Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

AP5A (P¹,P⁵-diadenosine pentaphosphate) belongs to the group of diadenosine polyphosphates (APnA) that act as humoral signal transducers and neurotransmitters and are coreleased with ATP and catecholamines from the adrenal medulla (Castillo et al., 1992; Sillero et al., 1994). They also are stored in and released from human platelets (Schlüter et al., 1994) and degradation half-time in the blood is considerably longer than that for ATP, the prototype purinergic agonist (Lüthje and Ogilvie, 1988). We evaluated the effects of AP5A in resistance arteries of the rat that possess a key position in the regulation of precapillary resistance (Mulvany and Halpern, 1977; Mulvany and Aalkjaer, 1990).

In a preceding study, we showed that in the superior epigastric resistance artery of the rat (SEA), APnA (n = 3-6) act as potent vasoconstrictors independent of the vessel tone, whereas in the mesenteric resistance artery (MrA) at basal vessel tone an initial rapid vasoconstriction is followed by a complete relaxation and at raised vessel tone with phenylephrine a marked vasodilation can be observed (Fig. 1). These reported effects were independent of the endothelium and from the different perivascular innervation of the two vessels (Stassen et al., 1997). Contractile and relaxing effects by APnA in the MrA had 1) comparable agonist rank order potencies; 2) could be reproduced by ,-methylene ATP (,-meATP); and 3) were blunted by pretreatment of the arteries with pyridoxal-phosphate-6-azophenyl-2',4'-disulfonic acid (PPADS) and ,-meATP, leading to the conclusion that APnA directly constricted rat resistance arterial smooth muscle through P2X-purinoceptors and subsequently triggered relaxation in some vascular beds through similar receptors or as a direct consequence of the initial contractile mechanism (Steinmetz et al., 2000a). Now we report further investigations performed to explain the observed regional heterogeneity of ApnA-induced vasoactivity and the dynamic dual effects in the mesenteric resistance artery. Specifically, the postulated role of K⁺ channels for APnA-induced vasorelaxation that could to be exerted via secondary activation of calcium-dependent K⁺ channels (Sumiyoshi et al., 1997) should be considered and the availability of the P2Y₁ purinoceptor antagonist ADP3'5' (Boyer et al., 1996) should be used. Thus, the yet incomplete elaboration of the involvement of purinoceptor subtypes in the diverse vascular effects of APnA should be addressed. Therefore, we tried to further identify the receptors responsible for the vascular effects of AP5A with various purinoceptor antagonists. Experiments were performed with the P2Y receptor agonist ADPS (Harden et al., 1998) and with ,-meATP (Delbro et al., 1985), the prototype P2X receptor agonist, to compare their vasoactivity and inhibitory profile with that of AP5A. AP5A was chosen as exemplary APnA because this agent displayed the highest potency of the APnA (n = 3-6) both in inducing vasorelaxation in the precontracted MrA and vasoconstriction in SEA and MrA (Steinmetz et al., 2000a). Reported putative targets of the purine nucleotide AP5A are purinoceptors (Harden et al., 1998; Humphrey et al., 1998). In the rat mesenteric vascular bed both P2X and P2Y receptors were reported operational (Ralevic and Burnstock, 1991). Numerous antagonists for the different purinoceptors have been described. Suramin was found to antagonize P2Y and P2X purinoceptors (Abbracchio and Burnstock, 1994) and PPADS was reported to act as selective P2X receptor antagonist (Windscheiff et al., 1994), although there are studies suggesting that it binds to P2Y₁ receptors as well (Communi et al., 1996). This assumption is remarkable because in the preceding study, PPADS blunted both contractile and dilatory effects in MrA. ADP3'5' was described as P2Y₁ purinoceptor antagonist (Boyer et al., 1996). Apart from these P2 receptor antagonists the P1 receptor antagonists 3,7-dimethyl-1-propargylxanthine (DMPX), an A₂ receptor antagonist, and 8-cyclopentyl-1,3-dipropylxanthine (DPCPX), an A₁ receptor antagonist (Fredholm et al., 1994), were used. To investigate the role of K⁺ channels for AP5A-induced vasorelaxtion we tried to inhibit the AP5A-induced vasorelaxation by inactivation of K⁺ channels with the Ca²⁺-dependent K⁺ channel blockers clotrimazole (Logsdon et al., 1997) and apamine (Moczydlowski et al., 1988) or the ATP-dependent K⁺ channel blocker glibenclamide (Robertson and Steinberg, 1990).

View larger version (14K): [in this window] [in a new window]   Fig. 1.   Original records of myograph experiments with SEA (top) and MrA (bottom). Phenylephrine (Phe; 10 µM) induces a stable tone for several minutes. At raised tone (10 µM Phe = 100% of force) AP5A (10 µM) causes a brief relaxation followed by a rapid vasoconstriction that maintained stable during several minutes in SEA but was followed by an almost complete relaxation in MrA.

Finally, the vascular effects of the pyrimidine UTP were evaluated. UTP is known to activate P2Y₂ and P2Y₆ receptors (Harden et al., 1998). Interactions of the desensitizing effects of AP5A and ,-meATP with UTP were investigated.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Vessel Isolation. Male Wistar rats (200-300 g) obtained from Charles River, Sulzfeld, Germany, were used. Rats had free access to water and standard rat diet. Rats were sacrificed by cervical dislocation and exsanguination. The abdominal skin and muscles as well as the mesentery were removed by blunt dissection. The abdominal wall was placed inside up in a Petri dish coated with sylgard (Dow Corning, Seneffe, Belgium) and filled with Krebs-Ringer-bicarbonate solution (KRB). Skeletal muscle overlaying the left SEA was carefully removed under a dissecting microscope and 2-mm-long segments of this vessel were isolated just below the diaphragm. From the mesentery third order side branches of the superior mesenteric artery were isolated. The luminal diameters of SEA and MrA were of the same order of magnitude (diameter 200-300 µm; Table 1). Both vessels are arterial anastomoses. The SEA interconnects the internal mammary artery to the inferior epigastric artery (side branches of the subclavian and common iliac artery, respectively) and gives rise to side branches perfusing the abdominal muscles. The MrA interconnects second order mesenteric artery side branches of the superior mesenteric artery and gives rise to side branches that penetrate into the ileum.

Tension Measurements. Arteries were mounted on two stainless steel wires (diameter 40 µm) as ring segments in an isometric myograph (model 410A; J.P. Trading, Aarhus, Denmark) between a force transducer (Kistler Morse DSC6, Seattle, WA) and a displacement device for recording of isometric force development (Mulvany and Aalkjaer, 1990). Arteries were stretched to their optimal luminal diameter with an active length tension protocol with 125 mM K⁺ as activating stimulus (De Mey and Brutsaert, 1984). During experimentation the vessels were kept in KRB that was maintained at 37°C and aerated with 95% O₂ and 5% CO₂.

Experimental Protocols. Previous studies showed that AP5A and ,-meATP effects could be reproduced only after 40 and 20 min, respectively. In view of this apparent desensitization and the transient nature of the contractile effects of the substances in MrA, a "single dose" concentration-response approach was used to determine agonist potencies. Vasorelaxing agonist effects in MrA were evaluated during contraction induced by 10 µM phenylephrine. Inhibitory effects of the purinoceptor antagonists were evaluated after the vessels had been incubated with the respective antagonist for 10 min. Effects of AP5A were antagonized with at least two different antagonist concentrations. The P2X agonist ,-meATP and the P2Y agonist ADPS usually were inhibited with only one antagonist concentration to be able to compare principally their qualitative inhibitory profile with that of AP5A. Inhibitory effects of K⁺ channel blockers were evaluated after a 5-min preincubation period. Adequate concentrations of K⁺ channel blockers were deduced from pilot experiments and experiments from our department when inhibiting K⁺ channel activity in other preparations such as kidney tubules or in patch-clamp studies with various cell types.

Compounds and Solutions. The composition of KRB was as follows: 118.5 mM NaCl, 4.7 mM KCl, 1.2 mM MgSO₄, 1.2 mM KH₂PO₄, 25.0 mM NaHCO₃, 2.5 mM CaCl₂, and 5.5 mM glucose. In high K⁺ solution (125 mM K⁺) all NaCl was replaced by an equimolar concentration of KCl. All agonists and pharmacological tools were obtained from Sigma Aldrich Fine Chemicals (Deisenhofen, Germany). PPADS, DMPX, and suramin were obtained from Biotrend Chemikalien (Cologne, Germany). Stock solutions were prepared on the day of use in double distilled water.

Data Analysis. Contractile reactivity was measured as active wall tension (active force divided by twice the length of the vessel segment) and expressed as a percentage of the tissue's contractile response to high K⁺ solution. From concentration-response curves potency (EC₅₀) was determined by least-squares sigmoidal curve fitting of individual curves (GraphPad Prism 1.00; GraphPad, San Diego, CA). Competitive antagonism was quantified by the pA2 value. The pD2' value expresses the affinity of noncompetitive antagonists. Differences were evaluated by the nonparametric Mann Whitney test, P < .05 denoting statistical significance. Data are shown as mean ± S.E. (n = 5 experiments if not stated otherwise).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Inhibition of Vasoconstriction in SEA and MrA. The arterial ring segments of SEA and MrA of the rats were of similar size and comparable in inducing contractile force (Table 1). Suramin did not inhibit the AP5A-induced vasoconstriction in SEA (log EC₅₀ = 5.65) at concentrations between 1 and 300 µM (Fig. 2, left). Likewise, there was no inhibition of ,-meATP-induced vasoconstriction (log EC₅₀ = 6.2) by suramin (10 and 100 µM) in SEA (data not shown). Different from SEA, in MrA there was a concentration-dependent significant shift to the right of the concentration-response curve of AP5A-induced vasoconstriction at 300 µM suramin, indicating a competitive inhibition (pA2 = 4.21; Fig. 2, right). Equally, suramin (10 µM) caused a significant shift to the right of the concentration-response curve of the ,-meATP-induced vasoconstriction (log EC₅₀ = 6.6), indicating a competitive inhibition by suramin (pA2 = 5.97; data not shown).

View larger version (21K): [in this window] [in a new window]   Fig. 2.   Reduction in AP5A-induced vasoconstriction in SEA (left) and MrA (right) by suramin. Wall tension is given in percentage of maximum tension induced by high K+ solution. Each point is the mean of five determinations ± S.E. Organ preparations of five animals.

View larger version (22K): [in this window] [in a new window]   Fig. 3.   Reduction in AP5A-induced vasoconstriction in SEA (left) and MrA (right) by PPADS. Wall tension is given in percentage of maximum tension induced by high K+ solution. Each point is the mean of five determinations ± S.E. Organ preparations of five animals.

View larger version (18K): [in this window] [in a new window]   Fig. 4.   Reduction ,-meATP-induced vasoconstriction in SEA (left) and MrA (right) by PPADS. Wall tension is given in percentage of maximum tension induced by high K+ solution. Each point is the mean of five determinations ± S.E. Organ preparations of five animals.

View larger version (21K): [in this window] [in a new window]   Fig. 5.   Effects of ADP3'5' (100 µM) on ,-meATP (3 µM)-induced vasoconstriction in MrA (left columns) and SEA (right columns) at basal vessel tone.

Inhibition of Vasorelaxation in MrA. Suramin concentration dependently caused a right shift of the concentration-response curve of AP5A-induced vasorelaxation (log EC₅₀ = 5.44), indicating a competitive inhibition. The pA2 value was 4.14 (Fig. 6 right). PPADS inhibited vasodilation by AP5A in a noncompetitive way (pD2' = 5.16; Fig. 6, left). ADP3'5' 100 µM caused a right shift of the concentration-response curve of AP5A-induced vasodilation, indicating a competitive inhibition. The pA2 value was 4.21 (Fig. 7, left). ADP3'5' (100 µM) also significantly inhibited the ADPS (prototype P2Y agonist)-induced vasodilation of MrA (log EC₅₀ = 4.67; data not shown).

View larger version (24K): [in this window] [in a new window]   Fig. 6.   Reduction in AP5A-induced vasorelaxation in MrA by PPADS (left) and suramin (right). Wall tension is given in percentage of maximum tension induced by 10 µM phenylephrine during precontraction. Each point is the mean of five determinations ± S.E. Organ preparations of five animals.

View larger version (23K): [in this window] [in a new window]   Fig. 7.   Reduction in AP5A-induced vasorelaxation in MrA by ADP3'5' (left) and DMPX (right). Wall tension is given in percentage of maximum tension induced by 105 M phenylephrine during precontraction. Each point is the mean of five determinations ± S.E. Organ preparations of five animals.

Inhibition of K⁺ Channels in MrA. Clotrimazole (1 µM) caused a transient constriction of MrA that lasted no longer than 10 s. It did not significantly influence vasorelaxation of precontracted arteries brought about by AP5A (data not shown). Apamine (0.1 and 1 µM) did not influence MrA at basal tone or the precontraction of MrA by phenylephrine (10 µM) nor did it inhibit the vasorelaxation by AP5A. Glibenclamide (100 µM) induced a vasorelaxation of the phenylephrine-precontracted MrA by 70 ± 9%. At 10 µM there was neither an effect on the precontracted arteries nor on the vasorelaxation induced by AP5A (data not shown).

Vascular Effects of UTP. UTP (1, 10, and 100 µM) induced a concentration-dependent stable vasoconstriction not only in SEA but also in MrA. This contraction was not influenced by prior desensitization of the resistance arteries with ,-meATP or AP5A nor did a prior UTP contraction blunt the vascular responses of AP5A or ,-meATP.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

In Steinmetz et al. (2000a) heterogenous vasoactivity of APnA in rat mesenteric (MrA) and epigastric (SEA) resistance arteries was described. Whereas in SEA APnA induced a stable vasoconstriction, in MrA only a transient vasoconstriction was observed. When the vessels were precontracted by phenylephrine, APnA induced a marked vasorelaxation in MrA but not in SEA. Collectively, the results of Steinmetz et al. (2000a) suggested that all of these vascular effects were mediated via activation of similar P2X purinoceptors or that APnA-induced vasorelaxation was a direct consequence of the preceding vasoconstriction. To pursue the hypothesis that this observed variety of vascular responses is due to a heterogenous distribution of purinoreceptors and to functionally identify the subtypes of receptors, the vascular effects of the most potent APnA, AP5A, were now examined in the presence of several purinoceptor antagonists. In several studies it had been found that P2X receptors are responsible for vasoconstriction induced by purinergic agonists (Burnstock and Kennedy, 1985; Ralevic and Burnstock, 1991; Bültmann and Starke, 1993; van der Giet et al., 1998). This assumption also seems to be correct for vasoconstriction induced by AP5A because the nonselective P2 receptor antagonist suramin and more specifically the P2X-selective antagonist PPADS (Windscheiff et al., 1994) inhibited AP5A-induced constriction of MrA and with lower potency in SEA. The absence of an effect of suramin in SEA indicates that probably different P2X receptors are expressed in these two resistance arteries. The antagonism of AP5A vasoactivity by PPADS was, however, noncompetitive. The fact that the antagonism of the prototype P2X receptor agonist ,-meATP with these two antagonists was similar to that of AP5A further supports the assumption that P2X receptors, possibly different subtypes, are responsible for the AP5A-induced vasoconstriction in SEA and MrA. Obviously, there are no P1 receptors contributing to the vasoconstrictor potency of AP5A because both DPCPX and DMPX failed to influence the contractile effects of AP5A.

Surprisingly, the putative selective P2X receptor antagonist PPADS also decreased the vasodilation induced by AP5A in precontracted vessels. This finding indicates that the selectivity of this antagonist might not be restricted to P2X receptors only. Similar conclusions had been drawn earlier suggesting a P2Y₁ receptor antagonism of PPADS as well (Vigne et al., 1998). In addition, in our own studies with anesthetized rats we could show that continuous i.v. infusion of AP5A significantly reduced blood pressure; this effect was dose dependently inhibited by PPADS. Continuous infusion of ,-meATP, however, induced a rise of blood pressure that also could be inhibited by PPADS, albeit in higher doses than necessary for inhibition of the hypotensive effect of AP5A (Steinmetz et al., 2000b).

Vasodilation induced by P2Y receptor activation has been described for the rat mesenteric arterial bed (Ralevic and Burnstock, 1991; Ralevic et al., 1995). We therefore examined whether the vasodilation seen with AP5A in precontracted MrA was due to activation of G-protein-coupled P2Y receptors. This hypothesis could be confirmed by the competitive antagonism of vasorelaxation with the nonselective P2 purinoreceptor antagonist suramin and even more by ADP3'5', which was described as selective P2Y₁ purinoceptor antagonist (Boyer et al., 1996). Additionally, the vasorelaxation induced by the selective P2Y₁ agonist ADPS (Harden et al., 1998), which mimicked the response of AP5A, could be inhibited with ADP3'5' as well.

The selectivity of the herein described purinoceptor antagonists appears limited because the putative P2X selective antagonist PPADS also inhibited AP5A-induced vasorelaxation and similarly the putative selective P2Y₁ receptor agonist ADP3'5' slightly reduced ,-meATP-induced vasoconstriction in SEA (indicating again that there are other receptors in SEA responsible for contraction than in MrA).

In a recent study with isolated perfused rat kidneys it could be demonstrated that vasodilation by AP3A was due to A₂ adenosine receptor activation (van der Giet et al., 1997). Adenosine, however, in concentrations up to 100 µM did not induce any significant vasorelaxation in our preparation, also excluding the assumption that adenosine as degradation product of AP5A is responsible for the vasorelaxation. To show that the absence of an effect of adenosine was not due to rapid degradation of this substance the more degradation resistant and specific A_2A receptor agonist CGS 21680 (Fredholm et al., 1998) was used; however, it did not exert any effect on MrA either. Surprisingly, the A_2A receptor antagonist DMPX (Fredholm et al., 1994) inhibited the vasorelaxation by AP5A in MrA. Because this inhibition occurred only at high concentrations of DMPX and in a noncompetitive way, this might reflect a nonspecific effect of this A₂ antagonist.

Activation of P2X purinoceptors might lead to an increase in the activity of Ca²⁺-dependent K⁺ channels because it will lead to an increase in intracellular Ca²⁺ (Pacaud et al., 1996; Schachter et al., 1996, 1997; Humphrey et al., 1998). Such an involvement of K⁺ channels in the vasodilation of vessels has been reported (Sumiyoshi et al., 1997). Thus, a K⁺ channel activation as well could be the reason for the AP5A-induced relaxation of MrA and therefore offers another explanation for the differences in the responses of MrA and SEA to AP5A. The inhibitors of such Ca²⁺-dependent K⁺ channels, clotrimazole or apamin, however, did not reduce AP5A-induced vasodilation. Similar inhibition of ATP-dependent K⁺ channels by glibenclamide was without effect. Consequently, an activation of K⁺ channels by AP5A that would lead to hyperpolarization of cell membrane potential can be excluded as mechanism responsible for AP5A-induced vasorelaxation.

Interestingly, another P receptor agonist, UTP, induced a stable vasoconstriction both in SEA and MrA, and these vascular responses remained completely unchanged after prior desensitization by ,-meATP or AP5A. This indicates that a further P receptor type is present in the rat vasculature.

In summary, the presented antagonistic profile of the heterogenous and dynamic vascular effects of AP5A indicates a regionally different distribution of different P2X and of P2Y₁ purinoceptors in resistance arteries from different vascular beds. A similar regionally different receptor expression is known, for example, for vascular - and -adrenergic receptors (Bylund et al., 1994).