Mechanical Transduction of Nitric Oxide Synthesis in the Beating Heart
Abstract NO alters contractile and relaxant properties of the heart. However, it is not known whether changes in ventricular loading conditions affect cardiac NO synthesis. To understand this potential contractile-relaxant autoregulatory mechanism, production of cardiac NO in response to mechanical stimuli was measured in vivo using a porphyrinic sensor placed in the left ventricular myocardium. The beating rabbit heart exhibited cyclic changes in [NO], peaking at 2.7±0.1 μmol/L near the endocardium and 0.93±0.20 μmol/L in the midventricular myocardium (concentrations were 15±4% lower in the rat heart). In the present study, we demonstrate for the first time that increasing or decreasing ventricular preload in vivo is followed by parallel changes in [NO], which may represent a novel autoregulatory mechanism to adjust cardiac performance or perfusion on a beat-to-beat basis. To quantify the relationship between applied force and NO synthesis, intermittent compressive or distending forces applied to ex vivo nonbeating hearts were shown to cause bursts of NO synthesis, with peak [NO] linearly related to ventricular transmural pressure. Experiments in which denuding cardiac endothelial and endocardial cells abrogated the NO signal indicate that these cells transduce mechanical stimulation into NO production in the heart. Taken together, these studies may help explain load-dependent relaxation, cardiac memory for mechanical events of preceding beats, diseases associated with myocardial distension, autoregulation of myocardial perfusion, and protection from thrombosis in the turbulent flow environment within the beating heart.
The beating heart responds rapidly to local changes in loading conditions, with increased cardiac output observed after simple volume infusion.1 2 Even when denervated after cardiac transplantation, the heart maintains its ability to autoregulate cardiac performance in response to increased venous return. Although several mechanisms underlying increased cardiac contractility in response to passive stretch have been identified,3 investigators have posited the existence of rapidly acting autoregulatory mechanisms to explain certain aspects of the beat-to-beat regulation of cardiac performance.4 In particular, the mechanism(s) underlying cardiac memory for mechanical events of preceding beats and load-dependent relaxation is incompletely understood.5
NO may be produced within the heart by either constitutive or inducible NO synthase.6 7 8 9 Recent studies have shown that endogenous NO agonists, exogenous NO donors, or increasing intracellular cGMP will hasten myocardial relaxation (positive lusitropy).10 11 In addition to their relaxation-hastening effects, NO and cGMP appear to directly depress ventricular contractility (negative inotropy).12 13
Conversely, the effects of changes in physiological load on instantaneous cardiac NO synthesis remain unknown. To understand this potentially important cardiac autoregulatory mechanism that may modulate both myocardial lusitropy, inotropy, and flow on a beat-to-beat basis, we considered whether altering loading conditions could modulate endogenous cardiac NO synthesis. In the present study, we demonstrate for the first time that altering ventricular filling in beating hearts in vivo or altering mechanical force on ex vivo hearts is followed by a parallel increase or decrease in cardiac NO synthesis. This appears to represent a novel autoregulatory mechanism that may be relevant to cardiac physiology as well as to clinical conditions associated with pathological myocardial distension.
Materials and Methods
[NO] was measured using a catheter-protected porphyrinic microsensor prepared in a previously described manner.14 15 16 Two techniques for measuring [NO], DPV and CA, were performed with a PAR model 273 voltammetric analyzer interfaced with an IBM 80486 computer with Lab-View software, which sampled at 4 kHz and amplified and recorded the analytical signal. DPV was used to measure the basal [NO]. Briefly, in the DPV method, current versus potential curves are generated as current is plotted against a potential (range between 0.40 and 0.70 V versus SSCE). The DPV peak current at the peak potential characteristic for NO oxidation (0.63 V) was found to be directly proportional to the [NO] in the immediate vicinity of the sensor. CA, fixed at the peak potential for the oxidation of NO versus SSCE, was used for fast (resolution time, 0.1 to 1 ms) and continuous measurement of the changes of [NO] from its basal level with time. Both techniques were performed using the three-electrode mode: a catheter-protected porphyrinic microsensor working electrode, a platinum wire counter electrode, and a reference SSCE.
The volume sampled is approximately equal to the volume of the sensor (10−10 to 10−12 L). Therefore, the [NO] measured was a local or surface concentration (not a bulk or global concentration). The porphyrinic microsensor had a response time of 0.1 ms at micromolar [NO] and 10 ms at the detection limit of 1 nmol/L. Linear calibration curves were constructed for each sensor from 5×10−9 to 2×10−5 mol/L NO, before and after in vivo or in vitro measurements, with aliquots of saturated NO prepared as described previously.17
Measurement of NO amidst the dynamic in vivo conditions of cyclic breathing and heart beating is a challenging task. In order to overcome these potential interferences and to record a reliable NO signal, we had to modify the porphyrinic sensor from our previously published16 in vivo design in two ways. First, the active sensor tip was shortened to 50 to 60 μm from the previously described 3 to 5 mm.16 Also, the truncated needle from which the active sensor tip emerges was cut 50 to 60 μm shorter than its protective catheter so that the tip of the sensor was completely recessed within the ventilated catheter tip, rather than protruding from an unventilated catheter tip as previously described.16 The completed prototype appeared very similar to a previously published illustration15 ; it was tested under different experimental conditions to ensure that the sensor did not generate an analytic signal when subjected to conditions of fluid pressure, mechanical deformation of the sensor tip, or intrinsic electrical activity within the heart.
New Zealand White rabbits (4 kg) or Wistar-Kyoto rats (300 g) were anesthetized (50 mg/kg ketamine and 5 mg/kg xylazine for rabbits; 100 mg/kg ketamine and 10 mg/kg xylazine for rats), intubated, and ventilated with room air using a Harvard small animal ventilator (tidal volume of 25 mL and rate of 70 breaths/min for rabbits; tidal volume of 2.5 mL and rate of 100 breaths/min for rats). After a median sternotomy was performed, cardiac [NO] was measured as follows: To implant the porphyrinic NO sensor, ventricular tissue was pierced with a standard 0.8-mm-diameter angiocatheter needle (clad with its catheter with two 100-μm ventilation holes near the tip). The catheter/needle unit was advanced to a desired place in the heart. The position of the catheter was secured, and the placement needle was removed and quickly replaced with a porphyrinic NO sensor mounted in a truncated needle. Confirmation of myocardial localization of the active tip of the sensor was by postmortem sectioning of the heart.
Cardiac index was calculated from cardiac output measurements obtained using a Doppler flow probe (Transonics) positioned at the aortic root. A 5F micromanometer-tipped catheter (Millar Instruments) was inserted into the left ventricle via the right carotid artery under constant pressure monitoring. This was calibrated using a transducer control unit (TC-510, Millar Instruments) and zeroed to pressure at the level of the left ventricle. The left ventricular pressure signal was sampled at 4 kHz and amplified and recorded with Lab-View software (IBM computer). In certain experiments, to increase preload, saline was given as a rapid intravenous bolus (10 mL, 37°C), and myocardial NO levels were recorded in the beating heart. After several minutes, when the heart had returned to baseline conditions, preload was reduced by snaring the venae cavae for 5 to 8 seconds. For certain experiments, L-NMMA (Sigma Chemical Co) was administered intravenously or into the proximal left anterior descending artery at the indicated doses.
Measurement of NO Release and Ca2+ Flux From Isolated Cells
Endocardial cells were mechanically removed from the surface of the left ventricle. Coronary vascular endothelial cells were flushed from the coronary vasculature with lactated Ringer’s solution (Baxter) after being loosened with collagenase (Sigma) applied in a brief pulse down the coronary arteries.18 These two primary isolates were plated on round coverslips maintained in HEPES-buffered saline (Sigma) at 37°C for later experiments. To measure [NO], an L-shaped thermally sharpened14 porphyrinic microsensor (diameter, 2 to 3 μm; length, 5 to 7 μm) was positioned parallel to the cell surface within a distance of 2 to 3 μm from the membrane surface, using a stereotactic micromanipulator under direct visual guidance. Mechanical force was generated by a pulse of saline solution injected from a computer-controlled nanoliter injector positioned at a fixed vertical distance from the cell surface. Applied force was measured with a piezoelectric microtransducer.
Ca2+ was measured using a fura 2-AM fluorescent method as described previously.19 The cells were loaded with 0.5 μmol/L of the indicator fura 2-AM (Sigma) in physiological salt solution for 25 minutes. Typically, [Ca2+]i images were calculated from an average of two successive background-subtracted video frames recorded after 360- and 380-nm excitation.
Ex Vivo Measurement of [NO] in the Nonbeating Heart in Response to Applied Compression and Endothelial Denudation
A heart was rapidly excised from a 2.5-kg male New Zealand White rabbit after pentobarbital anesthesia and placed between two Lucite plates connected to a force transduction system, which was immersed in Hanks’ balanced salt solution without phenol red (Sigma) at 37°C. [NO] was measured, with the intermittent application of mechanical compression, using a catheter-protected porphyrinic microsensor placed deeply between septal trabeculae and CA recording (at 0.63 V) using a PAR model 264A voltammetric analyzer. For certain experiments, cardiac endothelial cells were denuded before [NO] measurement by giving a brief (<20-s) pulse of the detergent Triton X-100 (Sigma) down the coronary arteries, followed by a lactated Ringer’s flush. This treatment has been shown to denude coronary microvascular endothelial cells, leaving adjacent cells (including myocytes) intact.20 Histological confirmation of the effects of the Triton X-100 pulse were obtained by fixing the heart in 10% formalin, embedding in paraffin, and performing thin sectioning followed by hematoxylin and eosin staining.
Measurement of NO Release in Response to Stretch
Male Sprague-Dawley rats (250 g) were euthanized and heparinized intravenously (300 U heparin, Sigma), followed by rapid cardiectomy. Hearts were incised, and a catheter-protected porphyrinic microsensor was embedded deeply between septal trabeculae in an orientation tangential to the applied force and immersed in 100 mL of lactated Ringer’s solution (Baxter) at 37°C. The edges of the opened left ventricle were secured with serrated arms attached to a strain-gauge transducer for application and measurement of applied stretch. The stretch was applied and released at the time points indicated. In the last 60 s of each 120-s interval, a DPV voltammogram (current versus potential) was recorded at a scan rate of 5 mV/s from 0.40 to 0.70 V. Peak current was observed at 0.63 V, which is the characteristic potential for NO oxidation on the sensor.9 14 15
Release of [NO] in response to circumferential stretch was measured by obtaining rat hearts as described above, inserting a cannula into the cross-clamped aortic root, and flushing the coronary vasculature free of blood using an infusion of lactated Ringer’s solution (1.6 mL/min controlled with a roller pump). A small incision was made in the left atrium, and a balloon-tipped catheter was introduced into the left ventricle through the mitral valve. The catheter was connected to a pressure manometer attached to a valved balloon inflation device, which enabled left ventricular end-diastolic pressures to be maintained and monitored at desired levels. The coronary arteries were perfused with lactated Ringer’s solution under constant flow conditions (1.6 mL/min controlled with a roller pump); [NO] was measured in the coronary sinus effluent by DPV using a porphyrinic sensor without the protective catheter.
Ca2+ Dependence of Load-Dependent Cardiac [NO] Release
Rat hearts were prepared for ex vivo measurement of [NO] release in response to circumferential stretch (as described above). Intracavitary pressure was varied by inflating the left ventricular balloon. After baseline measurements were obtained, a Ca2+ ionophore (A23187, 47.5 μmol, Sigma) was introduced into the coronary perfusate. In separate experiments, to determine the effects of Ca2+ chelation on load-dependent cardiac NO synthesis, intracellular and extracellular Ca2+ chelation was accomplished by adding EGTA (200 μmol/L, Sigma) and MAPTAM (200 μmol/L, Molecular Probes)21 to the coronary perfusate, after which the load (intracavitary pressure) was increased by inflating the left ventricular balloon.
Results are expressed as mean±SEM. In each set of experiments, n is the number of animals studied. Statistical analysis was performed using an unpaired Student’s t test or by ANOVA followed by Scheffé’s F test. Means were considered significantly different at P<.05.
By use of a porphyrinic microsensor to detect rapid fluctuations in [NO],16 in vivo measurements were performed in a beating rabbit or rat heart. Although this sensor has been used for both in vitro and in vivo measurements in prior published studies,9 14 16 several characteristics of the sensor were tested to ensure that recordings in the beating heart would reflect authentic [NO] rather than artifacts due to intramyocardial placement and consequent piezoelectric artifacts due to pressure changes or mechanical deformation of the sensor tip.
To demonstrate that the sensor does not respond with an electrical signal to changes in pressure, recordings were made in an electrolyte-filled sealed chamber, within which pressure was rapidly altered. As can be seen in Fig 1a⇓, alterations in pressure have no measurable effect on the amperometric current generated by the sensor, which was immersed in the confined electrolyte solution.
To rule out potential piezoelectric electrical interference due to mechanical deformation of the sensor tip, a micromanipulator was used to deform the sensor tip at a physiologically relevant frequency (3 Hz). This generated only small electrical noises, six times smaller than the signal recorded by the sensor at low (50 nmol/L) [NO] (Fig 1b⇑).
To eliminate the possibility that the sensor might respond to the electrical current generated within the intrinsic cardiac conduction system or within the depolarizing myocardium, the porphyrinic sensor was used as the exploring electrode connected to a standard ECG. As is apparent from Fig 1c⇑, when used in this manner, the porphyrinic sensor is barely able to detect an electrical signal. In these experiments defining the analytic response of the sensor to potential piezoelectric or in vivo electrical current interferants, the measured peak [NO] signals were in most cases at least 30 to 100 times larger and temporally shifted from the conservatively estimated background noises.
An anesthetized rabbit was used to measure local fluctuations in [NO] in the beating heart, with the sensor placed (via apical puncture) in the apical left ventricular endocardium or myocardium, as is shown schematically in Fig 2a⇓. Controlling the depth of penetration and postmortem confirmation of the location of the sensor tip were more easily accomplished in rabbit than in rat hearts, and relatively accurate data were produced in relation to the distribution of [NO] in the wall of the rabbit heart. However, this method was not precise enough for measurement of [NO] distribution in the relatively thin wall of the rat heart. Therefore, most of the in vivo [NO] measurements that required higher precision in relation to the localization of the sensor were performed in the rabbit heart. Rapid changes in cardiac [NO] related to the cardiac cycle were observed in both rabbit and rat hearts. In the rabbit heart, each cardiac cycle (period, 326±16 ms) began and ended with an intercycle [NO] of 0.67±0.16 μmol/L (n=20) near the endocardium (Fig 2b⇓, upper tracing). During early systole, [NO] reached a basal level of 0.62±0.05 μmol/L (n=20), followed by a slow increase (20±1 nmol · L−1 · ms−1) to a semiplateau. Early diastolic filling was accompanied by a brisk rise of [NO] (46±3 nmol · L−1 · ms−1), with a peak diastolic [NO] of 2.7±0.1 μmol/L (n=20) that was attained at 239±17 ms into the cardiac cycle. After this peak, there was a sharp decay (−30±3 nmol · L−1 · ms−1, n=20) to the intercycle [NO].
A similar cyclic fluctuation of [NO] was observed in the myocardium 1 to 2 mm beneath the endocardium (Fig 2b⇑, middle tracing). However, the intercycle [NO] at this depth within the myocardium was smaller (0.26±0.08 μmol/L, n=20), with a lower peak diastolic [NO] of 0.93±0.2 μmol/L (n=20) and a slower decay rate (−5±1 nmol · L−1 · ms−1, n=20). A similar pattern of [NO] fluctuation was observed in the rat heart with the sensor placed in the midventricular myocardium (Fig 2c⇑). The intercycle [NO] was 0.22±0.03 μmol/L (n=20), with a lower peak diastolic [NO] of 0.79±0.04 μmol/L (n=20). Both of these concentrations were 15±4% (P<.05) lower than that observed in the myocardium of the rabbit. The [NO] signal recorded by sensors in both rabbit and rat hearts disappeared when monitored at a potential 0.40 V (which is 230 mV below the peak potential of NO oxidation), clearly indicating that the signal measured at 0.63 V is due to NO, not mechanical noises, which are applied independent of potential (Fig 2b⇑, lower tracing). To demonstrate the relationship between intracavitary left ventricular pressure, the ECG signal, and instantaneous [NO], synchronous recordings of each of these were performed as described in “Materials and Methods.” Fig 3⇓ represents a simultaneous display of these three measured parameters.
To confirm that these measurements reflected authentic [NO], L-NMMA, an inhibitor of NO synthase, was administered intravenously (Fig 4a⇓). Both basal and peak [NO] were gradually (over several beats) but significantly reduced; administration of L-NMMA (40 mg/kg) diminished the peak [NO] in the endocardium to 0.95±0.07 μmol/L (n=20). In the underlying myocardium, peak [NO] was also reduced (to 0.33±0.05 μmol/L, n=20). Intravenous infusion of L-NMMA produced a consistent rise in mean arterial pressure (from 76±2 to 89±3 mm Hg) and duration of the cardiac cycle (from 326±16 to 363±18 ms) and a consistent fall in the cardiac index (from 123±7 to 109±5 mL · min−1 · kg−1), left ventricular end-diastolic pressure (from 11±2 to 8±1 mm Hg), and left ventricular systolic pressure (from 92±3 to 77±3 mm Hg).
In order to separate the vasoconstrictive effect of L-NMMA in the peripheral vasculature from its effects on the heart itself, a separate experiment was performed: L-NMMA (20 μL, 200 μmol/L) was directly injected into the proximal left anterior descending coronary artery of the rabbit. The pattern of NO release as well as the peak [NO] was identical, within experimental error, to that observed after the intravenous infusion of L-NMMA: inhibition of NO production (from 1.57±0.06 to 0.90±0.07 μmol/L), decrease of cardiac index (from 123±7 to 108±4 mL · min−1 · kg−1), decrease of left ventricular end-diastolic pressure (from 11±1 to 8±1 mm Hg), and decrease of left ventricular systolic pressure (from 92±3 to 75±3 mm Hg) were observed. However, in contrast to the intravenous infusion of L-NMMA, mean arterial pressure in the first 30 s decreased from 76±3 to 65±2 mm Hg. These experiments suggest that L-NMMA, by inhibiting the release of NO, decreased the efficiency of the heart.
When preload was increased by a rapid intravenous bolus of physiological saline, there was a gradual but significant increase in both peak and basal [NO] in the beating heart to 3.9±0.3 (n=7) and 0.80±0.12 μmol/L (n=7), respectively (Fig 4b⇑). When ventricular filling was reduced by temporary ligation of the venae cavae, the [NO] decreased, also gradually, over several beats to 0.81±0.05 μmol/L for peak [NO] (n=7) and 0.10±0.05 μmol/L for basal [NO] (n=7) (Fig 4c⇑).
To determine which of the cell types within the heart that express constitutive NO synthase activity might be responsible for load-dependent cardiac [NO], additional studies were performed. Endothelial cells or endocardial cells produced peak [NO] of 0.97±0.25 μmol/L (n=9) and 0.50±0.18 μmol/L (n=9), respectively, after application of an external mechanical force of 15±3 dyne/cm2 normal to the cell membrane (as measured by a porphyrinic microsensor placed 2 to 3 μm from the cell membrane) (Fig 5a⇓ and 5b⇓, upper tracings). In these experiments, the [NO] decreased exponentially with increasing distance, becoming undetectable at a distance of 35 to 50 μm from the studied cell. Mechanically transduced [NO] release also decreased (50±18%) in the presence of L-NMMA (Fig 5b⇓, lower tracing). In experiments using fura 2–loaded endothelial cells to detect increases in [Ca2+]i, a rapid increase in [Ca2+]i was observed to precede the [NO] pulse (Fig 5a⇓, lower tracing). When cardiac endothelial cells were denuded in an ex vivo heart using a brief pulse of Triton X-100 (a procedure that leaves myocytes unscathed) (see Fig 6d⇓),20 cardiac NO synthesis in response to applied force was greatly diminished (compare Fig 6a⇓ with 6b). When a similar ex vivo heart was examined without endothelial denudation, L-NMMA significantly diminished the [NO] response to applied force, confirming the measurements to be authentic [NO] (compare Fig 6a⇓ with 6c). Taken together, these experiments strongly suggest that endothelial and endocardial cells within the heart are the transducing cells responsible for load-dependent cardiac [NO].
In order to quantify the relationship between mechanically applied forces and cardiac [NO], ex vivo studies were performed. Initial observations were made in freshly excised nonbeating rabbit hearts subjected to mechanical deformation, with local [NO] measured with a sensor placed deep between septal trabeculae (Fig 6a⇑). Each incremental application of force, normal to the exterior of the heart (compressive force mimicking that of systole), was followed by a CA signal after 50 ms, indicating a burst of NO. A linear relationship between peak [NO] and applied force was observed, with an increase of peak [NO] (from 0.14±0.02 to 1.10±0.08 μmol/L) (n=7) over the range of applied force (from 4.70×10−2 to 28.2×10−2 N/cm2). Also, cardiac NO production was significantly inhibited by L-NMMA (Fig 6b⇑). In the presence of 100 μmol/L L-NMMA, [NO] decreased by 80±3% to 70±5% (n=7) over the range of applied forces from 4.70×10−2 to 28.2×10−2 N/cm2. As a control, cardiac endothelial cells were denuded,20 and subsequent cardiac NO bursts in response to applied forces were greatly diminished (Fig 6c⇑). The efficiency of the denudation process was tested with acetylcholine. Acetylcholine (1 μmol/L)–stimulated NO release decreased by 85±15%.
To characterize the cardiac NO synthesis that occurs in response to the circumferential stretch (mimicking forces during diastole), a balloon was inserted into the left ventricle through the left atrium/mitral valve. Left ventricular transmural pressure was varied by inflating the balloon, and [NO] was measured directly in the coronary effluent (to estimate the portion of [NO] that was released into the blood stream) with the porphyrinic microsensor. During constant flow perfusion with lactated Ringer’s solution (1.6 mL/min), circumferential stretch was accompanied by an increase in [NO] (Fig 7a⇓ and 7b⇓). This increase in [NO] was linearly related to transmural pressure over the range of pressures studied (40±10 nmol/L of NO [n=5] at 0 mm Hg to 167±20 nmol/L [n=5] at 60 mm Hg). Spiking the perfusate with the [Ca2+]o chelator EGTA and the membrane-permeable [Ca2+]i chelator MAPTAM21 caused a decrease of [NO], both at baseline and after application of transmural pressure (Fig 7c⇓). In experiments designed to facilitate [Ca2+]o entry, with ventricular transmural pressure maintained at 0 mm Hg, addition of the Ca2+ ionophore A23187 to the perfusate increased the [NO] more than that observed after the application of maximal circumferential stretch (compare the rightmost tracings of Fig 7a⇓ and 7c⇓), suggesting maximal stimulation of Ca2+-dependent NO synthesis. Taken together, these results suggest a role for [Ca2+]i in the mechanical transduction of load-dependent cardiac NO synthesis.
There have been a growing number of recent publications addressing the influence of NO on the contractile (inotropic) and relaxation (lusitropic) properties of cardiac myocytes and the heart.7 10 12 22 Endothelial shear stress has been shown to increase [NO] by stimulating a constitutive NO synthase.18 23 However, to date, there is no evidence that indicates that altering cardiac loading conditions independent of coronary blood flow affects NO synthesis in vivo. The present study demonstrates for the first time that NO is released in pulsatile fashion from the beating heart and that its synthesis is directly related to ventricular loading conditions in vivo. These results support the existence of a novel cardiac autoregulatory mechanism that may facilitate ventricular filling in times of increased demand. Furthermore, these data may help to explain certain aspects of the beat-to-beat regulation of cardiac performance and flow and provide insights into the pathophysiology of diseases associated with increased myocardial distension, such as valvular heart disease or heart failure.
If the heart were to function as a purely systolic (extrusion) pump, without the ability to transduce changes in preload into parallel changes in [NO], then within certain physiological limits, abrupt changes in preload should produce abrupt changes in cardiac output. In the present study, we offer an explanation for why this is not so: we demonstrate that abruptly increasing or decreasing ventricular preload in vivo is gradually (not abruptly) followed by parallel changes in [NO] over several cardiac cycles (Fig 4⇑). This model may also provide an explanation for several deviations observed from the Frank-Starling model (a relatively static systolic pump model, which does not reflect the dynamic interplay between systole and diastole).1 2 24
There is also likely to be an additional beneficial reason for increased cardiac [NO] under conditions of increased mechanical stimulation. The turbulent blood flow within the beating heart provides a potent stimulus for platelet adhesion and aggregation.25 Since both platelet adhesion to endocardial cells and platelet aggregation are inhibited by NO,26 the unusually high load-dependent local [NO] (Fig 2⇑) observed near the endocardium is likely to be of considerable importance in inhibiting local platelet adhesion and aggregation in the highly turbulent blood flow in the heart. It is possible that a lack of this effect in poorly contracting hearts and artificial hearts and on the surface of prosthetic cardiac valves may contribute to the prothrombotic diathesis observed under these conditions.
Although different cell types within the heart express constitutive NO synthase activity, cardiac microvascular endothelial cells appear to be the predominant source of load-dependent cardiac NO synthesis. They are abundant within the heart and are located close to all cardiac myocytes. The endocardium itself also contributes significantly to load-dependent cardiac NO synthesis. However, rapid effects due to endocardial NO in the heart wall are only likely to occur in a narrow zone of subjacent myocardium because of diffusion constraints.14 27 Therefore, most of the endocardium-derived NO is rapidly dissipated into the intracavitary flow of ventricular blood, where it inhibits local platelet aggregation; this situation may be relevant to the possibility that hemoglobin may serve as a systemic carrier for NO.28
Myocytes also possess a constitutive Ca2+-dependent NO synthase,7 although in the present experiments, they do not represent a significant source of NO synthesis, at least in response to an applied load. These data are in good agreement with previous data reported for spontaneously beating neonatal ventricular myocytes, which do not release detectable [NO] in culture in the absence of adrenergic stimulation.29 These data are not surprising, given that the large [Ca2+]i transients during excitation and contraction are likely to dwarf increases in [Ca2+]o entry via mechanically gated cation channels.30 Furthermore, stimulation of the constitutive NO synthase in intact hearts causes lusitropic effects that are not observed in similarly stimulated ventricular papillary muscle preparations,10 suggesting a role for cardiac endothelial cells in NO-dependent effects on cardiac muscle.
There are likely to be several physiological consequences of mechanical transduction of cardiac NO synthesis. Cells within the heart are subjected to tremendous mechanical deformation during filling and beating. In addition, there is substantial evidence that NO directly affects mechanical properties of cardiac myocytes. NO acts directly on myocytes via increases in cGMP to facilitate relaxation and to mediate an acetylcholine-stimulated decrease in contractility (negative inotropy).13 Coincidentally, previously published data clearly show that cGMP concentration in myocytes fluctuates during each cardiac cycle, reaching a sustained maximum just after the T wave.31 This is in excellent agreement with the present study, which observes the peak [NO] just after the T wave (Fig 3⇑).
The present study is the first to identify load-dependent mechanical transduction of NO synthesis in the beating heart in vivo and to identify phasic changes in [NO] that occur during the cardiac cycle. Because NO and cGMP have negative inotropic and positive lusitropic actions on the heart, these studies suggest that mechanical transduction of cardiac NO synthesis and its short-term accumulation may provide an important autoregulatory mechanism for the modulation of myocardial contractility and relaxation in response to abrupt changes in preload.
Selected Abbreviations and Acronyms
|DPV||=||differential pulse voltametry|
|MAPTAM||=||1,2-bis-methyl-aminophenoxyethane- N,N,N′,N′-tetraacetoxymethyl acetate|
|SSCE||=||silver/silver chloride electrode|
This study was supported in part by grants from the National Aeronautics and Space Administration, the American Heart Association, and the Public Health Service (HL-55397). We thank Y. Naka and S.T. Pinsky for expert technical assistance with these studies. E. Kubaszewski is on sabbatical leave from Poznan University of Technology.
- Received November 18, 1996.
- Accepted June 11, 1997.
- © 1997 American Heart Association, Inc.
Starling EH. Linacre Lecture on the Law of the Heart. London, UK: Longmans, Green and Co; 1918.
Parmley WW, Chuck L. Length-dependent changes in myocardial contractile state. Am J Physiol. 1973;224:1195-1199.
Slinker BK. Cardiac cycle length modulates cardiovascular regulation that is dependent on previous beat contraction history. Circ Res. 1991;69:2-11.
Zile MR, Gaasch WH. Load-dependent left ventricular relaxation in conscious dogs. Am J Physiol. 1991;261:H691-H699.
Balligand JL, Kelly RA, Marsden PA, Smith TW, Michel T. Control of cardiac muscle cell function by an endogenous nitric oxide signaling system. Proc Natl Acad Sci U S A. 1993;90:347-351.
Pinsky DJ, Oz MC, Koga S, Taha Z, Broekman MJ, Marcus AJ, Liao H, Naka Y, Brett J, Malinski T, Nowygrod R, Stern DM. Cardiac preservation is enhanced in a heterotopic rat transplant model by supplementing the nitric oxide pathway. J Clin Invest. 1994;93:2291-2297.
Grocott-Mason R, Anning P, Evans H, Lewis MJ, Shah AM. Modulation of left ventricular relaxation in isolated ejecting heart by endogenous nitric oxide. Am J Physiol. 1994;267:H1804-H1813.
Shah AM, Spurgeon HA, Sollott SJ, Talo A, Lakatta EG. 8-Bromo-cGMP reduces the myofilament response to Ca2+ in intact cardiac myocytes. Circ Res. 1994;74:970-978.
Finkel MS, Oddis CV, Jacob TD, Watkins SC, Hattler BG, Simmons RL. Negative inotropic effects of cytokines on the heart mediated by nitric oxide. Science. 1992;257:387-389.
Malinski T, Czuchajowski L. Nitric oxide measurements by electrochemical methods. In: Feelisch M, Stamler JS, eds. Methods of Nitric Oxide Research. Chichester, UK: John Wiley & Sons Ltd; 1996:319-339.
Jia L, Furchgott RF. Inhibition by sulfhydryl compounds of vascular relaxation induced by nitric oxide and endothelium-derived relaxing factor. J Pharmacol Exp Ther. 1993;267:371-378.
Buga GM, Gold ME, Fukuto JM, Ignarro LJ. Shear stress–induced release of nitric oxide from endothelial cells grown on beads. Hypertension. 1991;17:187-193.
Blatter LA, Taha Z, Mesaros S, Shacklock PS, Wier WG, Malinski T. Simultaneous measurements of Ca2+ and nitric oxide in bradykinin-stimulated vascular endothelial cells. Circ Res. 1995;76:922-924.
Nava E, Noll G, Lüscher TF. Increased activity of constitutive nitric oxide synthase in cardiac endothelium in spontaneous hypertension. Circulation. 1995;91:2310-2313.
Poch E, Botey A, Gaya J, Cases A, Rivera F, Revert L. Intracellular calcium mobilization and activation of the Na+/H+ exchanger in platelets. Biochem J. 1993;290:617-622.
Shah AM, Lewis MJ. Modulation of myocardial contraction by endocardial and coronary vascular endothelium. Trends Cardiovasc Med. 1993;3:98-103.
Kanai AJ, Strauss HC, Truskey GA, Crews AL, Grunfeld S, Malinski T. Shear stress induces ATP-independent transient nitric oxide release from vascular endothelial cells, measured directly with a porphyrinic microsensor Circ Res. 1995;77:284-293.
Lancaster JR. Simulation of the diffusion and reaction of endogenously produced nitric oxide. Proc Natl Acad Sci U S A. 1994;91:8137-8141.
Kanai AJ, Mesaros S, Finkel MS, Oddis CV, Strauss HC, Malinski T. Nitric oxide release measured directly with a porphyrinic microsensor reveals adrenergic control of constitutive nitric oxide synthase in cardiac myocytes. Circulation. 1995;92(suppl I):I-563. Abstract.