| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Center for Bioengineering, University of Washington, Seattle, Washington 98195
 |
ABSTRACT |
|---|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
The hydrolysis of AMP to adenosine during acute coronary underperfusion is temporarily beneficial to myocardial survival yet may cause tissue injury during sustained underperfusion because of depletion of adenine nucleotides. We hypothesized that the enzyme mediating AMP hydrolysis, 5'-nucleotidase (5'-NT), is downregulated during sustained coronary underperfusion to prevent excessive loss of nucleotides. Langendorff-perfused rabbit hearts were subjected to two successive, identical 45-min periods of underperfusion (4-5% of baseline flow) separated by 20 min of reperfusion. Although coronary venous lactate efflux was comparable in the two periods, total coronary purine efflux during the second period of underperfusion was attenuated by 75%. Phosphorus nuclear magnetic resonance data showed that ATP fell 46% in the first period but fell only another 10% in the second period. Phosphocreatine levels fell comparably (75-78%) during both periods of underperfusion. Analysis using a mathematical model describing the kinetics of myocardial energetics revealed that the combined data set was best described by a lower activity of 5'-NT (52% decrease in maximal reaction velocity) during the second period of underperfusion. Additional time course experiments showed that the decrease in 5'-NT activity was slow in onset, requiring ~20 min of underperfusion. The decrease in 5'-NT activity during sustained underperfusion may benefit tissue survival by limiting the depletion of myocardial adenine nucleotides. In conclusion, at the onset of coronary underperfusion, there is a high activity of 5'-NT, but later during sustained underperfusion, 5'-NT is downregulated, resulting in decreased AMP hydrolysis to adenosine.
nucleotides; ischemia; nuclear magnetic resonance; adenosine; phosphoenergetics
| |
INTRODUCTION |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
DURING MYOCARDIAL HYPOXIA or ischemia, there is a net hydrolysis of ATP to ADP and then AMP, which in turn is hydrolyzed to membrane-permeable adenosine. The release of adenosine during ischemia is beneficial by providing receptor-mediated vasodilatory protection and has also been shown to mediate ischemic preconditioning; i.e., a brief period of ischemia increases myocardial viability during subsequent prolonged ischemia (7, 8, 17). New research also indicates that AMP hydrolysis to adenosine in the myocyte is temporarily beneficial during compromised energy supply through an improvement of the ATP phosphorylation potential (22). During coronary underperfusion the kinetics of phosphocreatine (PCr) and ATP could only be explained by mass action effects of AMP hydrolysis to adenosine, creating a sink for cytosolic ADP (22). AMP hydrolysis, therefore, mediates an open adenylate system. However, AMP hydrolysis also leads to a loss of purines during sustained underperfusion and, therefore, to the depletion of nucleotide pools (15), contributing to the mechanical dysfunction of the myocardium during reperfusion (30). Because a high rate of AMP hydrolysis benefits tissue survival during the onset of underperfusion but will cause tissue injury during sustained underperfusion, the regulation of AMP hydrolysis may be of utmost importance to the survival of the myocardium during compromised flow.
In heart the primary enzymatic pathway responsible for the production
of adenosine is the dephosphorylation of 5'-AMP by the cytosolic
form of 5'-nucleotidase (5'-NT, E.C. 3.1.3.5, AMP 
adenosine + Pi) (4). The current
literature, however, is unclear regarding the regulation of
5'-NT in normal and ischemic heart. It has been suggested that
rat heart 5'-NT is regulated by the adenylate energy charge (14)
and that the free concentration of ADP regulates rabbit (33) and dog
(4) heart 5'-NT. Pi (27) and
pH (2) are also possible regulators of 5'-NT during ischemia. Studies examining ischemic preconditioning have
indicated that the hydrolysis of AMP during ischemia is
increased or decreased by a preceding brief period of ischemia
(10, 19, 25). However, a confounding factor in these studies in the
intact heart is that the production of myocardial purines during
ischemia depends on the activity of 5'-NT and the
cytosolic concentration of AMP, i.e., on the disturbance of the
myocardial energy balance. Therefore, the observation of decreased loss
of myocardial purines due to ischemic preconditioning may be due to
decreased net ATP breakdown rather than decreased 5'-NT activity.
Thus the activity and regulation of 5'-NT in the ischemic intact
heart remain to be determined.
The purpose of this study was to determine whether the activity of 5'-NT is downregulated during sustained coronary underperfusion. To investigate the regulation of 5'-NT during ischemia in the intact heart, it is necessary to distinguish between the effects of altered 5'-NT activity and effects of energy imbalance (net ATP hydrolysis and cytosolic AMP concentration). We have, therefore, applied a physiologically realistic mathematical model for analyzing simultaneously obtained data of myocardial high-energy phosphates and coronary venous purines. This new, integrated approach allowed us to account for changes in energy metabolism during sustained underperfusion in describing the kinetics of 5'-NT. The results show that, during sustained coronary underperfusion, there is a downregulation of the activity of 5'-NT that may benefit tissue survival by preserving the ATP pool.
| |
MATERIALS AND METHODS |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
Isolated Perfused Heart Preparation
New Zealand White rabbits (2.2-3.0 kg) were sedated with acetylpromazine (0.8 mg/kg sc) and anesthetized with ketamine (40 mg/kg iv) plus xylazine (5 mg/kg iv). After initial anesthesia the rabbits were treated with heparin (200 U iv) and 8-sulfophenyltheophylline (10 mg/kg iv) to prevent blood coagulation and adenosine receptor activation during the surgery required for isolation of the heart. After 5 min the heart was rapidly excised, rinsed in ice-cold saline, and immediately mounted by the aorta on a Langendorff apparatus. The hearts were perfused at 37°C at a constant pressure of 80-100 mmHg with a nonrecirculating modified Krebs-Henseleit bicarbonate buffer consisting of (in mM) 118 NaCl, 3.8 KCl, 1.2 KH2PO4, 0.7 MgSO4, 2.1 CaCl2, 0.1 EDTA, 25 NaHCO3, 11 glucose, and 5 pyruvate and 0.1% bovine serum albumin and equilibrated with 95% O2-5% CO2 using a membrane oxygenator, resulting in a pH of 7.35-7.45. A fluid-filled latex balloon was placed in the left ventricle, connected to a pressure transducer, and inflated to yield a systolic pressure between 70 and 90 mmHg, with an end-diastolic pressure <5 mmHg. Pacing electrodes were secured to the epicardial surface, and the hearts were paced at 180 beats/min.To collect coronary venous effluent samples in the nuclear magnetic
resonance (NMR) experiments, the azygous vein and inferior vena cava
were ligated. Nested outer (Tygon, 
th-in. OD) and inner
(PE-90) tubes were threaded through the superior vena cava into the
right ventricle and out the pulmonary artery. Coronary venous effluent
was withdrawn from the right ventricle through side holes in the inner
tube using a peristaltic pump. When coronary flow was subsequently
decreased, ~90% of the coronary arterial inflow volume could be
collected in this manner. The hearts were then submerged in 37°C
perfusate in a 3-cm-diameter plastic, temperature-controlled,
watertight cylinder that was encircled with a solenoid-style
radio-frequency NMR coil. The heart setup was placed inside the magnet,
the radio-frequency coil was tuned to 81 MHz and matched, the gradient
coils were shimmed, and a fully relaxed NMR spectrum was acquired. On
completion of each experiment, wet ventricular weight was determined.
Experimental Protocols
Protocol 1. 5'-NT activity during two identical sequential
periods of underperfusion.
Seven rabbits were used to investigate 5'-NT activity during two
sequential and identical periods of underperfusion, each lasting 45 min. After baseline measurements, coronary perfusion (pump flow) was
decreased to ~0.24
ml · g
1 · min1
(~4-5% baseline) and held constant for 45 min, then flow was restored to baseline for 20 min. The flow was then again reduced to the
same level as during the first period of underperfusion for another 45 min, and after a 20-min reperfusion period the experiment was
terminated. NMR spectra and venous samples were acquired identically
during both periods of underperfusion.
Protocol 2. Time course experiments: effect of different lengths of underperfusion on 5'-NT activity. Fifteen rabbits were used in non-NMR experiments to investigate the effect of a 0-, 5-, 10-, 20-, or 40-min period of underperfusion on 5'-NT activity during a subsequent 40-min test period of underperfusion. The two sequential periods of underperfusion were separated by a 20-min period of reperfusion. The duration of the first period of underperfusion was 5 min (5/40, n = 3 hearts), 10 min (10/40, n = 3), 20 min (20/40, n = 3), or 40 min (40/40, n = 6). In all hearts the duration of the second (test) period of underperfusion was 40 min. The effect of 0 min of underperfusion in the first period (0/40) was studied using the first period of 40 min in the 40/40 group as the test period.
Protocol 3. Time control experiments. Four rabbits were used in NMR experiments to study the effect of time on coronary venous purine release and myocardial high-energy phosphates. Protocol 1 was followed, except that during the first 45-min period, perfusion was maintained at constant baseline levels. Thus a single 45-min period of underperfusion was studied after a 65-min (45 min + 20 min of reperfusion) period of baseline perfusion.
Protocol 4. Indicator-dilution experiments to determine venous
sampling delays.
Three rabbits were used for separate indicator-dilution experiments to
derive a transfer function that would accurately describe the delay and
dispersion of the venules, veins, right atrium, and the cannula on the
kinetics of the venous purine measurements. Briefly, a bolus of 50 µl
of indocyanine green dye (2.5 mg/ml), an impermeant tracer, was
injected into the aorta during low flow (~0.25
ml · g1 · min1).
Controls were performed to test for the influence of the bolus volume
on flow. The dye-dilution curve was detected with a densitometer that
had been calibrated as described previously (6). A mathematical model
that describes dispersive vascular flow (VASCOP, BTEX40) (16) was used
to derive an empirical transfer function from the dye curves with the
assumption of a spike input function. This transfer function was then
used to derive capillary purine concentrations for model analysis by
deconvoluting the venous effluent purine concentrations. This
transformation of the data was quite minor, because the mean transit
time of the vascular system was ~45 s, whereas the mean time of the
venous purine curves was much greater (~15 min).
Protocol 5. Effect of adenosine kinase and adenosine deaminase
inhibitors.
Three rabbits were used to examine the influence of adenosine kinase
(adenosine AMP) and adenosine deaminase (adenosine inosine) on
venous purine efflux during two sequential and identical periods of
severe coronary underperfusion. Protocol
1 was followed, except 5 µM
erythro-9-[2-hydroxy-3-nonyl]adenine (EHNA) and 2 µM
iodotubercidin were present in the perfusate during the entire protocol
to inhibit adenosine deaminase and adenosine kinase, respectively (21).
NMR Spectroscopy
High-energy phosphate data were obtained using a 4.7-T magnet (Bruker) and a CSI spectrometer (GE-Omega), as previously described (22). In every experiment, one fully relaxed spectrum was acquired before the protocol began by summing 40 free induction decays obtained every 20 s. The areas of the PCr and ATP peaks in the fully relaxed spectrum were obtained by integration using Omega software and averaging the
,
, and 
peaks of ATP. Partially saturated
31P spectra were acquired every 88 s by summing 32 free induction decays obtained every 2.7 s. The
partially saturated spectra were analyzed using an automated fitting
routine (12), and peak areas were expressed relative to the average
baseline values and to the area of an internal phenylphosphonic acid
standard. Intracellular Pi was
determined, and intracellular pH and the free intracellular Mg2+ concentration were calculated
as described previously (22). Because of interference by extracellular
Pi in the perfusion medium, baseline and reperfusion intracellular pH were assumed to equal 7.1 (24). Cytosolic free AMP concentration was calculated using creatine
kinase and adenylate kinase equilibrium expressions
(Eq. 1) adjusted to the measured
H+ and
Mg2+ concentrations
(Eq. 2). It was assumed that the
total concentration of PCr + creatine (Cr) decreased by 7% during each
of the two periods of underperfusion, as observed previously (22)
![[AMP] = <IT>K</IT><SUP>MK</SUP><SUB>eq</SUB> ⋅ [ATP] ⋅ <FENCE><FR><NU>[Cr]</NU><DE><IT>K</IT><SUP>CK</SUP><SUB>eq</SUB> ⋅ [PCr]</DE></FR></FENCE><SUP>2</SUP>](/content/vol274/issue2/fulltext/H529/img002.gif)
|
(1) |
|
(2) |
Determination of Venous Purines and Lactate
Baseline samples were collected every 2 min for an 8-min period. During underperfusion, samples of the effluent were initially collected every 2 min and, after 15 min, every 5 min, into ice-cold vials containing 10 µM EHNA to prevent enzymatic degradation of adenosine to inosine (21) during analysis. Because of a detection limit of 100 nM, baseline samples (15-20 ml) were first desalted and concentrated 85 times using Sep-Pak cartridges (Waters). The samples were passed through the cartridges and then eluted first with 2 ml of 70% methanol and then with 2 ml of phosphate buffer (10 mM, pH 5.4). The eluent (4 ml) was then evaporated and reconstituted with water to 200 µl. Effluent samples collected during underperfusion were diluted 1:5 with 10 µM EHNA because of the high purine concentrations. Samples (75 µl) were injected onto a C18 reverse-phase column (Upchurch, 5 µm, 250 × 3 mm) using a Hewlett-Packard high-performance liquid chromatography system. The mobile phase contained 26 mM ammonium acetate and 5% methanol, with pH adjusted to 5.0 with 2 N acetic acid. Two pumps were programmed for gradient elution with 70% methanol at a flow of 0.3 ml/min over 20 min. Absorbance of the column eluent was continuously monitored using a photodiode array detector at 258 nm and recorded, using 450 nm as the reference wavelength. Peaks were identified and quantified by comparison with retention times of external standards. Adenosine, inosine, and hypoxanthine were reliably detectable to 20 pmol. Total purine release was calculated by summing the purines released during each 2- or 5-min period for the entire 45-min period of underperfusion. The lactate concentration of the effluent was measured with a YSI Glucose Analyzer equipped with a lactate membrane.Model Analysis
A mathematical model of myocardial phosphoenergetics and nucleotide metabolism, as depicted in Fig. 1, was used to analyze the data. The model describes the intracellular concentrations of PCr, Cr, ATP, ADP, AMP, Pi, adenosine, and inosine, the enzymes creatine kinase, myokinase, cytosolic 5'-NT, adenosine kinase, and adenosine deaminase, the membrane transport of adenosine and inosine, and Pi and Cr in exchange with an interstitial region. The model consists of a set of ordinary differential equations based on the conservation of mass. The model, which describes parenchymal cell (cardiomyocyte), interstitial space, endothelial cell, and capillary regions, is an extension of a previous model that described only parenchymal cell and extracellular regions (22). Because it is not feasible at this stage of our knowledge of the underlying biology to accurately describe all processes whereby ATP is synthesized (oxidative phosphorylation, glycolysis) and hydrolyzed (e.g., Ca2+ and ion pumps, myofibril contraction, adenosinetriphosphatases, homeostasis), a simplification was used. A continuous function,
rATP, was therefore
defined in the parenchymal cell region as
|
(3) |
|
rATP = 0, the rates of
ATP synthesis and hydrolysis are precisely matched. rATP provides a
flexible means for describing the time course and extent of the net
energy imbalance experienced during underperfusion and reperfusion
(22). The parameters describing rATP were estimated empirically
using optimization procedures described below. It was assumed that,
under baseline conditions, rATP = 0. Thus the integral of rATP
over time in the intact system is equivalent to the total net
hydrolysis (if negative) of high-energy phosphate bonds during the
interval of the integration.
|
Permeability-surface area (PS)
products were used to describe linear membrane exchange of adenosine,
inosine, Pi, and creatine with an
interstitial region and transport between capillary and interstitial
regions. Enzyme dissociation constants and
PS products were taken from literature
values (22), with the exception of the Michaelis constant
(Km) and
maximal reaction velocity
(Vmax) of
5'-NT. We chose to use accepted
PS values, which are based on normal
flow, because exploratory modeling indicated that reduced PS values could fit the purine data
only when adenosine deaminase activities were unrealistically elevated,
whereas the 5'-NT and rATP parameters were minimally
influenced. The present model incorporated the recently observed
inhibition of adenosine kinase due to increased
Pi (5, 9) [i.e., inhibition
constant (Ki) of adenosine kinase for Pi = 3 mM].
To fit the model to the NMR and purine data from
protocol 1, an automated
least-squares optimization routine (SIMPLEX) was used to
simultaneously fit the PCr, ATP, adenosine, and inosine curves by
adjusting the Km
and Vmax of
5'-NT and the parameters of the rATP function. All other model
parameters were held constant during fitting or were changed according
to direct measurements (flow, intracellular pH,
Mg2+). Because the model does
not include hypoxanthine, the coronary venous hypoxanthine data were
added to the inosine data for fitting with the model inosine data.
Means ± SE of the parameter estimates were obtained by fitting the
data from each individual experiment. Initial concentrations of ATP
(6.08 mM), PCr (10.6 mM), and Cr (13.2 mM) used in the modeling were
taken from biochemical measurements of freeze-clamped hearts performed
previously in this laboratory (22). The initial intracellular
concentration of Pi was assumed to
be 0.4 mM (11), and the initial pH was assumed to be 7.1 (24). The
equilibrium constant for the creatine kinase reaction was adjusted on
the basis of NMR measurements of intracellular pH and
Mg2+ (23).
To analyze the data from protocol 2,
where only purine data and no NMR data were obtained, the venous purine
effluent data from the second (test) period of underperfusion were
deconvoluted, as described above, to derive capillary concentrations of
inosine (+hypoxanthine) and adenosine. The inosine and adenosine curves were then fit by the model by optimizing for the parameters describing 5'-NT kinetics and the rATP function. The robustness of this procedure was confirmed by testing the fit obtained by analyzing the
purine data only from the protocol 1 experiments.
Statistical Analysis
Values are means ± SE. Mean values of the model results (i.e., Vmax and Km of 5'-NT and therATP parameters) were determined from the separate model
solutions for each experiment. Statistical significance of differences
(P 
0.05) was determined by using the nonparametric Mann-Whitney test for statistical variation.
| |
RESULTS |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
Attenuation of Purine Efflux During a Second Period of Underperfusion
The NMR, purine, and metabolic data for the two successive and identical periods of underperfusion (5% baseline flow) in the protocol 1 experiments are shown in Fig. 2. PCr fell rapidly to 2.3 mM after the onset of the first period of underperfusion and to 2.7 mM in the second period, recovering quickly toward baseline during both periods of reperfusion. The ATP concentration fell from 6.1 to 3.3 mM in the first period but only to 2.6 mM in the second period and did not recover toward baseline values during either reperfusion period. Therefore, the decrease in ATP was much reduced in the second period of underperfusion compared with the first period. During the first period of underperfusion, coronary effluent purine release rate (i.e., venous concentration × flow) peaked within the first 10 min at a rate of 36 nmol · g1 · min1,
resulting in a total purine release of 934 ± 161 nmol/g during the
entire 45 min of underperfusion. Purine release was severely attenuated
(75% decrease, total purine release = 196 ± 26 nmol/g) during the
second period of underperfusion and showed different kinetics, with a
slower peak after 12 min, which remained elevated. The relative
contributions of the various purines (adenosine, inosine, hypoxanthine)
to the total purine efflux, however, remained quite similar. Calculated
cytosolic ADP rose to 290 µM during the first period and to 120 µM
during the second period of underperfusion. Cytosolic AMP rose sharply
to 14 µM during the first period of underperfusion and only to 4 µM
during the second period. During both periods,
Pi rose to similar levels,
intracellular pH fell to 6.5-6.6, and lactate release rate was
similar. These data indicate that the glycolytic rate was comparable
during both periods of underperfusion. Left ventricular pressure fell
to undetectable levels within 2 min of underperfusion during both
periods, accompanied by an incomplete recovery during reperfusion.
|
Model Analysis
The decreased purine efflux in the second period of underperfusion is suggestive of decreased 5'-NT activity. However, because ATP fell to a lesser degree and ADP and AMP concentrations were lower in the second period than in the first period, this would also be expected to cause decreased purine efflux by mass action. Analysis with a mathematical model was, therefore, performed to differentiate between these two effects. Simultaneous model fits to the PCr, ATP, adenosine, and inosine data using the four-region model of myocardial energetics and enzymatic kinetics are given in Fig. 3. Separate fits were obtained for the two periods of underperfusion, resulting in different values for the parameters of therATP
function (ATP synthesis 
ATP hydrolysis) and values for the
Vmax and
Km for the enzyme
5'-NT, whereas all other model parameters were held constant
(Table 1). The integral of rATP, reported in Table 1, represents the total net high-energy phosphate breakdown during the period of underperfusion. The analysis indicates that the decreased purine efflux in the second period of underperfusion was caused by decreased 5'-NT activity as well as decreased AMP concentration.
|
|
The relation between calculated cytosolic AMP concentration and measured total coronary purine efflux (adenosine + inosine + hypoxanthine) for the two periods of underperfusion is shown in Fig. 4, top. Minutes of underperfusion are indicated by the numbers for the first period of underperfusion. The relations form loops (hysteresis) because of membrane transport, diffusion, and convection of the purines, which cause delay and dispersion. Although AMP concentrations increased to higher levels during the first period of underperfusion, in the range where AMP concentrations were similar (<4 µM), purine release was three- to sevenfold higher in the first than in the second period of underperfusion. The model predictions of the relationship between cytosolic free AMP and the vascular purine release are in good agreement with the data, passing within the measurement uncertainty bounds for almost every point (Fig. 4, middle). The results show that the observed degree of hysteresis is to be expected, given the relatively rapid kinetics of cytosolic AMP and realistic values describing membrane transport, diffusion, and convection of the purines. Because the model solutions in Figs. 3 and 4 are identical, representing the best overall fit to the global data set, the fit is powerfully constrained by the requirement for internal self-consistency.
|
Kinetics of Cytosolic 5'-NT
The rate of the cytosolic 5'-NT reaction during the first and second periods of underperfusion, as calculated by the model, is shown in Fig. 4, bottom. The high-energy phosphate and purine data from the first period of underperfusion were best fit with a Vmax of 140 nmol · g1 · min1
and a Km of 2.3 µM for the enzyme 5'-NT. The second period was best fit with a
decreased (P 0.05)
Vmax of 67 nmol · g1 · min1
and a Km of 1.3 µM (Table 1). The model solutions shown in Figs. 3 and 4 were
obtained by fitting the pooled data set. However, for statistical
comparison, the parameter estimates described in the text and Table 1
were based on fits of the individual experiments.
Time Course of Downregulation of 5'-NT
The observation of decreased 5'-NT activity (Vmax) in the second period of underperfusion is interpreted as indicating that the decrease in 5'-NT activity occurred during the latter portion of the first period of underperfusion and that the decrease in activity persisted through the intervening period of reperfusion. To determine the length of underperfusion necessary to induce a decrease in the Vmax of 5'-NT, special time course experiments were performed (protocol 2). Here, the duration of the first period of underperfusion (95% flow reduction) was varied (0, 5, 10, 20, 40 min), while venous purines were measured in the standardized second (test) period of underperfusion (40 min; Fig. 5). No NMR data were obtained. A 5-min period of underperfusion decreased total purine efflux during the test 40-min period of underperfusion by 8% (no significant difference). A 10-min period significantly (P 0.05) decreased total purine efflux by 40%, and a 20-min period
significantly decreased total purine efflux by 51%. Model analysis of
the venous purine data from the time course experiments showed a
significantly decreased
Vmax of
5'-NT after a 20-min period of underperfusion (Fig.
6). Although substrate AMP concentrations were not measured in these experiments, model analysis suggested that
there was decreased AMP after 10 min of underperfusion, accounting for
the decreased purine efflux (Fig. 5). However, only after 20 min of
underperfusion was there a decrease in the
Vmax of
5'-NT (Fig. 6).
|
|
To test the unlikely possibility that 1 h of perfusion itself caused a decrease in 5'-NT activity, separate time control experiments (protocol 3) were performed. A single 45-min period of underperfusion following a 65-min period of normal perfusion exhibited a purine release of 782 ± 96 nmol/g, which was not statistically different from that during a single period of underperfusion without an extended period of normal perfusion (i.e., 934 ± 161 nmol/g). Model analysis showed no decrease in the activity of 5'-NT in the time control experiments (Table 1). These experiments clearly indicate that the decreased activity of 5'-NT observed in the second period of underperfusion, as found in protocol 1, was due to underperfusion and was not simply due to time.
5'-NT Downregulation or Activation of Adenosine Kinase?
Studies of guinea pig heart and rabbit cardiomyocytes have shown a high rate of substrate cycling between AMP and adenosine (Fig. 1) catalyzed by the enzymes 5'-NT and adenosine kinase (1, 21, 32). During normoxia the recycling of adenosine to AMP via adenosine kinase is essential for the maintenance of low, nonvasodilatory concentrations of adenosine. New findings show that endogenous inhibition of adenosine kinase plays an important role in elevating adenosine levels during ATP depletion (5, 9). To test the unlikely possibility that the decrease in venous purines in the second period of underperfusion was unexpectedly due to an increase in the activity of adenosine kinase, we performed identical successive underperfusion experiments (95% flow decrease) in the absence and presence of iodotubercidin and EHNA, specific inhibitors of adenosine kinase (26) and adenosine deaminase (21), respectively. The results, shown in Table 2, indicate that, in the absence of iodotubercidin and EHNA, purine release was attenuated by 75% during the second period of underperfusion and by 69% in the presence of iodotubercidin. If there was an unexpected increase in adenosine recycling via activation of adenosine kinase during the second period of underperfusion, then there should have been reduced attenuation of the purine release in the presence of iodotubercidin. However, the results show no detectable difference. These data show that the attenuated purine release during the second period of underperfusion was not due to an upregulation of adenosine kinase.
|
| |
DISCUSSION |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
The major finding of the present study was a decreased activity (Vmax) of cytosolic 5'-NT in the second of two identical sequential periods of coronary underperfusion. The downregulation of 5'-NT was slow in onset, requiring 20 min of underperfusion, and persistent during a 20-min period of reperfusion. Although previous studies reported decreased AMP hydrolysis during ischemia, they did not distinguish between decreased activity of 5'-NT and decreased substrate AMP concentrations, which also occurs. Here we report the first quantitative description of the downregulation of 5'-NT in vivo while accounting for the potentially confounding effects of decreased AMP concentration. Downregulation of 5'-NT may be important for preserving myocardial adenine nucleotide stores during sustained coronary underperfusion.
Energy Imbalance During Underperfusion
The metabolic data on PCr, Pi, pH, and lactate presented in Fig. 2 were similar for both periods of underperfusion. However, ATP depletion and venous purine concentration, as well as the calculated intracellular ADP and AMP concentrations, were lower in the second than in the first period of underperfusion. These findings indicate that the degree of energy imbalance was less during the second than during the first period. The model estimate was that the total high-energy phosphate depletion (rATP) in the second
period of underperfusion was 66% of that of the first period (Table
1).
The estimate of the integral of rATP of 18.9 mM (Table 1)
means that there was a net hydrolysis of high-energy phosphate bonds of
18.9 mM during the first period of underperfusion. The total number of
high-energy phosphate bonds available is equal to [PCr] + 3 × [ATP] + 2 × [ADP] + [AMP] = 10.6 + 3 × 6.08 + 2 × 0.07 + 0.0007 = 29.0 mM during baseline conditions (22), where [PCr],
[ATP], [ADP], and [AMP] represent
PCr, ATP, ADP, and AMP concentrations. Thus 65% of the total
high-energy phosphate bonds were hydrolyzed by the end of the first
period of underperfusion. On reperfusion, the high-energy phosphate
bonds were partly restored, and during the second period of
underperfusion there was a net hydrolysis of 40% of the phosphate
bonds.
Because the oxygen delivery rate was the same for both periods of underperfusion and the rates of glycolysis, as evidenced by the lactate release rates and the pH time courses (Fig. 2), were similar, the rates for ATP synthesis were probably similar for the two periods. Therefore, it is likely that the rate of ATP hydrolysis was lower in the second than in the first period of underperfusion. Because Murry et al. (25) also reported slower utilization of energy as a result of a previous ischemic episode, this may be a general phenomenon. If this is so, then purine production during ischemia will normally be reduced as a result of a preceding period of ischemia because of lower substrate AMP concentrations. This effect may have confounded previous studies of 5'-NT activity during ischemia (see Relation to Other Studies of Ischemia).
Downregulation of 5'-NT
The direct evidence for the downregulation of 5'-NT is the vascular purine efflux-AMP relationship, which is depicted in Fig. 4; it shows that, at similar intracellular concentrations of AMP, coronary purine efflux was reduced in the second period of underperfusion. This qualitative finding is entirely model independent. However, without the model analysis, it would not be possible to arrive at a quantitative estimate of the degree of inhibition of 5'-NT. Because the model assumes Michaelis-Menten kinetics for 5'-NT, enzyme velocity curves of the form shown in Fig. 4, bottom, were expected.The model assumes constant values for the
Vmax and
Km of 5'-NT
in each period of underperfusion, and on the basis of this assumption,
accurate fits of the data were obtained (Figs. 3 and 4). However, the
results in Fig. 6 imply that the
Vmax of
5'-NT was decreased after 20 min of underperfusion in the first
period. The probable explanation for this apparent discrepancy is that during the first period of underperfusion, in the
protocol 1 experiments, the data
provide little sensitivity for changes in the activity of 5'-NT
after 20 min, since by that time, AMP concentrations were decreased to
low values. Therefore, the decrease in
Vmax of
5'-NT after 20 min of underperfusion, in the
protocol 1 experiments, would be
expected to have little effect on the model fit or the data. This was
confirmed by modifying the model to impose a linear decrease in the
Vmax of
5'-NT after 10 min of underperfusion. When this version of the
model was used to fit the data, there were no appreciable changes in
the estimates of the parameters describing 5'-NT kinetics and the
function rATP.
The model also assumed that the enzyme adenosine kinase (adenosine AMP) was inhibited during underperfusion, as has been demonstrated
during hypoxia (5, 9). The mechanism of inhibition was assumed to be
increased Pi concentration on the
basis of in vitro enzyme assay studies (9). The absolute values of
Vmax and
Km for
5'-NT estimated in the present study depend on the degree of
inhibition of adenosine kinase. However, even when the modeling assumed
no inhibition of adenosine kinase, there was a decrease in 5'-NT
activity in the second period of underperfusion similar to that
reported in Table 1 (results not shown).
It is interesting to note that the relation between AMP concentration and venous purine release showed such marked hysteresis. That such hysteresis is also predicted by the model indicates that the hysteresis can be explained by membrane and endothelial cell transport delays. This result demonstrates the importance of collecting complete time course data in studies of the regulation of myocardial adenosine formation. It appears that assumptions of a steady-state response to ischemia should be avoided.
Possible Regulatory Mechanisms
The primary pathway for 5'-AMP hydrolysis to adenosine in heart is by cytosolic 5'-NT (27). Darvish et al. (4) report the activity of extracted cytosolic 5'-NT in the presence of 0.1 mM AMP and physiological concentrations of ADP and Mg2+ to be ~150 nmol · min1 · g1,
which is in good agreement with the estimated in situ
Vmax of 140 nmol · min1 · g1
reported here. Other in vitro studies have provided evidence that
5'-NT is inhibited by high concentrations of protons
(H+) (4) and
Pi (14, 27) and that ADP increases
the apparent affinity of 5'-NT for its substrate, AMP (33). An
innovative study by Bak and Ingwall (2) showed the effects of acidosis on purine efflux by subjecting isolated hearts to hypoxia or global ischemia and concluded that
H+ is a potent inhibitor of
5'-NT. In the present study the activity of 5'-NT was
downregulated after 20 min of underperfusion. Because the intracellular
concentrations of H+ and
Pi were elevated during this
period, it is possible that either of these acts as a regulatory
effector for cytosolic 5'-NT. It would seem unlikely, however,
that inhibition by protonation or elevated
Pi levels would be slow in onset,
since H+ and
Pi rose quickly at the onset of
underperfusion. It is also unlikely that the reperfusion period plays a
regulatory role in the downregulation of 5'-NT, since the time
course experiments (protocol 2) had
identical periods of reperfusion. Each group was exposed to a 20-min
period of reperfusion, yet 5'-NT remained normally active for the
5/40 and 10/40 groups.
Whereas previous in vitro studies implicated allosteric interactions for the regulation of 5'-NT, it is possible that 5'-NT downregulation during sustained underperfusion is not mediated by the known allosteric regulators, since it is slow in onset and persistent, even through a 20-min period of reperfusion. One may speculate that an inhibitory substance may be binding tightly to the enzyme, thereby decreasing its catalytic availability, or that enhanced turnover/depressed de novo synthesis may be diminishing the amount of reactive 5'-NT.
Relation to Other Studies of Ischemia
Because the production of myocardial purines during ischemia depends on 5'-NT and adenosine kinase activity and the cytosolic AMP concentration, it has been difficult to describe the regulation of myocardial purine production. For example, Reimer et al. (28) observed no cumulative depletion of the adenine nucleotide pool after repeated occlusions and listed four possible explanations for the preservation of ATP. These included a depressed utilization of high-energy phosphates and the loss or inhibition of 5'-NT. Their data suggested that both phenomena were involved, yet at that stage it was impossible to distinguish and quantify the two effects.It has been reported that ischemic preconditioning is associated with an increase as well as a decrease in 5'-NT activity. On the basis of enzyme extraction measurements, Kitakaze et al. (17, 18, 20) found that ischemic preconditioning increases 5'-NT activity during ischemia and, thus, leads to an elevated adenosine release during reperfusion, thereby causing beneficial effects. Others, however, dispute this view and claim rather that 5'-NT is downregulated by preconditioning. For example, a number of studies in the intact heart (13, 28, 29, 31) have reported less ATP breakdown and less AMP and nucleoside accumulation during repetitive ischemic periods than during prolonged ischemic periods. In addition, Murry et al. (25) observed decreased tissue accumulation of ADP, AMP, and adenosine during 40 min of ischemia in preconditioned hearts and reported decreased energy metabolism due to ischemic preconditioning. Furthermore, Bradamante et al. (3) observed stable ATP levels during repetitive short periods of ischemia in rat heart concomitant with steadily decreasing rates of purine efflux. Because all these studies tend to show decreased high-energy phosphate depletion during repetitive periods of ischemia, decreased purines are to be expected because of lower AMP concentrations. This effect would tend to confound studies in which decreased purines in repetitive ischemia were attributed to inhibition of 5'-NT. Therefore, we believe that the present study is the first to rigorously demonstrate downregulation of 5'-NT due to repetitive ischemia, independently of the altered energetic state.
In conclusion, we are able to report the downregulation of 5'-NT during sustained severe coronary underperfusion. Because a potentially confounding factor in the study of the regulation of 5'-NT in the intact heart is the degree of energetic imbalance, it was necessary to obtain high-energy phosphate and purine efflux data and analyze these with an integrative model to clearly demonstrate decreased activity of 5'-NT during underperfusion. Analysis of the data indicates a regulatory mechanism that has a slow 20-min onset and persists, even through a 20-min period of reperfusion. Downregulation of 5'-NT may be advantageous for long-term tissue survival during sustained underperfusion by preserving adenine nucleotide stores.
| |
ACKNOWLEDGEMENTS |
|---|
The authors thank Rodney Gronka for invaluable expertise in conducting the NMR experiments, Tim Krell and Judy Pierce for expert assistance in performing the indicator-dilution experiments and analysis, and Greg Crowther for experimental assistance.
| |
FOOTNOTES |
|---|
Deceased 15 July 1997.
This study was supported by National Institutes of Health Grants HL-51152 and RR-01243.
Address for reprint requests: L. A. Gustafson, Laboratory for Physiology, Vrije Universiteit, Van der Boechorststraat 7, 1081 BT Amsterdam, The Netherlands.
Received 18 July 1997; accepted in final form 16 October 1997.
| |
REFERENCES |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
1.
Arch, J. R. S.,
and
E. A. Newsholme.
The control of the metabolism and the hormonal role of adenosine.
Essays Biochem.
14:
82-123,
1978[Medline].
2.
Bak, M. I.,
and
J. S. Ingwall.
Acidosis during ischemia promotes adenosine triphosphate resynthesis in postischemic rat heart. In vivo regulation of 5'-nucleotidase.
J. Clin. Invest.
93:
40-49,
1994.
3.
Bradamante, S.,
F. Piccinini,
C. Delu,
M. Janssen,
and
J. W. de Jong.
NMR evaluation of changes in myocardial high energy metabolism produced by repeated short periods of ischemia.
Biochim. Biophys. Acta
1243:
1-8,
1995[Medline].
4.
Darvish, A.,
R. W. Pomerantz,
P. G. Zografides,
and
P. J. Metting.
Contribution of cytosolic and membrane-bound 5'-nucleotidase to cardiac adenosine production.
Am. J. Physiol.
271 (Heart Circ. Physiol. 40):
H2162-H2167,
1996
5.
Decking, U. K.,
G. Schlieper,
K. Kroll,
and
J. Schrader.
Hypoxia-induced inhibition of adenosine kinase potentiates adenosine release.
Circ. Res.
81:
154-164,
1997
6.
Deussen, A.,
and
J. B. Bassingthwaighte.
Modeling [15O]oxygen tracer data for estimating oxygen consumption.
Am. J. Physiol.
270 (Heart Circ. Physiol. 39):
H1115-H1130,
1996
7.
Downey, J. M.,
G. S. Liu,
and
J. D. Thorton.
Adenosine and the anti-infarct effects of preconditioning.
Cardiovasc. Res.
27:
3-8,
1993
8.
Forman, M. B.,
C. E. Velasco,
and
E. K. Jackson.
Adenosine attenuates reperfusion injury following regional myocardial ischemia.
Cardiovasc. Res.
27:
9-17,
1993
9.
Gorman, M. W.,
M. X. He,
C. S. Hall,
and
H. V. Sparks.
Inorganic phosphate as a regulator of adenosine formation in the isolated guinea pig heart.
Am. J. Physiol.
272 (Heart Circ. Physiol. 41):
H913-H920,
1997
10.
Goto, M.,
M. V. Cohen,
D. G. L. van Wylen,
and
J. M. Downey.
Attenuated purine production during subsequent ischemia in preconditioned rabbit myocardium is unrelated to the mechanism of protection.
J. Mol. Cell Cardiol.
28:
447-454,
1996[Medline].
11.
He, M. X.,
M. W. Gorman,
G. D. Romig,
R. A. Meyer,
and
H. V. Sparks.
Adenosine formation and energy status during hypoperfusion and 2-deoxyglucose infusion.
Am. J. Physiol.
260 (Heart Circ. Physiol. 29):
H917-H926,
1991
12.
Heineman, F. W.,
J. Eng,
B. A. Berkowitz,
and
R. S. Balaban.
NMR spectral analysis of kinetic data using natural lineshapes.
Magn. Reson. Med.
13:
490-497,
1990[Medline].
13.
Henrichs, K. J.,
H. Matsuoka,
and
J. Schaper.
Influence of repetitive coronary occlusions on myocardial adenine nucleosides, high energy phosphates and ultrastructure.
Basic Res. Cardiol.
82:
557-565,
1987[Medline].
14.
Itoh, R.,
J. Oka,
and
H. Ozasa.
Regulation of rat heart cytosol 5'-nucleotidase by adenylate energy charge.
Biochem. J.
235:
847-851,
1986[Medline].
15.
Jennings, R. B.,
H. K. Hawkins,
J. E. Lowe,
M. L. Hill,
S. Klotman,
and
K. A. Reimer.
Relation between high energy phosphate and lethal injury in myocardial ischemia in the dog.
Am. J. Pathol.
92:
203-216,
1978.
16.
King, R. B.,
A. Deussen,
G. R. Raymond,
and
J. B. Bassingthwaighte.
A vascular transport operator.
Am. J. Physiol.
265 (Heart Circ. Physiol. 34):
H2196-H2208,
1993
17.
Kitakaze, M.,
M. Hori,
and
T. Kamada.
Role of adenosine and its interaction with -adrenoceptor activity in ischaemic and reperfusion injury of the myocardium.
Cardiovasc. Res.
27:
18-27,
1993[Medline].
18.
Kitakaze, M.,
M. Hori,
T. Morioka,
T. Minamino,
S. Takashima,
H. Sata,
Y. Shinozaki,
M. Chujo,
H. Mori,
and
M. Inoue.
1-Adrenoceptor activation mediates the infarct size-limiting effect of ischemic preconditioning through augmentation of 5'-nucleotidase activity.
J. Clin. Invest.
93:
2197-2205,
1994.
19.
Kitakaze, M.,
M. Hori,
S. Takashima,
H. Sato,
M. Inoue,
and
T. Kamada.
Ischemic preconditioning increases adenosine release and 5'-nucleotidase activity during myocardial ischemia and reperfusion in dogs. Implications for myocardial salvage.
Circulation
87:
208-215,
1993
20.
Kitakaze, M.,
T. Minamino,
K. Node,
K. Komamura,
and
M. Hori.
Activation of ecto-5'-nucleotidase and cardioprotection by ischemic preconditioning.
Basic Res. Cardiol.
91:
23-26,
1996[Medline].
21.
Kroll, K.,
U. K. M. Decking,
K. Dreikorn,
and
J. Schrader.
Rapid turnover of the AMP-adenosine metabolic cycle in the guinea pig heart.
Circ. Res.
73:
846-856,
1993
22.
Kroll, K.,
D. J. Kinzie,
and
L. A. Gustafson.
Open system kinetics of myocardial phosphoenergetics during coronary underperfusion.
Am. J. Physiol.
272 (Heart Circ. Physiol. 41):
H2563-H2576,
1997
23.
Mallet, R. T.,
Y. H. Kang,
N. Mukohara,
and
R. Bunger.
Use of cytosolic metabolite patterns to estimate free magnesium in normoxic myocardium.
Biochim. Biophys. Acta
1139:
239-247,
1992[Medline].
24.
Matherne, G. P.,
J. P. Headrick,
S. Berr,
and
R. M. Berne.
Metabolic and functional responses of immature and mature rabbit hearts to hypoperfusion, ischemia, and reperfusion.
Am. J. Physiol.
264 (Heart Circ. Physiol. 33):
H2141-H2153,
1993
25.
Murry, C. E.,
V. J. Richard,
K. A. Reimer,
and
R. B. Jennings.
Ischemic preconditioning slows energy metabolism and delays ultrastructural damage during a sustained ischemic episode.
Circ. Res.
66:
913-931,
1990
26.
Newby, A. C.
The interaction of inhibitors with adenosine metabolizing enzymes in intact isolated cells.
Biochem. Pharmacol.
30:
2611-2615,
1981[Medline].
27.
Newby, A. C.
The pigeon heart 5'-nucleotidase responsible for ischaemia-induced adenosine formation.
Biochem. J.
253:
123-130,
1988[Medline].
28.
Reimer, K. A.,
C. E. Murry,
I. Yamasawa,
M. L. Hill,
and
R. B. Jennings.
Four brief periods of myocardial ischemia cause no cumulative ATP loss or necrosis.
Am. J. Physiol.
251 (Heart Circ. Physiol. 20):
H1306-H1315,
1986
29.
Swain, J. L.,
R. L. Sabina,
J. J. Hines,
J. C. Greenfield, Jr.,
and
E. W. Holmes.
Repetitive episodes of brief ischaemia (12 min) do not produce a cumulative depletion of high energy phosphate compounds.
Cardiovasc. Res.
18:
264-269,
1984[Medline].
30.
Vary, T. C.,
E. T. Angelakos,
and
S. W. Schaffer.
Relationship between adenine nucleotide metabolism and irreversible ischemic tissue damage in isolated perfused rat heart.
Circ. Res.
45:
218-225,
1979[Abstract].
31.
Vuorinen, K.,
K. Ylitalo,
K. Peuhkurinen,
P. Raatikainen,
A. Ala Rami,
and
I. E. Hassinen.
Mechanisms of ischemic preconditioning in rat myocardium. Roles of adenosine, cellular energy state, and mitochondrial F1F0-ATPase.
Circulation
91:
2810-2818,
1995
32.
Wagner, D. R.,
F. Bontemps,
and
G. van den Berghe.
The AMP-adenosine cycle is active during normoxia and impaired in ATP depletion in isolated rabbit cardiomyocytes.
Adv. Exp. Med. Biol.
370:
323-326,
1994[Medline].
33.
Yamazaki, Y.,
V. L. Truong,
and
J. M. Lowenstein.
5'-Nucleotidase I from rabbit heart.
Biochemistry
30:
1503-1509,
1991[Medline].
This article has been cited by other articles:
|
|
 |
|
  D. Merkus, B. Houweling, M. van Vliet, and D. J. Duncker Contribution of KATP+ channels to coronary vasomotor tone regulation is enhanced in exercising swine with a recent myocardial infarction Am J Physiol Heart Circ Physiol, March 1, 2005; 288(3): H1306 - H1313. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. V. Gourine, Q. Hu, P. R. Sander, A. I. Kuzmin, N. Hanafy, S. A. Davydova, D. V. Zaretsky, and J. Zhang Interstitial purine metabolites in hearts with LV remodeling Am J Physiol Heart Circ Physiol, February 1, 2004; 286(2): H677 - H684. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. P. Headrick, B. Hack, and K. J. Ashton Acute adenosinergic cardioprotection in ischemic-reperfused hearts Am J Physiol Heart Circ Physiol, November 1, 2003; 285(5): H1797 - H1818. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. F. Torchiana, A. J. Vine, K. O. Shebani, H. L. Kantor, J. S. Titus, C.-Z. Lu, W. M. Daggett, and G. A. Geffin Cardioplegia and ischemia in the canine heart evaluated by 31P magnetic resonance spectroscopy Ann. Thorac. Surg., July 1, 2000; 70(1): 197 - 205. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. I. Kuzmin, V. L. Lakomkin, V. I. Kapelko, and G. Vassort Interstitial ATP level and degradation in control and postmyocardial infarcted rats Am J Physiol Cell Physiol, September 1, 1998; 275(3): C766 - C771. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
) and ATP (
).
B: calculated cytosolic ADP (
) and
AMP (
), adenosine (
) data in protocol 1 experiments using 4-region model of myocardial energetics and enzymatic
kinetics (Fig. 
