- Email: [email protected]
Significance of release of adenosine triphosphate and adenosine induced by hypoxia or adrenaline in perfused rat heart
j Mol Cell Cardio119, 187-197 (1987)
Significance of Release of Adenosine Triphosphate and Adenosine Induced by Hypoxia or Adrenaline in Perfused Rat Heart C. Vial, 1 P. Owen, L. H. Opie,* and D. Posel
MRC Ischaemic Heart Disease Research Unit, Department of Medicine, University of Cape Town Medical School, Observatory, 7925, South Africa and 1Department of Biochemistry, University of Lyon, France (Received 17 October 1984, acceptedin revisedform 3 November 1986) C. VIAL, P. OWEN, L. H. OP1E AND D. POSEL. Significance of Release of Adenosine Triphosphate and Adenosine Induced by Hypoxia or Adrenaline in Perfused Rat Heart. Journalof Molecularand CellularCardiology(1987) 19, 187 197. The status ofATP as a possible coronary vasodilator remains poorly understood. The onset ofhypoxia induced a rapid and transient increase of the ATP concentration in the coronary effluent of the isolated perfused rat heart from 0.8 q- 0.2 nM to the average peak value of 1.3 _+ 0.2 nM (P < 0.01) at 2 4- 0.5 rain; at the same time the coronary flow increased 2-fold so that the rate of ATP release increased from 10.2_ 2.9 to 21.4 ___4.2 pmol/g/min (P < 0.005). Hypoxia also produced a peak rate release of adenosine of 93 __ 5 nu/g/ rain occurring only after the peak increase of coronary flow and also after the peak release of ATP; at peak coronary flow, however, the adenosine concentration was sufficient for vasodilation (0.31 4- 0.19/tu). Peak release of ATP and of adenosine preceded that of lactate dehydrogenase. 10 -6 M adrenaline induced a rapid increase of coronary flow and release of ATP, the concentration of which rose from 0.9 4- 0.3 nM to an average peak of 1.7 _+ 0.2 nM (P < 0.01) at 2 4- 0.3 min. The rate of increase of ATP in the coronary effluent paralleled the rate of early rise of coronary flow, yet adenosine had also risen to vasodilatory values (0.28 4- 0.5 tiM). The absolute changes in the measured concentrations of ATP in the coronary effluent were more variable and 1000 x less in concentration than those of adenosine. Hence coronary dilation could be explained by adenosine without involving ATP, although an additional vasodilatory role for ATP could not be excluded, especially in the early phases of vasodilation. In one condition, hypoxic K-arrested hearts, the increase in coronary flow could not be linked to release of either adenosine or ATP. The changes in concentrations of potential vasodilators measured in the coronary effluent do not necessarily reflect changes in the interstitial fluid. KEy WORDS: ATP, Hypoxia; Vasodilation; Adenosine; Adrenaline.
Introduction A l t h o u g h it is w i d e l y h e l d t h a t cell m e m branes are impermeable to a d e n o s i n e - 5 ' t r i p h o s p h a t e ( A T P ) ( A n t o n i et al. 1960), several workers have detected ATP in the coronary effluent of isolated perfused hearts (Clemens and Forrester, 1980; Doyle and F o r r e s t e r , 1 9 8 5 ; N a y l e r et al., 1979; P a d d l e a n d B u r n s t o c k , 1974). F u r t h e r m o r e , P a d d l e a n d B u r n s t o c k (1974) a n d C l e m e n s a n d F o r r e s t e r (1980) h a v e p o s t u l a t e d t h a t t h e A T P r e l e a s e f r o m t h e m y o c a r d i u m is p a r t o f t h e c o r o n a r y v a s o d i l a t o r y r e s p o n s e to h y p o x i a . The origin of the ATP might be either from 'purinergic nerves' (Paddle and Burnstock, 1974) o r f r o m m y o c y t e s ( F o r r e s t e r a n d W i l l i a m s , 1977) o r c a p i l l a r y e n d o t h e l i a l ceils (Nees a n d G e r l a c h , 1983). T h e q u e s t i o n arises
whether or not ATP release has a physiological f u n c t i o n o r is j u s t a c o n s e q u e n c e o f a g e n e r a l i z e d loss o f m e m b r a n e i n t e g r i t y . I n t h e l a t t e r case, A T P r e l e a s e c o u l d o c c u r i n a pattern different from that of an intracellular m a r k e r e n z y m e s u c h as l a c t a t e d e h y d r o g e n a s e ( L D H ) . A s e c o n d q u e s t i o n is w h e t h e r A T P might have a vasodilator function; hence r a t e s o f r e l e a s e o f A T P i n t o t h e c o r o n a r y efflue n t w e r e c o m p a r e d w i t h t h e i n c r e a s e o f coro n a r y flow in r e s p o n s e to h y p o x i a o r adrenaline. The pattern of release of ATP was compared with that of adenosine, an establ i s h e d c o r o n a r y v a s o d i l a t o r ( B e r n e , 1980).
Materials and Methods H e a r t s f r o m m a l e L o n g - E v a n s r a t s (250 to 350 g, fresh h e a r t w t 0.9 to 1.2 g) w e r e
* To whom all correspondence should be addressed. 0022-2828/87/020187 + 11 $03.00/0
9 1987 Academic Press Inc. (London) Limited
188
C. V i a l et al.
arrested in icecold saline and then perfused by the Langendorff technique with KrebsHenseleit bicarbonate buffer pH 7.4 (millimolar concentrations: NaC1 118.5, N a H C O 3 25, KC1 4.7, K H z P O 4 1.2, MgSO4 1.2, CaC1 z 2.5, glucose 11.1) equilibrated with a mixture of 95% oxygen and 5% COz (Po2 from 500 to 600 mmHg) at 37~ and a constant perfusion pressure of 100 c m H 2 0 . Hypoxia was initiated by switching over to buffer equilibrated with 95% Nz and 5% CO2, the aortic Po2 fell to 25 to 70 m m H g within 15 s. In some experiments, hearts were arrested from the start of the perfusion by increasing the K + concentration to 24 mM and correspondingly decreasing the Na + concentration. Adrenaline (Merck) kept under nitrogen was dissolved in dilute hydrochloric acid (pH 2.0) to make a 2 mM solution which was protected from light. Immediately before the experiment, the solution was further diluted to 10 -6 i with preoxygenated warmed Krebs buffer and was protected from light with aluminium foil during the experiment. For ATP and adenosine measurements the coronary effluent collected during a 10 s period was immediately frozen in liquid nitrogen. Lactate dehydrogenase (LDH) activity measured in the perfusate (Wroblewski and La Due, 1955) was expressed as milliunits per gram wet weight per minute at 25~ ATP in the perfusate was estimated by the luciferinluciferase method using an Aminco-Bowman spectrofluorometer. A 0.15 ml sample was added to 0.5 ml of the assay mixture so that the final concentrations of the reactants in the cuvette were: 17.6 m i Tris-HC1, 0.175 mM EDTA, 4.4 mM magnesium acetate, 0.4 mra luciferin (Sigma), 25 #g/ml luciferase (Sigma), pH 7.75. Adenosine was measured by high performance liquid chromatography, by a modification of the method of Ely et al. (1983). Lowest concentration detectable was 0.1 # i . Coronary flow was measured by a timed collection of the coronary effluent.
Myocardial mechanical performance In an additional series of hearts, left ventricular (LV) pressure was measured by a ventricular catheter piercing the LV wall (Neely
et al., 1967). The catheter was connected to a pressure transducer and left ventricular diastolic and systolic pressure and heart rate recorded on a Devices recorder (type MV 216). ATP and adenosine were not measured in these experiments because of the possible effect of tissue trauma (LV penetration by catheter).
Statistical methods Results are expressed as m e a n s +__S.E.M. (number of hearts). Both the Student's paired and unpaired t tests with two-tailed tests of significance were used to assess statistical differences.
Results
First, the pattern of release of ATP, adenosine and L D H into the coronary effluent after excision of the heart was established. Initially both ATP and LDH were released at high rates which decreased exponentially (r = 0.98 and r = 0.99 respectively) until both rates of release became minimal between 50 and 60 rain after the start of cardiac perfusion (Fig. 1). ATP and L D H release during this 6O ,Ib
2OO
:
150
E
c~
~
40
E
ioo ~ 50
go_
-1-
123 J
-o
0
t
I0 Storl of
perfusion
20
30 40 Time (rain)
5'0
60
F I G U R E 1. Time-course of release of ATP, adenosine and LD H (lactate dehydrogenase) into the coronary effluent after excision of the heart. Perfusion with KrebsHenseleit buffer commenced at time zero as soon as the hearts were mounted on the aortic cannula. Samples of coronary effluent were analysed for ATP and adenosine concentration and lactate dehydrogenase activity (Mean _ s.E.~, of nine experiments).
R e l e a s e o f ATP f r o m P e r f u s e d Rat H e a r t
189
oF~
initial period is probably due to tissue damage incurred during the excision of the heart, which releases both L D H and ATP. Adenosine was not detected during this period. To study ATP release due to various interventions, a stabilization period of 50 min was allowed in subsequent experiments.
,0,
]
g
Hypoxia : release of A TP, L D H and adenosine During the first minute ofhypoxia (Fig. 2) the coronary flow rose from 12.9 __+0.5 ml/g/min to 21.0 -t- 0.9 ml/g/min. The rate of ATP release increased considerably (10.2-t-3.6 to 21.4 _ 4.2 pmol/g/min). The rate offall-offof ATP corresponded well with the rate of falloff in coronary flow. During the first 4 rain after the onset of hypoxia, there was no change in the rate of LDH release. After 10 to 15 min the amount of L D H released rose to a peak by which time the rate of ATP release was already decreasing. During the first minute of hypoxia adenosine increased rapidly from unmeasurable values to 0.3 -t- 0.2 ~M, while coronary flow rose markedly. Peak rates of adenosine release were very high and found 4 to 12 rain after the onset of hypoxia when the coronary flow rate had already passed its peak and was at the prehypoxic levels in this series of experiments. The peak concentration of adenosine released was 6.1 + 0.6/~M at 7 _.+0.8 min (Table 1); variability between hearts is shown in Fig. 3(a).
~,
4L Hypoxia
20
"-
15
Hyp0xia
F I G U R E 3. Adenosine [Fig. 3(a)] and ATP [Fig. 3 (b)] concentration in coronary effluent before and after the onset of hypoxia, taking the peak value after intervention. Note the increases in adenosine concentrations, which are approximately one thousand times greater than those of ATP.
In contrast the concentration of ATP was 1000 x lower. It increased from 0.8 + 0.2 nM to an average peak value of 1.3_+ 0.2 nM (P < 0.01, paired t test) at 2 _+ 0.5 rain after the onset of hypoxia (Table 1) ; there was considerable variability in the individual hearts [Fig. 3(b)]. Figure 4 shows the results of an individual heart exposed to hypoxia. Note the close relationship between ATP release and coronary flow increase. Peak adenosine ,~176
25
-~
Time (min)
T 75
-'~
o
~g
,oo
Start of hypoxia
~176 5
g
q
0 -I0
0
I0 Time (min)
2
30
F I G U R E 2. Patterns of release of ATP, adenosine and LDH in response to sustained hypoxia in the isolated perfused rat heart. Note the close correspondence between the pattern of release of ATP and the increase of coronary flow, with the increase of adenosine occurring at a time when coronary flow is falling or has returned to normal. Release of LDH occurred at an even later stage. Note the return of coronary flow to pre-hypnxic values, while hypoxia is sustained and adenosine high. (Mean _+ S.E.M. of six experiments)
190
(3. V i ~ l et al.
+1+1+1+1
+1~+1+1
+1+1+1+1
+1+1+1+1
O
b
O E
O
& +1+1+1+1
+~+1~+1
+1+1+1+1
~+1+1
~ +1+1+1
~ +1+1+1
o
~ +1+1+1
o+1+1+1
~+1+1~
~+1+1o
+1+1+~+1
+l+i~+l
+1+1+1+1
~+1+1
O
.<
+1+1+1
+1+1o
O
o.
88 O~
R O.
"n u =
O O~
..r
§ o
..~ 0
+l+l+l-H
-H-H-H+I
-H+l+l+i
>
-H-H+I~4
~R
88 "~ ,.~
~R
.~-~-~ ~ .~-~-~ ~ . ~ - ~ ~
-~~
~2 ,..-: ,.-1
<
~ . ~ o
R.~
R e l e a s e o f ATP f r o m P e r f u s e d Rat Heart
191 120
25-
I
/
_l
A c
90
c
O
60
,2.5
o~
==
>,
~
30 Start of T hypoxia J
0
o'
o
~o .5
(
-I0
I0
o
2o
.30
Time (rain) F I G U R E 4. Shows the release of ATP and adenosine in an individual heart in response to sustained hypoxia. Peak coronary flow and ATP occur simultaneously. Peak adenosine occurred 4 rain after ATP when coronary flow had returned to pre-hypoxic values.
al. 1969). The increase in coronary flow after
occurred when coronary flow had returned to prehypoxic values.
Hypoxia ptus potassium arrest [ Table I and Figures 5, 6(a), 6(b)] To avoid bursts of contractile activity during the hypoxic period and to obtain better reproducibility of the experimental conditions, one group of hearts was arrested by perfusion with 24 mM K + from the time of mounting of the heart. The pre-hypoxic values of coronary flow were lower (either because there was no contractile work, or because of a vasoconstrictive effect of high K§ In comparison to nonarrested hypoxic hearts, the pre-hypoxic release of L D H was higher (P < 0.05) presumably because of the high K + (de Leiris et
onset of hypoxia was less but still highly significant ( P < 0 . 0 0 0 1 , paired t test); peak values were attained between 4 and 8 min after the onset of hypoxia. During the first 6 min of hypoxia and peak coronary flow no adenosine was detectable (Fig. 5). Peak values, 25 _+ 3 nmol/g/min, were obtained 24 rain after onset of hypoxia (Table 1). Figure 6(a) shows the individual increase in concentration of adenosine from unrecordable values to peak values; average peak was 2 . 8 _ 0.2 /tM (Table 1). The rate of A T P release increased from 4.4 _+ 2.5 to 10.7 _ 7.8 pmol/g/rnin (P < 0.02) after l0 rain of hypoxia (Table 1). However, the concentration of A T P was not significantly
i~ ,o
x
"6 E 40 ~ |
L
oo.
8~
hypoxio oL -I0
T~TI~ ~o~=~ ,~
0
~,
g,-, ,
,
I0 20 30 Time (min) F I G U R E 5. Patterns of release of ATP and adenosine in response to sustained hypoxia in potassium (24 mM) arrested hearts. Note that neither ATP nor adenosine showed close correspondence to the increase in coronary flow. qote the rise in diastolic tension in response to sustained hypoxia. (Mean 4- S.E.M. of 6--10 experiments)
C. V i a l
192
onary flow, A T P and L D H release (Table 1), thus indicating that the responses to hypoxia alone were not due to fl-adrenoceptor stimulation.
i(b)
(o)
t
::L 3
t
r
:
1
Adrenergic stimulation [Figures 7, 8( a) and 8( b) and Table 1)]
v"/, 8 o
I
0
I
I
I
I
I0 20 30
et el,
I
0 I0 20 ,f Hypoxia
Time (rain) Hypoxi0 F I G U R E 6. Adenosine [Fig. 6(a)] and ATP ['Fig. 6(b)] concentration in coronary effluent before and after the onset of hypoxia in potassium arrested hearts, taking the peak value after the intervention. Note that all hearts except one showed increased ATP concentration but the difference was not significant due to the large difference between values.
increased even though all hearts except one showed an increase [Fig. 6(b)]. Enzyme release was lower and later than in nonarrested hearts but still significant.
fl-Adrenoceptor blockage ( Table 1) Pretreatment of hearts with propranolol 10 -6 M for 10 min before and during hypoxia did not alter the effects of hypoxia on cor-
Since some effects of hypoxia might be mediated by adrenergic stimulation, we also studied the effect of adrenaline on the release of ATP, adenosine and L D H (Fig. 7). The concentration of 10-6M adrenaline was selected as promoting high rates of release of lactate dehydrogenase in isolated working rat hearts (Horak and Opie, 1983). In normoxic Langendorff hearts perfused with adrenaline (Fig. 7), there was a rapid increase in the rate of A T P release [5.0 _+ 1.3 to 17.1 _+ 2.8 pmol/ g/min (P < 0.001)] similar to that observed during hypoxia, the increase of coronary flow was less but highly significant, adenosine was released with peak values occurring after that of ATP. However at peak coronary flow sufficient adenosine was released (0.3 + 0.05 #M) to cause vasodilation. Figures 8(a) and 8(b) show the individual increases in concentration from zero time to peak value for adenosine and A T P ; average peak values were 0.8 _+ 0.1 #M adenosine and 1.7 _ 0.2 nM A T P (Table 1). However, in contrast to hypoxia alone, a small albeit significant release of L D H [33 +_ 6 to 58 _+ 7 mU/g/min
150
t0
i
I00
c
E
z_ E
o
E
c
3 ~ I0 v
E
o
-I50 g
: r ~ 0(2_
5
c) ~a
r
...~ o
0
5
i
-I0
i
0 I0 ~' #' Adrenotine
i
20 Time (min)
i
c
JO
30
10-6M F I G U R E 7. Effect of 10 -6 M adrenaline on coronary flow, rate of release of ATP and lactate dehydrogenase (LDH) in normoxia. After 50 min of normal perfusion hearts were perfused for 10 rain with adrenaline added to the buffer (Mean _+ S.E.M. of 6-9 experiments).
Release of ATP from Perfused Rat Heart 1.5
3
(a)
(b)
f /
1.0
0.5 o
o
o 0
I
I
i
I
o
5
Adrenaline 10- 6 u
i
5
Discussion
FIGURE 8. Adenosine [Fig. 8(a)] and ATP [Fig. 8(b)] concentration in coronary effluent beforeand after adrenaline 10-6M in normoxic heart, taking the peak value after the intervention.
(P < 0.002)] was elicited by adrenaline infusion only during the first 2 min of adrenaline infusion (Fig. 7).
Mechanical effects of hypoxia During normoxia, hearts developed systolic intraventricular pressures of about 90 m m H g and the diastolic pressure was as expected, zero. With the onset of hypoxia (Fig. 9), the systolic pressure rapidly fell and the diastolic pressure rose until both were about
=
0
60 m m H g . Thus there was the rapid development of substantial hypoxia-induced contracture, no further increase in diastolic tension was obtained after 30 min of hypoxia (data n o t shown). Release of adenosine occurred before the onset of hypoxic contracture. The heart rate was stable during the normoxic period ( - 2 0 rain 266 • 9 and 0 min 261 4- 7 beats/min). During the first 10 rain ofhypoxia the heart rate fell to 51 4- 20 and remained thereabouts. In the potassium arrested hypoxic hearts (Fig. 5), the increase in the diastolic tension and release of adenosine occurred later than in the non-arrested hypoxic hearts.
Adrenaline I0 -e M
Time (rain)
-I0
193
I0
20
30
Time (rain)
FIGURE 9. Systolic and diastolic tension before and after the onset of hypoxia. Note the relationship between peak coronary flow at 1 min, peak adenosine release at 6 min, and the rise in diastolic tension, peak at 3 rain. (Mean _+S.E.M.ofsix experiments)
Patterns of release of A TP We show that A T P can be released from the isolated perfused Langendorff ("nonworking") rat heart at rates similar to those reported from the hypoxic rabbit heart (Nayler et al. 1979) and the isolated working frog heart (Doyle and Forrester, 1985). During hypoxia, the A T P concentration reached in the coronary effluent was similar to that found in the effluent from the isolated working rat heart (Clemens and Forrester, 1980). Peak rates of release of A T P were somewhat lower in our preparation because of the lower coronary flow rate (compare our Fig. 2 with Figs 4 and 5 in Clemens and Forrester, 1980). O u r findings show that A T P can be released by at least two mechanisms: (i) through tissue trauma (Fig. 1) when both A T P and lactate dehydrogenase are released simultaneously; and (ii) in response to stimuli such as hypoxia and adrenaline, both of which were associated with early coronary vasodilation (Figs 2 and 7). In response to hypoxia, A T P was rapidly released, starting within 1 min and lasting for about 5 to 8 min, in keeping with previous findings in other preparations (see Nayler et al., 1979; Paddle and Burnstock, 1974). Nayler et al. (1979) found that an initial burst of A T P release in response to hypoxia in the isolated perfused rabbit heart occurred before any release of myoglobin, a marked release of myoglobin occurring during prolonged hypoxia preceded a secondary rise of A T P release. In contrast,
194
C. V i a l et al.
the initial release of lactate dehydrogenase peaks at about 10 to 15 rain (Fig. 2). It is possible that this initial release was limited by the onset of hypoxic contracture in our model. Origin of and stimulus to releasable A T P
The origin of the readily releasable A T P is unknown. The following possibilities exist: firstly, the A T P may come from intramural purinergic nerves (Burnstock, 1980); secondly, A T P may be released from vascular endothelial or smooth muscle cells (Pearson and Gordon, 1985; Sparks and Bardenheuer, 1986); thirdly, A T P may be released from myocardial cells (Forrester and Williams, 1977). I f A T P is released from myocardial cells, it may come from a pool of A T P superficially located on the cell membrane (Williams and Forrester, 1983). A T P can act via Pz-purinceptors located on vascular endothelial cells to cause release of an endothelium derived relaxing factor which diffuses to the vascular smooth muscle and induces vasodilation. The main source of intraluminal A T P is likely to be endothelial cells (Burnstock and Kennedy, 1985). The nature of the stimulus giving rise to the liberation of A T P is also unclear: Clemens and Forrester (1980) suggest that such release may be coupled in some way to the myocardial oxygen demand and energy metabolism. In support of their proposal, we show that inhibition of energy production by hypoxia caused increased A T P release, while lowered myocardial oxygen demand by K + arrested hypoxic hearts decreased the rate and extent of A T P release. Conversely an increased myocardial oxygen demand by increased work during adrenaline stimulation caused increased release of A T P in normoxic hearts (Fig. 7). To assess the possible role of A T P as vasodilator first requires an evaluation of the contribution of adenosine.
Adenosine as a vasodilator
The most widely accepted mediator of coronary flow increase in response to vasodilatory stimuli is adenosine (Rubio and Berne, 1969) ; it can be formed from intracellular Sadenosylhomocysteine or from A M P by the
action of cytosolic and/or membrane-bound 5'-nucleotidase (Berne, 1986). However there are certain inconsistencies (Afonso et al., 1972 ; Berne, 1980; Bunger et al., 1975; Giles and Wilcken, 1977; Rothaul and Broadley, 1981 ; Saito et al., 1981 ; De Witt et al., 1983), especially the failure to detect adenosine in the effluent perfusate of the isolated guinea-pig heart subject to an increased workload (Bardenheuer and Schrader, 1983), the apparent differences in time between the onset of coronary vasodilation and adenosine release in response to hypoxia (Ishibashi et al., 1985), and the failure of aminophylline to reduce peak flow response to reactive hyperemia (Giles and Wilcken, 1977). In our experiments adenosine rose from unrecordable levels in the coronary effluent to peak concentrations of between 2 and 8/~M. These concentrations are considerably higher than those found by Schrader et al. (1977) in the guinea-pig heart, but in the same range as those of Ishibashi et al. (1985) in the rat heart. The initial adenosine concentrations in samples taken 1 min after the onset ofhypoxia were 0.31 #M, i.e. adequate to achieve maximal vasodilation (Schrader et al., 1977). Seemingly, the subsequent pattern of adenosine release can be dissociated from the pattern of change of coronary flow because at the time when adenosine was highest, coronary flow had virtually returned to normal after the initial hypoxic increase (Fig. 2 and Fig. 4). The explanation for this discrepancy could be the development of hypoxic contracture. Hypoxic contracture
During hypoxia there was the rapid development ofcontracture as shown by a rise in diastolic pressure 5 min after the onset of hypoxia (Fig. 9) with maximum pressure obtained 25 min later. As hypoxic contracture developed, the coronary flow fell so that prehypoxic values were regained. Hypoxic contracture, like ischemic contracture (Bricknell and Opie, 1978), is likely to compress coronary arteries and mechanically impede coronary flow, thereby eliminating the initial vasodilatory effect of hypoxia. Hypoxic contracture, an energy-requiring process, is also likely to lead to substantial increases in the rate of break-
R e l e a s e o f ATP f r o m P e r f u s e d Rat H e a r t
down of myocardial tissue high-energy phosphates with release of adenosine. The high peak rates of release of adenosine may at least in part be caused by breakdown of myocardial A T P stores during the initial phases of development of hypoxic contracture. In the hypoxic hearts, maximum hypoxic contracture occurred some 20 min after the peak release of adenosine in the hypoxic hearts. However, during hypoxia plus high K, the peak hypoxic contracture occurred much earlier (5 min) and about the same time as the adenosine release. Similarly the early fall in coronary flow appeared to be unaffected by the rate of development of hypoxic contracture present (Fig. 9). Therefore, the pattern of fall of coronary flow and rise in adenosine is not fully explained by the development of hypoxic contracture.
Significance of A T P in coronary effluent The A T P concentration in the coronary effluent is a complicated balance between the rate of release from the cells, the rate of breakdown by ecto-ATPases (Pearson et al., 1980) and the rate of washout by the coronary flow. This is borne out by the great variability found in the change in A T P concentrations before and after the various interventions [Figs 3(b), 6(b), 8(b)]. Ninety percent to 99% of exogenous A T P is broken down by ecto-ATPases during a single passage through the coronary vasculature (Paddle and Burnstock, 1974; Baer and Drummond, 1968). Even in a salineperfused isolated heart more than 97% of infused A T P is converted in one passage to AMP, adenosine, inosine and uric acid (Ronca-Testoni and Borghini, 1982). Therefore: (i) the actual A T P concentration in the perivascular space might have been considerably higher (up to 100-fold) although difficult to assess quantitatively and (ii) adenosine will inevitably rise by at least 10-fold from the breakdown of A T P ; such a rise of adenosine occurs as the coronary flow increases (RoncaTestoni and Borghini, 1982). Although the A T P concentration in the coronary effluent only increases between 2 to 3-fold during hypoxia (Fig. 3), A T P appears to be about 4 x more potent a vasodilator than adenosine (Wolf and Berne, 1956). However, in our experiments, the .peak concentration of aden-
195
osine was more than 1000 x higher than that of ATP. Paddle and Burnstock (1974) showed that sustained hypoxia caused a modest increase of A T P formation as coronary vasodilation took place (their Fig. 1). Repetitive bouts of hypoxia caused successively smaller releases of A T P until the fourth hypoxic bout caused virtually no A T P release despite similar degrees of coronary vasodilation [their Fig. 2(a)]. Therefore hypoxia-induced release of A T P from the perfused heart is not essential for hypoxic-induced vasodilation. This conclusion is especially true for hypoxic K-arrested hearts in which only a small and variable increase of A T P concentration in the venous effluent could be found [Figs 5 and 6(b)]. During adrenaline stimulation there was no increase in diastolic pressure, i.e. no hypoxic contracture. The fall off in release rates of A T P (Fig. 7) could therefore not be explained by mechanical limitation of coronary flow. The maintenance of increased coronary flow 5 min after adrenaline infusion terminated, when A T P release and adenosine release had returned to normal, could therefore only be explained by postulating other vasodilator influences. The release of A T P in response to adrenaline was accompanied by release of L D H (Fig. 7) so that the A T P could have originated from an intracellular site. Activation of fl-adrenergic receptors may be invoked in the myocardial response to hypoxia as suggested by an increased A T P release in response to adrenaline 10 -6 M. However, a high fl-adrenoceptor blocking dose of propranolol (10 -6 M) did not inhibit the hypoxia-induced release of A T P or of adenosine, nor the increase in coronary flow. Therefore vasodilation is neither due to the direct stimulation of the fl2-adrenoceptors of coronary arteries nor to the release of a vasoactive metabolite secondary to fll-adrenoceptor stimulation of the myocardial cells (Parratt and Wadsworth, 1972). Adenosine-mediated vasodilation is also not decreased by fl-adrenoceptor inhibition (Buckley, 1970).
Significance of sampling site Our measurements on the coronary effluent cannot exclude an earlier accumulation of adenosine at a vascular binding site, nor pro-
C. Vial et al.
196
longed activity of A T P on a vascular b i n d i n g site even when concentrations of the c o r o n a r y effluent have r e t u r n e d to p r e - h y p o x i c levels. I n a d d i t i o n A T P a n d adenosine values measured in the c o r o n a r y effluent do not necessarily reflect the concentrations present in the interstitial fluid (Ishibashi et al., 1985). T h e concentration of adenosine in the interstitial fluid is a b o u t 10 to 20 x that in the c o r o n a r y venous effluent ( K a m m e r m e i e r a n d Decking, 1986), a l t h o u g h only the initial rise in both sites can be linked to hypoxic c o r o n a r y vasodilation. Likewise, A T P in the interstitial fluid could be m u c h higher a n d r a p i d l y broken down to adenosine, so that the values of A T P at the vascular r e c e p t o r sites could be considerably higher t h a n that m e a s u r e d in the c o r o n a r y effluent.
Other reservations to the study W e are using a n o n - b l o o d perfused m e d i u m in which the o x y g e n a t i o n is thought to be adeq u a t e (Opie, 1984). Nevertheless the p r e p a r a t i o n has an a b n o r m a l l y high resting oxygen tension a n d c o r o n a r y flow rate. Therefore e x t r a p o l a t i o n from the proposed vasodilatory roles of A T P a n d adenosine in this p r e p a r a t i o n to those of the h e a r t in situ can only be u n d e r t a k e n with reserve. O t h e r reservations to our study a r e the absence o f direct p r o o f t h a t A T P acted as a vasodilator. T h e release of A T P was less consistent t h a n that o f adenosine. F u r t h e r work c o m p a r i n g the release of A T P a n d adenosine a n d increase in c o r o n a r y flow d u r i n g the first few seconds o f
h y p o x i a need to be studied. T h e ideal sampling site for both studies would be the myocardial insterstitial fluid.
Conclusion T h e concentrations of both A T P a n d adenosine rise in the c o r o n a r y effluent as p a r t of the response to early vasodilation in most b u t not all of our experiments. N e i t h e r A T P nor adenosine, b y themselves, fully explain the p a t t e r n of sustained c o r o n a r y vasodilation d u r i n g hypoxia. Release of A T P is not required to explain the initial vasodilatory response to hypoxia. W e c a n n o t exclude the possibility t h a t A T P released from the vascular e n d o t h e l i u m could be an early response to h y p o x i a to be followed b y A T P conversion to adenosine (Sparks a n d Bardenheuer, 1986). H o w e v e r , in most experiments the p a t t e r n of release of adenosine could fully explain the initial v a s o d i l a t o r y response. T h e existence of o t h e r vasodilators is shown by the discrepancy between the overall v a s o d i l a t o r y response to h y p o x i a a n d the release of A T P a n d adenosine, a n d also by the vasodilatory response to h y p o x i a plus high K, which is not explained b y changes in either A T P or adenosine.
Acknowledgements T h e F o u n d a t i o n Del Duca, the M e d i c a l Research Council of South Africa a n d the Chris B a r n a r d F u n d are t h a n k e d for their financial support, a n d O w e n Bricknell for technical assistance.
References AFONSO,S., ANSFIELD,T.J., BERNDT,T. B., RowE, G. C. Coronary vasodilation response to hypoxia before and after aminophylline. J Physio1221,589--600 ( 1972). ANTONI,H., ENGSTFELD,G., FLECXENSTE~N,A. Inotrope Effekte yon ATP und Adrenalin am hypodynamen Froschmyokard nach elektromechanischer Entkoppelung durch Ca + +-Entzug. Pfliigers Arch 272, 91-106 (1960). BAER,H. P., DRUMMOND,G. I. Catabolism of adenine nucleotides by the isolated perfused rat heart. Proc Soc Expt Biol Med 127, 33-36 (1968). BAnDENHEUER,H., SeHItADER,J. Relationship between myocardial oxygen consumption, coronary flow, and adenosine release in an improved isolated working heart preparation of guinea pigs. Cire Res 51,263-271 (1983). BERNE,R. M. The role of adenosine in the regulation of coronary blood flow. Circ Res 47, 807 813 (1980). BERNE,R. M. Adenosine in the local regulation of blood flow. Pfliigers Arch 107 [Suppl 1], S 16 (1986). BRIeKNELL,O. L., Ol'IE, L. H. Effects ofsubstrates on tissue metabolic changes in the isolated rat heart during underperfusion and on release of lactate dehydrogenase and arrhythmias during reperfusion. Circ Res 43, 102-115 (1978). BUCKLEY, N. M. Cardiac effects of nucleotides after propranolol treatments of isolated hearts. Am J Physiol 218, 1399-1405 (1970). BUNCEI~, R., HADDY,F. J., GERLACH,E. Coronary responses to dilating substances and competitive inhibition by theophylline on the isolated guinea-pig heart. PfliigersArch 358, 213-224 (1975). BVRNSTOeK,G. Purinergic receptors in the heart. Circ Res [Suppl 1] 46, I-175-187 (1980).
R e l e a s e o f ATP f r o m P e r f u s e d Rat Heart
197
BURNSTOCK,G., KENNEDY, C. A dual function for adenosine 5-triphosphate in regulation of vascular tone. Circ Res 58, 31%330 (1985). CLEMENS, M. G., FORRESTER, T. Appearance of adenosine triphosphate in the coronary sinus effluent from isolated working rat heart in response to hypoxia.`] Physio1312, 143-158 (1980). DE LEII~IS,J., BRETON,D., FEUVRAY,D., CORABOrUF,E. Lactic dehydrogenase release from perfused rat heart under the effect of abnormal media. Arch Int Physiol Biochim 77, 749-762 (1969). D~ WITT, D. F., WANOLER, R. D., THOMPSON, C. I., SPARKS, H. V. Phasic release of adenosine during steady-state metabolic stimulation in the isolated guinea-pig heart. Circ Res 53, 636-643 (1983). DOYLE, T. B., FORRmTER, T. Appearance of adenosine triphosphate in the perfnsate from working frog heart. Pfliigers Arch 405, 80-82 (1985). ELY, S. W., KNAna, R. M., BACCHUS,A. N., Rumo, R., BERNe, R. M. Measurements of coronary plasma and pericardial infusate adenosine concentrations during exercise in conscious dog: Relationship to myocardial oxygen consumption and coronary blood flow.,] Mol Cell Cardio115, 673 683 (1983). FORRESTeR, T., WILLIAMS, C. A. Release of adenosine triphosphate from isolated adult heart cells in response to hypoxia.J Physio1268, 371-390 (1977). GILES, R. W., WILCKEN, D. E. L. Reactive hyperaemia in the dog heart: interrelations between adenosine, ATP and aminophylline and the effect ofindomethacin. Cardiovasc Res 11, 113-121 (1977). HORAK, A. R., OPIE, L. H. Energy metabolism of the heart in catecholamine-induced myocardial injury. In : Advancesin Myocardiology,vol. 4, edited by E. Chazov, V. Saks, G. Rona, New York, Plenum Publishing, pp. 2343 (1983). ISHIBASHI,T., ICHIHARA,K., ABIKO,Y. Difference in the time course between increases in coronary flow and in effluent adenosine concentration during anoxia in the perfnsed rat heart. Jpn CircJ 49, 1090-1098 (1985). KAMMERMEIER, H., DECKING, U. Interstitial concentration of adenosine and inosine in isolated rat hearts of rats and guinea-pigs. Pfliigers Arch 407 [Suppl 1], 550 (1986). NAVLER, W. G., POOLE-WILSON,P. A., WILLIAMS,A. Hypoxia and calcium.,J Mol Cell Cardiol 11,683-706 (1979). NEeLY, J. R., LIEnERMeISTeR, H., BATTeRSBY, E.J., MORGAN, H. E. Effect of pressure development on oxygen consumption by isolated rat heart. Am J Physio1212, 804-814 (1967). Nero, S., GERLACH, E. Adenine nucleotide and adenosine metabolism in cultured coronary endothelial cells. Formation and release of adenine compounds and possible functional implication. In: RegulatoryFur~tion of Adenosine, edited by R. M. Berne, T. W. Rail, R. Rubio. The Hague, Martinus Nijhoff, pp. 347-360 (1983). Ovle, L. H. Adequacy of oxygenation of isolated perfnsed rat heart. Basic Res Cardio179, 300-306 (1984). PADDLE, B. M., BURNSTOCK,G. Release of ATP from perfused heart during coronary vasodilation. Blood Vessels 11, 11(~119 (1974). PARRATT,J. R., WADSWORTI4,R. M. The effects of dipyridamole on the myocardial vasodilator actions of noradrenaline, isoprenaline and adenosine. BrJ Pharmaco146, 585-593 (1972). PeAaSON, J. D., CARt.ETON,`]. S., GORDON,`]. L. Metabolism of adenine nucleotides by ectoenzymes of vascular endothelial and smooth muscle cells in culture. BiochemJ 190, 421429 (1980). PeARSON,J. D., GORDON,J. L. Nucleotide metabolism by endothelium. Ann Res Physio147, 617~27 (1985). RONCA-TEsTONI, S., BORGHINI,F. Degradation of perfused adenine compounds up to uric acid in isolated rat heart..J Mol Cell Cardiol 14, 177-180 (1982). ROTI~AUL, A. L., BROADLY,K.J. Use ofhexobendine to examine whether the coronary vasodilator metabolite released from guinea-pig isolated hearts is adenosine. Cardiovasc Res 15, 599-610 (1981). RUBIO, R., BERNE, R. M. Release of adenosine by the normal myocardium in dogs and its relationship to the regulation of coronary resistance. Circ Res 25, 4 0 7 4 15 (1969). SAITO, D., STeINHART, C. R., NIXON, D. G., OLSSON, R. A. Intracoronary adenosine deaminase reduces canine myocardial reactive hyperemia. Circ Res 49, 1262 1267 (1981). SCHRADeR,,]., HADDV, F.J., GERLACH, E. Release of adenosine, inosine and hypoxanthine from the isolated guinea-pig heart during hypoxia, flow-autoregulation and reactive hyperemia. Pflllgers Arch 369, 1-6 (1977). SPARKS, H. V., BARDeNHEUeR,H. Regulation of adenosine formation by the heart. Circ Res 58, 193-201 (1986). WILLIAMS, C. A., FORRESTeR, T. Possible source of adenosine triphosphate released from rat myocytes in response to hypoxia and acidosis. Cardiovasc Res 17, 301 312 (1983). WOLF, M. M., BERNE, M. D. Coronary vasodilator properties of purine and pyrimidine derivatives. Circ Res 4, 343-348 (1956). WROBI~eWSKI,F., LA Due,,]. S. Lacficodehydrogenase activity in blood. Proc Soc Expt Biol Med 20, 210-213 (1955).
Significance of Release of Adenosine Triphosphate and Adenosine Induced by Hypoxia or Adrenaline in Perfused Rat Heart C. Vial, 1 P. Owen, L. H. Opie,* and D. Posel
MRC Ischaemic Heart Disease Research Unit, Department of Medicine, University of Cape Town Medical School, Observatory, 7925, South Africa and 1Department of Biochemistry, University of Lyon, France (Received 17 October 1984, acceptedin revisedform 3 November 1986) C. VIAL, P. OWEN, L. H. OP1E AND D. POSEL. Significance of Release of Adenosine Triphosphate and Adenosine Induced by Hypoxia or Adrenaline in Perfused Rat Heart. Journalof Molecularand CellularCardiology(1987) 19, 187 197. The status ofATP as a possible coronary vasodilator remains poorly understood. The onset ofhypoxia induced a rapid and transient increase of the ATP concentration in the coronary effluent of the isolated perfused rat heart from 0.8 q- 0.2 nM to the average peak value of 1.3 _+ 0.2 nM (P < 0.01) at 2 4- 0.5 rain; at the same time the coronary flow increased 2-fold so that the rate of ATP release increased from 10.2_ 2.9 to 21.4 ___4.2 pmol/g/min (P < 0.005). Hypoxia also produced a peak rate release of adenosine of 93 __ 5 nu/g/ rain occurring only after the peak increase of coronary flow and also after the peak release of ATP; at peak coronary flow, however, the adenosine concentration was sufficient for vasodilation (0.31 4- 0.19/tu). Peak release of ATP and of adenosine preceded that of lactate dehydrogenase. 10 -6 M adrenaline induced a rapid increase of coronary flow and release of ATP, the concentration of which rose from 0.9 4- 0.3 nM to an average peak of 1.7 _+ 0.2 nM (P < 0.01) at 2 4- 0.3 min. The rate of increase of ATP in the coronary effluent paralleled the rate of early rise of coronary flow, yet adenosine had also risen to vasodilatory values (0.28 4- 0.5 tiM). The absolute changes in the measured concentrations of ATP in the coronary effluent were more variable and 1000 x less in concentration than those of adenosine. Hence coronary dilation could be explained by adenosine without involving ATP, although an additional vasodilatory role for ATP could not be excluded, especially in the early phases of vasodilation. In one condition, hypoxic K-arrested hearts, the increase in coronary flow could not be linked to release of either adenosine or ATP. The changes in concentrations of potential vasodilators measured in the coronary effluent do not necessarily reflect changes in the interstitial fluid. KEy WORDS: ATP, Hypoxia; Vasodilation; Adenosine; Adrenaline.
Introduction A l t h o u g h it is w i d e l y h e l d t h a t cell m e m branes are impermeable to a d e n o s i n e - 5 ' t r i p h o s p h a t e ( A T P ) ( A n t o n i et al. 1960), several workers have detected ATP in the coronary effluent of isolated perfused hearts (Clemens and Forrester, 1980; Doyle and F o r r e s t e r , 1 9 8 5 ; N a y l e r et al., 1979; P a d d l e a n d B u r n s t o c k , 1974). F u r t h e r m o r e , P a d d l e a n d B u r n s t o c k (1974) a n d C l e m e n s a n d F o r r e s t e r (1980) h a v e p o s t u l a t e d t h a t t h e A T P r e l e a s e f r o m t h e m y o c a r d i u m is p a r t o f t h e c o r o n a r y v a s o d i l a t o r y r e s p o n s e to h y p o x i a . The origin of the ATP might be either from 'purinergic nerves' (Paddle and Burnstock, 1974) o r f r o m m y o c y t e s ( F o r r e s t e r a n d W i l l i a m s , 1977) o r c a p i l l a r y e n d o t h e l i a l ceils (Nees a n d G e r l a c h , 1983). T h e q u e s t i o n arises
whether or not ATP release has a physiological f u n c t i o n o r is j u s t a c o n s e q u e n c e o f a g e n e r a l i z e d loss o f m e m b r a n e i n t e g r i t y . I n t h e l a t t e r case, A T P r e l e a s e c o u l d o c c u r i n a pattern different from that of an intracellular m a r k e r e n z y m e s u c h as l a c t a t e d e h y d r o g e n a s e ( L D H ) . A s e c o n d q u e s t i o n is w h e t h e r A T P might have a vasodilator function; hence r a t e s o f r e l e a s e o f A T P i n t o t h e c o r o n a r y efflue n t w e r e c o m p a r e d w i t h t h e i n c r e a s e o f coro n a r y flow in r e s p o n s e to h y p o x i a o r adrenaline. The pattern of release of ATP was compared with that of adenosine, an establ i s h e d c o r o n a r y v a s o d i l a t o r ( B e r n e , 1980).
Materials and Methods H e a r t s f r o m m a l e L o n g - E v a n s r a t s (250 to 350 g, fresh h e a r t w t 0.9 to 1.2 g) w e r e
* To whom all correspondence should be addressed. 0022-2828/87/020187 + 11 $03.00/0
9 1987 Academic Press Inc. (London) Limited
188
C. V i a l et al.
arrested in icecold saline and then perfused by the Langendorff technique with KrebsHenseleit bicarbonate buffer pH 7.4 (millimolar concentrations: NaC1 118.5, N a H C O 3 25, KC1 4.7, K H z P O 4 1.2, MgSO4 1.2, CaC1 z 2.5, glucose 11.1) equilibrated with a mixture of 95% oxygen and 5% COz (Po2 from 500 to 600 mmHg) at 37~ and a constant perfusion pressure of 100 c m H 2 0 . Hypoxia was initiated by switching over to buffer equilibrated with 95% Nz and 5% CO2, the aortic Po2 fell to 25 to 70 m m H g within 15 s. In some experiments, hearts were arrested from the start of the perfusion by increasing the K + concentration to 24 mM and correspondingly decreasing the Na + concentration. Adrenaline (Merck) kept under nitrogen was dissolved in dilute hydrochloric acid (pH 2.0) to make a 2 mM solution which was protected from light. Immediately before the experiment, the solution was further diluted to 10 -6 i with preoxygenated warmed Krebs buffer and was protected from light with aluminium foil during the experiment. For ATP and adenosine measurements the coronary effluent collected during a 10 s period was immediately frozen in liquid nitrogen. Lactate dehydrogenase (LDH) activity measured in the perfusate (Wroblewski and La Due, 1955) was expressed as milliunits per gram wet weight per minute at 25~ ATP in the perfusate was estimated by the luciferinluciferase method using an Aminco-Bowman spectrofluorometer. A 0.15 ml sample was added to 0.5 ml of the assay mixture so that the final concentrations of the reactants in the cuvette were: 17.6 m i Tris-HC1, 0.175 mM EDTA, 4.4 mM magnesium acetate, 0.4 mra luciferin (Sigma), 25 #g/ml luciferase (Sigma), pH 7.75. Adenosine was measured by high performance liquid chromatography, by a modification of the method of Ely et al. (1983). Lowest concentration detectable was 0.1 # i . Coronary flow was measured by a timed collection of the coronary effluent.
Myocardial mechanical performance In an additional series of hearts, left ventricular (LV) pressure was measured by a ventricular catheter piercing the LV wall (Neely
et al., 1967). The catheter was connected to a pressure transducer and left ventricular diastolic and systolic pressure and heart rate recorded on a Devices recorder (type MV 216). ATP and adenosine were not measured in these experiments because of the possible effect of tissue trauma (LV penetration by catheter).
Statistical methods Results are expressed as m e a n s +__S.E.M. (number of hearts). Both the Student's paired and unpaired t tests with two-tailed tests of significance were used to assess statistical differences.
Results
First, the pattern of release of ATP, adenosine and L D H into the coronary effluent after excision of the heart was established. Initially both ATP and LDH were released at high rates which decreased exponentially (r = 0.98 and r = 0.99 respectively) until both rates of release became minimal between 50 and 60 rain after the start of cardiac perfusion (Fig. 1). ATP and L D H release during this 6O ,Ib
2OO
:
150
E
c~
~
40
E
ioo ~ 50
go_
-1-
123 J
-o
0
t
I0 Storl of
perfusion
20
30 40 Time (rain)
5'0
60
F I G U R E 1. Time-course of release of ATP, adenosine and LD H (lactate dehydrogenase) into the coronary effluent after excision of the heart. Perfusion with KrebsHenseleit buffer commenced at time zero as soon as the hearts were mounted on the aortic cannula. Samples of coronary effluent were analysed for ATP and adenosine concentration and lactate dehydrogenase activity (Mean _ s.E.~, of nine experiments).
R e l e a s e o f ATP f r o m P e r f u s e d Rat H e a r t
189
oF~
initial period is probably due to tissue damage incurred during the excision of the heart, which releases both L D H and ATP. Adenosine was not detected during this period. To study ATP release due to various interventions, a stabilization period of 50 min was allowed in subsequent experiments.
,0,
]
g
Hypoxia : release of A TP, L D H and adenosine During the first minute ofhypoxia (Fig. 2) the coronary flow rose from 12.9 __+0.5 ml/g/min to 21.0 -t- 0.9 ml/g/min. The rate of ATP release increased considerably (10.2-t-3.6 to 21.4 _ 4.2 pmol/g/min). The rate offall-offof ATP corresponded well with the rate of falloff in coronary flow. During the first 4 rain after the onset of hypoxia, there was no change in the rate of LDH release. After 10 to 15 min the amount of L D H released rose to a peak by which time the rate of ATP release was already decreasing. During the first minute of hypoxia adenosine increased rapidly from unmeasurable values to 0.3 -t- 0.2 ~M, while coronary flow rose markedly. Peak rates of adenosine release were very high and found 4 to 12 rain after the onset of hypoxia when the coronary flow rate had already passed its peak and was at the prehypoxic levels in this series of experiments. The peak concentration of adenosine released was 6.1 + 0.6/~M at 7 _.+0.8 min (Table 1); variability between hearts is shown in Fig. 3(a).
~,
4L Hypoxia
20
"-
15
Hyp0xia
F I G U R E 3. Adenosine [Fig. 3(a)] and ATP [Fig. 3 (b)] concentration in coronary effluent before and after the onset of hypoxia, taking the peak value after intervention. Note the increases in adenosine concentrations, which are approximately one thousand times greater than those of ATP.
In contrast the concentration of ATP was 1000 x lower. It increased from 0.8 + 0.2 nM to an average peak value of 1.3_+ 0.2 nM (P < 0.01, paired t test) at 2 _+ 0.5 rain after the onset of hypoxia (Table 1) ; there was considerable variability in the individual hearts [Fig. 3(b)]. Figure 4 shows the results of an individual heart exposed to hypoxia. Note the close relationship between ATP release and coronary flow increase. Peak adenosine ,~176
25
-~
Time (min)
T 75
-'~
o
~g
,oo
Start of hypoxia
~176 5
g
q
0 -I0
0
I0 Time (min)
2
30
F I G U R E 2. Patterns of release of ATP, adenosine and LDH in response to sustained hypoxia in the isolated perfused rat heart. Note the close correspondence between the pattern of release of ATP and the increase of coronary flow, with the increase of adenosine occurring at a time when coronary flow is falling or has returned to normal. Release of LDH occurred at an even later stage. Note the return of coronary flow to pre-hypnxic values, while hypoxia is sustained and adenosine high. (Mean _+ S.E.M. of six experiments)
190
(3. V i ~ l et al.
+1+1+1+1
+1~+1+1
+1+1+1+1
+1+1+1+1
O
b
O E
O
& +1+1+1+1
+~+1~+1
+1+1+1+1
~+1+1
~ +1+1+1
~ +1+1+1
o
~ +1+1+1
o+1+1+1
~+1+1~
~+1+1o
+1+1+~+1
+l+i~+l
+1+1+1+1
~+1+1
O
.<
+1+1+1
+1+1o
O
o.
88 O~
R O.
"n u =
O O~
..r
§ o
..~ 0
+l+l+l-H
-H-H-H+I
-H+l+l+i
>
-H-H+I~4
~R
88 "~ ,.~
~R
.~-~-~ ~ .~-~-~ ~ . ~ - ~ ~
-~~
~2 ,..-: ,.-1
<
~ . ~ o
R.~
R e l e a s e o f ATP f r o m P e r f u s e d Rat Heart
191 120
25-
I
/
_l
A c
90
c
O
60
,2.5
o~
==
>,
~
30 Start of T hypoxia J
0
o'
o
~o .5
(
-I0
I0
o
2o
.30
Time (rain) F I G U R E 4. Shows the release of ATP and adenosine in an individual heart in response to sustained hypoxia. Peak coronary flow and ATP occur simultaneously. Peak adenosine occurred 4 rain after ATP when coronary flow had returned to pre-hypoxic values.
al. 1969). The increase in coronary flow after
occurred when coronary flow had returned to prehypoxic values.
Hypoxia ptus potassium arrest [ Table I and Figures 5, 6(a), 6(b)] To avoid bursts of contractile activity during the hypoxic period and to obtain better reproducibility of the experimental conditions, one group of hearts was arrested by perfusion with 24 mM K + from the time of mounting of the heart. The pre-hypoxic values of coronary flow were lower (either because there was no contractile work, or because of a vasoconstrictive effect of high K§ In comparison to nonarrested hypoxic hearts, the pre-hypoxic release of L D H was higher (P < 0.05) presumably because of the high K + (de Leiris et
onset of hypoxia was less but still highly significant ( P < 0 . 0 0 0 1 , paired t test); peak values were attained between 4 and 8 min after the onset of hypoxia. During the first 6 min of hypoxia and peak coronary flow no adenosine was detectable (Fig. 5). Peak values, 25 _+ 3 nmol/g/min, were obtained 24 rain after onset of hypoxia (Table 1). Figure 6(a) shows the individual increase in concentration of adenosine from unrecordable values to peak values; average peak was 2 . 8 _ 0.2 /tM (Table 1). The rate of A T P release increased from 4.4 _+ 2.5 to 10.7 _ 7.8 pmol/g/rnin (P < 0.02) after l0 rain of hypoxia (Table 1). However, the concentration of A T P was not significantly
i~ ,o
x
"6 E 40 ~ |
L
oo.
8~
hypoxio oL -I0
T~TI~ ~o~=~ ,~
0
~,
g,-, ,
,
I0 20 30 Time (min) F I G U R E 5. Patterns of release of ATP and adenosine in response to sustained hypoxia in potassium (24 mM) arrested hearts. Note that neither ATP nor adenosine showed close correspondence to the increase in coronary flow. qote the rise in diastolic tension in response to sustained hypoxia. (Mean 4- S.E.M. of 6--10 experiments)
C. V i a l
192
onary flow, A T P and L D H release (Table 1), thus indicating that the responses to hypoxia alone were not due to fl-adrenoceptor stimulation.
i(b)
(o)
t
::L 3
t
r
:
1
Adrenergic stimulation [Figures 7, 8( a) and 8( b) and Table 1)]
v"/, 8 o
I
0
I
I
I
I
I0 20 30
et el,
I
0 I0 20 ,f Hypoxia
Time (rain) Hypoxi0 F I G U R E 6. Adenosine [Fig. 6(a)] and ATP ['Fig. 6(b)] concentration in coronary effluent before and after the onset of hypoxia in potassium arrested hearts, taking the peak value after the intervention. Note that all hearts except one showed increased ATP concentration but the difference was not significant due to the large difference between values.
increased even though all hearts except one showed an increase [Fig. 6(b)]. Enzyme release was lower and later than in nonarrested hearts but still significant.
fl-Adrenoceptor blockage ( Table 1) Pretreatment of hearts with propranolol 10 -6 M for 10 min before and during hypoxia did not alter the effects of hypoxia on cor-
Since some effects of hypoxia might be mediated by adrenergic stimulation, we also studied the effect of adrenaline on the release of ATP, adenosine and L D H (Fig. 7). The concentration of 10-6M adrenaline was selected as promoting high rates of release of lactate dehydrogenase in isolated working rat hearts (Horak and Opie, 1983). In normoxic Langendorff hearts perfused with adrenaline (Fig. 7), there was a rapid increase in the rate of A T P release [5.0 _+ 1.3 to 17.1 _+ 2.8 pmol/ g/min (P < 0.001)] similar to that observed during hypoxia, the increase of coronary flow was less but highly significant, adenosine was released with peak values occurring after that of ATP. However at peak coronary flow sufficient adenosine was released (0.3 + 0.05 #M) to cause vasodilation. Figures 8(a) and 8(b) show the individual increases in concentration from zero time to peak value for adenosine and A T P ; average peak values were 0.8 _+ 0.1 #M adenosine and 1.7 _ 0.2 nM A T P (Table 1). However, in contrast to hypoxia alone, a small albeit significant release of L D H [33 +_ 6 to 58 _+ 7 mU/g/min
150
t0
i
I00
c
E
z_ E
o
E
c
3 ~ I0 v
E
o
-I50 g
: r ~ 0(2_
5
c) ~a
r
...~ o
0
5
i
-I0
i
0 I0 ~' #' Adrenotine
i
20 Time (min)
i
c
JO
30
10-6M F I G U R E 7. Effect of 10 -6 M adrenaline on coronary flow, rate of release of ATP and lactate dehydrogenase (LDH) in normoxia. After 50 min of normal perfusion hearts were perfused for 10 rain with adrenaline added to the buffer (Mean _+ S.E.M. of 6-9 experiments).
Release of ATP from Perfused Rat Heart 1.5
3
(a)
(b)
f /
1.0
0.5 o
o
o 0
I
I
i
I
o
5
Adrenaline 10- 6 u
i
5
Discussion
FIGURE 8. Adenosine [Fig. 8(a)] and ATP [Fig. 8(b)] concentration in coronary effluent beforeand after adrenaline 10-6M in normoxic heart, taking the peak value after the intervention.
(P < 0.002)] was elicited by adrenaline infusion only during the first 2 min of adrenaline infusion (Fig. 7).
Mechanical effects of hypoxia During normoxia, hearts developed systolic intraventricular pressures of about 90 m m H g and the diastolic pressure was as expected, zero. With the onset of hypoxia (Fig. 9), the systolic pressure rapidly fell and the diastolic pressure rose until both were about
=
0
60 m m H g . Thus there was the rapid development of substantial hypoxia-induced contracture, no further increase in diastolic tension was obtained after 30 min of hypoxia (data n o t shown). Release of adenosine occurred before the onset of hypoxic contracture. The heart rate was stable during the normoxic period ( - 2 0 rain 266 • 9 and 0 min 261 4- 7 beats/min). During the first 10 rain ofhypoxia the heart rate fell to 51 4- 20 and remained thereabouts. In the potassium arrested hypoxic hearts (Fig. 5), the increase in the diastolic tension and release of adenosine occurred later than in the non-arrested hypoxic hearts.
Adrenaline I0 -e M
Time (rain)
-I0
193
I0
20
30
Time (rain)
FIGURE 9. Systolic and diastolic tension before and after the onset of hypoxia. Note the relationship between peak coronary flow at 1 min, peak adenosine release at 6 min, and the rise in diastolic tension, peak at 3 rain. (Mean _+S.E.M.ofsix experiments)
Patterns of release of A TP We show that A T P can be released from the isolated perfused Langendorff ("nonworking") rat heart at rates similar to those reported from the hypoxic rabbit heart (Nayler et al. 1979) and the isolated working frog heart (Doyle and Forrester, 1985). During hypoxia, the A T P concentration reached in the coronary effluent was similar to that found in the effluent from the isolated working rat heart (Clemens and Forrester, 1980). Peak rates of release of A T P were somewhat lower in our preparation because of the lower coronary flow rate (compare our Fig. 2 with Figs 4 and 5 in Clemens and Forrester, 1980). O u r findings show that A T P can be released by at least two mechanisms: (i) through tissue trauma (Fig. 1) when both A T P and lactate dehydrogenase are released simultaneously; and (ii) in response to stimuli such as hypoxia and adrenaline, both of which were associated with early coronary vasodilation (Figs 2 and 7). In response to hypoxia, A T P was rapidly released, starting within 1 min and lasting for about 5 to 8 min, in keeping with previous findings in other preparations (see Nayler et al., 1979; Paddle and Burnstock, 1974). Nayler et al. (1979) found that an initial burst of A T P release in response to hypoxia in the isolated perfused rabbit heart occurred before any release of myoglobin, a marked release of myoglobin occurring during prolonged hypoxia preceded a secondary rise of A T P release. In contrast,
194
C. V i a l et al.
the initial release of lactate dehydrogenase peaks at about 10 to 15 rain (Fig. 2). It is possible that this initial release was limited by the onset of hypoxic contracture in our model. Origin of and stimulus to releasable A T P
The origin of the readily releasable A T P is unknown. The following possibilities exist: firstly, the A T P may come from intramural purinergic nerves (Burnstock, 1980); secondly, A T P may be released from vascular endothelial or smooth muscle cells (Pearson and Gordon, 1985; Sparks and Bardenheuer, 1986); thirdly, A T P may be released from myocardial cells (Forrester and Williams, 1977). I f A T P is released from myocardial cells, it may come from a pool of A T P superficially located on the cell membrane (Williams and Forrester, 1983). A T P can act via Pz-purinceptors located on vascular endothelial cells to cause release of an endothelium derived relaxing factor which diffuses to the vascular smooth muscle and induces vasodilation. The main source of intraluminal A T P is likely to be endothelial cells (Burnstock and Kennedy, 1985). The nature of the stimulus giving rise to the liberation of A T P is also unclear: Clemens and Forrester (1980) suggest that such release may be coupled in some way to the myocardial oxygen demand and energy metabolism. In support of their proposal, we show that inhibition of energy production by hypoxia caused increased A T P release, while lowered myocardial oxygen demand by K + arrested hypoxic hearts decreased the rate and extent of A T P release. Conversely an increased myocardial oxygen demand by increased work during adrenaline stimulation caused increased release of A T P in normoxic hearts (Fig. 7). To assess the possible role of A T P as vasodilator first requires an evaluation of the contribution of adenosine.
Adenosine as a vasodilator
The most widely accepted mediator of coronary flow increase in response to vasodilatory stimuli is adenosine (Rubio and Berne, 1969) ; it can be formed from intracellular Sadenosylhomocysteine or from A M P by the
action of cytosolic and/or membrane-bound 5'-nucleotidase (Berne, 1986). However there are certain inconsistencies (Afonso et al., 1972 ; Berne, 1980; Bunger et al., 1975; Giles and Wilcken, 1977; Rothaul and Broadley, 1981 ; Saito et al., 1981 ; De Witt et al., 1983), especially the failure to detect adenosine in the effluent perfusate of the isolated guinea-pig heart subject to an increased workload (Bardenheuer and Schrader, 1983), the apparent differences in time between the onset of coronary vasodilation and adenosine release in response to hypoxia (Ishibashi et al., 1985), and the failure of aminophylline to reduce peak flow response to reactive hyperemia (Giles and Wilcken, 1977). In our experiments adenosine rose from unrecordable levels in the coronary effluent to peak concentrations of between 2 and 8/~M. These concentrations are considerably higher than those found by Schrader et al. (1977) in the guinea-pig heart, but in the same range as those of Ishibashi et al. (1985) in the rat heart. The initial adenosine concentrations in samples taken 1 min after the onset ofhypoxia were 0.31 #M, i.e. adequate to achieve maximal vasodilation (Schrader et al., 1977). Seemingly, the subsequent pattern of adenosine release can be dissociated from the pattern of change of coronary flow because at the time when adenosine was highest, coronary flow had virtually returned to normal after the initial hypoxic increase (Fig. 2 and Fig. 4). The explanation for this discrepancy could be the development of hypoxic contracture. Hypoxic contracture
During hypoxia there was the rapid development ofcontracture as shown by a rise in diastolic pressure 5 min after the onset of hypoxia (Fig. 9) with maximum pressure obtained 25 min later. As hypoxic contracture developed, the coronary flow fell so that prehypoxic values were regained. Hypoxic contracture, like ischemic contracture (Bricknell and Opie, 1978), is likely to compress coronary arteries and mechanically impede coronary flow, thereby eliminating the initial vasodilatory effect of hypoxia. Hypoxic contracture, an energy-requiring process, is also likely to lead to substantial increases in the rate of break-
R e l e a s e o f ATP f r o m P e r f u s e d Rat H e a r t
down of myocardial tissue high-energy phosphates with release of adenosine. The high peak rates of release of adenosine may at least in part be caused by breakdown of myocardial A T P stores during the initial phases of development of hypoxic contracture. In the hypoxic hearts, maximum hypoxic contracture occurred some 20 min after the peak release of adenosine in the hypoxic hearts. However, during hypoxia plus high K, the peak hypoxic contracture occurred much earlier (5 min) and about the same time as the adenosine release. Similarly the early fall in coronary flow appeared to be unaffected by the rate of development of hypoxic contracture present (Fig. 9). Therefore, the pattern of fall of coronary flow and rise in adenosine is not fully explained by the development of hypoxic contracture.
Significance of A T P in coronary effluent The A T P concentration in the coronary effluent is a complicated balance between the rate of release from the cells, the rate of breakdown by ecto-ATPases (Pearson et al., 1980) and the rate of washout by the coronary flow. This is borne out by the great variability found in the change in A T P concentrations before and after the various interventions [Figs 3(b), 6(b), 8(b)]. Ninety percent to 99% of exogenous A T P is broken down by ecto-ATPases during a single passage through the coronary vasculature (Paddle and Burnstock, 1974; Baer and Drummond, 1968). Even in a salineperfused isolated heart more than 97% of infused A T P is converted in one passage to AMP, adenosine, inosine and uric acid (Ronca-Testoni and Borghini, 1982). Therefore: (i) the actual A T P concentration in the perivascular space might have been considerably higher (up to 100-fold) although difficult to assess quantitatively and (ii) adenosine will inevitably rise by at least 10-fold from the breakdown of A T P ; such a rise of adenosine occurs as the coronary flow increases (RoncaTestoni and Borghini, 1982). Although the A T P concentration in the coronary effluent only increases between 2 to 3-fold during hypoxia (Fig. 3), A T P appears to be about 4 x more potent a vasodilator than adenosine (Wolf and Berne, 1956). However, in our experiments, the .peak concentration of aden-
195
osine was more than 1000 x higher than that of ATP. Paddle and Burnstock (1974) showed that sustained hypoxia caused a modest increase of A T P formation as coronary vasodilation took place (their Fig. 1). Repetitive bouts of hypoxia caused successively smaller releases of A T P until the fourth hypoxic bout caused virtually no A T P release despite similar degrees of coronary vasodilation [their Fig. 2(a)]. Therefore hypoxia-induced release of A T P from the perfused heart is not essential for hypoxic-induced vasodilation. This conclusion is especially true for hypoxic K-arrested hearts in which only a small and variable increase of A T P concentration in the venous effluent could be found [Figs 5 and 6(b)]. During adrenaline stimulation there was no increase in diastolic pressure, i.e. no hypoxic contracture. The fall off in release rates of A T P (Fig. 7) could therefore not be explained by mechanical limitation of coronary flow. The maintenance of increased coronary flow 5 min after adrenaline infusion terminated, when A T P release and adenosine release had returned to normal, could therefore only be explained by postulating other vasodilator influences. The release of A T P in response to adrenaline was accompanied by release of L D H (Fig. 7) so that the A T P could have originated from an intracellular site. Activation of fl-adrenergic receptors may be invoked in the myocardial response to hypoxia as suggested by an increased A T P release in response to adrenaline 10 -6 M. However, a high fl-adrenoceptor blocking dose of propranolol (10 -6 M) did not inhibit the hypoxia-induced release of A T P or of adenosine, nor the increase in coronary flow. Therefore vasodilation is neither due to the direct stimulation of the fl2-adrenoceptors of coronary arteries nor to the release of a vasoactive metabolite secondary to fll-adrenoceptor stimulation of the myocardial cells (Parratt and Wadsworth, 1972). Adenosine-mediated vasodilation is also not decreased by fl-adrenoceptor inhibition (Buckley, 1970).
Significance of sampling site Our measurements on the coronary effluent cannot exclude an earlier accumulation of adenosine at a vascular binding site, nor pro-
C. Vial et al.
196
longed activity of A T P on a vascular b i n d i n g site even when concentrations of the c o r o n a r y effluent have r e t u r n e d to p r e - h y p o x i c levels. I n a d d i t i o n A T P a n d adenosine values measured in the c o r o n a r y effluent do not necessarily reflect the concentrations present in the interstitial fluid (Ishibashi et al., 1985). T h e concentration of adenosine in the interstitial fluid is a b o u t 10 to 20 x that in the c o r o n a r y venous effluent ( K a m m e r m e i e r a n d Decking, 1986), a l t h o u g h only the initial rise in both sites can be linked to hypoxic c o r o n a r y vasodilation. Likewise, A T P in the interstitial fluid could be m u c h higher a n d r a p i d l y broken down to adenosine, so that the values of A T P at the vascular r e c e p t o r sites could be considerably higher t h a n that m e a s u r e d in the c o r o n a r y effluent.
Other reservations to the study W e are using a n o n - b l o o d perfused m e d i u m in which the o x y g e n a t i o n is thought to be adeq u a t e (Opie, 1984). Nevertheless the p r e p a r a t i o n has an a b n o r m a l l y high resting oxygen tension a n d c o r o n a r y flow rate. Therefore e x t r a p o l a t i o n from the proposed vasodilatory roles of A T P a n d adenosine in this p r e p a r a t i o n to those of the h e a r t in situ can only be u n d e r t a k e n with reserve. O t h e r reservations to our study a r e the absence o f direct p r o o f t h a t A T P acted as a vasodilator. T h e release of A T P was less consistent t h a n that o f adenosine. F u r t h e r work c o m p a r i n g the release of A T P a n d adenosine a n d increase in c o r o n a r y flow d u r i n g the first few seconds o f
h y p o x i a need to be studied. T h e ideal sampling site for both studies would be the myocardial insterstitial fluid.
Conclusion T h e concentrations of both A T P a n d adenosine rise in the c o r o n a r y effluent as p a r t of the response to early vasodilation in most b u t not all of our experiments. N e i t h e r A T P nor adenosine, b y themselves, fully explain the p a t t e r n of sustained c o r o n a r y vasodilation d u r i n g hypoxia. Release of A T P is not required to explain the initial vasodilatory response to hypoxia. W e c a n n o t exclude the possibility t h a t A T P released from the vascular e n d o t h e l i u m could be an early response to h y p o x i a to be followed b y A T P conversion to adenosine (Sparks a n d Bardenheuer, 1986). H o w e v e r , in most experiments the p a t t e r n of release of adenosine could fully explain the initial v a s o d i l a t o r y response. T h e existence of o t h e r vasodilators is shown by the discrepancy between the overall v a s o d i l a t o r y response to h y p o x i a a n d the release of A T P a n d adenosine, a n d also by the vasodilatory response to h y p o x i a plus high K, which is not explained b y changes in either A T P or adenosine.
Acknowledgements T h e F o u n d a t i o n Del Duca, the M e d i c a l Research Council of South Africa a n d the Chris B a r n a r d F u n d are t h a n k e d for their financial support, a n d O w e n Bricknell for technical assistance.
References AFONSO,S., ANSFIELD,T.J., BERNDT,T. B., RowE, G. C. Coronary vasodilation response to hypoxia before and after aminophylline. J Physio1221,589--600 ( 1972). ANTONI,H., ENGSTFELD,G., FLECXENSTE~N,A. Inotrope Effekte yon ATP und Adrenalin am hypodynamen Froschmyokard nach elektromechanischer Entkoppelung durch Ca + +-Entzug. Pfliigers Arch 272, 91-106 (1960). BAER,H. P., DRUMMOND,G. I. Catabolism of adenine nucleotides by the isolated perfused rat heart. Proc Soc Expt Biol Med 127, 33-36 (1968). BAnDENHEUER,H., SeHItADER,J. Relationship between myocardial oxygen consumption, coronary flow, and adenosine release in an improved isolated working heart preparation of guinea pigs. Cire Res 51,263-271 (1983). BERNE,R. M. The role of adenosine in the regulation of coronary blood flow. Circ Res 47, 807 813 (1980). BERNE,R. M. Adenosine in the local regulation of blood flow. Pfliigers Arch 107 [Suppl 1], S 16 (1986). BRIeKNELL,O. L., Ol'IE, L. H. Effects ofsubstrates on tissue metabolic changes in the isolated rat heart during underperfusion and on release of lactate dehydrogenase and arrhythmias during reperfusion. Circ Res 43, 102-115 (1978). BUCKLEY, N. M. Cardiac effects of nucleotides after propranolol treatments of isolated hearts. Am J Physiol 218, 1399-1405 (1970). BUNCEI~, R., HADDY,F. J., GERLACH,E. Coronary responses to dilating substances and competitive inhibition by theophylline on the isolated guinea-pig heart. PfliigersArch 358, 213-224 (1975). BVRNSTOeK,G. Purinergic receptors in the heart. Circ Res [Suppl 1] 46, I-175-187 (1980).
R e l e a s e o f ATP f r o m P e r f u s e d Rat Heart
197
BURNSTOCK,G., KENNEDY, C. A dual function for adenosine 5-triphosphate in regulation of vascular tone. Circ Res 58, 31%330 (1985). CLEMENS, M. G., FORRESTER, T. Appearance of adenosine triphosphate in the coronary sinus effluent from isolated working rat heart in response to hypoxia.`] Physio1312, 143-158 (1980). DE LEII~IS,J., BRETON,D., FEUVRAY,D., CORABOrUF,E. Lactic dehydrogenase release from perfused rat heart under the effect of abnormal media. Arch Int Physiol Biochim 77, 749-762 (1969). D~ WITT, D. F., WANOLER, R. D., THOMPSON, C. I., SPARKS, H. V. Phasic release of adenosine during steady-state metabolic stimulation in the isolated guinea-pig heart. Circ Res 53, 636-643 (1983). DOYLE, T. B., FORRmTER, T. Appearance of adenosine triphosphate in the perfnsate from working frog heart. Pfliigers Arch 405, 80-82 (1985). ELY, S. W., KNAna, R. M., BACCHUS,A. N., Rumo, R., BERNe, R. M. Measurements of coronary plasma and pericardial infusate adenosine concentrations during exercise in conscious dog: Relationship to myocardial oxygen consumption and coronary blood flow.,] Mol Cell Cardio115, 673 683 (1983). FORRESTeR, T., WILLIAMS, C. A. Release of adenosine triphosphate from isolated adult heart cells in response to hypoxia.J Physio1268, 371-390 (1977). GILES, R. W., WILCKEN, D. E. L. Reactive hyperaemia in the dog heart: interrelations between adenosine, ATP and aminophylline and the effect ofindomethacin. Cardiovasc Res 11, 113-121 (1977). HORAK, A. R., OPIE, L. H. Energy metabolism of the heart in catecholamine-induced myocardial injury. In : Advancesin Myocardiology,vol. 4, edited by E. Chazov, V. Saks, G. Rona, New York, Plenum Publishing, pp. 2343 (1983). ISHIBASHI,T., ICHIHARA,K., ABIKO,Y. Difference in the time course between increases in coronary flow and in effluent adenosine concentration during anoxia in the perfnsed rat heart. Jpn CircJ 49, 1090-1098 (1985). KAMMERMEIER, H., DECKING, U. Interstitial concentration of adenosine and inosine in isolated rat hearts of rats and guinea-pigs. Pfliigers Arch 407 [Suppl 1], 550 (1986). NAVLER, W. G., POOLE-WILSON,P. A., WILLIAMS,A. Hypoxia and calcium.,J Mol Cell Cardiol 11,683-706 (1979). NEeLY, J. R., LIEnERMeISTeR, H., BATTeRSBY, E.J., MORGAN, H. E. Effect of pressure development on oxygen consumption by isolated rat heart. Am J Physio1212, 804-814 (1967). Nero, S., GERLACH, E. Adenine nucleotide and adenosine metabolism in cultured coronary endothelial cells. Formation and release of adenine compounds and possible functional implication. In: RegulatoryFur~tion of Adenosine, edited by R. M. Berne, T. W. Rail, R. Rubio. The Hague, Martinus Nijhoff, pp. 347-360 (1983). Ovle, L. H. Adequacy of oxygenation of isolated perfnsed rat heart. Basic Res Cardio179, 300-306 (1984). PADDLE, B. M., BURNSTOCK,G. Release of ATP from perfused heart during coronary vasodilation. Blood Vessels 11, 11(~119 (1974). PARRATT,J. R., WADSWORTI4,R. M. The effects of dipyridamole on the myocardial vasodilator actions of noradrenaline, isoprenaline and adenosine. BrJ Pharmaco146, 585-593 (1972). PeAaSON, J. D., CARt.ETON,`]. S., GORDON,`]. L. Metabolism of adenine nucleotides by ectoenzymes of vascular endothelial and smooth muscle cells in culture. BiochemJ 190, 421429 (1980). PeARSON,J. D., GORDON,J. L. Nucleotide metabolism by endothelium. Ann Res Physio147, 617~27 (1985). RONCA-TEsTONI, S., BORGHINI,F. Degradation of perfused adenine compounds up to uric acid in isolated rat heart..J Mol Cell Cardiol 14, 177-180 (1982). ROTI~AUL, A. L., BROADLY,K.J. Use ofhexobendine to examine whether the coronary vasodilator metabolite released from guinea-pig isolated hearts is adenosine. Cardiovasc Res 15, 599-610 (1981). RUBIO, R., BERNE, R. M. Release of adenosine by the normal myocardium in dogs and its relationship to the regulation of coronary resistance. Circ Res 25, 4 0 7 4 15 (1969). SAITO, D., STeINHART, C. R., NIXON, D. G., OLSSON, R. A. Intracoronary adenosine deaminase reduces canine myocardial reactive hyperemia. Circ Res 49, 1262 1267 (1981). SCHRADeR,,]., HADDV, F.J., GERLACH, E. Release of adenosine, inosine and hypoxanthine from the isolated guinea-pig heart during hypoxia, flow-autoregulation and reactive hyperemia. Pflllgers Arch 369, 1-6 (1977). SPARKS, H. V., BARDeNHEUeR,H. Regulation of adenosine formation by the heart. Circ Res 58, 193-201 (1986). WILLIAMS, C. A., FORRESTeR, T. Possible source of adenosine triphosphate released from rat myocytes in response to hypoxia and acidosis. Cardiovasc Res 17, 301 312 (1983). WOLF, M. M., BERNE, M. D. Coronary vasodilator properties of purine and pyrimidine derivatives. Circ Res 4, 343-348 (1956). WROBI~eWSKI,F., LA Due,,]. S. Lacficodehydrogenase activity in blood. Proc Soc Expt Biol Med 20, 210-213 (1955).