Potent effects of APSA and AP4A on coronary resistance and autacoid release of intact rabbit hearts ULRICH POHL, ADALING OGILVIE, DANIEL LAMONTAGNE, AND RUDI BUSSE Department of Applied Physiology, University of Freiburg, D-7800 Freiburg; Institut fib Biochemie (Medizinische Fakulttit) der Universittit Erlangen-Ntirnberg, D-8500 Erlangen, Federal Republic of Germany
exhibit considerable vasodilator activities at physiological concentrations (6). In these vessels, both dinucleotides acted as uncleaved compounds and did not act Physiol. 260 (Heart Circ. Physiol. 29): Hl692-H1697, 1991.through their hydrolysis products on endothelial recepWe investigated effects of platelet-derived dinucleotides dia- tors to release endothelium-derived relaxing factor denosine 5’,5”’ -P1,P3-triphosphate (AP3A) and diadenosine (EDRF; AP4A) and/or on smooth muscle receptors 5’5”-P1,P4-tetraphosphate (AP4A) on coronary vasculature. In isolated rabbit hearts, saline perfused at constant flow (36 (AP3A; see Ref. 6). It is not known, however, whether 2 3 ml/min), AP3A and AP4A induced dose-dependent de- these compounds are also active in the vascular bed of creases in coronary perfusion pressure. Dose-effect curves of an intact organ and especially in resistance arteries. The surface of a peripheral vascular bed APaA [-log M mean effective concentration (EC& 6.2 * 0.11 large endothelial and APdA (E& 6.4 & 0.2) were identical and not significantly may rapidly catabolize these compounds and also rapidly different from those of adenosine, ADP, and ATP (n = 4-8). clear their degradation products. Moreover, it is unThere were, however, distinct differences between both dinu- known whether AP3A and APhA have the potency to cleotides: pretreatment with endothelium-derived relaxing fac- release EDRF and prostaglandin (PG)Iz, autacoids that tor (EDRF)-inhibitors oxyhemoglobin (6 PM, n = 6) and pare important for both inhibition of platelet activation nitro=L-arginine (30 PM, n = 6) significantly reduced AP4A(1, 5, 18) and vasodilation (2) in an intact vascular bed. induced dilation by 44 and 42% but did not affect vasomotor Here we report that both dinucleotides are potent dilaeffects of AP3A or of sodium nitroprusside, adenosine, ATP, and ADP. Concentration of the stable hydrolysis product of tors of the coronary resistance vessels by mechanisms of PGIg release and only in part due to prostaglandin (PG)I*, 6-keto-PGFla, increased by 173 * 25% independent in coronary effluent (n = 23) during infusion of APaA (1 PM). EDRF release (APdA). Compared with ATP their surThis increase was significantly higher than during infusion of vival during a passage through the coronary bed is ~10 equimolar concentrations of AP4A (38 * lo%), ATP (23 * 5%), times higher. Differences with respect to the stimulatory adenosine (20 & lo%), or an equimolar combination of AMP effect on PGIz and EDRF release between several adeand ADP (52 k 25%), the hydrolysis products of APaA. Luminine nucleotides suggest that both uncleaved dinucleonometric and high-performance liquid chromatography analy- tides have direct effects on endothelial and/or vascular sis showed a nearly complete (94 & 3%) degradation of ATP smooth muscle cells of the coronary system.
POHL, ULRICH, ADALING OGILVIE, DANIEL LAMONTAGNE, AND RUDI BUSSE. Potent effects of AP3A and AP4A on coronau resistance and autacoid release of intact rabbit hearts. Am. J.
during passage through the coronary bed while significant amounts of AP3A (31 k 5%) and APdA (33 * 6%) remained uncleaved. We conclude that both dinucleotides are potent dilators of coronary resistance vessels. Compared with ATP, their survival during a passage through the coronary bed is -10 times higher. Their specific effects on PGIz and EDRF release suggest that both uncleaved dinucleotides have direct effects on cells of the vascular wall. endothelium-derived relaxing factor; endothelium-derived nitric oxide; nucleotides; adenosine triphosphb; &mmi~~; prostaglandin Iz; Langenclorff hearts
CONSIDERABLE
AMOUNTS
of the dinucleotides
sine 5’,5”’
diadeno-
-P1,P3-triphosphate (AP3A) and diadenosine 5’,5”‘-P’,P*-tetraphosphate (APdA) are stored in the dense granules of human platelets and are almost completely released during thrombin-induced aggregation (8, 13). These compounds have been shown to occur in the plasma (17) and may function as antagonistic modulators of platelet aggregation (11). Moreover, studies on isolated rabbit mesenteric arteries revealed that AP3A and APdA H1692
0363-6135/91
$1.50
METHODS
Mongrel rabbits of either sex (1.5-2 kg body wt) were anesthetized with pentobarbital sodium (30 mg/kg) and anticoagulated with heparin. One carotid artery was transected, and the animal was bled. After a subsequent lethal dose of pentobarbital sodium (30 mg/kg) the heart was rapidly excised and perfused through a cannula inserted into the aortic stump (Langendorff method). The perfusion solution consisted of a modified KrebsHenseleit buffer containing (in mM) 118 NaCl, 4 KCl, 2.5 CaClz, 1 KHzP04, 1.2 MgSOJ, 25 NaHC03, 5 glucose, and 2 pyruvate dissolved in double-distilled water. The perfusate was gassed with 95% 02.5% CO2 to maintain a pH of 7.4 and kept at a constant temperature of 37’C. Constant flow perfusion was established by means of a roller pump (Heidolph RGL 85, Kelheim, FRG). Coronary perfusion pressure was measured with a pressure transducer (Gould P 2310, Oxnard, CA) connected to a sidearm of the perfusion cannula. Isovolumetric left perfusion pressure was measured by a fluid-filled balloon
Copyright @ 1991 the American Physiological
Society
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 16, 2019.
AP3A
AND
AP4A
CAUSE
VASODILATION
inserted into the left ventricle and connected to a second pressure transducer. Heart rate was derived from the left ventricular pressure signal. All parameters were recorded on a multichannel pen recorder (Rikadenki R 50). Vasodilators were infused through a Y-connector in the aortic perfusion line at 1% of the coronary flow rate by means of a constant flow perfusor (Braun, Secura, Melsungen, FRG) using concentrated stock solutions to produce the final concentrations detailed in &perimental protocols. ExperimentaZ protocols. In a first series, dose-response
relationships for the vasodilator test agents were obtained in a noncumulative manner after an equilibration period of 30 min. Each agent was infused for at least 1 min, during which period a steady-state dilation was obtained. At least two different nucleotide dilators were used in one experiment together with standard doses (final concentration: 1 PM each) of the endotheliumindependent vasodilator sodium nitroprusside (SNP) and -the endothelium-dependent dilator acetylcholine (ACh). In three further series, standard dosages (final concentrations: 1 PM) of the vasodilators were infused before and after treatment of the coronary vasculature with the EDRF-inhibiting compounds (final concentrations) oxyhemoglobin (6 PM), fl-nitro+arginine (L-NNA, 30 PM) or the cyclooxygenase inhibitor indomethacin (10 PM). These substances were infused through a separate infusion line into the aortic cannula at 1% of the coronary flow rate. Continuous perfusion of the inhibitor was started 30 min before repetition of the vasodilator infusion and maintained throughout till the end of the experiment. In some experiments, instead of the inhibitor, only the solute (NaCl) was infused. To determine the release of PGIg, the coronary effluent was collected for 20 s immediately before in .fusion of the vasodilators as well as after a steady-state reduction of coronary perfusion pressure had been reached. The samples were frozen and kept at -7O’C until determination of 6-keto-PGFla, the stable metabolite of prostacyclin, by means of a specific radioimmunoassay (4). Degradation of nucleotides. In some experiments, during steady-state dilation induced by the various vasodilators, the coronary effluent was collec ted for 20 s to determine the degradation of adenine nucleotides during passage through the coronary vascular bed. To this end, the nucleotide-containing Krebs-Henseleit solutions were analyzed after having passed through rabbit hearts and were compared with the stock solution infused. Analysis of nucleotide catabolism performance liquid chromatography
by means of high(HPLC). The re-
verse-phase technique was applied as described with the modification that one column was used of two; this technique yielded shorter elution times with sti 11 sufficient resolution of the potential degradation products. Krebs-Henseleit solution containing the nucleotides (50-200 ~1) was directly loaded onto the column and eluted with a continuous gradient of methanol (O6% during 20 min) in 0.1 M KHZPOd, at pH 5.1. Bioluminometry. Analysis of nucleotide concentrations (ATP, APaA, AP*A) with bioluminescence techniques has been performed as described in detail previously (20,
AND
AUTACOID
RELEASE
H1693
21). Briefly, luciferin-luciferase produces light with ATP that is generated from APdA after hydrolysis with snake venom phosphodiesterase. The assay for APsA additionally contained phosphoenolpyruvate (1.9 mM) and pyruvate kinase (0.1 mg/ml). The assay contained, in a final volume of 0.3 ml, 30 ~1 of luciferin-luciferase (LKB, Pharmacia, Freiburg, FRG), 3.8 mM MgCIZ, 25 mM N2-hydroxyethylpiperazine-W-2-ethanesulfonic acid (pH 7.75), and l-10 ~1 of perfusate. The reaction was started at room temperature by the addition of snake venom phosphodiesterase (0.1 pg/rnl final concentration). The increase of luminescence was monitored with a photomultiplier (Lumacounter model 2080; Lumac Syitems, Basel, Switzerland). The light readings were monitored every 30 s (integration mode) until reaching the maximum value after 3-5 min. Drugs. ACh hydrochloride, adenosine (Ado), AMP, ADP, ATP, APaA, AP*A, bovine hemoglobin, and indomethacin were obtained from Sigma (Deisenhofen, FRG). Unless otherwise stated all reagents were dissolved in Tyrode solution. Hemoglobin and indomethatin were prepared as described earlier (6, 25). L-NNA was obtained from Serva (Heidelberg, FRG). It was dissolved in Tyrode solution under vigorous stirring at a temperature of 60°C. SNP (Sigma) was also dissolved in sodium acetate, diluted with Tyrode solution, and protected from light throughout the experiments. Statistics. All values are presented as means & SE. Differences in the vasodilator effects were tested by means of t tests for paired and unpaired data. Vasodilation (i.e., decrease in coron .ary perfision pressure) was expressed in percent of the actual contractile tone [i.e., actual coronary perfusion pressure - minimal coronary perfusion pressure obtained by supramaximal (XOW5 M) doses of Ado or SNP] for dose-effect curves or as decrease of the actual coronary perfusion pressure (comparison of standard doses of vasodilators). P c 0.05 was considered significant. RESULTS
The mean coronary flow rate as measured in a total of 42 experiments amounted to 36 & 3 ml/min, and the resulting coronary perfusion pressure amounted to 62 & 6 mmHg. The heart rate was 138 k 6 s-l, and the developed left ventricular pressure was 110 & 11 mmHg. Figure 1 depicts the dose-effect curves of the various vasodilators studied. There were no significant differences between the dose-effect curves of Ado, ADP, ATP, APsA, and APdA. The corresponding mean effective concentration (EC&) values amounted to 6.39 k 0.18 (Ado), 6.15 & 0.17 (ADP), 6.21 & 0.11 (ATP), 6.20 & 0.08 (APsA), and 6.36 & 0.21 (APdA). Only the EC50 for AMP was significantly (5.68 & 0.02; P < 0.05) different from those of the other compounds. Infusion of the EDRF-inhibitor hemoglobin or the stereospecific inhibitor of EDRF formation L-NNA increased the coronary perfusion pressure by 35 & 11 and 42 + 12%. Both compounds inhibited or reversed the dec;ase in coronary perfusion pressure induced by ACh but did not affect the vasodilator response to ATP or APsA. The response to ADO and ADP was also not
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 16, 2019.
Hl694
AP3A AND AP4A CAUSE VASODILATION AND AUTACOID RELEASE
m ADO v AMP A ADP 0 ATP
07 A -v j-7
0 control m L-NNA
AP4A
ACh
AP3A
SNP
30N
ACPP VI0 0
n control m Hb
.
APjA
AP4A
ATP
ACh
01 .
1 agonist [~I41
10
FIG. 1. Dilator effects of various adenine compounds on coronary resistance vessels. Dose-effect curves were obtained in a noncumulative manner (steady-state infusions of single doses). “Dilation” refers to decrease in coronary perfusion pressure. Maximal dilation (minimal coronary perfusion pressure) was obtained by a supramaximal dose of adenosine (Ado) or papaverine; n = 4-8 vessels. For sake of clarity, SD are not indicated. APaA, diadenosine 5’,5”‘-P1,P3-triphosphate; AP4A, diadenosine 5’,5”’ -P1,P4-tetraphosphate.
affected by either inhibitor. In contrast, the decrease in coronary perfusion pressure induced by APdA was reduced by 44 & 10 and 42 & 12% (Fig. 2). Infusion of the solute alone did not alter the AP*A-induced decrease in coronary perfusion pressure (not shown). Unlike the EDRF inhibitors, indomethacin did not affect selectively the response of coronary perfusion pressure to any vasodilator (Table 1). Release of PG&. In 23 experiments, the effluent from the heart was collected immediately before and during the steady-state dilation induced by infusion of standard doses (1 PM) of ACh, SNP, Ado, or adenine nucleotides (at least 4 different compounds in each single experiment) for determination of the concentration of 6-ketoPGFl&, the stable hydrolysis product of PGIz. The mean release of PGIZ from these hearts under unstimulated conditions amounted to 15.3 & 2.2 ng/min. There was a considerable variance between the individual hearts with values ranging from 2.7 ng/min to 34.9 ng/min. Figure 3 depicts the percent increase of the release of PGIz during infusion of the vasodilators. In contrast to the widely uniform vasodilator effects, the stimulator effects on the PGIZ release showed remarkable differences. The most potent agents were ACh and APSA. Pretreatment of the hearts with indomethacin significantly reduced the basal release of PGIz (0.5 * 0.001 ng/min, n = 3) and completely inhibited the increased release in response to these agents (not shown). For the most potent stimulating nucleotide, APsA, it was tested whether the same stimulator effect could be
FIG. 2. Effects of 1 PM concentrations of acetylcholine (ACh), AP4A, APaA, ATP, and sodium nitroprusside (SNP) on coronary perfusion pressure (CPP) before and after treatment with endotheliumderived relaxing factor (EDRF)-inhibitors p-nitro+arginine (LNNA; n = 6) or oxyhemoglobin (Hb; n = 6). ACPP indicates a decrease in perfusion pressure. Only dilator responses to ACh and AP4A are affected. **p < 0.01; NS, difference not significant.
TABLE 1. Effect of indomethacin on nucleotide-induced decrease in coronary perfusion pressure Dilator Responses (% of response to SNP, ADP
Control Indomethacin 00
ATP
100.7&12.0 127.Ok31.1
95.Wl6.0 159.8k55.5
1 PM)
AP3A
90.5212.1 155.lk53.6
APdA
93.8kl2.3 170.9k68.4
PM)
Values are means * SE; n = 4 hearts. There are no significant differences between controls and values after indomethacin treatment. SNP, sodium nitroprusside; APaA, diadenosine 5’,5’ ’ ‘-P1,P3-triphosphate; AP4A, diadenosine 5’,5”‘-P1,P4-tetraphosphate.
A PG&L/l 0
0
200-
O-
ACh AP3A AP4A ATP ADO
FIG. 3. Stimulator effects of ACh (n = 18), APSA (n = 23), AP4A (n = 19), ATP (n = Id), and adenosine (n = 13) on release of prostacyclin (PGIg) from coronary vascular bed. APaA was a significantly more potent stimulator than AP4A. APGIZ,increase in percent of control sample, obtained immediately before infusion of compound. *P < 0.05; *** p < 0.001 compared with APZA.
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 16, 2019.
APsA AND APdA CAUSE
I-
-
*1
VASODILATION
AND AUTACOID
A
T
H1695
RELEASE
B
ns,
1
A
40 0ml 30 A CPP 0 VI0
1
0I
l-
n.s.,lr
TI
ns.r-l
. T:
ADP
T I
AMP + ADP
AP3A
FIG. 4. Comparison of PG12 stimulator and vasodilator effects of APaA (1 PM) with an equimolar combination of its hydrolysis products AMP + ADP or ADP alone. Although all 3 interventions induced similar decreases in perfusion pressure, amount of PG12 released in response to APaA was significantly higher. (*P c 0.01; n = 5)
AP3A
A[A
ATP
FIG. 5. Survival of nucleotides during perfusion of Langendorff hearts (n = 5). Hearts have been perfused with Krebs-Henseleit buffer containing nucleotide to be investigated (APsA, APbA, ATP) at an initial concentration of 1 PM. Concentration of nucleotides in perfusates have been determined using bioluminometry methods. Data are means & SE.
obtained after simultaneous centrations of its immediate and ADP. Neither substance of both stimulated PG12 in (Fig. 4).
ofATP,
infusion of equimolar condegradation products AMP alone nor the combination a similar manner to APSA
To assess the degradation of the nucleotides during passage of the coronary vascular bed, the concentrations of ATP, AP3A, Biological stability
AP&,
40 min 40 min 20 FIG. 6. High-performance liquid chromatography tracings demonstrating hydrolysis of nucleotides in perfusate after passage through rabbit heart. ATP, APdA, APsA, and adenosine were dissolved in KrebsHenseleit solution at a final concentration of 1 PM. A-D: top tracings, chromatograms of solutions before perfusion; bottom trucings, corresponding solutions after passage through heart (at 4 min after starting perfusion). Analysis of adenosine (A ), ATP (B), APdA (C), and APA (D); the shorter elution times in D were obtained after exchanging column. Identification of nucleoside peaks (A, adenosine; I, inosine) as well as of nucleotides was performed by cochromatography with standard compounds.
20
and AP&.
and AP4A were determined in the effluent of the heart. Samples were obtained during steady-state perfusion (50 s after start) with Krebs-Henseleit buffer at initial 1 PM concentrations of the respective compounds. Applying the highly sensitive bioluminescence assay, it was possible to determine a residual ATP concentration of ~5% of the initial value in the coronary effluent during infusion of this compound (Fig. 5). In contrast, both APaA and APdA revealed an -lo-fold higher biological stability as reflected by effluent concentrations >30% of the initial values (Fig. 5). HPLC analysis of the degradation products in the effluent confirmed the higher stability of the dinucleotides (Fig. 6, C and D) compared with ATP (Fig. 6B). The pattern of degradation products was very similar for all nucleotides, showing AMP, Ado, and inosine as the only detectable compounds. Figure 6 depicts HPLC tracings of one perfusion experiment. Virtually identical tracings were obtained in four further experiments. DISCUSSION
Both dinucleotides, APsA and AP*A, acted as potent vasodilators of resistance vessels in the intact heart.
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 16, 2019.
H1696
APZA AND APdA CAUSE VASODILATION
Maximal dilation was obtained at even lower concentrations than previously observed in isolated mesenteric arteries (6). Distinct dilations were obtained already at concentrations of 0.1 PM, which is even below the concentration range that is likely reached during local activation of platelets. Based on the intraplatelet nucleotide content (12), it can be calculated that the complete release of APSA and AP*A during platelet activation would lead to a plasma concentration in the range of 0.5-l PM. In the surroundings of aggregating platelets, especially the platelet-endothelial cell-interface, even much higher concentrations must be assumed. Both dinucleotides apparently activate different receptor-dependent transduction cascades in endothelial and vascular smooth muscle cells. The partial inhibitory action of the EDRF-inhibitors L-NNA and hemoglobin on the APdA-induced dilation indicates that this compound induces EDRF release in coronary resistance vessels. In contrast to our previous findings in isolated mesenteric arteries (6), a significant part of the dilator effect of AP*A does not seem to be mediated by EDRF. There was also no indication of EDRF release in response to ATP, which in other vascular beds of the same species in vivo (26) and in vitro (6, 9) stimulates EDRF release by acting through endothelial PzY receptors (3). As in mesenteric arteries, EDRF does not mediate the dilation induced by APSA. Although APaA was, in contrast to the other nucleotides, a potent stimulator of PGIZ release, it is unlikely that PGIZ was a significant mediator of AP3Ainduced dilation, because treatment with indomethacin did not selectively inhibit the dilator effect of APSA compared with all other nucleotides. However, it is tempting to speculate that the high release of the antiaggregating PGIZ in response to APaA has functional significance as a simultaneous antagonistic principle to the proaggregatory action of APaA on platelets (14). This fits into the general principle of simultaneous activation of dual control mechanisms in hemostatic processes (19). The analysis of the degradation of the dinucleotides in the intact coronary vascular bed revealed that a considerable amount of APSA and APAA remained uncleaved. Therefore, in contrast to ATP, APSA and APdA can be considered as long-lived substances in the coronary vasculature. The rate-limiting step of degradation presumably is the asymmetric cleavage of the dinucleotides by pyrophosphohydrolases (15, 16), which are found to be located as ectoenzymes on endothelial (23) as well as on vascular smooth muscle cells (22). The relative rates of degradation to AMP and ADP/ATP during heart passage were in a similar range as found on freshly isolated or cultured endothelial and smooth muscle cells (22, 23). Further degradation of these cleavage products was obviously controlled by the cascade of ectonucleotidases on both types of vascular cells (24). The activity of 5’nucleotidase appears to be the second rate-limiting step. With respect to ATP, our findings are in good agreement with recent investigations in perfused rat hearts (7). In view of the considerable vascular activity of the whole spectrum of degradation products (lo), the question arises whether APsA and APdA exerted their dilator and PGIZ-stimulating activity per se or only after hydrolytic cleavage to mononucleotides or nucleosides. There
AND AUTACOID
RELEASE
are, however, several findings that suggest that APsA and APdA have stimulating effects on endothelial and vascular smooth muscle cells as uncleaved parent compounds. First, there was no difference in the dose-effect curves obtained for the dilations induced by the dinucleotides compared with ATP and Ado. As a matter of fact, the dose-effect curve for AMP was even shifted to the right, which is not consistent with the assumption that the main degradation products, AMP and Ado, were the main mediators of vasodilation. Second, the production of PGIZ induced by APSA could not be mimicked by either cleavage product alone or by the equimolar combination of the immediate hydrolysis products ADP and AMP. Likewise, AP*A induced a distinct EDRF-mediated dilation that was not observed in response to any of the potential degradation products. Further support for a direct action of both dinucleotides on the vascular wall is derived from our previous findings in isolated mesenteric segments. In these segments, exhibiting only a small endothelial surface in relation to the perfused volume over time, no degradation of APsA and APdA could be detected (6). Nevertheless, both compounds induced significant vasodilation. The unique effect of APsA on PGIZ release and that of APdA on EDRF release, which were not shared by any other nucleotide or nucleoside tested, suggest that both compounds may act on specific subclasses of PZ receptors. However, their characterization and localization on endothelial and/or smooth muscle cells remain to be established. In conclusion, our studies demonstrate that the platelet-derived dinucleotides APaA and APdA are long-lived and potent vasodilators of resistance vessels of the rabbit heart. They may represent a novel class of physiological regulators of vascular tone and platelet activity at sites distal to thrombus formation, which may be subject to an enhanced risk of thrombus propagation and embolization. This study was supported by the Deutsche Forschungsgemeinschaft Grant PO 307/l-Z, Og 2/3-3. D. Lamontagne is a research fellow from the Heart and Stroke Foundation of Canada. Address for reprint requests: U. Pohl, Institut fur Physiologic, Medizinische Universitat zu Lubeck, Ratzeburger Allee 160, D-2400 Lubeck 1, FRG. Received 10 September 1990; accepted in final form 11 January 1991. REFERENCES 1. AZUMA, H., M. ISHIKAWA, AND S. SEKIZAKI. Endothelium-dependent inhibition of platelet aggregation. Br. J. Phurmacol. 88: 4ll415,1986. 2. BASSENGE, E., AND R. BUSSE. Endothelial modulation of coronary tone. Prog. Cardiovasc. Dis. 30: 349-380, 1988. G., AND C. KENNEDY. A dual function for adenosine 3. BURNSTOCK, 5’-triphosphate in the regulation of vascular tone. Excitatory cotransmitter with noradrenaline from perivascular nerves and locally released inhibitory intravascular agent. Circ. Res. 58: 319330,1985. 4. BUSSE, R., U. F~RSTERMANN, H. MATSUDA, AND U. POHL. The role of prostaglandins in the endothelium-mediated vasodilatory response to hypoxia. PfZuegers Arch. 401: 77-83, 1984. AND E. BASSENGE. Endothelium-derived 5. BUSSE, R., A. LUCKHOFF, relaxant factor inhibits platelet activation. Naunyz-Schmiedeberg’s Arch. Pharmacol. 336: 566-571, 1987. AND U. POHL. Vasomotor activity of 6. BUSSE, R., A. OGILVIE, diadenosine triphosphate and diadenosine tetraphosphate in iso-
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 16, 2019.
APaA AND APdA CAUSE VASODILATION lated arteries. Am. J. Physiol.
254 (Heart
Circ.
Physiol.
23):
H828-
H832,1988. 7 FLEETWOOD,
G., S. B. COADE, J. L. GORDON, AND J. D. PEARSON. Kinetics of adenine nucleotide catabolism in coronary circulation of rats. Am. J. Physiol. 256 (Heart Circ. Physiol. 25): Hl565Hl572,1989. 8. FLODGAARD, H., AND H. KLENOW. Abundant amounts of diadenosine 5’,5’ ‘I- P1,P4-tetraphosphate are present and releasable, but metabolically inactive, in human platelets. Biochem. J. 208: 737742,1982. 9. FURCHGOW, 10. 11.
12.
13.
R. F. Role of endothelium in responses of vascular smooth muscle. Circ. Res. 53: 557-573, 1983. GERLACH, E., AND B. F. BECKER. Topics and Perspectiues in Adenosine Research. Berlin: Springer-Verlag, 1987. LUTHJE, J., J. BARINGER, AND A. OGILVIE. Effects of adenosine triphosphate (APZA) and adenosine tetraphosphate (AP4A) on platelet aggregation in unfractionated human blood. BZut 51: 405413,1985. LUTHJE, J., D. MILLER, AND A. OGILVIE. Unproportionally high concentrations of diadenosine triphosphate (APaA) and diadenosine tetraphosphate (AP4A) in heavy platelets. Consequences for in vitro studies with human platelets. BZut 54: 193-200, 1987. LUTHJE, J., AND A. OGILVIE. The presence of diadenosine 5’,5”‘Biophys. P1,P3-triphosphate (APaA) in human platelets. Biochem. Res. Commun.
115: 253-260,1983.
14. LUTHJE, J., AND A. OGILVIE. Diadenosine triphosphate (AP3A) mediates human platelet aggregation by liberation of ADP. Biochem.
Biophys.
Res. Commun.
118: 704-709,1984.
15 LUTHJE, J., AND A. OGILVIE. Catabolism of AP4A and AP3A in human plasma: purification and characterization of a glycoprotein complex with 5’-nucleotide phosphodiesterase activity. Eur. J. 16.
Biochem. LUTHJE,
149: US-127,1985. J., AND A. OGILVIE.
human serum: identification acterization. Eur. J. Biochem.
Catabolism of AP4A and AP3A in of isoenzymes and their partial char169: 385-388.
1987.
AND AUTACOID
RELEASE
H1697
17. LUTHJE, J., AND A. OGILVIE. Catabolism of APhA and APZA in whole blood the dinucleotides are long-lived signal molecules in the blood ending up as intracellular ATP in the erythrocytes. &r. J. Biochem. 18. MONCADA,
173: 241-245,1988.
S., E. A. HIGGS, AND J. R. VANE. Human arterial and venous tissues generate prostacyclin (prostaglandin X), a potent inhibitor of platelet aggregation. Lancet 1: 18-21, 1977. 19. NAWROTH, P. P., AND D. M. STERN. Endothelial cell procoagulant properties and the host response. Semin. Thromb. Hemostasis 13:
391-397,1987. 20. OGILVIE, A.
Determination of diadenosine tetraphosphate (AP4A) levels in subpicomole quantities by a phosphodiesterase luciferinluciferase coupled assay: application as a specific assay for diadenosine tetraphosphatase. Anal. B&hem. 115: 302-307,198l. A., AND P. JAKOB. Diadenosine 5’,5”‘-P1,P3-triphos21. OGILVIE, phate in eukaryotic cells: identification and quantitation. Anal
Biochem. 22. OGILVIE,
134: 382-392,1983.
A., AND J. LUTHJE. Diadenosine tetraphosphate (AP4A) and diadenosine triphosphate (AP3A) are degraded by ectoenzymes of arterial smooth muscle cells. Biol. Chem. Hoppe-Seyler 369: 887-
888,1988. 23. OGILVIE,
A., J. LUTHJE, U. POHL, AND R. BUSSE. Identification and characterization of an AP4A hydrolase on intact bovine aortic endothelial cells. Biochem. J. 259: 97-103, 1989. 24. PEARSON, J. D., J. S. CARLETON, AND J. L. GORDON. Metabolism of adenine nucleotides by ectoenzymes of vascular endothelial and smooth muscle cells in culture. Biochem. J. 190: 421-429, 1980. 25. POHL, U., AND R. BUSSE. EDRF increases cyclic GMP in platelets during passage through the coronary vascular bed. Circ. Res. 65:
1798-1803,198s. 26. POHL, U., L. DIIZSI,
B. SIMON, AND R. BUSSE. Selective inhibition of endothelium-dependent dilation in resistance-sized vessels in 253 (Heart Circ. Physiol. 22): H234-H239, vivo. Am. J. Physiol. 1987.
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 16, 2019.
exhibit considerable vasodilator activities at physiological concentrations (6). In these vessels, both dinucleotides acted as uncleaved compounds and did not act Physiol. 260 (Heart Circ. Physiol. 29): Hl692-H1697, 1991.through their hydrolysis products on endothelial recepWe investigated effects of platelet-derived dinucleotides dia- tors to release endothelium-derived relaxing factor denosine 5’,5”’ -P1,P3-triphosphate (AP3A) and diadenosine (EDRF; AP4A) and/or on smooth muscle receptors 5’5”-P1,P4-tetraphosphate (AP4A) on coronary vasculature. In isolated rabbit hearts, saline perfused at constant flow (36 (AP3A; see Ref. 6). It is not known, however, whether 2 3 ml/min), AP3A and AP4A induced dose-dependent de- these compounds are also active in the vascular bed of creases in coronary perfusion pressure. Dose-effect curves of an intact organ and especially in resistance arteries. The surface of a peripheral vascular bed APaA [-log M mean effective concentration (EC& 6.2 * 0.11 large endothelial and APdA (E& 6.4 & 0.2) were identical and not significantly may rapidly catabolize these compounds and also rapidly different from those of adenosine, ADP, and ATP (n = 4-8). clear their degradation products. Moreover, it is unThere were, however, distinct differences between both dinu- known whether AP3A and APhA have the potency to cleotides: pretreatment with endothelium-derived relaxing fac- release EDRF and prostaglandin (PG)Iz, autacoids that tor (EDRF)-inhibitors oxyhemoglobin (6 PM, n = 6) and pare important for both inhibition of platelet activation nitro=L-arginine (30 PM, n = 6) significantly reduced AP4A(1, 5, 18) and vasodilation (2) in an intact vascular bed. induced dilation by 44 and 42% but did not affect vasomotor Here we report that both dinucleotides are potent dilaeffects of AP3A or of sodium nitroprusside, adenosine, ATP, and ADP. Concentration of the stable hydrolysis product of tors of the coronary resistance vessels by mechanisms of PGIg release and only in part due to prostaglandin (PG)I*, 6-keto-PGFla, increased by 173 * 25% independent in coronary effluent (n = 23) during infusion of APaA (1 PM). EDRF release (APdA). Compared with ATP their surThis increase was significantly higher than during infusion of vival during a passage through the coronary bed is ~10 equimolar concentrations of AP4A (38 * lo%), ATP (23 * 5%), times higher. Differences with respect to the stimulatory adenosine (20 & lo%), or an equimolar combination of AMP effect on PGIz and EDRF release between several adeand ADP (52 k 25%), the hydrolysis products of APaA. Luminine nucleotides suggest that both uncleaved dinucleonometric and high-performance liquid chromatography analy- tides have direct effects on endothelial and/or vascular sis showed a nearly complete (94 & 3%) degradation of ATP smooth muscle cells of the coronary system.
POHL, ULRICH, ADALING OGILVIE, DANIEL LAMONTAGNE, AND RUDI BUSSE. Potent effects of AP3A and AP4A on coronau resistance and autacoid release of intact rabbit hearts. Am. J.
during passage through the coronary bed while significant amounts of AP3A (31 k 5%) and APdA (33 * 6%) remained uncleaved. We conclude that both dinucleotides are potent dilators of coronary resistance vessels. Compared with ATP, their survival during a passage through the coronary bed is -10 times higher. Their specific effects on PGIz and EDRF release suggest that both uncleaved dinucleotides have direct effects on cells of the vascular wall. endothelium-derived relaxing factor; endothelium-derived nitric oxide; nucleotides; adenosine triphosphb; &mmi~~; prostaglandin Iz; Langenclorff hearts
CONSIDERABLE
AMOUNTS
of the dinucleotides
sine 5’,5”’
diadeno-
-P1,P3-triphosphate (AP3A) and diadenosine 5’,5”‘-P’,P*-tetraphosphate (APdA) are stored in the dense granules of human platelets and are almost completely released during thrombin-induced aggregation (8, 13). These compounds have been shown to occur in the plasma (17) and may function as antagonistic modulators of platelet aggregation (11). Moreover, studies on isolated rabbit mesenteric arteries revealed that AP3A and APdA H1692
0363-6135/91
$1.50
METHODS
Mongrel rabbits of either sex (1.5-2 kg body wt) were anesthetized with pentobarbital sodium (30 mg/kg) and anticoagulated with heparin. One carotid artery was transected, and the animal was bled. After a subsequent lethal dose of pentobarbital sodium (30 mg/kg) the heart was rapidly excised and perfused through a cannula inserted into the aortic stump (Langendorff method). The perfusion solution consisted of a modified KrebsHenseleit buffer containing (in mM) 118 NaCl, 4 KCl, 2.5 CaClz, 1 KHzP04, 1.2 MgSOJ, 25 NaHC03, 5 glucose, and 2 pyruvate dissolved in double-distilled water. The perfusate was gassed with 95% 02.5% CO2 to maintain a pH of 7.4 and kept at a constant temperature of 37’C. Constant flow perfusion was established by means of a roller pump (Heidolph RGL 85, Kelheim, FRG). Coronary perfusion pressure was measured with a pressure transducer (Gould P 2310, Oxnard, CA) connected to a sidearm of the perfusion cannula. Isovolumetric left perfusion pressure was measured by a fluid-filled balloon
Copyright @ 1991 the American Physiological
Society
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 16, 2019.
AP3A
AND
AP4A
CAUSE
VASODILATION
inserted into the left ventricle and connected to a second pressure transducer. Heart rate was derived from the left ventricular pressure signal. All parameters were recorded on a multichannel pen recorder (Rikadenki R 50). Vasodilators were infused through a Y-connector in the aortic perfusion line at 1% of the coronary flow rate by means of a constant flow perfusor (Braun, Secura, Melsungen, FRG) using concentrated stock solutions to produce the final concentrations detailed in &perimental protocols. ExperimentaZ protocols. In a first series, dose-response
relationships for the vasodilator test agents were obtained in a noncumulative manner after an equilibration period of 30 min. Each agent was infused for at least 1 min, during which period a steady-state dilation was obtained. At least two different nucleotide dilators were used in one experiment together with standard doses (final concentration: 1 PM each) of the endotheliumindependent vasodilator sodium nitroprusside (SNP) and -the endothelium-dependent dilator acetylcholine (ACh). In three further series, standard dosages (final concentrations: 1 PM) of the vasodilators were infused before and after treatment of the coronary vasculature with the EDRF-inhibiting compounds (final concentrations) oxyhemoglobin (6 PM), fl-nitro+arginine (L-NNA, 30 PM) or the cyclooxygenase inhibitor indomethacin (10 PM). These substances were infused through a separate infusion line into the aortic cannula at 1% of the coronary flow rate. Continuous perfusion of the inhibitor was started 30 min before repetition of the vasodilator infusion and maintained throughout till the end of the experiment. In some experiments, instead of the inhibitor, only the solute (NaCl) was infused. To determine the release of PGIg, the coronary effluent was collected for 20 s immediately before in .fusion of the vasodilators as well as after a steady-state reduction of coronary perfusion pressure had been reached. The samples were frozen and kept at -7O’C until determination of 6-keto-PGFla, the stable metabolite of prostacyclin, by means of a specific radioimmunoassay (4). Degradation of nucleotides. In some experiments, during steady-state dilation induced by the various vasodilators, the coronary effluent was collec ted for 20 s to determine the degradation of adenine nucleotides during passage through the coronary vascular bed. To this end, the nucleotide-containing Krebs-Henseleit solutions were analyzed after having passed through rabbit hearts and were compared with the stock solution infused. Analysis of nucleotide catabolism performance liquid chromatography
by means of high(HPLC). The re-
verse-phase technique was applied as described with the modification that one column was used of two; this technique yielded shorter elution times with sti 11 sufficient resolution of the potential degradation products. Krebs-Henseleit solution containing the nucleotides (50-200 ~1) was directly loaded onto the column and eluted with a continuous gradient of methanol (O6% during 20 min) in 0.1 M KHZPOd, at pH 5.1. Bioluminometry. Analysis of nucleotide concentrations (ATP, APaA, AP*A) with bioluminescence techniques has been performed as described in detail previously (20,
AND
AUTACOID
RELEASE
H1693
21). Briefly, luciferin-luciferase produces light with ATP that is generated from APdA after hydrolysis with snake venom phosphodiesterase. The assay for APsA additionally contained phosphoenolpyruvate (1.9 mM) and pyruvate kinase (0.1 mg/ml). The assay contained, in a final volume of 0.3 ml, 30 ~1 of luciferin-luciferase (LKB, Pharmacia, Freiburg, FRG), 3.8 mM MgCIZ, 25 mM N2-hydroxyethylpiperazine-W-2-ethanesulfonic acid (pH 7.75), and l-10 ~1 of perfusate. The reaction was started at room temperature by the addition of snake venom phosphodiesterase (0.1 pg/rnl final concentration). The increase of luminescence was monitored with a photomultiplier (Lumacounter model 2080; Lumac Syitems, Basel, Switzerland). The light readings were monitored every 30 s (integration mode) until reaching the maximum value after 3-5 min. Drugs. ACh hydrochloride, adenosine (Ado), AMP, ADP, ATP, APaA, AP*A, bovine hemoglobin, and indomethacin were obtained from Sigma (Deisenhofen, FRG). Unless otherwise stated all reagents were dissolved in Tyrode solution. Hemoglobin and indomethatin were prepared as described earlier (6, 25). L-NNA was obtained from Serva (Heidelberg, FRG). It was dissolved in Tyrode solution under vigorous stirring at a temperature of 60°C. SNP (Sigma) was also dissolved in sodium acetate, diluted with Tyrode solution, and protected from light throughout the experiments. Statistics. All values are presented as means & SE. Differences in the vasodilator effects were tested by means of t tests for paired and unpaired data. Vasodilation (i.e., decrease in coron .ary perfision pressure) was expressed in percent of the actual contractile tone [i.e., actual coronary perfusion pressure - minimal coronary perfusion pressure obtained by supramaximal (XOW5 M) doses of Ado or SNP] for dose-effect curves or as decrease of the actual coronary perfusion pressure (comparison of standard doses of vasodilators). P c 0.05 was considered significant. RESULTS
The mean coronary flow rate as measured in a total of 42 experiments amounted to 36 & 3 ml/min, and the resulting coronary perfusion pressure amounted to 62 & 6 mmHg. The heart rate was 138 k 6 s-l, and the developed left ventricular pressure was 110 & 11 mmHg. Figure 1 depicts the dose-effect curves of the various vasodilators studied. There were no significant differences between the dose-effect curves of Ado, ADP, ATP, APsA, and APdA. The corresponding mean effective concentration (EC&) values amounted to 6.39 k 0.18 (Ado), 6.15 & 0.17 (ADP), 6.21 & 0.11 (ATP), 6.20 & 0.08 (APsA), and 6.36 & 0.21 (APdA). Only the EC50 for AMP was significantly (5.68 & 0.02; P < 0.05) different from those of the other compounds. Infusion of the EDRF-inhibitor hemoglobin or the stereospecific inhibitor of EDRF formation L-NNA increased the coronary perfusion pressure by 35 & 11 and 42 + 12%. Both compounds inhibited or reversed the dec;ase in coronary perfusion pressure induced by ACh but did not affect the vasodilator response to ATP or APsA. The response to ADO and ADP was also not
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 16, 2019.
Hl694
AP3A AND AP4A CAUSE VASODILATION AND AUTACOID RELEASE
m ADO v AMP A ADP 0 ATP
07 A -v j-7
0 control m L-NNA
AP4A
ACh
AP3A
SNP
30N
ACPP VI0 0
n control m Hb
.
APjA
AP4A
ATP
ACh
01 .
1 agonist [~I41
10
FIG. 1. Dilator effects of various adenine compounds on coronary resistance vessels. Dose-effect curves were obtained in a noncumulative manner (steady-state infusions of single doses). “Dilation” refers to decrease in coronary perfusion pressure. Maximal dilation (minimal coronary perfusion pressure) was obtained by a supramaximal dose of adenosine (Ado) or papaverine; n = 4-8 vessels. For sake of clarity, SD are not indicated. APaA, diadenosine 5’,5”‘-P1,P3-triphosphate; AP4A, diadenosine 5’,5”’ -P1,P4-tetraphosphate.
affected by either inhibitor. In contrast, the decrease in coronary perfusion pressure induced by APdA was reduced by 44 & 10 and 42 & 12% (Fig. 2). Infusion of the solute alone did not alter the AP*A-induced decrease in coronary perfusion pressure (not shown). Unlike the EDRF inhibitors, indomethacin did not affect selectively the response of coronary perfusion pressure to any vasodilator (Table 1). Release of PG&. In 23 experiments, the effluent from the heart was collected immediately before and during the steady-state dilation induced by infusion of standard doses (1 PM) of ACh, SNP, Ado, or adenine nucleotides (at least 4 different compounds in each single experiment) for determination of the concentration of 6-ketoPGFl&, the stable hydrolysis product of PGIz. The mean release of PGIZ from these hearts under unstimulated conditions amounted to 15.3 & 2.2 ng/min. There was a considerable variance between the individual hearts with values ranging from 2.7 ng/min to 34.9 ng/min. Figure 3 depicts the percent increase of the release of PGIz during infusion of the vasodilators. In contrast to the widely uniform vasodilator effects, the stimulator effects on the PGIZ release showed remarkable differences. The most potent agents were ACh and APSA. Pretreatment of the hearts with indomethacin significantly reduced the basal release of PGIz (0.5 * 0.001 ng/min, n = 3) and completely inhibited the increased release in response to these agents (not shown). For the most potent stimulating nucleotide, APsA, it was tested whether the same stimulator effect could be
FIG. 2. Effects of 1 PM concentrations of acetylcholine (ACh), AP4A, APaA, ATP, and sodium nitroprusside (SNP) on coronary perfusion pressure (CPP) before and after treatment with endotheliumderived relaxing factor (EDRF)-inhibitors p-nitro+arginine (LNNA; n = 6) or oxyhemoglobin (Hb; n = 6). ACPP indicates a decrease in perfusion pressure. Only dilator responses to ACh and AP4A are affected. **p < 0.01; NS, difference not significant.
TABLE 1. Effect of indomethacin on nucleotide-induced decrease in coronary perfusion pressure Dilator Responses (% of response to SNP, ADP
Control Indomethacin 00
ATP
100.7&12.0 127.Ok31.1
95.Wl6.0 159.8k55.5
1 PM)
AP3A
90.5212.1 155.lk53.6
APdA
93.8kl2.3 170.9k68.4
PM)
Values are means * SE; n = 4 hearts. There are no significant differences between controls and values after indomethacin treatment. SNP, sodium nitroprusside; APaA, diadenosine 5’,5’ ’ ‘-P1,P3-triphosphate; AP4A, diadenosine 5’,5”‘-P1,P4-tetraphosphate.
A PG&L/l 0
0
200-
O-
ACh AP3A AP4A ATP ADO
FIG. 3. Stimulator effects of ACh (n = 18), APSA (n = 23), AP4A (n = 19), ATP (n = Id), and adenosine (n = 13) on release of prostacyclin (PGIg) from coronary vascular bed. APaA was a significantly more potent stimulator than AP4A. APGIZ,increase in percent of control sample, obtained immediately before infusion of compound. *P < 0.05; *** p < 0.001 compared with APZA.
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 16, 2019.
APsA AND APdA CAUSE
I-
-
*1
VASODILATION
AND AUTACOID
A
T
H1695
RELEASE
B
ns,
1
A
40 0ml 30 A CPP 0 VI0
1
0I
l-
n.s.,lr
TI
ns.r-l
. T:
ADP
T I
AMP + ADP
AP3A
FIG. 4. Comparison of PG12 stimulator and vasodilator effects of APaA (1 PM) with an equimolar combination of its hydrolysis products AMP + ADP or ADP alone. Although all 3 interventions induced similar decreases in perfusion pressure, amount of PG12 released in response to APaA was significantly higher. (*P c 0.01; n = 5)
AP3A
A[A
ATP
FIG. 5. Survival of nucleotides during perfusion of Langendorff hearts (n = 5). Hearts have been perfused with Krebs-Henseleit buffer containing nucleotide to be investigated (APsA, APbA, ATP) at an initial concentration of 1 PM. Concentration of nucleotides in perfusates have been determined using bioluminometry methods. Data are means & SE.
obtained after simultaneous centrations of its immediate and ADP. Neither substance of both stimulated PG12 in (Fig. 4).
ofATP,
infusion of equimolar condegradation products AMP alone nor the combination a similar manner to APSA
To assess the degradation of the nucleotides during passage of the coronary vascular bed, the concentrations of ATP, AP3A, Biological stability
AP&,
40 min 40 min 20 FIG. 6. High-performance liquid chromatography tracings demonstrating hydrolysis of nucleotides in perfusate after passage through rabbit heart. ATP, APdA, APsA, and adenosine were dissolved in KrebsHenseleit solution at a final concentration of 1 PM. A-D: top tracings, chromatograms of solutions before perfusion; bottom trucings, corresponding solutions after passage through heart (at 4 min after starting perfusion). Analysis of adenosine (A ), ATP (B), APdA (C), and APA (D); the shorter elution times in D were obtained after exchanging column. Identification of nucleoside peaks (A, adenosine; I, inosine) as well as of nucleotides was performed by cochromatography with standard compounds.
20
and AP&.
and AP4A were determined in the effluent of the heart. Samples were obtained during steady-state perfusion (50 s after start) with Krebs-Henseleit buffer at initial 1 PM concentrations of the respective compounds. Applying the highly sensitive bioluminescence assay, it was possible to determine a residual ATP concentration of ~5% of the initial value in the coronary effluent during infusion of this compound (Fig. 5). In contrast, both APaA and APdA revealed an -lo-fold higher biological stability as reflected by effluent concentrations >30% of the initial values (Fig. 5). HPLC analysis of the degradation products in the effluent confirmed the higher stability of the dinucleotides (Fig. 6, C and D) compared with ATP (Fig. 6B). The pattern of degradation products was very similar for all nucleotides, showing AMP, Ado, and inosine as the only detectable compounds. Figure 6 depicts HPLC tracings of one perfusion experiment. Virtually identical tracings were obtained in four further experiments. DISCUSSION
Both dinucleotides, APsA and AP*A, acted as potent vasodilators of resistance vessels in the intact heart.
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 16, 2019.
H1696
APZA AND APdA CAUSE VASODILATION
Maximal dilation was obtained at even lower concentrations than previously observed in isolated mesenteric arteries (6). Distinct dilations were obtained already at concentrations of 0.1 PM, which is even below the concentration range that is likely reached during local activation of platelets. Based on the intraplatelet nucleotide content (12), it can be calculated that the complete release of APSA and AP*A during platelet activation would lead to a plasma concentration in the range of 0.5-l PM. In the surroundings of aggregating platelets, especially the platelet-endothelial cell-interface, even much higher concentrations must be assumed. Both dinucleotides apparently activate different receptor-dependent transduction cascades in endothelial and vascular smooth muscle cells. The partial inhibitory action of the EDRF-inhibitors L-NNA and hemoglobin on the APdA-induced dilation indicates that this compound induces EDRF release in coronary resistance vessels. In contrast to our previous findings in isolated mesenteric arteries (6), a significant part of the dilator effect of AP*A does not seem to be mediated by EDRF. There was also no indication of EDRF release in response to ATP, which in other vascular beds of the same species in vivo (26) and in vitro (6, 9) stimulates EDRF release by acting through endothelial PzY receptors (3). As in mesenteric arteries, EDRF does not mediate the dilation induced by APSA. Although APaA was, in contrast to the other nucleotides, a potent stimulator of PGIZ release, it is unlikely that PGIZ was a significant mediator of AP3Ainduced dilation, because treatment with indomethacin did not selectively inhibit the dilator effect of APSA compared with all other nucleotides. However, it is tempting to speculate that the high release of the antiaggregating PGIZ in response to APaA has functional significance as a simultaneous antagonistic principle to the proaggregatory action of APaA on platelets (14). This fits into the general principle of simultaneous activation of dual control mechanisms in hemostatic processes (19). The analysis of the degradation of the dinucleotides in the intact coronary vascular bed revealed that a considerable amount of APSA and APAA remained uncleaved. Therefore, in contrast to ATP, APSA and APdA can be considered as long-lived substances in the coronary vasculature. The rate-limiting step of degradation presumably is the asymmetric cleavage of the dinucleotides by pyrophosphohydrolases (15, 16), which are found to be located as ectoenzymes on endothelial (23) as well as on vascular smooth muscle cells (22). The relative rates of degradation to AMP and ADP/ATP during heart passage were in a similar range as found on freshly isolated or cultured endothelial and smooth muscle cells (22, 23). Further degradation of these cleavage products was obviously controlled by the cascade of ectonucleotidases on both types of vascular cells (24). The activity of 5’nucleotidase appears to be the second rate-limiting step. With respect to ATP, our findings are in good agreement with recent investigations in perfused rat hearts (7). In view of the considerable vascular activity of the whole spectrum of degradation products (lo), the question arises whether APsA and APdA exerted their dilator and PGIZ-stimulating activity per se or only after hydrolytic cleavage to mononucleotides or nucleosides. There
AND AUTACOID
RELEASE
are, however, several findings that suggest that APsA and APdA have stimulating effects on endothelial and vascular smooth muscle cells as uncleaved parent compounds. First, there was no difference in the dose-effect curves obtained for the dilations induced by the dinucleotides compared with ATP and Ado. As a matter of fact, the dose-effect curve for AMP was even shifted to the right, which is not consistent with the assumption that the main degradation products, AMP and Ado, were the main mediators of vasodilation. Second, the production of PGIZ induced by APSA could not be mimicked by either cleavage product alone or by the equimolar combination of the immediate hydrolysis products ADP and AMP. Likewise, AP*A induced a distinct EDRF-mediated dilation that was not observed in response to any of the potential degradation products. Further support for a direct action of both dinucleotides on the vascular wall is derived from our previous findings in isolated mesenteric segments. In these segments, exhibiting only a small endothelial surface in relation to the perfused volume over time, no degradation of APsA and APdA could be detected (6). Nevertheless, both compounds induced significant vasodilation. The unique effect of APsA on PGIZ release and that of APdA on EDRF release, which were not shared by any other nucleotide or nucleoside tested, suggest that both compounds may act on specific subclasses of PZ receptors. However, their characterization and localization on endothelial and/or smooth muscle cells remain to be established. In conclusion, our studies demonstrate that the platelet-derived dinucleotides APaA and APdA are long-lived and potent vasodilators of resistance vessels of the rabbit heart. They may represent a novel class of physiological regulators of vascular tone and platelet activity at sites distal to thrombus formation, which may be subject to an enhanced risk of thrombus propagation and embolization. This study was supported by the Deutsche Forschungsgemeinschaft Grant PO 307/l-Z, Og 2/3-3. D. Lamontagne is a research fellow from the Heart and Stroke Foundation of Canada. Address for reprint requests: U. Pohl, Institut fur Physiologic, Medizinische Universitat zu Lubeck, Ratzeburger Allee 160, D-2400 Lubeck 1, FRG. Received 10 September 1990; accepted in final form 11 January 1991. REFERENCES 1. AZUMA, H., M. ISHIKAWA, AND S. SEKIZAKI. Endothelium-dependent inhibition of platelet aggregation. Br. J. Phurmacol. 88: 4ll415,1986. 2. BASSENGE, E., AND R. BUSSE. Endothelial modulation of coronary tone. Prog. Cardiovasc. Dis. 30: 349-380, 1988. G., AND C. KENNEDY. A dual function for adenosine 3. BURNSTOCK, 5’-triphosphate in the regulation of vascular tone. Excitatory cotransmitter with noradrenaline from perivascular nerves and locally released inhibitory intravascular agent. Circ. Res. 58: 319330,1985. 4. BUSSE, R., U. F~RSTERMANN, H. MATSUDA, AND U. POHL. The role of prostaglandins in the endothelium-mediated vasodilatory response to hypoxia. PfZuegers Arch. 401: 77-83, 1984. AND E. BASSENGE. Endothelium-derived 5. BUSSE, R., A. LUCKHOFF, relaxant factor inhibits platelet activation. Naunyz-Schmiedeberg’s Arch. Pharmacol. 336: 566-571, 1987. AND U. POHL. Vasomotor activity of 6. BUSSE, R., A. OGILVIE, diadenosine triphosphate and diadenosine tetraphosphate in iso-
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 16, 2019.
APaA AND APdA CAUSE VASODILATION lated arteries. Am. J. Physiol.
254 (Heart
Circ.
Physiol.
23):
H828-
H832,1988. 7 FLEETWOOD,
G., S. B. COADE, J. L. GORDON, AND J. D. PEARSON. Kinetics of adenine nucleotide catabolism in coronary circulation of rats. Am. J. Physiol. 256 (Heart Circ. Physiol. 25): Hl565Hl572,1989. 8. FLODGAARD, H., AND H. KLENOW. Abundant amounts of diadenosine 5’,5’ ‘I- P1,P4-tetraphosphate are present and releasable, but metabolically inactive, in human platelets. Biochem. J. 208: 737742,1982. 9. FURCHGOW, 10. 11.
12.
13.
R. F. Role of endothelium in responses of vascular smooth muscle. Circ. Res. 53: 557-573, 1983. GERLACH, E., AND B. F. BECKER. Topics and Perspectiues in Adenosine Research. Berlin: Springer-Verlag, 1987. LUTHJE, J., J. BARINGER, AND A. OGILVIE. Effects of adenosine triphosphate (APZA) and adenosine tetraphosphate (AP4A) on platelet aggregation in unfractionated human blood. BZut 51: 405413,1985. LUTHJE, J., D. MILLER, AND A. OGILVIE. Unproportionally high concentrations of diadenosine triphosphate (APaA) and diadenosine tetraphosphate (AP4A) in heavy platelets. Consequences for in vitro studies with human platelets. BZut 54: 193-200, 1987. LUTHJE, J., AND A. OGILVIE. The presence of diadenosine 5’,5”‘Biophys. P1,P3-triphosphate (APaA) in human platelets. Biochem. Res. Commun.
115: 253-260,1983.
14. LUTHJE, J., AND A. OGILVIE. Diadenosine triphosphate (AP3A) mediates human platelet aggregation by liberation of ADP. Biochem.
Biophys.
Res. Commun.
118: 704-709,1984.
15 LUTHJE, J., AND A. OGILVIE. Catabolism of AP4A and AP3A in human plasma: purification and characterization of a glycoprotein complex with 5’-nucleotide phosphodiesterase activity. Eur. J. 16.
Biochem. LUTHJE,
149: US-127,1985. J., AND A. OGILVIE.
human serum: identification acterization. Eur. J. Biochem.
Catabolism of AP4A and AP3A in of isoenzymes and their partial char169: 385-388.
1987.
AND AUTACOID
RELEASE
H1697
17. LUTHJE, J., AND A. OGILVIE. Catabolism of APhA and APZA in whole blood the dinucleotides are long-lived signal molecules in the blood ending up as intracellular ATP in the erythrocytes. &r. J. Biochem. 18. MONCADA,
173: 241-245,1988.
S., E. A. HIGGS, AND J. R. VANE. Human arterial and venous tissues generate prostacyclin (prostaglandin X), a potent inhibitor of platelet aggregation. Lancet 1: 18-21, 1977. 19. NAWROTH, P. P., AND D. M. STERN. Endothelial cell procoagulant properties and the host response. Semin. Thromb. Hemostasis 13:
391-397,1987. 20. OGILVIE, A.
Determination of diadenosine tetraphosphate (AP4A) levels in subpicomole quantities by a phosphodiesterase luciferinluciferase coupled assay: application as a specific assay for diadenosine tetraphosphatase. Anal. B&hem. 115: 302-307,198l. A., AND P. JAKOB. Diadenosine 5’,5”‘-P1,P3-triphos21. OGILVIE, phate in eukaryotic cells: identification and quantitation. Anal
Biochem. 22. OGILVIE,
134: 382-392,1983.
A., AND J. LUTHJE. Diadenosine tetraphosphate (AP4A) and diadenosine triphosphate (AP3A) are degraded by ectoenzymes of arterial smooth muscle cells. Biol. Chem. Hoppe-Seyler 369: 887-
888,1988. 23. OGILVIE,
A., J. LUTHJE, U. POHL, AND R. BUSSE. Identification and characterization of an AP4A hydrolase on intact bovine aortic endothelial cells. Biochem. J. 259: 97-103, 1989. 24. PEARSON, J. D., J. S. CARLETON, AND J. L. GORDON. Metabolism of adenine nucleotides by ectoenzymes of vascular endothelial and smooth muscle cells in culture. Biochem. J. 190: 421-429, 1980. 25. POHL, U., AND R. BUSSE. EDRF increases cyclic GMP in platelets during passage through the coronary vascular bed. Circ. Res. 65:
1798-1803,198s. 26. POHL, U., L. DIIZSI,
B. SIMON, AND R. BUSSE. Selective inhibition of endothelium-dependent dilation in resistance-sized vessels in 253 (Heart Circ. Physiol. 22): H234-H239, vivo. Am. J. Physiol. 1987.
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 16, 2019.
