myocardial ischemia.
In conclusion, we have uncovered the presence of renin in sympathetic nerve terminals isolated from the human heart and have demonstrated its marked and rapid activation in ischemic conditions. This suggests a novel autocrine/paracrine mechanism by which Ang II, formed at adrenergic endings in myocardial ischemia, elicits carrier-mediated NE release by activating adjacent AT1R coupled to neuronal NHE.
Acknowledgments
We thank our colleagues Drs. Daniel F. Catanzaro and Randi B. Silver for helpful suggestions and criticism, and Dr. B. Simoneau for donating the renin inhibitor, BILA 2157BS. We also gratefully acknowledge the help of the surgical and nursing staff of the Department of Cardiothoracic Surgery, New York Presbyterian Weill-Cornell Medical Center, in providing us with surgical specimens of human right atrium.
Footnotes
- Abbreviations:
- Ang II
- angiotensin II
- NE
- norepinephrine
- AT1
- angiotensin II type 1
- AT1R
- AT1 receptors
- Ang I
- angiotensin I
- ACE
- angiotensin-converting enzyme
- KHS
- Krebs-Henseleit solution
- HBS
- HEPES-buffered saline solution
- TBS
- Tris-buffered saline
- DMI
- desipramine hydrochloride
- EIPA
- 5-(N-ethyl-N-isopropyl)-amiloride
- EXP 3174
- 2-N-butyl-4-chloro-1-[2′-(1H-tetrazol-5-yl)biphenyl-4-yl]methyl]imidazole-5-carboxylic acid
- NHE
- Na+/H+ exchanger
- ANOVA
- analysis of variance
- OD
- optical density
- Received October 4, 2001.
- Accepted April 29, 2002.
- The American Society for Pharmacology and Experimental Therapeutics