Characterization of thromboxane and prostacyclin effects on pulmonary vascular resistance.

Characterization of thromboxane and prostacyclin effects on pulmonary vascular resistance JOSEPH Department

W. BARNARD,

ROBERT

A. WARD,

BARNARD,JOSEPHW., ROBERTA. WARD, W. KEITH ADKINS, AND AUBREYE. TAYLOR. Characterization of thromboxane and prostacyclin effects on pulmonary vascular resistance. J. Appl. Physiol. 72(5): 18451853, 1992.-Although thromboxane and prostacyclin (PGI,) have long been described as major controllers of pulmonary vascular resistance, little has been reported on the characteristics of the interactions between the two arachidonic acid products. The current study uses segmental vascular resistance and compliance measurements to evaluate the actions of thromboxane and PGI, in isolated blood-perfused rat lung. The thromboxane analogue U-46619 increases pulmonary vascular resistance by increasing only small artery resistance and decreases pulmonary vascular compliance in the middle compartment. Among the vascular effects of U-46619 are a maximum increase in resistance (Rmaxu-4s619) of 60.3 t 15.6 cmH,O 1-l min. 100 g-’ and a concentration required for 50% of maximum increase (K,, u-46619) of 1.60 t 0.85 nM for small artery resistance, a minimum vascular compliance (CminU-46619 ) of -0.93 t 0.58 cmH,O, and a K,, u-46619of 1.10 t 1.60 nM for middle compartment compliance: ‘Similar results were obtained for total resistance and total compliance. The effects of PGI, on thromboxane-induced resistance and compliance changes were evaluated using K 0.5,PGIg 7 RmaxpG12, and at each dose of thromboxane. PGI, was more effeccmaxPGI, tive in reversing the thromboxane constriction at higher concentrations of thromboxane. These data show that the absolute concentration of PGI, and thromboxane and not a simple ratio of thromboxane to PGI, determines vascular tone. l

isolated

W. KEITH

ADKINS,

AND AUBREY

E. TAYLOR

of Physiology, Collegeof Medicine, University of South Alabama, Mobile, Alabama 36688

l

lung

THE PULMONARY VASCULAR RESISTANCE isthoughtto

be controlled through a dynamic interaction between several vasoconstrictors and vasodilators. A large number of vasoactive substances derived from arachidonic acid have been proposed as possible regulators of pulmonary vascular resistance, especially the maintenance of a low pulmonary vascular resistance. The cyclooxygenase pathway forms several vasoactive substances from arachidonic acid. Of these compounds, thromboxane A, is a potent pulmonary vasoconstrictor, and prostacyclin (PGI,) is a potent vasodilator. Therefore, thromboxane A, and PGI, have been postulated as a yin and a yang of vascular smooth muscle tone, especially in the lung. Thromboxane A, causes a pulmonary hypertension with endotoxin shock (33) I) after burn wounds (12), 2) with ischemia-reperfusion (4), 3) with airway smoke exposure (32), and 4) after lung challenge with oleic acid (35), paraquat (30), endothelin-1 (7, l3), neutrophil and phorbol

myristate acetate (11, 25), or hydrogen peroxide (9, 10). 0161-7567/92

$2.00

In these studies, PGI, always increased when thromboxane was released, presumably acting as a protective mechanism to buffer the pulmonary vasoconstriction. In some studies, such as administration of acetylcholine or endothelin-1 in dog lungs, or xanthine/xanthine oxidase in rat lungs, a greater vasoconstriction occurred when the cyclooxygenase pathway was blocked, suggesting that PGI, was opposing some of the vasoconstrictive actions of these compounds (5, 7, 8). These findings certainly implicate thromboxane and PGI, as important controllers of pulmonary vascular tone. The ratio of the concentrations of thromboxane to PGI, is believed by many to determine the level of vascular resistance (1517,22,36-39), not only in the lung but in vascular beds of other tissues. This assumption has several ramifications. First, if the ratio of thromboxane to PGI, does control vascular tone, high concentrations of PGI, would be required to reverse the vasoconstriction induced by high concentrations of thromboxane. Second, the two would necessarily be competitive either for the same receptor or for a population of antagonist receptors that exhibited similar properties, i.e., they would produce the same amount of resistance change at the same concentration. Third, they must act on the same portion of the vasculature in the lung, because constriction could occur in one segment of the vascular bed whereas dilation might affect another segment of the circulation. Therefore any study with total vascular resistance (RT) changes will not totally explain the mechanisms responsible for the total resistance changes. Despite these questions concerning the nature of the interaction between thromboxane A, and PGI, in many forms of lung injury, the specifics of their interactions in the pulmonary circulation have only been partially evaluated. The effects and interactions of these cyclooxygenase products can be evaluated with the use of several indicators of pre- and postcapillary resistance changes. Because an important factor associated with vasoactive substances is fluid balance, a venoconstrictor will increase capillary pressure (Ppc), which promotes edema formation. Although thromboxane is a pulmonary venoconstrictor in canine, ovine, and feline lungs, the site of action of PGI, in the pulmonary circulation has not been determined in any of these species. We have recently shown that endothelin-1 causes thromboxane A, release in the isolated rat lung, which constricts the small artery segment (7). In this study endothelin-1 also caused thromboxane A, release in the isolated dog lung, but the

Copyright 0 1992 the American Physiological Society

1845

Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 19, 2019.

1846

INTERACTIONS

OF

THROMBOXANE

vasoconstriction was confined to the small and large vein segments of the pulmonary circulation. Preliminary work from this laboratory shows that thromboxane administration to African Green Monkey lungs also constricts the small artery vascular segments (1). The current study evaluates the effects of thromboxane and PGI, on the segmental vascular resistances and compliances in isolated perfused rat lungs. These effects were evaluated by measuring vascular resistances and compliances at different concentrations of thromboxane and PGI,. The kinetic characteristics were determined for both thromboxane and PGI, by measuring classical dose-response curves of the change in segmental vascular resistance and compliance in response to changes in the substances’ concentration. Double reciprocal plots of resistance increases or compliance decreases as a function of thromboxane concentration were used to determine 1) the maximum resistance or compliance change (Rmaxu-,,,,, and Cminu-&; 2) the concentration required for 50% of the maximum resistance increase or compliance decrease (K, *5,u-46619), a measure of receptor activity; 3) PGI,‘s antagonism of thromboxane’s resistance increase or compliance decrease (Rmax,,,, , the permeability cmaxPGI, 7 and & 5 PG12 ). In addition, effects of PGI, were evaluated by measuring the capillary filtration coefficient (K,,), in all studies. METHODS

Lung Isolation Male Sprague-Dawley rats (250-350 g; n = 20) were anesthetized with pentobarbital sodium (50 mg/kg ip), and the lungs isolated following these procedures (7): The trachea was cannulated, and a thoracotomy was performed. This was followed immediately by ventilation on a rodent respirator (Harvard) with 95% O,-5% CO,. Three hundred units of heparin were infused directly into the left ventricle of the heart. Five minutes later, the pulmonary artery was cannulated and perfusion was initiated, immediately followed by cannulation of the left atrium through the left ventricle. Lungs were perfused with blood obtained from donor rats by carotid bleeding to provide a circulating volume of ~20 ml. The blood was pumped by a peristaltic pump (Gilson Minipuls 2) from a venous reservoir through a water-jacketed heating coil (37°C) and through a bubble trap into the pulmonary artery. Venous outflow pressure (Ppv) could be set to any desired level by raising or lowering the reservoir and was initially set at 4-5 cmH,O in all preparations. Pulmonary arterial (Ppa), and Ppv were monitored with Statham pressure transducers (P23 De) attached by side arm to the appropriate cannula. To measure the vascular occlusion pressures, solenoid valves (General Valve) were placed in a position such that the Ppa and Ppv side arm cannulas were between the solenoid valves and the lungs. This allowed the continued recording of vascular pressure when either or both valves were closed. The heart and lungs were covered with a thin plastic sheet to prevent evaporation and suspended by the trachea from a force transducer (model FT03C, Grass) to monitor lung weight. A Grass model 7D polygraph recorded the pressures and lung weights throughout the experiment.

WITH

PROSTACYCLIN

Lungs were ventilated at a tidal volume of 10 ml/kg and a rate of 30 breaths/min with the use of 95% O,-5% CO, to maintain normal pH. Positive end-expiratory pressure was 2 cmH,O. The lungs were perfused under zone 3 constant-flow conditions (Ppa > pulmonary venous pressure > airway pressure), and flow was initially adjusted to attain an isogravimetric state to measure baseline values. Flow was maintained constant throughout the remainder of the experimental procedure and averaged 0.442 t 0.019 (SE) 1 min-’ 100 g lung wt? l

l

Measurements Segmental vascular resistance. Vascular occlusion methods were used to estimate segmental resistances within the pulmonary circulation (23,24,34). These techniques have been developed for rat, rabbit, and dog lungs (2,5,6,26,28) and will be described here briefly. Prevailing Ppc is easily measured by simultaneously occluding both the inflow and outflow perfusing catheters, because the vascular pressures rapidly equilibrate to Ppc in zone 3 lungs. RT was calculated using the arterial-venous pressure difference divided by flow (Q) and was normalized to 100 g lung wet wt. On rapid arterial occlusion, an initial sudden drop in arterial pressure occurs that is followed by a secondary slower decrease in arterial pressure. The pressure at the end of the rapid pressure drop is defined as arterial occlusion pressure (Ppa,o). When Ppa,o is used with Ppa and Ppc, the total arterial (precapillary) resistance can be divided into large artery (RLA = Ppa - Ppa,o/&) and small artery (RSA = Ppa,o - Ppc@) resistances. On sudden occlusion of the venous outflow cannula, venous pressure rises rapidly, followed by a more gradual increase in venous pressure. The venous pressure at the end of the rapid pressure rise is defined as the venous occlusion pressure (Ppv,o). When Ppv,o is used with Ppc and Ppv, the total venous (postcapillary) resistance can be divided into small vein (RSV = Ppc Ppv,o/Q) and large vein (RLV = Ppv,o - Ppv/Q) resistances. Segmental vascular compliances. Total pulmonary vascular compliance (CT) was calculated by measuring the slope of the slow component of the venous pressure rise (APlnt) obtained after venous occlusion and by measuring the prevailing blood flow (Q), with the use of the following equation (2 1) CT

= Q/(APlAt)

(1)

The following equation derived by Linehan et al. (24) was used to calculate the middle compartment compliance (CMC) = 4 Ppa,i - Ppv CT Ppa - Ppv it

-CMC

1 1 (2) _ o 75 ‘I2 ’

Ppa,i is the arterial pressure intercept obtained by extrapolating the arterial pressure rise occurring after venous occlusion to the time that occlusion occurred. The sum of the large vessel compliance (CLV) is then calculated as


INTERACTIONS CLV

= CT

-

CMC

OF

THROMBOXANE

WITH

0

IT,I

Filtration coefficient. The filtration coefficient (K,,) is an index of vascular endothelial permeability that measures the amount of fluid that crosses the capillary wall for a given imbalance in capillary and tissue forces as described by the following equation J v,c= Kfc[(pPc - PptiJ - OdCTpc - rpti)] (4) , This equation states that net fluid flux (J,,) occurring across the microcirculation is determined by’the K,,, the difference between Ppc and interstitial fluid pressure (Ppti) and the protein osmotic pressure gradient (difference between protein osmotic pressures of capillaries (7r& and tissues (Xpti)* od is the osmotic reflection coefficient and equals one if the capillary wall is impermeable to the proteins and is equal to zero if the proteins are freely permeable. When the other variables remain constant, & can be determined by measuring the increase in JvC produced by a known increase in Ppc. The rate of weight gain was plotted as a logarithmic function of time, and the zero-time rate of weight gain was obtained by extrapolating the slow component of weight gain to zero time. This measurement is done to estimate K,, before changes have occurred in the other forces in Ei. 4. The K,, is calculated by dividing the change in zero-time weight gain rate by the change in Ppc, and then normalized to 100 g wet lung wt. Z& (27,28) is unchanged after 3-4 h of perfusion in normal lungs and has been shown to increase up to IO-fold after the endothelium has been damaged (2, 4, 7, 18, 26-28). Receptor analysis. The Michaelis-Menten equation for enzyme kinetics has been adapted for receptor analysis. This equation describes the interaction between a receptor and its ligand as a hyperbolic relationship that rises to a maximum effect (Emax) and gives a concentration of ligand required to reach one-half that maximum effect (K, & in the following equation Emax X [ligand] (5) [ligand] + K, .5 In the current study, the effect was either an increase in resistance or a decrease in compliance, and the ligand was the thromboxane analogue U-46619. This analysis was used in the current study to explore the interaction between the thromboxane analogue U-46619 and its receptor. Although our experiments did not reach the maximal resistance increase predicted by the MichaelisMenton analysis of our data, the data points do extend into the plateau region of the response, and MichaelisMenton analysis does not require a peak response but rather predicts one on the basis of the data that are present for analysis. Experimental protocol. After a 30-min equilibration period in which the lung weight and hemodynamics became stable, the thromboxane analogue U-46619 was added as a bolus to the perfusing blood in increasing concentrations, with concentration calculated as the final circulating concentration. U-46619 was prepared as serial dilutions in 0.9% saline. After each concentration of U-46619, Ppa rose and stabilized. Segmental vascular resistances and compliances were then obtained bv the oceffect =

1847

PROSTACYCLIN

t t

0

1

+6.67pM Prostacyclin

+ 1.43nM U46619 5

10

25

30

35

Time (minutes) FIG. 1. Typical response to addition of 1.43 nM U-46619 (first arrow). Pulmonary arterial pressure increased from 23 to 32 cmH,O. Typical response to addition of 6.67 pM PGI, (second arrow). Pulmonary arterial pressure rapidly decreased from 32 to 27 cmH,O.

elusion techniques described above. The next concentration was administered, and the occlusion pressures were obtained again, until the maximum concentration of U-46619 had been administered. Then PGI, was administered as a bolus in increasing concentrations, and the same measurements of segmental vascular resistance and compliance were made. K,,c was measured under baseline conditions and after the highest concentration of PGI,. PGI, was prepared as serial dilutions in a glycine buffer with a pH of 10, because it chemically breaks down at normal pH within 3-4 min. No lung received ~200 ~1of the high pH buffer, and it did not change the pH of the perfusing blood. Reagents. U-46619 was provided by Upjohn Corporation (Kalamazoo, MI). PGI, was provided by Burroughs Wellcome (Research Triangle Park, NC). Statistical analysis. Data were analyzed with analysis of variance. If a significant difference was noted using analysis of variance, the Student-Newman-Keuls test was used to determine which groups were different (40). Statistical significance was defined as P < 0.05. RESULTS

Figure 1 shows the effect of infusing the thromboxane analogue U-46619 into an isolated rat lung (final concentration 1.43 nM). The Ppa response, shown at the first arrow, was rapid after U-46619 administration, and although it is not shown, the effect lasted for 90 min. Figure 1 also shows this lung’s Ppa response to PGI, (13.4 pM final concentration) at the second arrow. This concentration of PGI, almost completely reversed the pressure increase caused by a U-46619 dose of 1.43 nM. Thromboxane Effects The effect of U-46619 on the pulmonary circulation was evaluated by measuring segmental vascular resistances and compliance. Figure 2A shows the segmental vascular resistance changes induced by thromboxane at concentrations from 0.143 to 2.853 nM in five steps: I) refers to 0.143 nM (50 rig/ml); 2) refers to 0.357 nM (125 rig/ml); 3) refers to 0.713 nM (250 rig/ml); 4) refers to 1.427 nM (500 rig/ml); and 5) refers to 2.853 nM (1,000 rig/ml). The increased pulmonary vascular resistance occurs largely within the small artery segment (RsA). Accordingly, only the RT and RSA were analyzed for thromboxane responses in the following data. Figure 2B shows the segmental vascular compliance changes induced by thromboxane at concentrations from 0.14 to 2.85 nM. Because the vascular compliance decrease was also con-


1848

INTERACTIONS

1.2.3.4.5.

RT

123.45.

R LA

OF

THROMBOXANE

1.2.3.4.5.

1.2.3.4.5.

1.2.3.4.5.

R SA

R sv

R LV

CT

C MC

C LV

1. 2. 3. 4. 5.

1. 2. 3. 4. 5.

1. 2. 3. 4. 5.

WITH

PROSTACYCLIN

crease in RT is higher than that in RSA. This may be attributed to the significant increase in small vein resistance seen at the highest doses of U-46619. Note also that the decrease in CT can be completely ascribed to the decrease in CMC. Additionally, the U-46619 KoS5is similar for each segment analyzed. This suggests that the decrease in CMC is due to the increased RSA. To verify that the values for Rmax, 466l9, Cminu &6lg, and & 5 u &6lg derived from the curve-fitting program were reasonable, double reciprocal plots of the change in resistance against the concentration of U-46619 were made. The double reciprocal plot is a use of the Lineweaver-Burk transformation of the Michaelis-Menten equation. This plot produces a linear equation from the hyperbolic doseresponse curves. The effect-intercept, where l/concentration is zero equals l/Rmax, 466lg for resistance and for compliance. The substrate concenl/CminU 46619 tration-intercept, where the effect is zero equals -l/ K O.&U-46619 for either resistance or compliance. Figure 3C shows a plot of RT, and Fig. 30 shows a plot of CT. On each panel are the values for Rmaxu-4661gor Cminu-4661g, respectively. Each panel also shows the K. 5 u 466lgderived from the equation for the regression line-for each set of data points. As Fig. 3 shows, both the curve-fitting program from the change in resistance (or compliance) and the linear regression from the double reciprocal plot of resistance (or compliance) give similar values for the maximal effect and K, .5. PGI, Effects

FIG. 2. A: change in pulmonary

vascular resistance as U-46619 concentration increased from 0.14 to 2.85 nM. Successive histograms: 1) 0.14 nM, 2) 0.36 nM, 3) 0.71 nM, 4) 1.43 nM, and 5) 2.85 nM. Resistance, cmH,O 1-l. min. 100 g lung wet wt? RT,total resistance; RLA, large artery resistance; RSA,small artery resistance; RSV,small vein resistance; RLV, large vein resistance. *P < 0.05 different from baseline, tP < 0.05 different from other groups. B: change in pulmonary vascular compliance as U-46619 concentration increased from 0.14 to 2.85 nM. CT, total vascular compliance; CMC,middle compartment compliance; CLV, large vessel compliance. *P < 0.05 different from baseline.

To determine how thromboxane and PGI, interact to determine the overall pulmonary vascular tone, the receptor effects of PGI, were evaluated at each concentration of thromboxane. PGI, did not affect pulmonary vascular resistance at normal flow and vascular tone. If the lung was preconstricted with U-46619, then PGI, decreased tone in a concentration-dependent manner. The effects of PGI, were analyzed similarly to the analysis used for U-46619 but were evaluated relative to the fined to the middle vascular compartment, only the CT change that PGI, produced in RT or CT after the U-46619-induced changes. Figure 4A shows the effect of and CMC were calculated. Thromboxane’s effects were evaluated with the use of PGI, on the resistance change induced by U-46619 when the PGI, concentration was increased from 6.7 pM to receptor kinetics analysis. The effect of either resistance 1.335 nM. Each curve is the hyperbolic curve-fit of the increase or compliance decrease was plotted as a function of thromboxane concentration and a Rmaxu-,,,,, or data points. The figure shows that PGI, causes a larger change in resistance when U-46619 concentration is Cmin 1J-46619 required to produce - ----- 9 and the concentrations higher. Figure 4B shows the effect of PGI, on the compliK 0.5.U-46619 were also calculated. Figure 3A shows the increased RT produced by the thromboxane analogue at ance change induced by U-46619 as the PGI, concentration was increased from 6.7 pM to 1.335 nM. Each curve the five different concentrations. The mean t standard is the hyperbolic curve-fit of the data points. Figure 4 error of the mean change in resistance is plotted against each concentration of U-46619. These points were fit by a shows that PGI, causes a greater increase in compliance when U-46619 concentration is higher and that the effect hyperbolic curve (Eq. 5) by the use of a graphics program Sigmaplot 4.1 curve-fitting function. This procedure fit of PGI, on total compliance was less at 2.85 than at 1.43 nM U-46619. This may be due to the intense vasoconthe data with the use of a nonlinear regression analysis striction caused by U-46619 at high concentrations. (the Marquardt-Levenberg algorithm). The program produced an Rmax, &?,6lgor a Cminu 466lg and & 5u 466lg There was visible blanching of the lung in these cases, based on the best fit of the data to Eq.5. These values are and it is possible that the PGI, did not reach all of the areas where vasoconstriction had occurred, lessening its shown for RT in Fig. 3A and for CT in Fig. 3B. The abilities. mean t standard error data for RT, RSA, CT, and CMC vasodilatory and compliance-increasing are presented in Table 1. Note that the maximum inFrom the hyperbolic curves fit to the data, l


INTERACTIONS

OF

THROMBOXANE

WITH

PROSTACYCLIN

1849

Total Resistance

z-ii MAX u4619 (Total Resistance) 80 -- -isl-~wR . - - --- ------+ * 8 100 F 2

I 0.71

0.00

. = 1.34nM : K I 0.5,U46619 1 I I 2.14 2.85 1.43

0.002 0.004 0.006 0.008

U46619 (nM) B

0.00

0.71

L hllN.

6 0 a

1.43

2.14

l/ U46619 (PM) 2.85

D

-0.002 0.000 0.002 0.004 0.006 0.008

L hlIN.lM6lY = -0.871nl/c111 H,O -

1’46619

=- I . I 1 ml/cnl H,O

Total Compliance 3. A: effect of increasing U-46619 concentration on total vascular resistance. 0, Means + SE. Curve was generated by curve-fitting program, which fit values using least-squares analysis to Michaelis-Menton equation, which produces hyperbolic curve and generates maximum value for change in resistance and concentration required to produce l/2 that maximum value (&). B: effect of increasing U-46619 concentration on total vascular compliance. 0, Means + SE. Curve was generated using curve-fitting program of least-squares analysis to Michaelis-Menton equation, which produces hyperbolic curve and generates maximum value for change in compliance and concentration required to produce Koe5. C: double reciprocal transformation of total resistance against concentration of U-46619. y-Intercept is 1/Rmax,-,,,,, , and x-intercept is - 1 lK0.5,u-46619.0, Means + SE. D: double reciprocal transformation of total compliance against concentration of U-46619. y-Intercept is 1/Cminu-46619, and x-intercept is -1/K0.5,u-46619. 0, Means fi SE. FIG.

The net resis3 and KO 5 PG12 were determined. tance change values for total resistance are plotted on Fig. 5A. The closed circles represent the Rmax induced by PGI, subtracted from the U-46619-induced resistance increase at that concentration. To show the effectiveness of PGI, in reversing U-46619-induced vasoconstriction, the increase in resistance caused by U-46619 at each concentration is also plotted (0). Figure 5A also shows that PGI, can effectively reverse U-46619-induced resistance increases at the lower concentrations of U-46619. PGI, is not as effective for concentrations of U-46619 from 0.713 to 2.853 nM, but even at these very high concentrations of U-46619, PGI, reverses more than 50% of the resistance change induced by the thromboxane analogue, and CmaxPG12

1. Rmax, Cmin, and KOS5 for U-46619-induced changes in vascular resistances and compliances TABLE

Rmax and Cmin K 0.5

RT

RSA

CT

CMC

81.4k15.6 1.35kO.57

60.3k15.6 1.6OkO.85

-1.lkO.6 1.06k1.32

-0.93+0.58 1.10+1.60

DISCUSSION

Rmax for resistances are expressed as change from baseline resistance (cmH,O 1-l min. 100 g-l), and Cmin for compliances are expressed as change from baseline compliance (ml/cmH,O). Koe5 for both resistances and compliances are expressed as nM. l

l

the maximum effect on resistance produced by PGI, is higher. Figure 5B is a similar plot of the Cmax, PGI, at each concentration of U-46619. The net compliance changes are plotted in each case against the decrease in compliance caused by U-46619. Figure 5B shows that PGI, completely reverses the compliance decrease caused by U-46619 at each concentration of U-46619 except the highest (2.85 nM). The values for RmaxpGIz, Cmaxpoiz) and KO 5 PG12 for each vascular segment are given in Table 2. Table 2 shows that the PGI, KOa5 -_-does not change when the concentration of U-46619 changes. This shows that the ratio of thromboxane to PGI, may not determine the level of tone in the pulmonary circulation and suggests that the absolute concentration of PGI, is an important determinant of the tone. The K,,, was unchanged by PGI, in all preparations.

The thromboxane A, precursor U-46619 increases pulmonary vascular resistance and decreases pulmonary vascular compliance in the isolated rat lung. The resistance increase is most pronounced in the small artery segment, which is different from the effect measured in


1850

INTERACTIONS

A

e8n 8

Prostacyclin (PM) 0

668

1335

tn.@ ‘4

0

2

:

OF THROMBOXANE

19

0.14

l r(

$

-10

0.36 0.71

3=” E -20 ,o_

1.43 2.85

L Total Resistance

iance

U46619 (nW * 1.43

2.85 0.71

WITH

PROSTACYCLIN

The effects of thromboxane on vascular resistances have been studied using U-46619, U-44069, or an unnamed endoperoxide analogue (which may be either U-46619 or U-44069) in other experimental preparations, and dose-response relationships have been reported. In Table 3 the maximum effect of the thromboxane analogue and the J&s are presented. These values were calculated in the current study from the data presented in the cited publications. The studies include work with pulmonary vessel rings, lung parenchyma strips, and mesenteric arteries. The KO.~s calculated here are within an order of magnitude of the &,. 466l9 acquired from the isolated lungs reported in this’study. The differences from the current study are most probably due to species differences, the several different preparations used, and some difference between the potencies of the analogues of thromboxane used. It is possible that the effects of thromboxane and PGI, are due to platelet aggregation, because thromboxane causes platelet aggregation (29) and PGI, causes platelets to disaggregate (19). However, two buffer-perfused lungs (Earl’s balanced salt solution, with 5% albumin and sodium bicarbonate) were studied, and only an increase in RSA was measured, which was greater in magA

0.36

0.14

z

75 r Total Resistance AR U46619

Prostacyclin (PM) 4. A: effect of increasing prostacyclin (PGI,) on total resistance (RT) in presence of different concentrations of U-46619. 0, Effect of PGI, in presence of 0.14 nM U-46619. V, Effect of PGI, in presence of 0.36 nM U-46619.0, PGI, in presence of 0.71 nM U-46619. A, Effect of PGI, in presence of 1.43 nM U-46619. 0, Effect of PGI, in presence of 2.85 nM U-46619. All symbols, means k SE. Curves were generated using curve-fitting program and fitted individual values to hyperbolic equation. B: effect of increasing PGI, on total compliance in presence of different concentrations of U-46619. 0, Effect of PGI, in the presence of 0.14 nM U-46619. V, Effect of PGI, in presence of 0.36 nM U-46619. Cl, Effect of PGI, in presence of 0.71 nM U-46619. A, Effect of PGI, in presence of 1.43 nM U-46619. 0, Effect of PGI:, in presence of 2.85 nM U-46619. All symbols, means + SE. Curves were generated using curve-fitting program and fitted individual values to hyperbolic equation. PGI, was less effective in presence of 2.85 nM U-46619 than in presence of 1.43 nM U-46619.

AR MAX,PGI,

FIG.

AR NET

bh a

0.00

0.71

1.43

2.85

U46619 (nM) B

0.00

0.71

1.43

2.14

2.85

00 o^

l

AC NET

z?

dog lungs, where the resistance increase occurred only in the small and large vein segments of the pulmonary resistance (5). In the present study, associated with the increase in Psa was a decreased CMC. Thromboxane’s receptor dose-response relationships are characterized for both vascular resistances and compliances. The affinity is high, as demonstrated by the low K, 5. The maximal effect of U-46619 is also high, as demonstrated by the large Rmax and Cmin. Although thromboxane receptors could be present throughout all segments of the rat’s pulmonary vascular bed, the present study indicates that the functional receptors are present only in the small arteries and the vessels that produce the CMC.

2.14

AC MAX,PGI, AC U466 19

u a-*

15 L Total Compliance

FIG. 5. A: calculated RmaxPGIz at each concentration of U-46619 for RT (@, + SE). Plotted as reference is change in RT induced by U-46619 as its concentration is increased (0, + SE). B: calculated Cmaxpol, at each concentration of U-46619 for total compliance (CT) (0, + SE). Plotted as reference is change in CT induced by U-46619 as its concentration is increased (0, + SE).


INTERACTIONS

OF

THROMBOXANE

WITH

1851

PROSTACYCLIN

2. Calculated Rmax, Cmin, and KOe5 for prostacyclin-induced changes in vascular resistances or compliances at each U-46619 concentration TABLE

U-46619 Rmax

or Cmin

0.1427 nM 50 rig/ml 0.3566 nM 125 rig/ml 0.7133 nM 250 rig/ml 1.4265 nM 500 rig/ml 2.8531 nM 1.0 pg/ml

K 0.5 _.Rmax

or Cmax

K RKax

or Cmax

K Rgax r7

or Cmax

Ko.5

Rmax

RT

or Cmax

K 0.5

Rmax for resistances are expressed as cmH,O compliances are expressed as ml/cmH,O change expressed as pM.

4.2kl.l 127k130 14.4k1.8 59+35 16.4k1.4 22+11 23.9k2.2 44t19 32.123.5 86+44 -1-l min 100 g lung from U-46619-induced l

l

nitude than that seen in blood-perfused lungs. In addition, two blood-perfused lungs were performed with inflow from the venous side; in these lungs the constriction was also confined to the small arteries as seen in lungs perfused with arterial inflow (data not shown). If platelet aggregation were occurring in these vessels, one would expect that the small vein resistance would have increased, rather than the RSA. Furthermore, in bufferperfused lungs, there are essentially no platelets available to increase pulmonary resistance. Consequently, the resistance increase measured in the blood-perfused lungs is not due to platelet aggregation in the small arteries but to a direct effect of thromboxane and PGI, on pulmonary vascular smooth muscle. Some concern also exists that increasing tone by infusion of the thromboxane analogue would cause the release of PGI, and therefore attenuate the vasoconstriction, as well as cause an overestimate of the effectiveness of the exogenous PGI, in reversing the vasoconstriction. To’ evaluate this possibility, we pretreated two bloodperfused rat lungs with indomethacin (100 PM) and then challenged them with the thromboxane analogue (data not shown). These lungs exhibited the same vasoreactivity to the analogue as the lungs that had not been pretreated with indomethacin, which supports the probability that PGI, is not released in the rat lung as a consequence of increased tone by U-46619. In addition, Dr. Philip Kadowitz has reported an increased resistance for in situ cat lungs that has been pretreated with meclofenamate and then challenged by U-46619, prostaglandin Fza, prostaglandin D,, or serotonin, which was similar to

wt-’ change compliance

RSA

CT

2.8k1.7 218~1502 6.3k1.3 50251 10.6kl.2 27+17 14.4k1.7 33+19 16.5k1.5 77k33

0.09+0.02

CMC

0.05~0.03 117+265 0.13t0.02 151+91 0.34kO.09 242+209 0.67kO.10 28+20 0.42kO.07 82k52

1802174 0.15+0.02 66+38 0.38kO.09 228+198 0.78t0.09 30+17 0.48kO.07 101+60

from U-46619-induced resistance increase, and Cmax decrease. Koe5 for both resistances and compliances

for are

lungs that have not been cyclooxygenase inhibited. Meclofenamate did, however, completely inhibit the increase in resistance caused by arachidonic acid infusion (personal communication). PGI, partially or completely reversed the resistance increase caused by thromboxane at every concentration of thromboxane. In addition, PGI, completely reversed the compliance decrease at all but the highest thromboxane concentration, and at that concentration PGI, reversed -50% of the compliance decrease. The amount of PGI, needed to effectively reverse the resistance and compliance effects of thromboxane was the same, regardless of the concentration of thromboxane. This was demonstrated by the K,, values for PGI, in Table 2. The RmaxpoI, and Cmax,,, for PGI, increased as the dose of thromboxane increaied. Figure 4 shows that most of the resistance increase and compliance decrease caused by U-46619 is reversed by PGI,. At the higher concentrations of U-46619 the resistance change was not completely reversible by PGI,. This may be because the intense vasoconstriction caused by U-46619 prevented PGI, from reaching the constricted areas. PGI,‘s effects on thromboxane-induced pulmonary vasoconstriction have several ramifications: First, because similar concentrations of PGI, were effective at reversing thromboxane-induced vasoconstriction regardless of the concentration of thromboxane present, then the commonly made assumption that the ratio of PGI, to thromboxane present determines the RT is questionable. In fact, PGI, becomes a better vasodilator at higher doses of thromboxane! PGI, produces a similar effect on resis-

3. Calculated values for effective concentrations of thromboxane and thromboxane analogues and maximum effects in previous publications TABLE

Ref

Species

Doucet et al. (14) Kadowitz et al. (42, 51) Greenberg Aiken TxA,,

Equine Canine Bovine Canine

(20)

(3)

Canine thromboxane

Treatment U-44069 Endoperoxide Endoperoxide Endoperoxide Endoperoxide U-46619 TxA,

Preparation

analogue analogue analogue analogue

Lung parenchymal strips Intrapulmonary vein Intrapulmonary vein Pulmonary artery Pulmonary vein Mesenteric artery Mesenteric artery

Emax 200 0.3 2.3 8.8 0.06 23 100

182% max 120 g/cm2 83 g/cm2 83% max 88% max 2.3 g 2.0 g

A,.


1852

INTERACTIONS

OF

THROMBOXANE

tance and compliance regardless of the thromboxane level. At higher concentrations of thromboxane, the absolute change in resistance induced by a given amount of PGI, was much greater, even though the percent reduction in resistance induced by PGI, was about the same. If the ratio of thromboxane to PGI, did determine the pulmonary resistance, as previously believed, then high concentrations of PGI, would be required to reverse the resistance increase induced by high concentrations of thromboxane. The present data also suggest that relatively high concentrations of PGI, will not be more effective than relatively low concentrations in treating pulmonary hypertension. Because the K,, p(& values for PGI, are low regardless of the thromboxane concentration, administration of low concentrations of PGI, will be effective in reducing the vasoconstriction and reverse the vascular compliance decrease caused by elevated thromboxane levels. Address for reprint requests: MSB 3024, College of Medicine, AL 36688. Received

20 June 1991; accepted

J. W. Barnard, Dept. of Physiology, University of South Alabama, Mobile, in final

form

4 December

1991.

REFERENCES 1. ADKINS, W. K., J. W. BARNARD, A. G. BRADY, J. L. ARDELL, S. A. BARMAN, G. L. LONGENECKER, AND A. E. TAYLOR. Characterization of segmental vascular resistance in the isolated monkey lung (Abstract). FASEB J. 4: A1244, 1990. 2. ADKINS, W. K., AND A. E. TAYLOR. Role of xanthine oxidase and neutrophils in ischemia-reperfusion injury in rabbit lung. J. Appl. Physiol. 69: 2012-2018, 1990. 3. AIKEN, J. W. Prostaglandins. J. Cardiovasc. Pharmacol. 6: S413S420, 1984. 4. ALLISON, R. C., E. W. HERNANDEZ, V. R. PRASAD, M. B. GRISHAM, AND A. E. TAYLOR. Protective effects of 0, radical scavengers and adenosine in PMA-induced lung injury. J. Appl. Physiol. 64: 21752182, 1988. 5. BARMAN, S. A., E. SENTENO, S. SMITH, AND A. E. TAYLOR. Acetylcholine’s effect on vascular resistance and compliance in the pulmonary circulation. J. Appl. Physiol. 67: 14951503, 1989. 6. BARMAN, S. A., AND A. E. TAYLOR. Histamine’s effect on pulmonary vascular resistance and compliance at elevated tone. Am. J. Physiol. 257 (Heart Circ. Physiol. 26): H618-H625, 1989. 7. BARNARD, J. W., S. A. BARMAN, W. K. ADKINS, G. L. LONGENECKER, AND A. E. TAYLOR. Sustained effects of endothelin-1 on the rabbit, dog, and rat pulmonary circulation. Am. J. Physiol. 261 (Heart Circ. Physiol. 30): H479-H486, 1991. 8. BARNARD, J. W., C. E. PATTERSON, M. T. HULL, W. W. WAGNER, JR., AND R. A. RHOADES. The role of microvascular pressure in reactive oxygen induced lung edema. J. Appl. Physiol. 66: 14861493, 1989. 9. BURGHUBER, O., M. M. MATHIAS, I. F. MCMURTRY, J. T. REEVES, AND N. F. VOELKEL. Lung edema due to hydrogen peroxide is independent of cyclooxygenase products. J. Appl. Physiol. 56: 900-905, 1984. 10. BURGHUBER. O., R. J. STRIFE, J. ZIRROLLI, P. M. HENSON, J. E. HENSON, M. M. MATHIAS, J. T. REEVES, R. C. MURPHY, AND N. F. VOELKEL. Leukotriene inhibitors attenuate rat lung injury induced by hydrogen peroxide. Am. Rev. Respir. Dis. 131: 778-785, 1985. 11. CARPENTER-DEYO, L., AND R. A. ROTH. Cyclooxygenase inhibition in lungs or in neutrophils attenuates neutrophil-dependent edema in rat lungs perfused with phorbol myristate acetate. J. Pharmacol. Exp. Ther. 251: 983-991, 1989. 12. DEMLING, R. H., A. KATZ, C. LALONDE, P. RYAN, AND L. J. JIN. The immediate effect of burn wound excision on pulmonary function in sheep: the role of prostanoids, oxygen radicals, and chemoattractants. Surgery 101: 44-55, 1987. 13. DE NUCCI, G., R. THOMAS, P. D’ORLEANS-JUSTE, E. ANTUNES, C. WALDER, T. D . WARNER, ANI J. R. VANE. Pressor effects of circu-

WITH

PROSTACYCLIN

lating endothelin are limited by its removal in the pulmonary circulation and by the release of PGI, and endothelium-derived relaxing 14 factor. Proc. Natl. Acad. Sci. USA 85: 9797-9800, 1988. DOUCET, M. Y., T. R. JONES, AND A. W. FORD-HUTCHINSON. Re* sponses of equine trachealis and lung parenchyma to methacholine, histamine, serotonin, prostanoids, and leukotrienes in vitro. Can. J. Physiol. Pharmacol. 68: 379-383, 1990. 15. FARRUKH, I. S., J. R. MICHAEL, W. R. SUMMER, N. F. ADKINSON, JR., AND G. H. GURTNER. Thromboxane-induced pulmonary vasoconstriction: involvement of calcium. J. Appl. Physiol. 58: 34-44, 1985. 16. FITZGERALD, D. J., W. ROCKI, R. MURRAY, G. MAYO, AND G. A. FITZGERALD. Thromboxane A, synthesis in pregnancy-induced hypertension. Lancet 335: 751-754, 1990. FLEMING, W. H., L. B. SARAFIAN, LEUSCHEN, M. C. NEW17*LAND, E. M. KENNEDY, J. D. KUGLER,M. P.J. W. CHAPIN, B. J. HURLBERT, D. L. BOLAM, AND R. M. NELSON. Serum concentrations of PGI, and thromboxane in children before, during, and after cardiopulmonary bypass. J. Thorac. Cardiovasc. Surg. 92: 73-78, 1986. 18. GAAR, K. A., A. E. TAYLOR, L. J. OWENS, AND A. C. GUYTON. Pulmonary capillary pressure and filtration coefficient in the isolated perfused lung. Am. J. Physiol. 213: 910-914, 1967. 19. GORMAN, R. R., S. BUNTING, AND 0. V. MILLER. Modulation of human platelet adenylate cyclase by prostacyclin (PGX). Prosta-

glandins 13: 377-388, 1977. 20. GREENBERG, S. Effect of PGI, and 9a,l la-epoxymethanoprostaglandin Hz on calcium and magnesium fluxes and tension development in canine intralobar pulmonary arteries and veins. J. Pharmacol. Exp. Ther. 219: 326-337, 1981. 31 HAKIM, T. S., C. A. DAWSON, AND J. H. LINEHAN. Hemodynamic responses of dog lung lobe to lobar venous occlusion. J. Appl. Physiol. 47: 145-152, 1979. 22. KAUFMAN, R. P. JR., H. ANNER, L. KOBZIK, C. R. VALERI, D. SHEPRO, AND H. B. HECHTMAN. A high plasma prostaglandin to thromboxane ratio protects against renal ischemia. Surg. Gynecol. Obstet. 165: 404-409, 1987. 23. LINEHAN, J. H., C. A. DAWSON, AND D. A. RICKABY. Distribution of vascular resistance and compliance in a dog lung lobe. J. Appl. Physiol. 53: 158-168, 1982. 24. LINEHAN, J. H., C. A. DAWSON, D. A. RICKABY, AND T. A. BRONIKOWSKI. Pulmonary vascular compliance and viscoelasticity. J. Appl. Physiol. 61: 1802-1814, 1986. 25. MCDONALD, R. J., E. M. BERGER, AND J. E. REPINE. Neutrophilderived oxygen metabolites stimulate thromboxane release, pulmonary artery pressure increases, and weight gains in isolated perfused rat lungs. Am. Rev. Respir. Dis. 135: 957-959, 1987. 26. PERRY, M., AND A. E. TAYLOR. Phorbol myristate acetate-induced injury of isolated perfused rat lungs: neutrophil dependence. J. Appl. Physiol. 65: 2164-2169, 1988. 27. PERRY, M. L., S. G. KAYES, J. W. BARNARD, AND A. E. TAYLOR. Effects of phorbol myristate acetate-stimulated human neutrophils and mononuclear cells on rat lung. J. Appl. Physiol. 68: 235240, 1990. 28. RIPPE, B., R. C. ALLISON, J. C. PARKER, AND A. E. TAYLOR. Effects of histamine, serotonin, and norepinephrine on circulation of dog lungs. J. Appl. Physiol. 57: 223-232, 1984. 29. SAMUELSSON, B. Prostaglandin endoperoxides and thromboxanes. Role in platelets and in vascular and respiratory smooth muscle. Acta. Biol. Med. Ger. 35: 1055-1063, 1976. 30. SHIBAMOTO, T., AND T. KOBAYASHI. Acute effect of paraquat on lung fluid balance and prostanoid production in awake sheep. Am. Rev. Respir. Dis. 134: 1252-1257, 1986. 31. SHIBAMOTO, T., J. C. PARKER, A. E. TAYLOR, AND M. I. TOWNSLEY. Derecruitment of filtration surface area in paraquat-injured isolated dog lungs. J. Appl. Physiol. 68: 1581-1589, 1990. 32. SHINOZAWA, Y., C. HALES, W. JUNG, AND J. BURKE. Ibuprofen prevents synthetic smoke-induced pulmonary edema. Am. Rev. Respir. Dis. 134: 1145-1148, 1986. 33. SNAPPER, J. R., A. A. HUTCHISON, M. L. OGLETREE, AND K. L. BRIGHAM. Effects of cyclooxygenase inhibitors on the alterations in lung mechanics caused by endotoxemia in the unanesthetized sheep. J. Clin. Invest. 72: 63-76, 1983. 34. TOWNSLEY, M. I., R. J. KORTHIUS, B. RIPPE, J. C. PARKER, AND A. E. TAYLOR. Validation of double vascular occlusion method for Pc,i in lung and skeletal muscle. J. Appl. Physiol. 61: 127-132,1986. 35. TOWNSLEY, M. I., G. E. TAYLOR, R. J. KORTHIUS, AND A. E. TMYI.


INTERACTIONS

OF THROMBOXANE

LOR. Promethazine or DPPD pretreatment attenuates oleic acidinduced injury in isolated canine lungs. J. Appl. Physiol. 59: 39-46, 1985. 36. WALLENBURG,

H. C., N. ROTMANS. Prevention of recurrent idiopathic fetal growth retardation by low-dose aspirin and dipyridamole. Am. J. Obstet. Gynecol. 157: 1230-1235, 1987. 37. WALSH, S. W. Low-dose aspirin: treatment for the imbalance of increased thromboxane and decreased PGI, in preeclampsia. Am. J. Perinatol. 38. WEIR, M.

6: 124-132, 1989. R., D. K. KLASSEN,

N. HOOVER,

AND

F. L. DOUGLAS.

WITH

PROSTACYCLIN

1853

Preliminary observations of the acute effects of selective serum thromboxane inhibition and angiotensin converting enzyme inhibition on blood pressure and renal hemodynamics in hypertensive humans. J. Clin. Pharmacol. 29: 1108-1116, 1989. 39. WONG, D. G., J. D. SPENCE, L. LAMKI, D. FREEMAN, AND J. W. MCDONALD. Effect of nonsteroidal anti-inflammatory drugs on control of hypertension by beta-blockers and diuretics. Lancet 1: 997-1001,1986. 40. ZAR, J. H. Biostatistical Prentice Hall, 1984,

Analysis

p. 236-243.

(2nd ed.). Englewood Cliffs, NJ:


Pulmonary vascular responses to thromboxane and prostacyclin.

Thromboxane and prostacyclin metabolites in pulmonary hypertension.

Characterization of thromboxane receptor blocking effects of SQ 29548 in the feline pulmonary vascular bed.

An imbalance between the excretion of thromboxane and prostacyclin metabolites in pulmonary hypertension.

Excretion of metabolites of prostacyclin and thromboxane by rats with nephrotoxic nephritis: effects of interleukin-1.

thromboxane ratio in human breast cancer.

Characterization of the thromboxane receptor mediating prostacyclin release from cultured endothelial cells.

Effects of low-to-high doses of aspirin on platelet aggregability and metabolites of thromboxane A2 and prostacyclin.

Effects of increased intracranial pressure on pulmonary vascular resistance of fetal and neonatal goats.

Effects of nitrendipine and hypoxia on pulmonary vascular resistance in experimental emphysema.

Prostacyclin and thromboxane A2 formation is increased in human sepsis syndrome. Effects of cyclooxygenase inhibition.

Increased prostacyclin and thromboxane A2 formation in human varicose veins.

Thromboxane dominance and prostacyclin deficiency in habitual abortion.

Triene prostaglandins: prostacyclin and thromboxane biosynthesis and unique biological properties.

Prostacyclin and thromboxane synthesis by endometrial cancer and leiomyomas.

Comparative effects of immediate-release and extended-release aspirin on basal and bradykinin-stimulated excretion of thromboxane and prostacyclin metabolites.

The role of thromboxane and prostacyclin in ciclosporin-induced nephrotoxicity.

Effect of glutathione depletion on tissue and plasma prostacyclin and thromboxane in rats.

Effects of different cyclooxygenase inhibitors on the segmental distribution of pulmonary vascular resistance in the dog.

Effects of lung inflation on longitudinal distribution of pulmonary vascular resistance.

Aspirin, thromboxane, and prostacyclin in rats: A dilemma resolved?

Effects of non-steroidal anti-inflammatory drugs on prostacyclin and thromboxane biosynthesis in patients with mild essential hypertension.

Effects of hyperventilation on prostacyclin formation and on pulmonary vasodilation after group B beta-hemolytic streptococci-induced pulmonary hypertension.

Impact of lowering pulmonary vascular resistance on right and left ventricular deformation in pulmonary arterial hypertension.