- Email: [email protected]
Myocardial Oxygen Consumption Modulates Adenosine Formation by Canine Right Ventricle in Absence of Hypoxia
J Mol Cell Cardiol 32, 345–354 (2000) doi:10.1006/jmcc.1999.1077, available online at http://www.idealibrary.com on
Myocardial Oxygen Consumption Modulates Adenosine Formation by Canine Right Ventricle in Absence of Hypoxia Xiaoming Bian, Min Fu, Robert T. Mallet, Rolf Bu¨nger∗ and H. Fred Downey Department of Integrative Physiology, University of North Texas Health Science Center at Fort Worth, Fort Worth, TX 76107-2699, USA, ∗Department of Physiology, Uniformed Services University of the Health Sciences, Bethesda, MD 20814-4799, USA (Received 24 September 1999, accepted 5 November 1999) X. B, M. F, R. T. M, R. B¨ H. F. D. Myocardial Oxygen Consumption Modulates Adenosine Formation by Canine Right Ventricle in Absence of Hypoxia. Journal of Molecular and Cellular Cardiology (2000) 32, 345–354. Myocardial adenosine formation varies with myocardial oxygen consumption (MVO2), but whether concurrent hypoxia is required for adenosine formation is uncertain. Changes in right coronary (RC) perfusion pressure (RCP) produce directionally similar alterations in right ventricular (RV) MVO2 and in RC venous P2 (Pv2), an index of myocardial P2. RCP was varied in 10 anesthetized, open chest dogs to determine if, under these conditions, RV formation of adenosine would increase with MVO2 in absence of myocardial hypoxia. Dialysis probes were implanted in the mid myocardium of RV free wall for collecting dialysate samples for HPLC analyses to estimate interstitial adenosine and other purines. Coronary venous blood was sampled from a superficial vein draining the RC artery (RCA) perfusion territory. At 115±3 mmHg baseline RCP, RC blood flow (RCBF)= 0.51±0.04 ml/min/g, MVO2=4.6±0.5 ml/min/100 g, Pv2=34±1.5 mmHg, and dialysate adenosine= 0.27±0.03 l. When RCP was lowered to 61±1 mmHg by adjusting an occluder on the proximal RCA, RCBF decreased to 0.36±0.03 ml/min/g, MVO2 fell to 3.7±0.4 ml/min/100 g, lactate uptake remained positive, Pv2 fell to 30±1.7 mmHg, and dialysate adenosine decreased to 0.20±0.03 l. Reactive hyperemia of 1.25±0.13 ml/ min/g was observed when the RCA constriction was released, although dialysate adenosine had fallen. When RCP was elevated to 164±2 mmHg by inflating a balloon catheter in the descending aorta, RCBF increased to 0.70±0.06 ml/min/g, MVO2 increased to 5.8±1.0 ml/min/100 g, Pv2 rose to 39±2.3 mmHg, and dialysate adenosine increased to 0.33±0.04 l. These data indicate that (1) RV oxygen demand varies with RCP; (2) if RV ischemia is absent, myocardial adenosine formation is modulated by MVO2, with no requirement for hypoxia; (3) pressure–flow autoregulation is relatively ineffective in the RC circulation, where adenosine does not mediate 2000 Academic Press and may even blunt autoregulation. K W: Purine metabolism; Microdialysis; Oxygen consumption; Adenosine; Right ventricular myocardium; Coronary flow.
Introduction Since the adenosine hypothesis of coronary flow regulation was proposed in 1963,1,2 numerous studies have produced evidence that adenosine plays
an important role in the regulation of coronary blood flow.3 In many studies, a decreased O2 supply was the primary stimulus for adenosine production.4 However, studies of increased myocardial O2 demand under conditions of ample O2 supply also
Please address all correspondence to: Xiaoming Bian, Department of Integrative Physiology, University of North Texas Health Science Center at Fort Worth, 3500 Camp Bowie Boulevard, Fort Worth, TX 76107-2699, USA.
0022–2828/00/030345+10 $35.00/0
2000 Academic Press
346
X. Bian et al.
demonstrated an increase in adenosine production. These studies increased O2 demand by administration of inotropic agents,5 bilateral stellate ganglion stimulation,6 and cardiac paired pacing.7 However, the increase in adenosine production in all those studies was associated with a decline in coronary venous P2, an index of tissue O2 tension.8 These observations agree with the concept that myocardial hypoxia is a prerequisite for adenosine formation.9 On the other hand, Bu¨nger10 proposed that adenosine formation was regulated by the cytosolic phosphorylation potential as influenced by the rate of myocardial energy expenditure. According to this proposal, myocardial hypoxia is not absolutely required for increased adenosine formation. Instead, any condition that would lower the phosphorylation potential or raise free cytosolic AMP would physiologically produce adenosine, whether or not myocardial hypoxia occurs concomitantly.10,11 Autoregulation of coronary blood flow is remarkably ineffective in right ventricle (RV),12,13 so changes in right coronary (RC) pressure (RCP) alter RC blood flow appreciably. This ineffective RC autoregulation is associated with RCP induced changes in RV myocardial O2 consumption (MVO2). With moderate reductions in RCP, RC O2 demand falls sufficiently that RV ischemia is avoided.14,15 Another consequence of ineffective RC autoregulation is that RC venous P2 and, thus, RV myocardial P2 vary with RCP coincident with parallel changes in RV MVO2.12,14 Consequently, the RV offers a novel model to test the hypothesis that myocardial hypoxia is required for adenosine formation during increased myocardial O2 demand. In this investigation, RCP was varied in anesthetized, open chest dogs to determine if RV production of adenosine would increase with MVO2 in absence of myocardial hypoxia. RV adenosine was estimated by microdialysis. This is the first report on relationships between RV adenosine formation, RV MVO2, and RC flow. Results indicate that hypoxia is not required for adenosine formation.
Methods This investigation was approved by the Institutional Animal Care and Use Committee of the University of North Texas Health Science Center at Fort Worth and was conducted in accordance with the Guide for the Care and Use of Laboratory Animals (Institute of Laboratory Animal Resources, National Research Council, USA, 1996).
Animal preparation Ten adult healthy mongrel dogs of either sex (22–29 kg) were anesthetized with intravenous sodium pentobarbital (30 mg/kg). Additional sodium pentobarbital was given as needed to maintain adequate anesthesia throughout the experiment. The animals were intubated and mechanically ventilated with room air supplemented with oxygen to maintain normal arterial blood gases throughout the experiment. Core body temperature was monitored with a rectal temperature probe and maintained at 37.0–38.0°C with a heating pad. A catheter was inserted into a femoral artery to monitor aortic pressure, and the ipsilateral femoral vein was cannulated for infusion of supplemental fluids and anesthetics. A balloon catheter (20 cc, St Jude Medical) was inserted into the contralateral femoral artery and advanced into the thoracic aorta to obtain samples for arterial blood gas analyses and to increase the RCP by inflating the balloon as required in the protocols. A right thoracotomy was performed in the fourth intercostal space, and the heart was suspended in a pericardial cradle. Figure 1 illustrates the experimental preparation. A non-branching section of the RC artery was dissected free for 1–1.5 cm for placement of a single-crystal Doppler flow probe proximal to an adjustable occluder. To measure RCP distal to the occluder, a Micro-Renathane catheter (0.014 in ID, 0.033 in OD) was inserted into the RC artery through a small right atrial side branch. A Millar catheter-tip pressure transducer was inserted in the right atrial appendage and advanced across the tricuspid valve to measure right ventricular pressure. RV dP/dt was determined by electronic differentiation of the RV pressure signal. Saline-filled catheters were inserted into the main pulmonary artery and right atrium to measure pulmonary arterial and right atrial pressures. A 24-gauge catheter was inserted into a superficial vein draining the area perfused by the RC artery for sampling RC venous blood. Arterial and RC venous samples were collected anaerobically. RV MVO2 and lactate uptake were calculated from the product of RC blood flow times the respective regional arteriovenous differences. RC flow velocity was measured with a Triton Technology model 100 pulsed Doppler flowmeter. Heart rate (HR), aortic pressure (PAO), RCP, pulmonary artery pressure (PPA), RV pressure (PRV), RV dP/dt, right atrial pressure (PRA), and RC flow velocity signals were recorded continuously with a multichannel Grass model 7D polygraph.
Adenosine Formation and MVO2
347
initiation of the experimental protocol. During this time dialysate nucleoside concentrations fell and stabilized at concentrations which were taken as the baseline values. This recovery to basal nucleoside concentrations following implantation of a dialysis probe has been documented by Van Wylen et al.16 The outflows of the dialysis probes were collected in tared 250 ll plastic vials (Fig. 1). Each sample (30 ll) was weighed and diluted with distilled H2O to 70 ll, then immediately frozen in liquid N2 and stored at −80°C until analysis. Experimental protocol
Figure 1 In situ canine heart preparation. A portion of the proximal right coronary artery was dissected for placement of a flow probe and screw-clamp occluder. Two dialysis probes were implanted in the mid myocardium near the center of the right coronary perfusion territory to assess interstitial fluid purines. Catheters were placed in the right atrium, right ventricle, aortic arch, and pulmonary trunk to monitor right atrial (PRA), right ventricular (PRV), aortic (PAO) and pulmonary arterial (PPA) pressures, respectively. Right coronary pressure distal to the occluder was measured with a catheter inserted into the right coronary artery via a small side branch. Right coronary venous samples were collected anaerobically from a superficial vein draining the right coronary perfusion territory. Right coronary perfusion pressure was lowered by partially constricting the right coronary artery with the occluder. A balloon catheter in the thoracic aorta was inflated to increase aortic and, thus, right coronary perfusion pressure.
Cardiac microdialysis Myocardial interstitial adenosine and other purine concentrations were estimated with the cardiac microdialysis technique, as described previously by Van Wylen et al.16 Two microdialysis probes were implanted in the mid myocardium of the RC perfusion territory. The probes were inserted with the aid of an introducer needle, and each probe was gently pulled through the myocardium until the dialysis window rested entirely within the muscle wall. A 23-gauge needle had been previously glued to the inflow port of the dialysis tubing, and the hub of this needle was connected to a gas-tight syringe filled with Krebs–Henseleit buffer aerated with 95% N2 and 5% CO2.16 Buffer was perfused through the dialysis probes at 2 ll/min for the remainder of the protocol with a Harvard model 22 infusion pump. A 60-min stabilization period was allowed before
Beginning 60 min after the microdialysis probes were implanted, three consecutive dialysate samples, each 15 min in duration, were collected from each of the two probes to establish the baseline nucleoside concentrations. At the mid-point of each dialysate sample collection, hemodynamic variables were measured and duplicate arterial and RC venous blood samples were collected. RCP was reduced from baseline (approx. 115 mmHg) to 60 mmHg by constricting the RC artery with the adjustable occluder. After allowing 15 min for stabilization, three 15-min dialysate samples from each probe, hemodynamic readings, and blood samples were taken in the same manner as described for the baseline condition. After the last dialysate sample at 60 mmHg RCP was collected, the RC artery constriction was abruptly released, and the peak RC flow was recorded. After 30 min, RCP was increased to 160 mmHg by inflating the balloon catheter in the descending aorta. Following a 15min stabilization period, three 15-min dialysate samples were collected from each probe, and hemodynamic readings and blood samples were taken in the same manner as described for the baseline condition. At the end of each experiment, India ink was injected through the RC artery catheter to delineate the RC perfusion territory. The heart was excised, and the dyed area was resected and weighed. The internal diameter of the RC artery under the flow velocity probe was measured, and its cross-sectional area was calculated. RC blood flow (ml/min/g) was calculated as the product of RC flow velocity times the cross-sectional area divided by perfused tissue mass. Analytical methods Arterial and venous pH, P2, and P2 were determined by an Instrumentation Laboratory model
348
X. Bian et al.
Synthesis 30 blood gas analyzer. Lactate was measured by a Yellow Springs Instruments model 2300 Stat L-lactate analyzer. Purine nucleosides (adenosine and inosine) in the dialysate were measured by HPLC (Shimadzu model LC-600) as previously described.17 Duplicate samples collected simultaneously from the two probes were averaged.
Statistical analyses All values are expressed as mean±... Values of variables measured during three consecutive 15min periods during each RCP were averaged after repeated measures ANOVA detected no period-related differences among these values. Hemodynamic data, MVO2, Pv2, lactate uptake, and dialysate nucleoside concentrations were analyzed with a repeated measures ANOVA with RCP as the independent variable. Values of P<0.05 were taken to indicate statistical significance.
Results A hemodynamic recording from a typical experiment is shown in Figure 2, and hemodynamic variables are summarized in Table 1. RC blood flow varied with RCP (P<0.05), reflecting poor autoregulation in right coronary circulation. Heart rate, right atrial pressure, and RV systolic pressure, dP/dtmax and dP/dtmin were unaffected by altering RCP and were stable during the three consecutive 15-min periods at each RCP. These data indicate that global RV preload, afterload, contractility, and relaxation were constant throughout the protocol. In Figure 3, RV dialysate adenosine concentration is plotted as a function of RCP for each dog. In all experiments, dialysate adenosine decreased when RCP was reduced, and increased when RCP was elevated from baseline. RV dialysate adenosine, adenosine + inosine, and MVO2 at reduced, baseline, and elevated RCP are presented in Figure 4A. Corresponding values of RC venous P2 (Pv2) and lactate uptake are shown in Figure 4B. When RCP was lowered from 115±3 mmHg baseline to 61±1 mmHg, RV dialysate adenosine concentration fell from 0.27±0.03 l to 0.20±0.03 l (P<0.05). When RCP was elevated from the baseline to 164±2 mmHg, dialysate adenosine increased to 0.33±0.04 l (P<0.05 vs baseline). Similar changes in adenosine + inosine were observed as RCP was varied (Fig. 4A). RV MVO2 increased from 3.7±0.4 ml/min/100 g at 61 mmHg RCP to
5.8±1.0 ml/min/100 g at 164 mmHg RCP (P<0.05). Elevation of RCP in absence of effective pressure–flow autoregulation produced an increase in Pv2 from 30±1.8 mmHg at 61 mmHg RCP to 39±2.3 mmHg at 164 mmHg RCP (P<0.05; Fig. 4B). Over this broad range of RCP, lactate uptake increased from 0.22±0.03 to 0.42±0.06 lmol/ min/g (P<0.05; Fig. 4B), but the ratio of lactate uptake to MVO2 changed insignificantly from 0.069±0.014 to 0.076±0.011 lmol/ml. Thus, varying RCP produced parallel changes in RV MVO2, dialysate adenosine and adenosine + inosine concentrations, RC venous oxygen tension, and RV lactate uptake. In these experiments, RV dialysate adenosine concentration was greatest when RV MVO2 was at its highest value (Fig. 4A) and when RC Pv2 was also elevated (Fig. 4B). RC Pv2 was at its lowest value (Table 1) when RV dialysate adenosine concentration was greatest (Fig. 4A). Because Pv2 is inversely related to intracellular pH,10 this result ruled out intracellular acidosis as a trigger for adenosine formation. RV MVO2 fell as a response to decreased RCP rather than in response to a flow limitation of O2 delivery, since lactate uptake as a function of MVO2 was not compromised. Under these non-ischemic conditions at moderately decreased RCP, RV dialysate adenosine fell with RV MVO2 (Fig. 4A). Figure 5 shows RC blood flow measured at low and baseline RCP and corresponding concentrations of RV dialysate adenosine. Both variables fell significantly when RCP was lowered to 61 mmHg. When the RC artery constriction was abruptly released to restore RCP to baseline, a peak hyperemic response of 1.25±0.13 ml/min/g was observed. This hyperemia exceeded the baseline flow by more than two-fold (P<0.01). The presence of this significant reactive hyperemia demonstrated that the RC vasculature had dilated during the period of decreased RCP, despite the concomitant reduction in dialysate adenosine concentration.
Discussion This investigation tested the hypothesis that myocardial hypoxia is required for RV adenosine formation during RCP induced changes in RV myocardial O2 demand. In this poorly autoregulating model, changes in coronary perfusion pressure produce directionally similar changes in MVO2 and in Pv2, an index of myocardial P2,8 so this model was well suited for testing this hypothesis. The hypothesis was not supported, since
Adenosine Formation and MVO2
349
Figure 2 Typical tracing of hemodynamic variables at different right coronary perfusion pressures. Right coronary artery perfusion pressures (RCP) in the three depicted phases of this experiment were approximately 60, 115 and 160 mmHg (tracing 2). Measured variables included aortic pressure (PAO: tracing 1), right coronary blood flow (RCBF: tracing 3), pulmonary arterial pressure (PPA: tracing 4), right ventricular dP/dt (tracing 5), phasic right ventricular pressure (PRV: tracing 6) and right atrial pressure (PRA: tracing 7). The increase in PAO during the final phase of the experiment reflects inflation of an intra-aortic balloon to increase RCP.
350
X. Bian et al. Table 1 Systemic hemodynamic variables at different right coronary pressure RCP (mmHg) 61±1 HR (beats/min) PAO (mmHg) RCBF (ml/min/g) RCR (mmHg/ml/min/g) RVSP (mmHg) RPP (mmHg/beats/min×10) dP/dtmax (mmHg/s) dP/dtmin (mmHg/s) RAP (mmHg) GU (lmol/min/g) RC PCO2 (mmHg)
115±3
124±8 117±4 0.36±0.03 170±13 24±2 306±23 644±55 −519±25 3.3±0.2 0.13±0.02 47.7±1.5
121±9 119±2 0.51±0.04† 232±22† 24±2 298±23 668±61 −538±45 3.2±0.1 0.13±0.04 44.0±1.1†
164±2 124±9 166±3∗ 0.65±0.04∗ 259±19∗ 25±2 311±23 693±72 −586±66 3.4±0.2 0.30±0.05∗ 41±1.7†
Values are means±... RCP, right coronary pressure; HR, heart rate; PAO, aortic pressure; RCBF, right coronary blood flow; RCR, right coronary resistance; RVSP, right ventricular systolic pressure; RPP, rate–pressure product; dP/dtmax, maximum rate of right ventricular pressure development; dP/dtmin, maximum rate of decline of right ventricular pressure; RAP, right atrial pressure; GU, glucose uptake; RC P2, right coronary venous P2; ∗ P<0.05 vs baseline and lowered CPP. † P<0.05 vs lowered CPP.
Figure 3 Dialysate adenosine (Ado) concentration as function of right coronary perfusion pressure (RCP). Data from each experiment are shown.
adenosine formation increased coincidently with elevated RC Pv2. The investigation also showed that baseline RV dialysate adenosine levels and RC blood flow were about 50% of values previously reported for canine left ventricle.18–20
Cardiac microdialysate as an index of local myocardial adenosine production RV interstitial purines were indexed by concentrations in the microdialysate as RCP was varied
between approx. 60 and 160 mmHg. While there is controversy regarding the relationship between dialysate adenosine and the interstitial concentration of adenosine,21 the myocardial microdialysis procedure has been utilized by others to evaluate the effects of ischemia,16,19 hypoxia,18 inotropic interventions,16,20 and treatment with adenosine deaminase inhibitors.22,23 Changes in dialysate adenosine produced by these interventions were as expected and agreed with directional changes in interstitial adenosine obtained by other methods. Thus, myocardial microdialysis can be expected to reflect accurately directional changes in local purine nucleoside formation as required for this investigation. Dialysate inosine, the immediate deamination product of adenosine, was assayed to provide more complete information on adenosine production. Changes in adenosine and adenosine + inosine were consistent for each change in RCP.
Did reduction in right coronary perfusion pressure produce myocardial ischemia? Right ventricular MVO2 fell significantly when RCP was reduced to 61 mmHg. To interpret the current findings, it is necessary to determine if this reduction of MVO2 reflected a downregulation in O2 demand or a flow limited O2 supply, i.e. ischemia. Depressed myocardial lactate uptake may reflect increased anaerobic glycolytic lactate production and, thus, be a marker of ischemia. Net lactate uptake by RV myocardium declined at 61 mmHg RCP, but the ratio of lactate uptake to MVO2 was not significantly
Adenosine Formation and MVO2
351
Figure 5 Right coronary blood flow and dialysate adenosine at baseline and lowered right coronary pressures. Data are means±... from 10 experiments. Right coronary blood flow (RCBF; closed circles) was measured at the midpoint of dialysate collection. Adenosine (Ado) was measured in the dialysate (open bars). Peak RCBF measured after release of the partial right coronary artery occlusion is also depicted. ∗ P<0.05 v baseline.
Figure 4 Metabolic variables as influenced by right coronary perfusion pressure. Variables were measured during perfusion at the three different right coronary arterial perfusion pressures (RCP) as described in Methods. Data are means±... from 10 experiments. Panel A: dialysate purine concentrations (Ado: adenosine, black bars; Ado + Ino: adenosine + inosine, open bars) and right ventricular myocardial O2 consumption (MVO2, hatched bars). Panel B: right coronary venous PO2 (PvO2, open bars) and myocardial lactate uptake (filled bars). ∗ P<0.05 v 115 mmHg RCP baseline; † P<0.05 v 61 mmHg RCP.
changed, so that the physiological contribution of lactate to oxidative metabolism was preserved. Also, the decline in interstitial adenosine concentration as indexed by dialysate adenosine at lowered RCP is incompatible with local ischemia. Moreover, we previously found no decrease in RV cytosolic phosphorylation potential after reducing RCP from 100 to 60 mmHg, although a more severe reduction
of RCP to 30 mmHg lowered the phosphorylation potential and caused net lactate release.15 In the current study, the decrease in MVO2 observed at RCP of 61 mmHg appears, indeed, to reflect a downregulation of RV O2 demand. The decrease in dialysate adenosine observed at this moderately reduced RCP is, thus, consistent with the concept that adenosine production is modulated by the metabolic rate and not by changes in tissue oxygenation of non-ischemic myocardium.10,11
Mechanism of coronary pressure modulation of right ventricular O2 demand Global measures of RV hemodynamic performance and myocardial energy demand, including RV peak systolic and right atrial filling pressures, maximum and minimum RV dP/dt, and the product of heart rate and RV peak systolic pressure were unaltered by changes in RCP between 61 and 164 mmHg. Regional contractile function was not measured in these experiments because we were concerned that implantation of measuring devices might adversely affect the microdialysis procedure. However, earlier studies in our laboratory detected no RCP-induced changes in regional myocardial segment shortening over this RCP range.13–15,24,25 Although that index of external RV work was unchanged by RCP, we
352
X. Bian et al.
did detect a marked effect of RCP on RV systolic stiffness, an important component of RV internal work.14,25 When autoregulation is ineffective, as in the RC circulation, coronary perfusion pressure directly modulates coronary vascular volume and regional myocardial oxygen consumption; effective autoregulation dampens these changes.24 The marked perfusion pressure-dependent changes in coronary vascular engorgement that result from ineffective RC autoregulation alter the stiffness of the RV wall and modulate the internal work required to shorten the RC perfusion territory during systolic contraction.14,24,25 This dependence of RV internal work on RCP would account for the observed effects of RCP on RV MVO2.
Mechanism of coronary pressure modulation of purine formation in normoxic right ventricle When RCP was lowered from baseline 115 to 61 mmHg, dialysate adenosine and total purine nucleoside (adenosine + inosine) concentrations fell along with RV MVO2. Conversely, when RCP was increased to 164 mmHg, dialysate adenosine and total purine nucleoside concentrations increased along with RV MVO2. What mechanism was responsible for these parallel changes in purine formation and RV MVO2? Adenosine formation is known to be accelerated under the hypoxic conditions produced by myocardial ischemia.1–3 Enhanced adenosine formation has also been described under conditions of increased myocardial O2 demand with normal perfusion pressure and no restriction of coronary blood flow.5–7 However, these increases in adenosine production were accompanied by declines in coronary venous P2, an index of tissue O2 tension,8 so hypoxia might have been responsible for adenosine formation. The poor pressure–flow autoregulatory ability of the RC circulation12–14 results in parallel changes in RCP and RC Pv2.12,14 RV MVO2 also changes in parallel with RCP (Fig. 4A and references 12 and 14), so this model provided an opportunity to determine if myocardial adenosine formation would vary with RV MVO2 in absence of hypoxia. Since adenosine formation increased along with elevated Pv2 and lactate uptake, tissue hypoxia is ruled out as the mechanism responsible for this increase during the increased RV O2 demand produced by elevated RCP. An alternative to the hypoxia theory for adenosine formation was advanced by Bu¨nger10 and further described by Olsson and Bu¨nger.11 They
proposed that interventions which depress the cytosolic ATP phosphorylation potential or raise free cytosolic AMP would physiologically produce adenosine, even if tissue oxygen tension does not fall. This concept was supported by observations that adenosine release from normoxic, isolated hearts varied with cytosolic-free AMP and MVO2.10,26 The current observations provide the first direct evidence of hypoxia-independent, work-related parallel changes in MVO2 and adenosine formation in blood perfused, in-situ myocardium. Work-related changes in RV O2 demand can produce changes in cytosolic energetics of non-ischemic myocardium. For example, Schwartz et al.27 reported that pulmonary artery constriction doubled regional RV MVO2 and lowered the phosphocreatine:ATP concentration ratio, an index of cytosolic ATP phosphorylation potential, by about 11% in in situ pig hearts. In myocardium, decreased cytosolic phosphorylation potential is associated with increased free cytosolic concentration of AMP, the immediate precursor of adenosine.10,11,27 While the observed changes in dialysate purine nucleoside concentrations would be expected to reflect opposite directional changes in cytosolic phosphorylation potential, these changes might be small given the steep, inverse relationships between cytosolic ATP phosphorylation potential and adenosine formation reported in isolated perfused rat28 and guinea-pig hearts.27,29,30 Thus, even modest changes in cytosolic energetics can impact purine nucleoside formation. This may explain why we previously detected no change in cytosolic phosphorylation potential in RV myocardium after RCP was lowered from baseline to 60 mmHg.15 Regardless of the exact details of the underlying metabolic control mechanism of adenosine formation, the present in vivo results clearly support the concept that increases in myocardial O2 demand can modulate adenosine formation without the requirement for O2 supply:demand imbalance resulting in myocardial hypoxia.10,11
Relationship between dialysate adenosine and right coronary resistance Although RC autoregulation was relatively poor in these experiments, RC resistance did increase when RCP was increased above baseline and decrease when RCP was reduced. These expected responses to changes in RCP occurred despite increases in dialysate adenosine at high pressure and reduced adenosine at low pressure. Thus, adenosine, a potent coronary vasodilator,1–4 cannot be responsible
Adenosine Formation and MVO2
for the RC autoregulatory adjustments to changes in RCP. Degradation of interstitial adenosine30 and blockade of adenosine receptors32 lead to the same conclusion with regard to the role of adenosine in left coronary autoregulation. The present findings suggest, in addition, that the parallel changes in RV adenosine with RCP might act to attenuate RC autoregulation. To what extent this mechanism accounts for relatively poor RC autoregulation requires further study. Steady state dialysate adenosine concentration fell appreciably after RCP was lowered from 115 to 61 mmHg. Despite the decreased concentration of this vasodilator, substantial hyperemia occurred when RCP was restored from 61 to 115 mmHg. This is further evidence that factors other than adenosine31–33 must have dilated the RC vasculature during the period of decreased RCP. Although adenosine was not responsible for RCP induced changes in RC resistance, the possibility that adenosine affected baseline resistance requires comment. Baseline RC flow was about 50% of left coronary flow measured in similar canine preparations,12 so baseline RC resistance was about twice left coronary resistance. Interestingly, RV baseline dialysate concentrations of adenosine were about 50% of those reported for left ventricle.16,18,19 These data raise the intriguing possibility that different basal concentrations of interstitial adenosine may account for the difference in baseline left and right coronary resistances. The difference in right and left ventricular dialysate adenosine is indeed consistent with the different rates of ventricular MVO2,6,24,25 and, therefore, lends further credence to the theory that adenosine production is directly tied to the metabolic rate. Coronary blood flow is directly modulated by the metabolic rate,4,11 but further investigation is required to determine if a lesser RV interstitial concentration of adenosine is responsible for differences in RC and left coronary resistance.
Right ventricular glucose uptake Myocardial glucose uptake increased appreciably when RCP was raised above baseline. Adenosine has been reported to stimulate myocardial glucose uptake.34,35 Thus, the increase in interstitial adenosine, combined with increased internal work and substrate demand, may have been responsible for increased glucose uptake at 164 mmHg RCP. According to this mechanism, reduction of RCP from 115 to 61 mmHg would be expected to lower glucose uptake due to the concomitant reductions
353
in interstitial adenosine, internal myocardial work and energy demand, but glucose uptake did not fall. We36 and others37,38 have demonstrated a shift in myocardial fuel uptake from fatty acid to glucose during moderate coronary hypoperfusion, and in the present study, increased oxidation of the more oxygen-efficient fuel glucose appeared to contribute to increased oxygen utilization efficiency during coronary hypoperfusion. Such a shift to glucose combustion may have maintained glucose uptake despite the fall in dialysate adenosine.
Acknowledgments This work was supported by grants from the National Heart, Lung and Blood Institute to HFD (HL 35027) and RTM (HL 50441). Arthur G. Williams, Jr, B.S., and Jie Sun, B.S., provided expert assistance for animal experimentation and for high performance liquid chromatography, respectively.
References 1. B RM. Cardiac nucleotides in hypoxia: possible role in regulation of coronary blood flow. Am J Physiol 1963; 204: 317–322. 2. G E, D B, D RH. Der Nucleotidabbau im Herzmuskel bei Sauerstoffmangel und seine mo¨gliche Bedeutung fu¨r die Corona¨rdurchblutung. Naturwissenschaften 1963; 50: 228– 229. 3. B RM. The role of adenosine in the regulation of coronary blood flow. Circ Res 1980; 47: 807–813. 4. R R, B RM. Regulation of coronary blood flow. Prog Cardiovasc Dis 1975; 18: 105–122. 5. W VT, S LH. Effects of catecholamines, histamine, and nitroglycerin on flow, oxygen utilization, and adenosine production in the perfused guinea pig heart. Circ Res 1977; 41: 503–508. 6. M WL, B L, B A, F DH, R R, B RM. Canine myocardial adenosine and lactate production, oxygen consumption, and coronary blood flow during stellate ganglia stimulation. Circ Res 1979; 45: 708–718. 7. S D, N DG, V RB, O RA. Relationship of cardiac oxygen usage, adenosine content, and coronary resistance in dogs. Circ Res 1980; 47: 875–882. 8. T SM. A theoretical analysis of the relationship between venous blood and mean tissue oxygen pressure. Respir Physiol 1974; 20: 283–296. 9. D A, S J. Cardiac adenosine production is linked to myocardial P2. J Mol Cell Cardiol 1991; 23: 495–504. 10. B¨ R. Thermodynamic state of cytosolic adenylates in guinea pig myocardium. Energy-linked adaptive changes in free adenylates and purine nucleoside release. In: Gerlach E, Becker BF, eds. Topics
354
11. 12.
13.
14. 15.
16.
17.
18.
19. 20.
21. 22.
23.
24.
X. Bian et al.
and perspectives in adenosine research. New York: Springer-Verlag, 1987: 223–235. O RA, B¨ R. Metabolic control of coronary blood flow. Prog Cardiovasc Dis 1987; 29: 369–387. Y S, W N, C JL, G JF, D HF. Mechanism of attenuated pressure-flow autoregulation in the right coronary circulation of dogs. Circ Res 1987; 60: 133–141. B X, W AG J, G PA, D HF. Right coronary autoregulation in conscious, chronically instrumented dogs. Am J Physiol 1998; 275: H169–H175. B X, D HF. Right coronary pressure modulates right ventricular stiffness and oxygen consumption. Cardiovasc Res 1999; 42: 80–86. I M, M RT, G ZP, W AG J, D HF. Stability of high-energy phosphates in right ventricle: myocardial energetics during right coronary hypotension. Am J Physiol 1996; 271: H320–H328. V W DGL, W J, S J, W RJ, L RD, M RM J. Cardiac microdialysis to estimate interstitial adenosine and coronary blood flow. Am J Physiol 1990; 258: H1642–H1649. M RT, S J, F WL, K YH, B¨ R. Magnesium activated adenosine formation in intact perfused heart: predominance of ecto 5′-nucleotidase during hypermagnesemia. Biochim Biophys Acta 1996; 1290: 165–176. V W DGL, W AG J, D HF. Interstitial purine metabolites and lactate during regional myocardial hypoxia. Cardiovasc Res 1993; 27: 1498–1503. D JA, V W DGL. Endocardial and epicardial interstitial purines and lactate during graded ischemia. Am J Physiol 1994; 266: H1019–H1026. H JL, V W DGL, P RD, H CD, R CM, S WC. Myocardial interstitial purine metabolites and lactate with increased work in swine. Cardiovasc Res 1995; 30: 351–356. T Y, R KN, B RV, F EO. Role of adenosine in local metabolic coronary vasodilation. Am J Physiol 1999; 276: H1425–H1433. S PH, D D, V W DGL. Adenosine deaminase inhibition augments interstitial adenosine but does not attenuate myocardial infarction. Cardiovasc Res 1995; 29: 616–623. M SA, R CM, V W DGL. Dual cardiac microdialysis to assess drug-induced changes in interstitial purine metabolites: adenosine deaminase inhibition versus adenosine kinase inhibition. Cardiovasc Res 1998; 37: 171–178. B XJ, I T, W AG J, F WL, D HF. Coronary pressure-flow autoregulation protects myocardium from pressure-induced changes in oxygen consumption. Am J Physiol 1994; 266: H2359–H2368.
25. I T, B XJ, D HF. Coronary perfusionrelated changes in myocardial contractile force and systolic ventricular stiffness. Cardiovasc Res 1994; 28: 1331–1336. 26. K YH, M RT, B¨ R. Coronary autoregulation and purine release in normoxic heart at various cytoplasmic phosphorylation potentials: disparate effects of adenosine. Pflu¨gers Arch 1992; 421: 188–199. 27. S GG, G CR, W JA, G J, S S. Relation among regional O2 consumption, high-energy phosphates, and substrate uptake in porcine right ventricle. Am J Physiol 1994; 266: H521–H530. 28. H J. Adenosine production and energy metabolism in ischemic and metabolically stimulated rat heart. J Mol Cell Cardiol 1989; 21: 1089–1100. 29. H MX, W RD, D PF, R GD, S HV. Phosphorylation potential and adenosine release during norepinephrine infusion in guinea pig heart. Am J Physiol 1987; 253: H1184–H1191. 30. B¨ R, M RT, K YH. Guinea-pig cardiac free, bound and interstitial adenylates: energy-linked and energy-independent adenosine release. In: Imai S, Nakazawa M, eds. Role of adenosine and adenine nucleotides in the biological system. Amsterdam: Elsevier Science, 1991: 337–352. 31. H FL, G MT, S MB, H JI. Role of adenosine in coronary autoregulation. Am J Physiol 1986; 250: H558–H566. 32. S DW, K K, F EO. K+ATP channels and adenosine are not necessary for coronary autoregulation. Am J Physiol 1997; 273: H1299–H1308. 33. B TP, F EO. Role of myocardial oxygen and carbon dioxide in coronary autoregulation. Am J Physiol 1992; 262: H1231–H1237. 34. L WR, M PM. Adrenergic, insulin, and work interactions with adenosine’s effects on in situ myocardial glucose uptake. Cardiovasc Res 1995; 31: 691–698. 35. F HK, S C, S LJ, R RL, L WR. Cardiac contractile function during coronary stenosis in dogs: association of adenosine in glycolytic dependence. Am J Physiol 1997; 272: H2195– H2203. 36. H BJ, B X, M RT, D HF. The effects of hyperlipidemia and hypoglycemia on myocardial oxygen utilization efficiency but not contractile function during coronary hypoperfusion (Abstract). FASEB J 1999; 13: A765. 37. G A, D L. Fatty acid oxidation in the heart. J Cardiovasc Pharmacol 1996; 28 (Suppl. 1): S11–S17. 38. H SC, S. L JD, L JE, A- S. Free fatty acid metabolism during myocardial ischemia and reperfusion. Mol Cell Biochem 1997; 166: 85–94.
Myocardial Oxygen Consumption Modulates Adenosine Formation by Canine Right Ventricle in Absence of Hypoxia Xiaoming Bian, Min Fu, Robert T. Mallet, Rolf Bu¨nger∗ and H. Fred Downey Department of Integrative Physiology, University of North Texas Health Science Center at Fort Worth, Fort Worth, TX 76107-2699, USA, ∗Department of Physiology, Uniformed Services University of the Health Sciences, Bethesda, MD 20814-4799, USA (Received 24 September 1999, accepted 5 November 1999) X. B, M. F, R. T. M, R. B¨ H. F. D. Myocardial Oxygen Consumption Modulates Adenosine Formation by Canine Right Ventricle in Absence of Hypoxia. Journal of Molecular and Cellular Cardiology (2000) 32, 345–354. Myocardial adenosine formation varies with myocardial oxygen consumption (MVO2), but whether concurrent hypoxia is required for adenosine formation is uncertain. Changes in right coronary (RC) perfusion pressure (RCP) produce directionally similar alterations in right ventricular (RV) MVO2 and in RC venous P2 (Pv2), an index of myocardial P2. RCP was varied in 10 anesthetized, open chest dogs to determine if, under these conditions, RV formation of adenosine would increase with MVO2 in absence of myocardial hypoxia. Dialysis probes were implanted in the mid myocardium of RV free wall for collecting dialysate samples for HPLC analyses to estimate interstitial adenosine and other purines. Coronary venous blood was sampled from a superficial vein draining the RC artery (RCA) perfusion territory. At 115±3 mmHg baseline RCP, RC blood flow (RCBF)= 0.51±0.04 ml/min/g, MVO2=4.6±0.5 ml/min/100 g, Pv2=34±1.5 mmHg, and dialysate adenosine= 0.27±0.03 l. When RCP was lowered to 61±1 mmHg by adjusting an occluder on the proximal RCA, RCBF decreased to 0.36±0.03 ml/min/g, MVO2 fell to 3.7±0.4 ml/min/100 g, lactate uptake remained positive, Pv2 fell to 30±1.7 mmHg, and dialysate adenosine decreased to 0.20±0.03 l. Reactive hyperemia of 1.25±0.13 ml/ min/g was observed when the RCA constriction was released, although dialysate adenosine had fallen. When RCP was elevated to 164±2 mmHg by inflating a balloon catheter in the descending aorta, RCBF increased to 0.70±0.06 ml/min/g, MVO2 increased to 5.8±1.0 ml/min/100 g, Pv2 rose to 39±2.3 mmHg, and dialysate adenosine increased to 0.33±0.04 l. These data indicate that (1) RV oxygen demand varies with RCP; (2) if RV ischemia is absent, myocardial adenosine formation is modulated by MVO2, with no requirement for hypoxia; (3) pressure–flow autoregulation is relatively ineffective in the RC circulation, where adenosine does not mediate 2000 Academic Press and may even blunt autoregulation. K W: Purine metabolism; Microdialysis; Oxygen consumption; Adenosine; Right ventricular myocardium; Coronary flow.
Introduction Since the adenosine hypothesis of coronary flow regulation was proposed in 1963,1,2 numerous studies have produced evidence that adenosine plays
an important role in the regulation of coronary blood flow.3 In many studies, a decreased O2 supply was the primary stimulus for adenosine production.4 However, studies of increased myocardial O2 demand under conditions of ample O2 supply also
Please address all correspondence to: Xiaoming Bian, Department of Integrative Physiology, University of North Texas Health Science Center at Fort Worth, 3500 Camp Bowie Boulevard, Fort Worth, TX 76107-2699, USA.
0022–2828/00/030345+10 $35.00/0
2000 Academic Press
346
X. Bian et al.
demonstrated an increase in adenosine production. These studies increased O2 demand by administration of inotropic agents,5 bilateral stellate ganglion stimulation,6 and cardiac paired pacing.7 However, the increase in adenosine production in all those studies was associated with a decline in coronary venous P2, an index of tissue O2 tension.8 These observations agree with the concept that myocardial hypoxia is a prerequisite for adenosine formation.9 On the other hand, Bu¨nger10 proposed that adenosine formation was regulated by the cytosolic phosphorylation potential as influenced by the rate of myocardial energy expenditure. According to this proposal, myocardial hypoxia is not absolutely required for increased adenosine formation. Instead, any condition that would lower the phosphorylation potential or raise free cytosolic AMP would physiologically produce adenosine, whether or not myocardial hypoxia occurs concomitantly.10,11 Autoregulation of coronary blood flow is remarkably ineffective in right ventricle (RV),12,13 so changes in right coronary (RC) pressure (RCP) alter RC blood flow appreciably. This ineffective RC autoregulation is associated with RCP induced changes in RV myocardial O2 consumption (MVO2). With moderate reductions in RCP, RC O2 demand falls sufficiently that RV ischemia is avoided.14,15 Another consequence of ineffective RC autoregulation is that RC venous P2 and, thus, RV myocardial P2 vary with RCP coincident with parallel changes in RV MVO2.12,14 Consequently, the RV offers a novel model to test the hypothesis that myocardial hypoxia is required for adenosine formation during increased myocardial O2 demand. In this investigation, RCP was varied in anesthetized, open chest dogs to determine if RV production of adenosine would increase with MVO2 in absence of myocardial hypoxia. RV adenosine was estimated by microdialysis. This is the first report on relationships between RV adenosine formation, RV MVO2, and RC flow. Results indicate that hypoxia is not required for adenosine formation.
Methods This investigation was approved by the Institutional Animal Care and Use Committee of the University of North Texas Health Science Center at Fort Worth and was conducted in accordance with the Guide for the Care and Use of Laboratory Animals (Institute of Laboratory Animal Resources, National Research Council, USA, 1996).
Animal preparation Ten adult healthy mongrel dogs of either sex (22–29 kg) were anesthetized with intravenous sodium pentobarbital (30 mg/kg). Additional sodium pentobarbital was given as needed to maintain adequate anesthesia throughout the experiment. The animals were intubated and mechanically ventilated with room air supplemented with oxygen to maintain normal arterial blood gases throughout the experiment. Core body temperature was monitored with a rectal temperature probe and maintained at 37.0–38.0°C with a heating pad. A catheter was inserted into a femoral artery to monitor aortic pressure, and the ipsilateral femoral vein was cannulated for infusion of supplemental fluids and anesthetics. A balloon catheter (20 cc, St Jude Medical) was inserted into the contralateral femoral artery and advanced into the thoracic aorta to obtain samples for arterial blood gas analyses and to increase the RCP by inflating the balloon as required in the protocols. A right thoracotomy was performed in the fourth intercostal space, and the heart was suspended in a pericardial cradle. Figure 1 illustrates the experimental preparation. A non-branching section of the RC artery was dissected free for 1–1.5 cm for placement of a single-crystal Doppler flow probe proximal to an adjustable occluder. To measure RCP distal to the occluder, a Micro-Renathane catheter (0.014 in ID, 0.033 in OD) was inserted into the RC artery through a small right atrial side branch. A Millar catheter-tip pressure transducer was inserted in the right atrial appendage and advanced across the tricuspid valve to measure right ventricular pressure. RV dP/dt was determined by electronic differentiation of the RV pressure signal. Saline-filled catheters were inserted into the main pulmonary artery and right atrium to measure pulmonary arterial and right atrial pressures. A 24-gauge catheter was inserted into a superficial vein draining the area perfused by the RC artery for sampling RC venous blood. Arterial and RC venous samples were collected anaerobically. RV MVO2 and lactate uptake were calculated from the product of RC blood flow times the respective regional arteriovenous differences. RC flow velocity was measured with a Triton Technology model 100 pulsed Doppler flowmeter. Heart rate (HR), aortic pressure (PAO), RCP, pulmonary artery pressure (PPA), RV pressure (PRV), RV dP/dt, right atrial pressure (PRA), and RC flow velocity signals were recorded continuously with a multichannel Grass model 7D polygraph.
Adenosine Formation and MVO2
347
initiation of the experimental protocol. During this time dialysate nucleoside concentrations fell and stabilized at concentrations which were taken as the baseline values. This recovery to basal nucleoside concentrations following implantation of a dialysis probe has been documented by Van Wylen et al.16 The outflows of the dialysis probes were collected in tared 250 ll plastic vials (Fig. 1). Each sample (30 ll) was weighed and diluted with distilled H2O to 70 ll, then immediately frozen in liquid N2 and stored at −80°C until analysis. Experimental protocol
Figure 1 In situ canine heart preparation. A portion of the proximal right coronary artery was dissected for placement of a flow probe and screw-clamp occluder. Two dialysis probes were implanted in the mid myocardium near the center of the right coronary perfusion territory to assess interstitial fluid purines. Catheters were placed in the right atrium, right ventricle, aortic arch, and pulmonary trunk to monitor right atrial (PRA), right ventricular (PRV), aortic (PAO) and pulmonary arterial (PPA) pressures, respectively. Right coronary pressure distal to the occluder was measured with a catheter inserted into the right coronary artery via a small side branch. Right coronary venous samples were collected anaerobically from a superficial vein draining the right coronary perfusion territory. Right coronary perfusion pressure was lowered by partially constricting the right coronary artery with the occluder. A balloon catheter in the thoracic aorta was inflated to increase aortic and, thus, right coronary perfusion pressure.
Cardiac microdialysis Myocardial interstitial adenosine and other purine concentrations were estimated with the cardiac microdialysis technique, as described previously by Van Wylen et al.16 Two microdialysis probes were implanted in the mid myocardium of the RC perfusion territory. The probes were inserted with the aid of an introducer needle, and each probe was gently pulled through the myocardium until the dialysis window rested entirely within the muscle wall. A 23-gauge needle had been previously glued to the inflow port of the dialysis tubing, and the hub of this needle was connected to a gas-tight syringe filled with Krebs–Henseleit buffer aerated with 95% N2 and 5% CO2.16 Buffer was perfused through the dialysis probes at 2 ll/min for the remainder of the protocol with a Harvard model 22 infusion pump. A 60-min stabilization period was allowed before
Beginning 60 min after the microdialysis probes were implanted, three consecutive dialysate samples, each 15 min in duration, were collected from each of the two probes to establish the baseline nucleoside concentrations. At the mid-point of each dialysate sample collection, hemodynamic variables were measured and duplicate arterial and RC venous blood samples were collected. RCP was reduced from baseline (approx. 115 mmHg) to 60 mmHg by constricting the RC artery with the adjustable occluder. After allowing 15 min for stabilization, three 15-min dialysate samples from each probe, hemodynamic readings, and blood samples were taken in the same manner as described for the baseline condition. After the last dialysate sample at 60 mmHg RCP was collected, the RC artery constriction was abruptly released, and the peak RC flow was recorded. After 30 min, RCP was increased to 160 mmHg by inflating the balloon catheter in the descending aorta. Following a 15min stabilization period, three 15-min dialysate samples were collected from each probe, and hemodynamic readings and blood samples were taken in the same manner as described for the baseline condition. At the end of each experiment, India ink was injected through the RC artery catheter to delineate the RC perfusion territory. The heart was excised, and the dyed area was resected and weighed. The internal diameter of the RC artery under the flow velocity probe was measured, and its cross-sectional area was calculated. RC blood flow (ml/min/g) was calculated as the product of RC flow velocity times the cross-sectional area divided by perfused tissue mass. Analytical methods Arterial and venous pH, P2, and P2 were determined by an Instrumentation Laboratory model
348
X. Bian et al.
Synthesis 30 blood gas analyzer. Lactate was measured by a Yellow Springs Instruments model 2300 Stat L-lactate analyzer. Purine nucleosides (adenosine and inosine) in the dialysate were measured by HPLC (Shimadzu model LC-600) as previously described.17 Duplicate samples collected simultaneously from the two probes were averaged.
Statistical analyses All values are expressed as mean±... Values of variables measured during three consecutive 15min periods during each RCP were averaged after repeated measures ANOVA detected no period-related differences among these values. Hemodynamic data, MVO2, Pv2, lactate uptake, and dialysate nucleoside concentrations were analyzed with a repeated measures ANOVA with RCP as the independent variable. Values of P<0.05 were taken to indicate statistical significance.
Results A hemodynamic recording from a typical experiment is shown in Figure 2, and hemodynamic variables are summarized in Table 1. RC blood flow varied with RCP (P<0.05), reflecting poor autoregulation in right coronary circulation. Heart rate, right atrial pressure, and RV systolic pressure, dP/dtmax and dP/dtmin were unaffected by altering RCP and were stable during the three consecutive 15-min periods at each RCP. These data indicate that global RV preload, afterload, contractility, and relaxation were constant throughout the protocol. In Figure 3, RV dialysate adenosine concentration is plotted as a function of RCP for each dog. In all experiments, dialysate adenosine decreased when RCP was reduced, and increased when RCP was elevated from baseline. RV dialysate adenosine, adenosine + inosine, and MVO2 at reduced, baseline, and elevated RCP are presented in Figure 4A. Corresponding values of RC venous P2 (Pv2) and lactate uptake are shown in Figure 4B. When RCP was lowered from 115±3 mmHg baseline to 61±1 mmHg, RV dialysate adenosine concentration fell from 0.27±0.03 l to 0.20±0.03 l (P<0.05). When RCP was elevated from the baseline to 164±2 mmHg, dialysate adenosine increased to 0.33±0.04 l (P<0.05 vs baseline). Similar changes in adenosine + inosine were observed as RCP was varied (Fig. 4A). RV MVO2 increased from 3.7±0.4 ml/min/100 g at 61 mmHg RCP to
5.8±1.0 ml/min/100 g at 164 mmHg RCP (P<0.05). Elevation of RCP in absence of effective pressure–flow autoregulation produced an increase in Pv2 from 30±1.8 mmHg at 61 mmHg RCP to 39±2.3 mmHg at 164 mmHg RCP (P<0.05; Fig. 4B). Over this broad range of RCP, lactate uptake increased from 0.22±0.03 to 0.42±0.06 lmol/ min/g (P<0.05; Fig. 4B), but the ratio of lactate uptake to MVO2 changed insignificantly from 0.069±0.014 to 0.076±0.011 lmol/ml. Thus, varying RCP produced parallel changes in RV MVO2, dialysate adenosine and adenosine + inosine concentrations, RC venous oxygen tension, and RV lactate uptake. In these experiments, RV dialysate adenosine concentration was greatest when RV MVO2 was at its highest value (Fig. 4A) and when RC Pv2 was also elevated (Fig. 4B). RC Pv2 was at its lowest value (Table 1) when RV dialysate adenosine concentration was greatest (Fig. 4A). Because Pv2 is inversely related to intracellular pH,10 this result ruled out intracellular acidosis as a trigger for adenosine formation. RV MVO2 fell as a response to decreased RCP rather than in response to a flow limitation of O2 delivery, since lactate uptake as a function of MVO2 was not compromised. Under these non-ischemic conditions at moderately decreased RCP, RV dialysate adenosine fell with RV MVO2 (Fig. 4A). Figure 5 shows RC blood flow measured at low and baseline RCP and corresponding concentrations of RV dialysate adenosine. Both variables fell significantly when RCP was lowered to 61 mmHg. When the RC artery constriction was abruptly released to restore RCP to baseline, a peak hyperemic response of 1.25±0.13 ml/min/g was observed. This hyperemia exceeded the baseline flow by more than two-fold (P<0.01). The presence of this significant reactive hyperemia demonstrated that the RC vasculature had dilated during the period of decreased RCP, despite the concomitant reduction in dialysate adenosine concentration.
Discussion This investigation tested the hypothesis that myocardial hypoxia is required for RV adenosine formation during RCP induced changes in RV myocardial O2 demand. In this poorly autoregulating model, changes in coronary perfusion pressure produce directionally similar changes in MVO2 and in Pv2, an index of myocardial P2,8 so this model was well suited for testing this hypothesis. The hypothesis was not supported, since
Adenosine Formation and MVO2
349
Figure 2 Typical tracing of hemodynamic variables at different right coronary perfusion pressures. Right coronary artery perfusion pressures (RCP) in the three depicted phases of this experiment were approximately 60, 115 and 160 mmHg (tracing 2). Measured variables included aortic pressure (PAO: tracing 1), right coronary blood flow (RCBF: tracing 3), pulmonary arterial pressure (PPA: tracing 4), right ventricular dP/dt (tracing 5), phasic right ventricular pressure (PRV: tracing 6) and right atrial pressure (PRA: tracing 7). The increase in PAO during the final phase of the experiment reflects inflation of an intra-aortic balloon to increase RCP.
350
X. Bian et al. Table 1 Systemic hemodynamic variables at different right coronary pressure RCP (mmHg) 61±1 HR (beats/min) PAO (mmHg) RCBF (ml/min/g) RCR (mmHg/ml/min/g) RVSP (mmHg) RPP (mmHg/beats/min×10) dP/dtmax (mmHg/s) dP/dtmin (mmHg/s) RAP (mmHg) GU (lmol/min/g) RC PCO2 (mmHg)
115±3
124±8 117±4 0.36±0.03 170±13 24±2 306±23 644±55 −519±25 3.3±0.2 0.13±0.02 47.7±1.5
121±9 119±2 0.51±0.04† 232±22† 24±2 298±23 668±61 −538±45 3.2±0.1 0.13±0.04 44.0±1.1†
164±2 124±9 166±3∗ 0.65±0.04∗ 259±19∗ 25±2 311±23 693±72 −586±66 3.4±0.2 0.30±0.05∗ 41±1.7†
Values are means±... RCP, right coronary pressure; HR, heart rate; PAO, aortic pressure; RCBF, right coronary blood flow; RCR, right coronary resistance; RVSP, right ventricular systolic pressure; RPP, rate–pressure product; dP/dtmax, maximum rate of right ventricular pressure development; dP/dtmin, maximum rate of decline of right ventricular pressure; RAP, right atrial pressure; GU, glucose uptake; RC P2, right coronary venous P2; ∗ P<0.05 vs baseline and lowered CPP. † P<0.05 vs lowered CPP.
Figure 3 Dialysate adenosine (Ado) concentration as function of right coronary perfusion pressure (RCP). Data from each experiment are shown.
adenosine formation increased coincidently with elevated RC Pv2. The investigation also showed that baseline RV dialysate adenosine levels and RC blood flow were about 50% of values previously reported for canine left ventricle.18–20
Cardiac microdialysate as an index of local myocardial adenosine production RV interstitial purines were indexed by concentrations in the microdialysate as RCP was varied
between approx. 60 and 160 mmHg. While there is controversy regarding the relationship between dialysate adenosine and the interstitial concentration of adenosine,21 the myocardial microdialysis procedure has been utilized by others to evaluate the effects of ischemia,16,19 hypoxia,18 inotropic interventions,16,20 and treatment with adenosine deaminase inhibitors.22,23 Changes in dialysate adenosine produced by these interventions were as expected and agreed with directional changes in interstitial adenosine obtained by other methods. Thus, myocardial microdialysis can be expected to reflect accurately directional changes in local purine nucleoside formation as required for this investigation. Dialysate inosine, the immediate deamination product of adenosine, was assayed to provide more complete information on adenosine production. Changes in adenosine and adenosine + inosine were consistent for each change in RCP.
Did reduction in right coronary perfusion pressure produce myocardial ischemia? Right ventricular MVO2 fell significantly when RCP was reduced to 61 mmHg. To interpret the current findings, it is necessary to determine if this reduction of MVO2 reflected a downregulation in O2 demand or a flow limited O2 supply, i.e. ischemia. Depressed myocardial lactate uptake may reflect increased anaerobic glycolytic lactate production and, thus, be a marker of ischemia. Net lactate uptake by RV myocardium declined at 61 mmHg RCP, but the ratio of lactate uptake to MVO2 was not significantly
Adenosine Formation and MVO2
351
Figure 5 Right coronary blood flow and dialysate adenosine at baseline and lowered right coronary pressures. Data are means±... from 10 experiments. Right coronary blood flow (RCBF; closed circles) was measured at the midpoint of dialysate collection. Adenosine (Ado) was measured in the dialysate (open bars). Peak RCBF measured after release of the partial right coronary artery occlusion is also depicted. ∗ P<0.05 v baseline.
Figure 4 Metabolic variables as influenced by right coronary perfusion pressure. Variables were measured during perfusion at the three different right coronary arterial perfusion pressures (RCP) as described in Methods. Data are means±... from 10 experiments. Panel A: dialysate purine concentrations (Ado: adenosine, black bars; Ado + Ino: adenosine + inosine, open bars) and right ventricular myocardial O2 consumption (MVO2, hatched bars). Panel B: right coronary venous PO2 (PvO2, open bars) and myocardial lactate uptake (filled bars). ∗ P<0.05 v 115 mmHg RCP baseline; † P<0.05 v 61 mmHg RCP.
changed, so that the physiological contribution of lactate to oxidative metabolism was preserved. Also, the decline in interstitial adenosine concentration as indexed by dialysate adenosine at lowered RCP is incompatible with local ischemia. Moreover, we previously found no decrease in RV cytosolic phosphorylation potential after reducing RCP from 100 to 60 mmHg, although a more severe reduction
of RCP to 30 mmHg lowered the phosphorylation potential and caused net lactate release.15 In the current study, the decrease in MVO2 observed at RCP of 61 mmHg appears, indeed, to reflect a downregulation of RV O2 demand. The decrease in dialysate adenosine observed at this moderately reduced RCP is, thus, consistent with the concept that adenosine production is modulated by the metabolic rate and not by changes in tissue oxygenation of non-ischemic myocardium.10,11
Mechanism of coronary pressure modulation of right ventricular O2 demand Global measures of RV hemodynamic performance and myocardial energy demand, including RV peak systolic and right atrial filling pressures, maximum and minimum RV dP/dt, and the product of heart rate and RV peak systolic pressure were unaltered by changes in RCP between 61 and 164 mmHg. Regional contractile function was not measured in these experiments because we were concerned that implantation of measuring devices might adversely affect the microdialysis procedure. However, earlier studies in our laboratory detected no RCP-induced changes in regional myocardial segment shortening over this RCP range.13–15,24,25 Although that index of external RV work was unchanged by RCP, we
352
X. Bian et al.
did detect a marked effect of RCP on RV systolic stiffness, an important component of RV internal work.14,25 When autoregulation is ineffective, as in the RC circulation, coronary perfusion pressure directly modulates coronary vascular volume and regional myocardial oxygen consumption; effective autoregulation dampens these changes.24 The marked perfusion pressure-dependent changes in coronary vascular engorgement that result from ineffective RC autoregulation alter the stiffness of the RV wall and modulate the internal work required to shorten the RC perfusion territory during systolic contraction.14,24,25 This dependence of RV internal work on RCP would account for the observed effects of RCP on RV MVO2.
Mechanism of coronary pressure modulation of purine formation in normoxic right ventricle When RCP was lowered from baseline 115 to 61 mmHg, dialysate adenosine and total purine nucleoside (adenosine + inosine) concentrations fell along with RV MVO2. Conversely, when RCP was increased to 164 mmHg, dialysate adenosine and total purine nucleoside concentrations increased along with RV MVO2. What mechanism was responsible for these parallel changes in purine formation and RV MVO2? Adenosine formation is known to be accelerated under the hypoxic conditions produced by myocardial ischemia.1–3 Enhanced adenosine formation has also been described under conditions of increased myocardial O2 demand with normal perfusion pressure and no restriction of coronary blood flow.5–7 However, these increases in adenosine production were accompanied by declines in coronary venous P2, an index of tissue O2 tension,8 so hypoxia might have been responsible for adenosine formation. The poor pressure–flow autoregulatory ability of the RC circulation12–14 results in parallel changes in RCP and RC Pv2.12,14 RV MVO2 also changes in parallel with RCP (Fig. 4A and references 12 and 14), so this model provided an opportunity to determine if myocardial adenosine formation would vary with RV MVO2 in absence of hypoxia. Since adenosine formation increased along with elevated Pv2 and lactate uptake, tissue hypoxia is ruled out as the mechanism responsible for this increase during the increased RV O2 demand produced by elevated RCP. An alternative to the hypoxia theory for adenosine formation was advanced by Bu¨nger10 and further described by Olsson and Bu¨nger.11 They
proposed that interventions which depress the cytosolic ATP phosphorylation potential or raise free cytosolic AMP would physiologically produce adenosine, even if tissue oxygen tension does not fall. This concept was supported by observations that adenosine release from normoxic, isolated hearts varied with cytosolic-free AMP and MVO2.10,26 The current observations provide the first direct evidence of hypoxia-independent, work-related parallel changes in MVO2 and adenosine formation in blood perfused, in-situ myocardium. Work-related changes in RV O2 demand can produce changes in cytosolic energetics of non-ischemic myocardium. For example, Schwartz et al.27 reported that pulmonary artery constriction doubled regional RV MVO2 and lowered the phosphocreatine:ATP concentration ratio, an index of cytosolic ATP phosphorylation potential, by about 11% in in situ pig hearts. In myocardium, decreased cytosolic phosphorylation potential is associated with increased free cytosolic concentration of AMP, the immediate precursor of adenosine.10,11,27 While the observed changes in dialysate purine nucleoside concentrations would be expected to reflect opposite directional changes in cytosolic phosphorylation potential, these changes might be small given the steep, inverse relationships between cytosolic ATP phosphorylation potential and adenosine formation reported in isolated perfused rat28 and guinea-pig hearts.27,29,30 Thus, even modest changes in cytosolic energetics can impact purine nucleoside formation. This may explain why we previously detected no change in cytosolic phosphorylation potential in RV myocardium after RCP was lowered from baseline to 60 mmHg.15 Regardless of the exact details of the underlying metabolic control mechanism of adenosine formation, the present in vivo results clearly support the concept that increases in myocardial O2 demand can modulate adenosine formation without the requirement for O2 supply:demand imbalance resulting in myocardial hypoxia.10,11
Relationship between dialysate adenosine and right coronary resistance Although RC autoregulation was relatively poor in these experiments, RC resistance did increase when RCP was increased above baseline and decrease when RCP was reduced. These expected responses to changes in RCP occurred despite increases in dialysate adenosine at high pressure and reduced adenosine at low pressure. Thus, adenosine, a potent coronary vasodilator,1–4 cannot be responsible
Adenosine Formation and MVO2
for the RC autoregulatory adjustments to changes in RCP. Degradation of interstitial adenosine30 and blockade of adenosine receptors32 lead to the same conclusion with regard to the role of adenosine in left coronary autoregulation. The present findings suggest, in addition, that the parallel changes in RV adenosine with RCP might act to attenuate RC autoregulation. To what extent this mechanism accounts for relatively poor RC autoregulation requires further study. Steady state dialysate adenosine concentration fell appreciably after RCP was lowered from 115 to 61 mmHg. Despite the decreased concentration of this vasodilator, substantial hyperemia occurred when RCP was restored from 61 to 115 mmHg. This is further evidence that factors other than adenosine31–33 must have dilated the RC vasculature during the period of decreased RCP. Although adenosine was not responsible for RCP induced changes in RC resistance, the possibility that adenosine affected baseline resistance requires comment. Baseline RC flow was about 50% of left coronary flow measured in similar canine preparations,12 so baseline RC resistance was about twice left coronary resistance. Interestingly, RV baseline dialysate concentrations of adenosine were about 50% of those reported for left ventricle.16,18,19 These data raise the intriguing possibility that different basal concentrations of interstitial adenosine may account for the difference in baseline left and right coronary resistances. The difference in right and left ventricular dialysate adenosine is indeed consistent with the different rates of ventricular MVO2,6,24,25 and, therefore, lends further credence to the theory that adenosine production is directly tied to the metabolic rate. Coronary blood flow is directly modulated by the metabolic rate,4,11 but further investigation is required to determine if a lesser RV interstitial concentration of adenosine is responsible for differences in RC and left coronary resistance.
Right ventricular glucose uptake Myocardial glucose uptake increased appreciably when RCP was raised above baseline. Adenosine has been reported to stimulate myocardial glucose uptake.34,35 Thus, the increase in interstitial adenosine, combined with increased internal work and substrate demand, may have been responsible for increased glucose uptake at 164 mmHg RCP. According to this mechanism, reduction of RCP from 115 to 61 mmHg would be expected to lower glucose uptake due to the concomitant reductions
353
in interstitial adenosine, internal myocardial work and energy demand, but glucose uptake did not fall. We36 and others37,38 have demonstrated a shift in myocardial fuel uptake from fatty acid to glucose during moderate coronary hypoperfusion, and in the present study, increased oxidation of the more oxygen-efficient fuel glucose appeared to contribute to increased oxygen utilization efficiency during coronary hypoperfusion. Such a shift to glucose combustion may have maintained glucose uptake despite the fall in dialysate adenosine.
Acknowledgments This work was supported by grants from the National Heart, Lung and Blood Institute to HFD (HL 35027) and RTM (HL 50441). Arthur G. Williams, Jr, B.S., and Jie Sun, B.S., provided expert assistance for animal experimentation and for high performance liquid chromatography, respectively.
References 1. B RM. Cardiac nucleotides in hypoxia: possible role in regulation of coronary blood flow. Am J Physiol 1963; 204: 317–322. 2. G E, D B, D RH. Der Nucleotidabbau im Herzmuskel bei Sauerstoffmangel und seine mo¨gliche Bedeutung fu¨r die Corona¨rdurchblutung. Naturwissenschaften 1963; 50: 228– 229. 3. B RM. The role of adenosine in the regulation of coronary blood flow. Circ Res 1980; 47: 807–813. 4. R R, B RM. Regulation of coronary blood flow. Prog Cardiovasc Dis 1975; 18: 105–122. 5. W VT, S LH. Effects of catecholamines, histamine, and nitroglycerin on flow, oxygen utilization, and adenosine production in the perfused guinea pig heart. Circ Res 1977; 41: 503–508. 6. M WL, B L, B A, F DH, R R, B RM. Canine myocardial adenosine and lactate production, oxygen consumption, and coronary blood flow during stellate ganglia stimulation. Circ Res 1979; 45: 708–718. 7. S D, N DG, V RB, O RA. Relationship of cardiac oxygen usage, adenosine content, and coronary resistance in dogs. Circ Res 1980; 47: 875–882. 8. T SM. A theoretical analysis of the relationship between venous blood and mean tissue oxygen pressure. Respir Physiol 1974; 20: 283–296. 9. D A, S J. Cardiac adenosine production is linked to myocardial P2. J Mol Cell Cardiol 1991; 23: 495–504. 10. B¨ R. Thermodynamic state of cytosolic adenylates in guinea pig myocardium. Energy-linked adaptive changes in free adenylates and purine nucleoside release. In: Gerlach E, Becker BF, eds. Topics
354
11. 12.
13.
14. 15.
16.
17.
18.
19. 20.
21. 22.
23.
24.
X. Bian et al.
and perspectives in adenosine research. New York: Springer-Verlag, 1987: 223–235. O RA, B¨ R. Metabolic control of coronary blood flow. Prog Cardiovasc Dis 1987; 29: 369–387. Y S, W N, C JL, G JF, D HF. Mechanism of attenuated pressure-flow autoregulation in the right coronary circulation of dogs. Circ Res 1987; 60: 133–141. B X, W AG J, G PA, D HF. Right coronary autoregulation in conscious, chronically instrumented dogs. Am J Physiol 1998; 275: H169–H175. B X, D HF. Right coronary pressure modulates right ventricular stiffness and oxygen consumption. Cardiovasc Res 1999; 42: 80–86. I M, M RT, G ZP, W AG J, D HF. Stability of high-energy phosphates in right ventricle: myocardial energetics during right coronary hypotension. Am J Physiol 1996; 271: H320–H328. V W DGL, W J, S J, W RJ, L RD, M RM J. Cardiac microdialysis to estimate interstitial adenosine and coronary blood flow. Am J Physiol 1990; 258: H1642–H1649. M RT, S J, F WL, K YH, B¨ R. Magnesium activated adenosine formation in intact perfused heart: predominance of ecto 5′-nucleotidase during hypermagnesemia. Biochim Biophys Acta 1996; 1290: 165–176. V W DGL, W AG J, D HF. Interstitial purine metabolites and lactate during regional myocardial hypoxia. Cardiovasc Res 1993; 27: 1498–1503. D JA, V W DGL. Endocardial and epicardial interstitial purines and lactate during graded ischemia. Am J Physiol 1994; 266: H1019–H1026. H JL, V W DGL, P RD, H CD, R CM, S WC. Myocardial interstitial purine metabolites and lactate with increased work in swine. Cardiovasc Res 1995; 30: 351–356. T Y, R KN, B RV, F EO. Role of adenosine in local metabolic coronary vasodilation. Am J Physiol 1999; 276: H1425–H1433. S PH, D D, V W DGL. Adenosine deaminase inhibition augments interstitial adenosine but does not attenuate myocardial infarction. Cardiovasc Res 1995; 29: 616–623. M SA, R CM, V W DGL. Dual cardiac microdialysis to assess drug-induced changes in interstitial purine metabolites: adenosine deaminase inhibition versus adenosine kinase inhibition. Cardiovasc Res 1998; 37: 171–178. B XJ, I T, W AG J, F WL, D HF. Coronary pressure-flow autoregulation protects myocardium from pressure-induced changes in oxygen consumption. Am J Physiol 1994; 266: H2359–H2368.
25. I T, B XJ, D HF. Coronary perfusionrelated changes in myocardial contractile force and systolic ventricular stiffness. Cardiovasc Res 1994; 28: 1331–1336. 26. K YH, M RT, B¨ R. Coronary autoregulation and purine release in normoxic heart at various cytoplasmic phosphorylation potentials: disparate effects of adenosine. Pflu¨gers Arch 1992; 421: 188–199. 27. S GG, G CR, W JA, G J, S S. Relation among regional O2 consumption, high-energy phosphates, and substrate uptake in porcine right ventricle. Am J Physiol 1994; 266: H521–H530. 28. H J. Adenosine production and energy metabolism in ischemic and metabolically stimulated rat heart. J Mol Cell Cardiol 1989; 21: 1089–1100. 29. H MX, W RD, D PF, R GD, S HV. Phosphorylation potential and adenosine release during norepinephrine infusion in guinea pig heart. Am J Physiol 1987; 253: H1184–H1191. 30. B¨ R, M RT, K YH. Guinea-pig cardiac free, bound and interstitial adenylates: energy-linked and energy-independent adenosine release. In: Imai S, Nakazawa M, eds. Role of adenosine and adenine nucleotides in the biological system. Amsterdam: Elsevier Science, 1991: 337–352. 31. H FL, G MT, S MB, H JI. Role of adenosine in coronary autoregulation. Am J Physiol 1986; 250: H558–H566. 32. S DW, K K, F EO. K+ATP channels and adenosine are not necessary for coronary autoregulation. Am J Physiol 1997; 273: H1299–H1308. 33. B TP, F EO. Role of myocardial oxygen and carbon dioxide in coronary autoregulation. Am J Physiol 1992; 262: H1231–H1237. 34. L WR, M PM. Adrenergic, insulin, and work interactions with adenosine’s effects on in situ myocardial glucose uptake. Cardiovasc Res 1995; 31: 691–698. 35. F HK, S C, S LJ, R RL, L WR. Cardiac contractile function during coronary stenosis in dogs: association of adenosine in glycolytic dependence. Am J Physiol 1997; 272: H2195– H2203. 36. H BJ, B X, M RT, D HF. The effects of hyperlipidemia and hypoglycemia on myocardial oxygen utilization efficiency but not contractile function during coronary hypoperfusion (Abstract). FASEB J 1999; 13: A765. 37. G A, D L. Fatty acid oxidation in the heart. J Cardiovasc Pharmacol 1996; 28 (Suppl. 1): S11–S17. 38. H SC, S. L JD, L JE, A- S. Free fatty acid metabolism during myocardial ischemia and reperfusion. Mol Cell Biochem 1997; 166: 85–94.