INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

BOTH CYCLOOXYGENASE (COX)- and lipoxygenase-mediated products of arachidonic acid (AA) metabolism have been established as important participants in the regulation of pulmonary vascular tone (13, 29). In contrast, the role of products of the third pathway of AA metabolism, i.e., the cytochrome P-450 monooxygenase pathway, are less well characterized (12). Pulmonary AA metabolism via cytochrome P-450 monooxygenase activity to the four regioisomeric (5,6-; 8,9-; 11,12-; and 14,15-) cis-epoxyeicosatrienoic acids (EETs) has been observed in dogs, rabbits, guinea pigs, rats, and humans (3, 14, 15, 26, 32-35). We previously reported (25, 26) that in isolated canine pulmonary vascular rings contracted with either PGF_2a or U-46619, a thromboxane (TX)/PGH₂ (TP) receptor agonist, administration of 5,6-EET decreased active tension in a concentration-dependent manner. Similarly, in isolated, perfused canine lungs in which the pulmonary vascular resistance (PVR) was increased with serotonin or U-46619, 5,6-EET decreased PVR. Collectively, these results suggested a vasodilator action for 5,6-EET in the pulmonary circulation of the dog (25).

Of the EETs synthesized in the rabbit lung, the most abundant regioisomer formed was reported to be 5,6-EET (34). Similar to our observations in isolated pulmonary vascular rings from the dog (25, 26), Schwartzman et al. (22) reported that 5,6-EET elicited a concentration-dependent decrease in active tension in rabbit pulmonary artery (PA) rings. In contrast to reports that suggested that 5,6-EET was a pulmonary vasodilator, Zhu et al. (35) found that 5,6-EET, as well as the other EET regioisomers, constricted isolated rabbit PA. A possible explanation of these contradictory results might reside in the fact that Schwartzman et al. (22) studied effects of 5,6-EET on active tension in rings of large conduit pulmonary vessels, whereas Zhu et al. (35) described 5,6-EET-evoked constriction in smaller cannulated and pressurized PA. Thus the seemingly opposite responses obtained for 5,6-EET in the rabbit pulmonary vasculature in these two reports may be a consequence of differences in the size of the pulmonary vessels studied or the location from which they were obtained; for example, Park et al. (18) observed that large extralobar rabbit PA contracted to epinephrine whereas intralobar PA of the same diameter did not. However, both large extralobar and intralobar PA of the same diameter constricted to serotonin, whereas smaller intrapulmonary arteries did not (18).

In the present study, we propose that the effect of 5,6-EET on large extralobar PA segments in the rabbit is to decrease active tension via COX-dependent synthesis of a pulmonary vasodilator such as PGI₂, in a manner similar to that which we observed in the dog (25). Moreover, we propose that in smaller intralobar PA segments and resistance vessels, 5,6-EET increases active tension through activation of the TP receptor, resulting from increased synthesis of TX or a related COX-dependent endoperoxide. To test these hypotheses, we compared effects of 5,6-EET on active tension in extralobar, large (>2 mm)-outside diameter (OD) isolated rabbit PA rings with those observed in smaller (1- to 2-mm OD) extralobar and intralobar rings. To investigate the overall effect of 5,6-EET on the pulmonary resistance vessels, we measured 5,6-EET-mediated changes in PVR in isolated, perfused rabbit lungs (1, 5, 7). We examined whether the effects of 5,6-EET on vascular reactivity were dependent on COX activity, TX synthesis, or TP receptor activation.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animal preparation. Adult New Zealand White rabbits (2.4-3.0 kg) were anesthetized with pentobarbital sodium (15 mg/kg iv) 10 min after intramuscular administration of ketamine (8 mg/kg) and xylazine (2 mg/kg). A tracheostomy was performed for insertion of a tracheal cannula. The animals were ventilated via a fixed volume ventilator (Harvard) with room air (8-10 ml/kg tidal volume, 15 cycles/min). A catheter was inserted into a carotid artery for administration of heparin (1,000 units iv) 10 min before exsanguination of the animal. The protocol for animal use was approved by the Saint Louis University Institutional Animal Care and Use Committee.

Isolated vessel protocols. After exsanguination of the rabbit, extralobar and intralobar PA were dissected free of extravascular tissue and stored in cold (4°C) physiological salt solution (PSS) containing (in mM) 118.3 NaCl, 4.7 KCl, 2.5 CaCl₂, 1.2 MgSO₄, 1.2 KH₂PO₄, 25.0 NaHCO₃, 0.026 Na-EDTA, and 11.1 glucose (pH 7.4) saturated with 95% O₂-5% CO₂ as previously described for isolated canine vessels (25). Immediately before use, the pulmonary vessels were cut into rings 3-4 mm in length and suspended in water-jacketed tissue chambers containing 10 ml of PSS gassed with 95% O₂-5% CO₂ at 37°C. Each ring was mounted between two stainless steel support wires (smaller-gauge wires were used for the smallest vessels). Ring tension was measured from one of the support wires attached to an isometric force transducer (FT03, Grass) and was recorded continuously on a polygraph (model 7, Grass). To maximize length-tension relationships in the vascular ring preparations, each ring was placed under a basal tension determined to result in a maximal contractile response to KCl (60 mM). After an initial depolarizing contraction with KCl (60 mM), the rings were washed with PSS and allowed to return to basal tension over a period of 30 min. At basal tension, concentrations of 5,6-EET (1 × 10⁸-1 × 10⁵ M, each in 1 µl of absolute ethanol) were added individually or cumulatively to the rings. In different but identically prepared ring preparations, 5,6-EET was administered to PGF₂-contracted rings after incubation for 30 min with vehicle or indomethacin, a COX inhibitor (30 µM), a concentration shown previously to inhibit EET-stimulated eicosanoid synthesis (25). At the concentrations used, the ethanol vehicle did not alter the basal tension or active tension of these PA rings. PGF₂ (1-5 × 10⁶ M) was added to achieve a contraction that was 50-80% of that produced with KCl (60 mM). After administration of 5,6-EET, vessels were washed with PSS and recontracted with PGF₂ and acetylcholine (1 × 10⁶ M) was administered. Vessels in which acetylcholine did not decrease active tension >60% were considered to lack functional endothelium and were not used in the data analysis. To confirm that the effects of indomethacin on 5,6-EET-induced changes in active tension were due to COX inhibition, in a separate group of rings, 5,6-EET was added to PGF₂-contracted rings in which COX activity was inhibited for 30 min with meclofenamate (100 µM).

Enzyme immunoassay. Enzyme immunoassay (EIA) was performed for quantitative identification of 6-keto-PGF₁ (the stable degradation product of PGI₂) and TXB₂ (the stable degradation product of TXA₂) as described previously (20, 25). Briefly, enzymatic tracers consisted of 6-keto-PGF₁ or TXB₂ covalently linked to purified acetylcholinesterase as described previously (20). The sample (50 µl) was combined with 50 µl of enzymatic tracer in a well of a 96-well microtiter plate (Nunc) that had been precoated with 2 µg/well goat anti-rabbit IgG antibody (Calbiochem). The antiserum for 6-keto-PGF₁ or TXB₂ (Cayman) was then added (50 µl). The plates were incubated for 18-20 h at room temperature and washed three times with 500 µg of potassium phosphate buffer (1 × 10² M, pH 7.4; containing 0.05% Tween 20). After washing, 200 µl of Ellman's reagent was added to each well. Ellman's reagent consisted of 2 µg/ml acetylthiocholine iodide and 2.15 µg/ml 5,5'-dithiobis(2-nitrobenzoic acid) (DTNB) in 1 × 10² M potassium phosphate buffer. The reaction product (reduced DTNB) was monitored at 405 nm in a BIO-TEK (model EL-309) EIA plate reader. All samples and standards were run in duplicate. Sample unknowns were determined by comparison to standards with log-logit data transformation.

Isolated lungs. Rabbit lungs were isolated as described previously (23). Briefly, a midsternal thoracotomy was performed, and the heart and lungs were removed en bloc. Fluid-filled catheters were placed into the PA and the left atrium for lung perfusion and pressure measurements. The isolated lungs were ventilated at 10 ml/kg with 26-30 breaths/min of 15% O₂-6% CO₂-79% N₂ to achieve a perfusate pH of 7.33 ± 0.01, PCO₂ of 38.1 ± 2.8 mmHg, and PO₂ of 107.3 ± 5.9 mmHg. The lungs were perfused in a humidified chamber (34-37°C) in a recirculating manner with 150 ml of PSS containing (in mM) 118.3 NaCl, 4.7 KCl, 2.5 CaCl₂, 1.2 MgSO₄, 1.2 KH₂PO₄, 25.0 NaHCO₃, 0.026 Na-EDTA, and 11.1 glucose (pH 7.4) to which 5% dextran (mol wt 70,000) was added for maintenance of oncotic pressure. Pulmonary arterial (P_pa, inflow), pulmonary venous (P_la, outflow), and airway pressures were recorded continuously. Microvascular pressure (P_mv) was measured by the double occlusion method (6, 31). To prevent atelectasis, positive end-expiratory airway pressure (1-1.5 mmHg) was maintained throughout each experiment. Lungs were perfused under zone III conditions (P_pa > P_la > airway pressure) with a roller pump (Masterflex, Cole Parmer Instrument) at 100 ml/min. Outflow pressure was adjusted to 2-3 mmHg via a screw clamp on the outflow tubing. In those experiments in which antagonism of the TX receptor was required, ONO-3708, the TX receptor antagonist (ONO Pharmaceutical), was added to the perfusate reservoir 15 min before measurements of vascular pressures were obtained. Total PVR was calculated as (P_pa  P_la)/perfusate flow rate. Arterial PVR was calculated as (P_pa  P_mv)/perfusate flow rate, and venous PVR was calculated as (P_mv  P_la)/perfusate flow rate (25). In a separate group of experiments, lungs were perfused with autologous blood. In each group, 5,6-EET was added to the perfusate (in ethanol). The final ethanol concentration in the perfusate was <0.01%, a concentration that did not change PVR.

1

1

4

6

5

4

1

1

Statistical methods. All values are expressed as means ± SE. Differences between experimental groups were determined by ANOVA. If the F ratio indicated significant differences, a Tukey's protected t-test was used to establish differences between individual sample means. When appropriate, a Student's t-test for paired data was used. Values of P < 0.05 were considered to be statistically significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Effect of 5,6-EET on active tension in PGF₂-contracted PA rings. The dominant response of 5,6-EET administered to PGF₂-contracted isolated rings of large main (5.6 ± 0.3-mm OD) and extralobar branch (2.1 ± 0.1-mm OD) PA was a sustained concentration-dependent decrease in active tension (maximum relaxation occurred at ~3 min; Fig. 1). Before the relaxation a small contraction of short duration (within the 1st minute after 5,6-EET administration) was observed (Fig. 2). Both the 5,6-EET-induced relaxation and contraction responses were inhibited by indomethacin (Fig. 3). However, only the contraction response to 5,6-EET was inhibited by ONO-3708 (2 × 10⁵ M), a selective TP receptor antagonist, suggesting that this response was dependent on the synthesis of TX or another COX-dependent TP receptor agonist. In these rings, acetylcholine (1 × 10⁶ M) resulted in a decrease in active tension of 78.4 ± 4.4% in vehicle-treated rings and 73.6 ± 2.64% in indomethacin-treated rings (n = 5; not significant). The lack of effect of indomethacin on acetylcholine-induced relaxation indicated that the vessels contained functional endothelium and remained capable of relaxing. Identical vessel studies were performed with a chemically dissimilar inhibitor of COX activity, meclofenamate. Results of these studies were consistent with those obtained with indomethacin. Meclofenamate (100 µM) reduced the 5,6-EET-induced decrease in active tension from 65.2 ± 5.9% to 2.8 ± 1.7% at 10 µM 5,6-EET (n = 5; P < 0.01), confirming that a chemically dissimilar inhibitor of COX activity produced an identical effect. Effects of 5,6-EET on relaxation of PGF₂-contracted extralobar branch PA rings were similar to main PA, with significant but small differences occurring only at the highest concentrations of 5,6-EET administered (Fig. 1). However, compared with either main or extralobar branches, the concentration-response curve of intralobar PA rings (1.3 ± 0.1-mm OD) was shifted to the right. These results demonstrate the greater potency of 5,6-EET for decreasing active tension in rings of main and extralobar PA branches of dissimilar diameter than in intralobar PA with diameters similar to, although slightly smaller than, those of extralobar branches.

View larger version (18K): [in this window] [in a new window]   Fig. 1.   Relaxation responses to 5,6-epoxyeicosatrienoic acid (EET) (1 × 108-1 × 105 M) in isolated rabbit pulmonary arterial (PA) rings from main PA (n = 5), extralobar branch PA (n = 9), and intralobar PA (n = 8). 5,6-EET was administered to vessels contracted with PGF2 (1 × 106-5 × 106 M) to achieve 50-80% of the maximal contractile response to KCl (60 mM). *P < 0.05 compared with main PA; P < 0.05 compared with extralobar branch PA. o.d., Outside diameter.

View larger version (7K): [in this window] [in a new window]   Fig. 2.   Representative trace of the effects of 5,6-EET (3 × 107 M) on active tension in an isolated rabbit extralobar branch PA ring contracted with PGF2 (2 × 106 M). Increases and decreases in active tension were compared with the value recorded during the period of stable active tension immediately preceding the application of 5,6-EET.

View larger version (18K): [in this window] [in a new window]   Fig. 3.   Effects of 5,6-EET (3 × 107 M) on active tension in isolated rabbit main and extralobar branch PA rings (n = 4) contracted with PGF2 (3 × 107-2 × 106 M) alone (vehicle) or coincubated for 30 min with ONO-3708 (2 × 105 M) or indomethacin (Indo; 3 × 106 M). Initial measurements were recorded at the peak of the increase in active tension within 1 min of 5,6-EET administration. *P < 0.05 compared with vehicle for the same time period.

Effect of 5,6-EET on active tension in PA rings at basal tension. At basal tension, 5,6-EET increased active tension in intralobar PA rings in a concentration-dependent manner (Fig. 4). The concentration-response curve for 5,6-EET-induced increases in active tension in extralobar branches was shifted to the right of that for intralobar PA, demonstrating a lower sensitivity of extralobar branches to the contractile effect of 5,6-EET. Rings of main PA did not contract to 5,6-EET administered at basal tension.

View larger version (20K): [in this window] [in a new window]   Fig. 4.   Effect of 5,6-EET (1 × 108-1 × 105 M) in main (n = 5), extralobar branch (n = 9), and intralobar (n = 8) PA administered at basal tension. *P < 0.05 compared with basal tension; P < 0.05 compared with intralobar PA; P < 0.05 compared with extralobar branch PA.

Effect of 5,6-EET on PVR in isolated rabbit lungs. In lungs perfused with PSS, administration of 5,6-EET into the perfusate at concentrations of 1 × 10⁹ to 1 × 10⁵ M resulted in concentration-dependent increases in PVR (Fig. 5). The EC₅₀ for this response was 5.9 ± 1.7 × 10⁷ M. Administration of a single bolus of 5,6-EET (1 × 10⁵ M) was associated with an increase in total PVR resulting predominantly from the increase in PVR within the arterial segment (Fig. 6). To determine whether a dilator component of 5,6-EET would be unmasked in the presence of increased PVR in the intact lung, U-46619 was infused before 5,6-EET administration. When P_pa was raised from 13.4 ± 2.9 to 27.3 ± 6.1 mmHg with U-46619, administration of 5,6-EET resulted in a pressor response similar to that observed in the absence of U-46619. A depressor response following the pressor response, similar to that seen in large isolated PA rings (Fig. 2), was not observed in the intact lung preparation (data not shown). In lungs perfused with autologous blood, 5,6-EET (10 µM) increased PVR 4.6 ± 0.5-fold over PVR measured immediately before administration of 5,6-EET (Fig. 7). These results demonstrate that, when administered in blood, 5,6-EET is not so avidly bound to plasma proteins or taken up by cellular elements in the blood that responses to it are prevented.

View larger version (10K): [in this window] [in a new window]   Fig. 5.   Effect of 5,6-EET (1 × 109-1 × 105 M) administered cumulatively on pulmonary vascular resistance (PVR) in physiological salt solution (PSS)-perfused, isolated rabbit lungs (n = 4). *P < 0.05 compared with PVR in the absence of exogenously administered 5,6-EET.

View larger version (14K): [in this window] [in a new window]   Fig. 6.   Effect of 5,6-EET (1 × 105 M) on PVR in PSS-perfused, isolated rabbit lungs (n = 4). Total PVR represents [pulmonary arterial pressure (Ppa)  pulmonary venous pressure (Pla)]/flow; arterial PVR represents [Ppa  microvascular pressure (Pmv)]/flow; and venous PVR represents (Pmv  Pla)/flow. *P < 0.05 compared with control.

View larger version (10K): [in this window] [in a new window]   Fig. 7.   Effect of 5,6-EET (1 × 105 M) on PVR in autologous blood-perfused, isolated rabbit lungs (n = 5). PVR represents (Ppa  Pla)/flow. *P < 0.05 compared with control (Con).

Effect of 5,6-EET on TXA₂ and PGI₂ synthesis in PSS-perfused rabbit lungs. In PSS-perfused lungs, samples of perfusate collected immediately after the completion of each pressure measurement were analyzed for concentrations of 6-keto-PGF₁ and TXB₂, the stable degradation products of PGI₂ and TXA₂, respectively. In the absence of pharmacological inhibitors, administration of 5,6-EET (1 × 10⁵ M) increased concentrations of TXB₂ and 6-keto-PGF₁ in the lung perfusate (n = 4; Fig. 8). The concentration of TXB₂ increased from 9.9 ± 0.7 to 100.2 ± 13.4 pg/ml perfusate, a 10-fold increase from basal, whereas 6-keto-PGF₁ increased from 100.3 ± 8.3 to 206.4 ± 24.6 pg/ml, a 2-fold increase from basal.

View larger version (20K): [in this window] [in a new window]   Fig. 8.   Values of thromboxane (TX)B2 and 6-keto-PGF1 (6-kPGF1) measured in the perfusate of PSS-perfused, isolated rabbit lungs (n = 4) immediately before (Con) and 5 min after administration of 5,6-EET (1 × 105 M) into the perfusate. *P < 0.05 compared with control.

Effect of ONO-3708, a TP receptor antagonist, on 5,6-EET-induced changes in PVR. In PSS-perfused rabbit lungs, ONO-3708 attenuated the 5,6-EET-induced increases in PVR in a concentration-dependent manner (Fig. 9). At 200 µM, ONO-3708 completely prevented the 5,6-EET-induced increase in PVR. It was also at this concentration of ONO-3708 that the pulmonary vasculature no longer responded to a 10-µg bolus injection of U-46619, the TP receptor agonist (Fig. 9, inset), suggesting that TXA₂, or an endoperoxide-like TP receptor agonist, may mediate the 5,6-EET-induced increase in PVR.

View larger version (23K): [in this window] [in a new window]   Fig. 9.   Inhibition of the 5,6-EET (1 × 105 M)-mediated increase in PVR with ONO-3708 (2-200 µM) in PSS-perfused, isolated rabbit lungs (n = 4). Inset: inhibition of the U-46619 (10 µM)-mediated increase in PVR with ONO-3708 (2-200 µM). *P < 0.05 compared with control.

Effect of OKY-046 and indomethacin on 5,6-EET-induced changes in PVR in lungs perfused with PSS. Administration of OKY-046 (7 × 10⁴ M) into the perfusate 30 min before administration of 5,6-EET (1 × 10⁵ M) reduced the 5,6-EET-induced increase in TXB₂ by more than sixfold (n = 4; Fig. 10) from 113.7 ± 8.1 to 17.7 ± 1.8 pg/ml perfusate, nearly to control levels. However, the OKY-046-mediated decrease in TX did not prevent the 5,6-EET-induced increase in PVR (Fig. 11). OKY-046 only reduced the 5,6-EET-mediated increase in PVR by 12.75 ± 2.03%. In contrast, COX inhibition with indomethacin (100 µM) prevented the 5,6-EET-mediated increase in TX synthesis (Fig. 10) as well as the increase in PVR (Fig. 11).

View larger version (16K): [in this window] [in a new window]   Fig. 10.   Effect of OKY-046 (7 × 104 M; n = 4) or indomethacin (1 × 104 M; n = 4) on the increase in TXB2 measured in the perfusate of PSS-perfused, isolated rabbit lungs mediated by administration of 5,6-EET (1 × 105 M). *P < 0.05 compared with control; P < 0.05 compared with vehicle (Veh).

View larger version (22K): [in this window] [in a new window]   Fig. 11.   Effect of OKY-046 (7 × 104 M; n = 4) or indomethacin (1 × 104 M; n = 4) on the increase in PVR mediated by administration of 5,6-EET (1 × 105 M). *P < 0.05 compared with control; P < 0.05 compared with vehicle.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In the present study, we report differential responses to 5,6-EET administration in main, extralobar branch, and intralobar segments of the rabbit PA. In rings of large (>2 mm) PA isolated from the main, common PA, and extralobar portions of the right and left main branches contracted with PGF₂, administration of 5,6-EET resulted in a concentration-dependent decrease in active tension. Differences in the amount of relaxation exhibited by main PA and extralobar branches were small and limited to the highest concentrations of 5,6-EET administered. However, 5,6-EET was significantly less potent in relaxing intralobar PA than either extralobar segment. In contrast, 5,6-EET was more potent at increasing active tension in intralobar PA rings compared with extralobar PA branches. When administered at basal tension, 5,6-EET did not increase active tension in main PA at any concentration administered. The differential responses to 5,6-EET observed in these rings from main and intralobar rabbit PA mimic the differential response to hypoxia observed in large and small rabbit PA. Hypoxia was reported to decrease contraction in large isolated PA vessel segments (4, 7, 17) and to increase it in small PA (7, 16). It has been suggested that this differential response may provide a compensatory mechanism for modulation of the pulmonary arterial blood pressure, i.e., as small PA segments contract in response to hypoxia, large segments relax (4). A similar vasodilator mechanism to 5,6-EET in the large PA segments of the rabbit may modulate the pressure response to constriction of the small resistance segments.

Attempts to describe the responses of vasoactive compounds in the pulmonary circulation have often been limited to an examination of their vasoactive effects on either large conduit vessels or small resistance vessels. When effects of a variety of vasoactive compounds have been compared in large and small rabbit PA vessels, significant differences have been reported (18). Specifically, epinephrine and phenylephrine potently stimulate contraction of large extralobar and intralobar PA of similar size but produce little response in smaller intralobar PA, whereas histamine contracts large and small intralobar PA with little effect on extralobar PA (18). In the present study, we found 5,6-EET to be a potent constrictor of intralobar rabbit PA. 5,6-EET was significantly less potent in extralobar PA branches than in intralobar PA. Because 5,6-EET did not constrict extralobar main PA when administered at basal tension, the relaxation to 5,6-EET in PGF₂-contracted vessels would be unopposed by a 5,6-EET-mediated constrictor effect. Small extralobar branches with diameters more similar to intralobar than main PA relaxed when 5,6-EET was administered to PGF₂-contracted PA rings but contracted when it was administered to PA rings at basal tension.

Consistent with the results presented in the present study, Zhu et al. (35) described EETs as vasoconstrictors in small pressurized rabbit PA (300- to 500-µm OD). Their results support the dependence of vasoconstriction on either COX activity or activation of the TP receptor that we report in the present study. More recently, Yaghi et al. (32) described a similar vasoconstrictor action for EETs in small (200-400 µm) intralobar arterial rings of rats. The rat and rabbit respond similarly to 5,6-EET, whereas dogs respond to 5,6-EET with decreased active tension in isolated vessels and decreased PVR in the isolated, perfused lung, suggesting only a vasodilator role for 5,6-EET (25). Thus the response to 5,6-EET in the pulmonary circulation appears to differ among species. However, the dependence on COX activity for the vasodilator action of 5,6-EET in large rabbit PA is consistent with results reported previously for dog pulmonary vessels (25) and suggests that in both dog and rabbit the vasodilator effect of 5,6-EET depends on the synthesis of a COX-dependent pulmonary vasodilator. Because 5,6-EET increases intracellular calcium in endothelium and smooth muscle cells, it may activate phospholipases, resulting in the release of AA for synthesis of endogenous pulmonary vasodilators such as PGI₂ (25) or 20-hydroxyeicosatetraenoic acid (20-HETE) (3). In the present study, we report that in the pulmonary circulation, 5,6-EET increases the concentration of 6-keto-PGF₁, the nonenzymatic breakdown product of PGI₂, suggesting that PGI₂ could mediate the dilator response. However, the dilator activity of exogenously added 20-HETE was also reported to require COX activity (3), suggesting that 20-HETE is either metabolized to a pulmonary vasodilator prostanoid or stimulates the synthesis of an endogenous vasodilator prostaglandin such as PGI₂ (3). 5,6-EET itself can be metabolized by COX to vasodilator prostanoids (10). Therefore, the vasodilator activity of 5,6-EET in the rabbit pulmonary vasculature may depend on synthesis of endogenous COX product, such as PGI₂, or other COX-dependent signaling molecules.

In isolated, perfused rabbit lungs in which PVR was increased with U-46619, Tan et al. (28) reported previously that 5,6-EET resulted in dilation of the rabbit pulmonary vasculature. We attempted to reproduce those findings by administering 5,6-EET to rabbit lungs under conditions that appeared to be identical to those described by Tan et al. We did not, however, observe vasodilation with administration of 5,6-EET in the presence of increased PVR with U-46619. In our studies of isolated, perfused rabbit lungs in which PVR was increased with U-46619, administration of 5,6-EET resulted only in increased PVR. The discrepancy between our results and those of Tan et al. suggests that factors other than species, vessel location, or vessel size may determine whether the net effect of 5,6-EET on the pulmonary circulation is to increase or decrease PVR. These might include factors such as the level of vascular COX-2 expression (8) or vascular TX receptor density (19).

The finding that 5,6-EET decreased active tension in large (>2 mm)-conduit PA rings, precontracted with PGF₂, is supported by the findings of Schwartzman et al. (22) that EETs relaxed rabbit PA rings precontracted with phenylephrine. In the present study, the 5,6-EET-mediated decrease in active tension in these large PA rings was often accompanied by a small, transient increase in active tension that preceded the larger, sustained relaxation (Fig. 2). Although inhibition of COX activity with indomethacin attenuated the 5,6-EET-mediated decrease in active tension and the transient increase in active tension, ONO-3708 inhibited only the transient increase in tension. Together, these results suggest that in the rabbit the vasoconstrictor response to 5,6-EET predominates in the smaller PA and resistance vessels, whereas in the larger-conduit pulmonary vessels the 5,6-EET-mediated vasoconstriction is replaced with a dilator response mediated by a COX-dependent product such as PGI₂. The net effect of 5,6-EET on PVR when infused into the intact pulmonary circulation was an increase in PVR.

The activity of EETs in most vascular beds has been reported to result in vasodilation (10, 21). However, even in some systemic vascular preparations studied, EETs produce vasoconstriction; for example 5,6-EET was reported to constrict the vasculature of the rat kidney (11, 27). In the rat kidney, the vasoconstrictor effect of 5,6-EET was also decreased with an inhibitor of COX activity (11, 27) or antagonism of the TP receptor (11). In the present study, 5,6-EET was observed to preferentially stimulate the synthesis of TX (10-fold) compared with PGI₂ (2-fold) in the isolated rabbit lung, whereas 5,6-EET preferentially stimulated the synthesis of PGI₂ in the perfusate of the dog lung (25). This difference in the profile of COX products synthesized in the rabbit and dog pulmonary circulation in response to 5,6-EET stimulation led us to examine whether constriction resulting from administration of 5,6-EET in the rabbit may be mediated by TX. Although the 5,6-EET-stimulated increase in TX in lung perfusate was prevented with either indomethacin or OKY-046, inhibition of TX synthesis had only a small effect on the 5,6-EET-induced increase in PVR whereas COX inhibition prevented it. Therefore, it appears that not TX, but an endoperoxide metabolite of COX activity, is the primary mediator of the 5,6-EET-mediated vasoconstrictor effect in the rabbit pulmonary circulation. Alternately, as platelets have been reported to metabolize 5,6-EET via COX activity (2, 9) to an endoperoxide metabolite of 5,6-EET, a 5,6-EET-derived endoperoxide, rather than the endoperoxide PGH₂, might mediate the vasoconstrictor response of 5,6-EET in the rabbit pulmonary circulation.

The pulmonary circulation, like the systemic circulation, is not longitudinally homogeneous. In its progression from the right heart to the pulmonary capillaries, the pulmonary vasculature narrows and divides, its composition changing from large elastic to small muscular resistance arteries (30). Longitudinal variation in responses to hypoxia and other vasoactive agents is widely recognized, although the mechanistic basis for the differences often remains poorly defined. In the present study, we report that within the pulmonary vasculature of the rabbit, 5,6-EET can evoke either a vasoconstrictor or a vasodilator response, depending on the size of the pulmonary vessel or the location from which it was obtained. We have presented evidence supporting a TX- or COX-dependent endoperoxide mechanism for the 5,6-EET-induced constrictor response in the small arteries and resistance vessels and COX-dependent synthesis of a vasodilator prostaglandin, such as PGI₂, in the larger extralobar vessels. These mechanisms may be responsible for the difference in response to 5,6-EET between large and small rabbit pulmonary vessels.