Acta physiol. scand. 1975. 95. 95-101 From the Institute of Physiology, University of Oslo, Norway
The Pulmonary Vasoconstrictor Response to Hypoxia: Effects of Inhibitors of Prostaglandin Biosynthesis BY J. VAAGE,L. BJERTNRS and A. HAUGE Received 13 March 1975
Abstract VAAGE,J., L. BJERTNRS and A. HAUGE.The pulmonary vasoconstrictor response to hypoxia: Effects of inhibitors of prostaglandin biosynthesis. Acta physiol. scand. 1975.95. 95-101. The main purpose of the present work was to determine whether prostaglandins (PGs) synthetised in the lungs mediate the vasoconstrictor response to acute alveolar hypoxia. Isolated and ventilated lungs of rats were perfused a t 37°C with homologous blood at constant-volume, pulsatile inflow, and pressor responses to 3 min periods of standardized ventilation hypoxia recorded. Indomethacin, sodium meclofenamate and acetylsalicylic acid (all 100 pglml), which are potent inhibitors of PG biosynthesis, did not reduce the hypoxic vasoconstrictor response. Sometimes they even enhanced this response. We conclude that PGs do not mediate the hypoxia-induced vasoconstriction. We suggest that vasodilatory PGs might act to reduce and modify pulmonary arterial hypertension due to hypoxia.
The mechanism whereby acute alveolar hypoxia elicits pulmonary vasoconstriction still remains to be properly clarified. Since von Euler and Liljestrand (1946) demonstrated in cats that inhalation of subnormal concentrations of oxygen produced an increase in pulmonary arterial pressure, a number of vasoactive substances have been suggested-and discarded, as possible humoral transmitters of this response. During the last few years several investigators in this field have focused their interest on histamine and other vasoactive agents contained within mast cells situated near small muscular arteries in the lung (Hauge 1968 b, Hauge and Staub 1969, Sussmano and Carleton 1971, Haas and Bergofsky 1972, Kay et al. 1974). The role of histamine has, however, been disputed by some workers (Silove and Simcha 1973). In the present work we have focused our attention on another group of substances, namely the prostaglandins (PGs). The lungs have a large capacity for synthesis and liberation of PGs secondary to various mechanical and chemical stimuli (Piper and Vant; 1971, Piper 1974). Of the various PGs synthetised within the lung, PGF,, is a potent constrictor in the pulmonary circulation (Anggilrd and Bergstrom 1963). Furthermore, PGs have been suggested to control ventilation/perfusion ratios in the lung (Liljestrand 1967, Piper and Vane 1971, Smith 1973). PGs are released from lungs by ventilation (Berry, Edmonds and Wyllie 1971), and PG-like activity of the venous effluent from lungs has been reported to increase 95
96
J. VAAGE, L. BJERTNgS AND A. HAUGE
following deep lung inflations (Said, Kitamura and Vreim 1972, Piper and Vaage 1973) and during hypoxia (Said and Yoshida 1974). In order to answer the question whether PGs are mediators of the pulmonary pressor response to acute alveolar hypoxia, we have studied the effect of pharmacological inhibition of PG-biosynthesis in an isolated blood-perfused rat lung preparation.
Methods Experimenial preparations. Inbred albino rats (local strain), weighing between 180 and 280 g, were anesthetized with pentobarbitone sodium (3-4 mg/100 g b.wt., i.p.); a tracheotomy was performed, and the chest opened during positive pressure ventilation. The trachea, the lungs and the larger intrathoracic vessels were dissected free and heparin was injected (100 I.U. in 0.5 ml saline) into the left ventricle before the ligation of the caval veins. Stainless steel cannulas were placed in the pulmonary artery and the left atrium. The cannulas were fastened by a tight ligature around the heart, and the lung preparation was transferred to a humidified, constant-temperature perspex chamber. The lungs were freely suspended in a string fastened to the heart ligature and perfused with 20-25 ml heparinized (10 I.U./ml) blood, obtained by cardiac puncture of ether-anesthetized donor rats. At the onset of perfusion flow was adjusted so as to give a mean pulmonary arterial pressure of about 12-16 mm Hg. Perfusion was begun within 12 min of the interruption of the circulation. A Harvard Pulsatile Blood Pump (M-1405) giving constant volume inflow, was used. PPA was measured by a Statham P23DB pressure transducer connected to a Sanborn Model 320 D C amplifier recorder via a Sandborn model 350-1 100 B D C preamplifier. Mean pulmonary arterial pressure was calculated as the sum of the diastolic pressure and one-third of the pulse pressure. Controls as determined with electrical damping of the pressure gave corresponding results. The venous effluent was drained into an open, double-walled, thermostated blood reservoir with a bottom outlet. Since left atrial pressure and volume flow were kept constant, changes in pulmonary vascular resistance were directly reflected in changes in mean inflow pressure (AP,,). The method was a modification of that described elsewhere (Hauge 1968 a). Ventilation. After one or two gentle inflations of the lungs constant volume positive pressure ventilation was started, using a Starling “Ideal” ventilation pump ( C . F. Palmer, Ltd.). End-expiratory pressure was kept at about 2 cm H,O by means of a water seal. Tracheal pressure was recorded by a differential pressure transducer (Model 270, Hewlett-Packard) connected via a preamplifier to the Sanborn recorder. Stroke volume was adjusted so as to give a peak tracheal pressure of about 6 cm H,O at the onset. A fixed ventilation frequency of 80 per min was used for all the experiments. The standard gas mixture used for ventilation was 21 5: 02,5 % CO, and 7 4 % N,. Alveolar hypoxia was induced by ventilating the lungs with a gas mixture of 2 (%,0,, 5 % CO, and 93 % N, for standardized periods of 3 min with 5 min intervals (“hypoxic tests”). The pH of the perfusate was measured with a Radiometer acid-base analyzer (Model PHM 71). Drugs. The following inhibitors of PG synthesis were tested: Indomethacin, sodium meclofenamate and acetylsalicylic acid (Vane 1973, Ferreira and Vane 1974 a). Indomerhacin was prepared as follows: 27.218 g NaH,PO, was dissolved in 200 ml of sterile water (solution I). 50 ml of this solution was added to 46.1 ml 1 N NaOH. Sterile water was then added u p to 200 ml (solution 11). 85 ml of solution I1 were mixed with 2.5 g of indomethacin. 1 N NaOH was added until all indomethacin was solved at pH about 8.0. Finally solution I1 was added to a total volume of 100 ml. The prepared indomethacin solution (25 mg/ml) was stored dark in the refrigerator. Sodium meclofenamafe. The necessary amount was made up as required with water free of carbon dioxide. Acety/.sa/icy/ie acid. 20 mg acetylsalicylic acid was solved in 1 ml 96 :X, ethanol, and 19 ml of destilled water was added. KaNidin (Synthetic kallidin, Sandoz A.G.).
Results When 3 min periods of alveolar hypoxia are repeated at standardized intervals, the pressor response in isolated rat lungs follows a characteristic pattern (Hauge 1968 a, b): At the beginning of perfusion such “hypoxic tests” produce no o r very small pressor responses. After
91
ALVEOLAR HYPOXIA AND PROSTAGLANDINS
I
t
J . 1 '
1hr 0 min
lhr8min
'
. 1 1 ' lyhrl6min
lhrZCmin
J.
1min
Indomethoc in
Fig. 1. Pressor responses to 3 min periods of ventilation hypoxia between arrows before and after addition of indomethacin (100 ,ug/ml). Constant volume perfusion of isolated rat lungs. PPA is the pulmonary arterial pressure, P,, is the tracheal pressure.
a few tests the response gradually increases until there is a maximum effect for the given length of hypoxia. Several equal maximum responses can then be obtained before a gradual decline in responsiveness occurs. When Finally alveolar hypoxia no longer produces a pressor effect, the pulmonary vascular bed will still respond to vasoactive agents, e.g. kallidin. Since we expected inhibitors of PG-synthesis either to have no effect o r to reduce the pressor response to hypoxia, we tested the drugs, in all but one case, in periods when the responses were either increasing or were completely stable. Fig. 1 gives an example of the test procedure. First, two control pressor responses to hypoxia were obtained, whereafter indomethacin (2 mg) was added to the perfusate; then the two following pressor responses were recorded. Fig. 2 demonstrates in diagrammatic form all the experiments. In the left panel Roman numerals I and I1 indicate the last two responses (APFA) obtained before drug administration, whereas 111 and I V indicate the first two responses obtained afterwards. Zero level in
I
;:[
I
0 a + E +24
m
ga
o
w
c
-2
-
I
g
-4 0
I Betore
after drug odminislrotion
I
-.. I 1-11
11-Ill
Ill-IV
Fig. 2. Left panel demonstrates increases in inflow pressure (AP,,) to 3 min periods of ventilation hypoxia (in 8 expts.) before (0) and after ( 0 ) administration of inhibitors of prostaglandin biosynthesls. Numbers 1 to 8 refers to Table I where the type of drug given to each individual lung preparation is listed. Right panel demonstrates the difference between one hypoxic pressor response and the preceding response (A(AP,,)). This difference was increased by the administration of prostaglandin biosynthesis inhibitors, which were given between response I1 and 111 (3).
I - 155819
98
J. VAAGE, L. BJERTNKS AND A. HAUGE
TABLE I. Expt. no.
Drug (2 mg)
Volume of Flow perfusate (ml/min) (ml)
Before drug administration
Baseline
APpA** APpA
pPA*
I
I1
After drug administration Baseline APpA Pp* I11
APpA
11.0
IV
1
Acetylsalicylic acid
25
15.3
10.0
11.0
14.8
13.7
2
Indomethacin 21
17.7
12.3
12.0
17.3
17.0
19.0
3
Indomethacin 20
14
13.0
18.0
20.0
15.0
23.0
23.0
4
Acetylsalicylic acid
11.6
10.7
12.4
12.3
13.7
13.7
5
Indomethacin 20
10
16.0
8.3
3.0
16.0
10.0
13.3
6
Meclofenamate
11.6
4.7
6.0
12.0
7.3
8.7
7
Indomethacin 22
24
11.3
19.7
20.3
11.3
22.0
22.0
8
Meclofenamate
16
15.0
16.0
16.0
15.0
16.0
16.0
20
25
20
* Pulmonary art_erialpressure and pressure-changes are given in mm Hg. * * Peak rise in PpA during 3 min of ventilation hypoxia.
this panel represents baseline level of ppAbefore each individual hypoxic test. With the exception of exp. no. 3, Table I, the baseline level of PPAdid not change. The right hand panel in Fig. 2 demonstrates the change in APpz\(A(APpA)) from I to 11, from I1 to I11 and from I11 to IV. Thus, a continuous increase in the response to hypoxia would give positive values of A(APpA), and if the increase was linear the values would have been identical. A sudden increase in the response, in excess of a steady linear rise, would have increased A(APpA) compared with the preceding one. Following drug administration this is exactly what happened in the majority of preparations. The response to hypoxia was lifted to a higher plateau. When all the values of A(APpA) from I to I1 were regarded as one group and compared to the group of A(APPA) from I1 to 111, this last group was significantly increased (P i0.04, Wilcoxon two-sided test). In no case did inhibitors of PG synthesis cause a reduction in the pressor response to acute hypoxia. Undiminished responses could be elicited as long as 48 min after administration of indomethacin, which is in agreement with the normal response pattern in this preparation. Table I gives a general overview of preparation parameters such as volume of perfusate, flow and baseline inflow pressure (PpA).In addition the table shows which inhibitor drug that was given to each individual lung preparation and also the numeric values of pressor responses. None of the inhibitors of PG-synthesis used had any effect on the vasoconstrictor response to kallidin (100 pg) injected intra-arterially. Neither hypoxic tests nor the administration of inhibitors of PG biosynthesis caused consistent bronchomotor responses, judged from the observation of tracheal pressure.
ALVEOLAR HYPOWA AND PROSTAGLANDINS
99
Discussion Since inhibition of P G biosynthesis was not experimentally verified in the present study, we have relied upon the works of other investigators when deciding the dosage of inhibitors and the length of the observation periods. The use of nonsteroid anti-inflammatory drugs is a tool to test the involvement of PGs in biological systems (Ferreira and Vane 1974 b). Piper and Vane (1969) found that 0.1-0.5 yg/ml of indomethacin and 1-5 pg/ml of aspirin blocked PG synthesis in isolated guinea-pig lungs perfused with Krebs-Ringer solution. I n our experiments 2 mg of each agent tested was diluted in about 20 ml of blood, i.e. approximately 1 0 0 ,ug/ml. Since about 90 per cent of indomethacin is bound to plasma-proteins (Vane 1973), this is equivalent to an effective dose of this particular drug of some 10 pg/ml, i.e. about 100 times the minimum dose recommended by the above workers. Herbazynska-Cedro and Vane (1973) have also shown that renal PG-synthesis in vivo was inhibited by indomethacin 2 mg/kg. Plasma protein binding of acetyl salicylic acid is 5&80 per cent (Vane 1973). The fraction of meclofenamate which is bound to plasma protein is not known (Vane 1973). However, when evaluated on a weight basis as well as on a molar basis, this drug is the most potent inhibitor of PG-synthesis of the three used in the present work (Ferreira and Vane 1974 a). Perfused rat lungs have a considerable capacity to release PGs secondary to a wide variety of stimuli (Bakhle and Vane 1974). Whether intrapulmonary synthesis and release of PGs will cause a constrictor or dilator response depends on the balance between the different subgroups of PGs which act on vascular smooth muscle cells. PGF,, is a potent pulmonary vasoconstrictor agent (Anggird and Bergstrom 1963), whereas PGE, is able to dilate pulmonary resistance vessels, provided there is a certain degree of smooth muscle tone (Hauge et al. 1967). The release of PGs from lungs is dependent on a rapidly induced biosynthesis, since only negligible amounts of PGs are stored in lung tissue (Piper and Vane 1971). Accordingly, the present observation that pharmacological blockade of PG synthesis did not diminish the vasoconstrictor response to hypoxia, indicates that vasoconstrictor PGs are not mediators of this response. An unexpected finding was that nonsteroid, anti-inflammatory drugs sometimes induced a moderate potentiation of the vasoconstrictor effect caused by hypoxia. This observation is supported, however, by a short recent report from Weir et al. (1974), in which it is stated that meclofenamate and indomethacin augment the pulmonary pressor response to hypoxia in intact dogs. As an explanation it is tempting to speculate whether pulmonary arterial hypertension due to hypoxia might induce synthesis and release of some vasodilatory PGs. PG release has been induced by various chemical and mechanical stimuli with distortion of cell membranes (Piper and Vane 1971). The release of some vasodilatory PGs from lungs during severe pulmonary hypertension with subsequent moderation of vascular transmural pressure can be regarded as a functional and advantageous phenomenon modulating the response to hypoxia. Such a postulated mechanism might then be prevented by administration of PG synthesis inhibitors. A secondary release of this type may be the reason why other investigators have observed PG-like biologically active substances in the effluent perfusate from lungs during hypoxia (Said and Yoshida 1974).
100
J. VAAGE, L. BJERTNAZS AND A. HAUGE
Recently, Said and coworkers (1974) reported that the hypoxic pressor response of cat lungs in situ is reduced by aspirin. There are, however, several reasons why this particular work is difficult to assess. In these experiments aspirin by itself elevated pulmonary vascular resistance (PVR), whereas the PVR-level obtained during hypoxia was unaltered. When a reduction in the PVR-response to hypoxia was claimed, it was based on calculations in per cent of the new PVR base-line levels, assuming a linear system. Furthermore, no tests with a pulmonary vasoconstrictor agent was carried out in order to detect unspecific damping of vascular smooth muscle reactivity. Such general and non-specific depression of vascular smooth muscles has been described for rabbit lungs following the administration of other nonsteroid anti-inflammatory drugs, such as sodium salicylate and phenylbutazone (Hauge et al. 1966). Finally, we would like to point out that also Said and coworkers found that indomethacin (1 to 10 mg/kg) “sometimes even enhanced“ the pulmonary arterial pressor response in cats. L. B. and J. V. are Research Fellows of the Norwegian Council for Science an d the Humanities. This support and additional support, through the Institute of Physiology, from the Nansen Foundation and the Anders Jahre Foundation for the Promotion of Science together with the generous gifts of indomethacin (Dumex, Copenhagen, Denmark), acetylsalicylic acid (Nyegaard & Co., Oslo, Norway) and sodium meclofenamate (Parke-Davis, Pontypool, Mon., England) are all gratefully acknowledged.
References A N G G ~ R E. D , and S. BERGSTROM, Biological effects of an unsaturated trihydroxy acid (PGFaa) from normal swine lung. Acta physiol. scand. 1963. 58. 1-12. BAKHLE,Y. S. and J. R. VANE,Pharmacokinetic function of the pulmonary circulation. Physiol. Reo. 1974. 54. 1007-1045. BERRY,P. A., .I. F. EDMONDS and J. M. WYLLIE, Release of prostaglandin E, and unidentified factors from ventilated lungs. Brit. J . Surg. 1971. 58. 189-192. EULER,U. S. VON and G. LILJESTRAND, Observations on the pulmonary arterial blood pressure in the cat. Acta physiol. seand. 1946. 12. 301-320. FERREIRA, S. H. and J. R. VANE,New aspects of the mode of action of nonsteroid anti-inflammatory drugs. Ann. Rev. Pharmacol. 1974 a. 14. 57-73. FERREIRA, S. H. and J. R. VANE,Aspirin and prostaglandins, in “The prostaglandins”, edited by Peter W. Ramwell, Plenum Press, New York, 1974 b. Vol. 2, pp. 1-47. HAAS,F. and E. H. BERGOFSKY, Role of the mast cell in the pulmonary pressor response to hypoxia. J. clin. Invest. 1972. 51. 3154-3162. HAUGE,A., Conditions governing the pressor response to ventilation hypoxia in isolated, perfused rat lungs. Acta physiol. scand. 1968 a. 72. 33-44. HAUGE,A., Role of histamine in hypoxic pulmonary hypertension in the rat. I. Blockade and potentiation of endogenous amines, kinins and ATP. Circulat. Res. 1968 b. 22. 371-383. HAUGE, A., P. K. M. LUNDEand B. A. WAALER, The effect of bradykinin, kallidin and eledoisin upon the pulmonary vascular bed of an isolated blood-perfused rabbit lung preparation. Acra physiol. scand. 1966. 66. 269-277. HAUGE,A., P. K. M. LUNDEand B. A. WAALER, Effects of prostaglandin E, and adrenaline on the pulmonary vascular resistance (PVR) in isolated rabbit lungs. Life Sci. 1967. 6. 673-680. HAUGE,A. and N. C. STAUB,Prevention of hypoxic vasoconstriction in cat lung by histamine releasing agent 48/80. J. appl. Physiol. 1969. 26. 693-699. HERBAZYNSKA-CEDRO, K. and J. R. VANE,Contribution of intrarenal generation of prostaglandin to autoregulation of renal blood flow in the dog. Circulat. Res. 1973. 33. 428-36. KAY,J. M., J. C. WAYMIRE and R. T. GROVER, Lung mast cell hyperplasia and pulmonary histamine-forming capacity in hypoxic rats. Amer. J. Phjwiol. 1974. 226. 178-184.
ALVEOLAR HYPOXIA AND PROSTAGLANDINS
101
LILJESTRAND, G . , Discussion remarks to Anggard and Samuelsson: “The metabolism of prostaglandins in lung tissue’’ in Proc. 2nd Nobel Symp. Prostaglandins, edited by S . Bergstrom and B. Samuelsson. Almqvist & Wiksell, Stockholm 1967, pp. 107-108. PIPER,P. J., Release and metabolism of prostaglandins in lung tissue. Pol. J. Pharmacol. Pharmacy 1974. 26. 61-72. PIPER,P. J. and J. VAAGE,Unpublished observations. 1973. PIPER, P. J. and J. R. VANE,Release of additional factors in anaphylaxis and its antagonism by anti-inflammatory drugs. Nature (Lond.) 1969. 223. 29-37. PIPER, P. J. and J. R. VANE,The release of prostaglandin from lung and other tissues. Ann. N . Y. Acad. Sci. 1971. 180. 363-385. SAID,S . I., S. KITAMURA and C. VREIM,Prostaglandin release from the lung during mechanical ventilation at large tidal volumes. J. clin. Invest. 1972. 51. 83a. Release of prostaglandins and other humoral mediators during hypoxic breathSAID, S. I. and T. YOSHIDA, ing anli pulmonary edema. Chest 1974. 66. Suppl. (part 2), 12 A-I3 A. and C. VREIM,Pulmonary alveolar hypoxia: Release of prostaglanSAID,S. I., T. YOSHIDA,S. KITAMURA dins and other humoral mediators. Science 1974. 185. 1181-1183. SILOVE,E. D. and A. J. SIMCHA,Histamine-induced pulmonary vasodilatation in the calf: relationship to hypoxia. J. appl. Physiol. 1973. 35. 830-836. SMITH,A. P., “The prostaglandins”, edited by P. W. Ramsvell. Plenum Press, New York-London, 1973, vol. 1, pp. 203-218. SUSMANO, A. and R. A. CARLETON, Prevention of hypoxic pulmonary hypertension by chlorpheniramine. J. appl. Physiol. 1971. 31. 531-535. VANE,J. R., Prostaglandins and Aspirin-like drugs. Pharmacology and the future of man. Proc. 5th Int. congr. Pharmacology, San Francisco 1972, vol. 5. pp. 352-378 (Karger, Basel). 1973. WEIR,E. K., J. T. REEVES and R. F. GROVER,Meclofenamate and indomethacin augment the pulmonary pressor response to hypoxia and exogenous prostaglandin FZa.The Physiologist 1974. 17. 355.
FYSaGLQGlSKA INSTITUTIONEW I IiAROLIMSXA INSTITUTET .-.- .- STOCKHOLM
The Pulmonary Vasoconstrictor Response to Hypoxia: Effects of Inhibitors of Prostaglandin Biosynthesis BY J. VAAGE,L. BJERTNRS and A. HAUGE Received 13 March 1975
Abstract VAAGE,J., L. BJERTNRS and A. HAUGE.The pulmonary vasoconstrictor response to hypoxia: Effects of inhibitors of prostaglandin biosynthesis. Acta physiol. scand. 1975.95. 95-101. The main purpose of the present work was to determine whether prostaglandins (PGs) synthetised in the lungs mediate the vasoconstrictor response to acute alveolar hypoxia. Isolated and ventilated lungs of rats were perfused a t 37°C with homologous blood at constant-volume, pulsatile inflow, and pressor responses to 3 min periods of standardized ventilation hypoxia recorded. Indomethacin, sodium meclofenamate and acetylsalicylic acid (all 100 pglml), which are potent inhibitors of PG biosynthesis, did not reduce the hypoxic vasoconstrictor response. Sometimes they even enhanced this response. We conclude that PGs do not mediate the hypoxia-induced vasoconstriction. We suggest that vasodilatory PGs might act to reduce and modify pulmonary arterial hypertension due to hypoxia.
The mechanism whereby acute alveolar hypoxia elicits pulmonary vasoconstriction still remains to be properly clarified. Since von Euler and Liljestrand (1946) demonstrated in cats that inhalation of subnormal concentrations of oxygen produced an increase in pulmonary arterial pressure, a number of vasoactive substances have been suggested-and discarded, as possible humoral transmitters of this response. During the last few years several investigators in this field have focused their interest on histamine and other vasoactive agents contained within mast cells situated near small muscular arteries in the lung (Hauge 1968 b, Hauge and Staub 1969, Sussmano and Carleton 1971, Haas and Bergofsky 1972, Kay et al. 1974). The role of histamine has, however, been disputed by some workers (Silove and Simcha 1973). In the present work we have focused our attention on another group of substances, namely the prostaglandins (PGs). The lungs have a large capacity for synthesis and liberation of PGs secondary to various mechanical and chemical stimuli (Piper and Vant; 1971, Piper 1974). Of the various PGs synthetised within the lung, PGF,, is a potent constrictor in the pulmonary circulation (Anggilrd and Bergstrom 1963). Furthermore, PGs have been suggested to control ventilation/perfusion ratios in the lung (Liljestrand 1967, Piper and Vane 1971, Smith 1973). PGs are released from lungs by ventilation (Berry, Edmonds and Wyllie 1971), and PG-like activity of the venous effluent from lungs has been reported to increase 95
96
J. VAAGE, L. BJERTNgS AND A. HAUGE
following deep lung inflations (Said, Kitamura and Vreim 1972, Piper and Vaage 1973) and during hypoxia (Said and Yoshida 1974). In order to answer the question whether PGs are mediators of the pulmonary pressor response to acute alveolar hypoxia, we have studied the effect of pharmacological inhibition of PG-biosynthesis in an isolated blood-perfused rat lung preparation.
Methods Experimenial preparations. Inbred albino rats (local strain), weighing between 180 and 280 g, were anesthetized with pentobarbitone sodium (3-4 mg/100 g b.wt., i.p.); a tracheotomy was performed, and the chest opened during positive pressure ventilation. The trachea, the lungs and the larger intrathoracic vessels were dissected free and heparin was injected (100 I.U. in 0.5 ml saline) into the left ventricle before the ligation of the caval veins. Stainless steel cannulas were placed in the pulmonary artery and the left atrium. The cannulas were fastened by a tight ligature around the heart, and the lung preparation was transferred to a humidified, constant-temperature perspex chamber. The lungs were freely suspended in a string fastened to the heart ligature and perfused with 20-25 ml heparinized (10 I.U./ml) blood, obtained by cardiac puncture of ether-anesthetized donor rats. At the onset of perfusion flow was adjusted so as to give a mean pulmonary arterial pressure of about 12-16 mm Hg. Perfusion was begun within 12 min of the interruption of the circulation. A Harvard Pulsatile Blood Pump (M-1405) giving constant volume inflow, was used. PPA was measured by a Statham P23DB pressure transducer connected to a Sanborn Model 320 D C amplifier recorder via a Sandborn model 350-1 100 B D C preamplifier. Mean pulmonary arterial pressure was calculated as the sum of the diastolic pressure and one-third of the pulse pressure. Controls as determined with electrical damping of the pressure gave corresponding results. The venous effluent was drained into an open, double-walled, thermostated blood reservoir with a bottom outlet. Since left atrial pressure and volume flow were kept constant, changes in pulmonary vascular resistance were directly reflected in changes in mean inflow pressure (AP,,). The method was a modification of that described elsewhere (Hauge 1968 a). Ventilation. After one or two gentle inflations of the lungs constant volume positive pressure ventilation was started, using a Starling “Ideal” ventilation pump ( C . F. Palmer, Ltd.). End-expiratory pressure was kept at about 2 cm H,O by means of a water seal. Tracheal pressure was recorded by a differential pressure transducer (Model 270, Hewlett-Packard) connected via a preamplifier to the Sanborn recorder. Stroke volume was adjusted so as to give a peak tracheal pressure of about 6 cm H,O at the onset. A fixed ventilation frequency of 80 per min was used for all the experiments. The standard gas mixture used for ventilation was 21 5: 02,5 % CO, and 7 4 % N,. Alveolar hypoxia was induced by ventilating the lungs with a gas mixture of 2 (%,0,, 5 % CO, and 93 % N, for standardized periods of 3 min with 5 min intervals (“hypoxic tests”). The pH of the perfusate was measured with a Radiometer acid-base analyzer (Model PHM 71). Drugs. The following inhibitors of PG synthesis were tested: Indomethacin, sodium meclofenamate and acetylsalicylic acid (Vane 1973, Ferreira and Vane 1974 a). Indomerhacin was prepared as follows: 27.218 g NaH,PO, was dissolved in 200 ml of sterile water (solution I). 50 ml of this solution was added to 46.1 ml 1 N NaOH. Sterile water was then added u p to 200 ml (solution 11). 85 ml of solution I1 were mixed with 2.5 g of indomethacin. 1 N NaOH was added until all indomethacin was solved at pH about 8.0. Finally solution I1 was added to a total volume of 100 ml. The prepared indomethacin solution (25 mg/ml) was stored dark in the refrigerator. Sodium meclofenamafe. The necessary amount was made up as required with water free of carbon dioxide. Acety/.sa/icy/ie acid. 20 mg acetylsalicylic acid was solved in 1 ml 96 :X, ethanol, and 19 ml of destilled water was added. KaNidin (Synthetic kallidin, Sandoz A.G.).
Results When 3 min periods of alveolar hypoxia are repeated at standardized intervals, the pressor response in isolated rat lungs follows a characteristic pattern (Hauge 1968 a, b): At the beginning of perfusion such “hypoxic tests” produce no o r very small pressor responses. After
91
ALVEOLAR HYPOXIA AND PROSTAGLANDINS
I
t
J . 1 '
1hr 0 min
lhr8min
'
. 1 1 ' lyhrl6min
lhrZCmin
J.
1min
Indomethoc in
Fig. 1. Pressor responses to 3 min periods of ventilation hypoxia between arrows before and after addition of indomethacin (100 ,ug/ml). Constant volume perfusion of isolated rat lungs. PPA is the pulmonary arterial pressure, P,, is the tracheal pressure.
a few tests the response gradually increases until there is a maximum effect for the given length of hypoxia. Several equal maximum responses can then be obtained before a gradual decline in responsiveness occurs. When Finally alveolar hypoxia no longer produces a pressor effect, the pulmonary vascular bed will still respond to vasoactive agents, e.g. kallidin. Since we expected inhibitors of PG-synthesis either to have no effect o r to reduce the pressor response to hypoxia, we tested the drugs, in all but one case, in periods when the responses were either increasing or were completely stable. Fig. 1 gives an example of the test procedure. First, two control pressor responses to hypoxia were obtained, whereafter indomethacin (2 mg) was added to the perfusate; then the two following pressor responses were recorded. Fig. 2 demonstrates in diagrammatic form all the experiments. In the left panel Roman numerals I and I1 indicate the last two responses (APFA) obtained before drug administration, whereas 111 and I V indicate the first two responses obtained afterwards. Zero level in
I
;:[
I
0 a + E +24
m
ga
o
w
c
-2
-
I
g
-4 0
I Betore
after drug odminislrotion
I
-.. I 1-11
11-Ill
Ill-IV
Fig. 2. Left panel demonstrates increases in inflow pressure (AP,,) to 3 min periods of ventilation hypoxia (in 8 expts.) before (0) and after ( 0 ) administration of inhibitors of prostaglandin biosynthesls. Numbers 1 to 8 refers to Table I where the type of drug given to each individual lung preparation is listed. Right panel demonstrates the difference between one hypoxic pressor response and the preceding response (A(AP,,)). This difference was increased by the administration of prostaglandin biosynthesis inhibitors, which were given between response I1 and 111 (3).
I - 155819
98
J. VAAGE, L. BJERTNKS AND A. HAUGE
TABLE I. Expt. no.
Drug (2 mg)
Volume of Flow perfusate (ml/min) (ml)
Before drug administration
Baseline
APpA** APpA
pPA*
I
I1
After drug administration Baseline APpA Pp* I11
APpA
11.0
IV
1
Acetylsalicylic acid
25
15.3
10.0
11.0
14.8
13.7
2
Indomethacin 21
17.7
12.3
12.0
17.3
17.0
19.0
3
Indomethacin 20
14
13.0
18.0
20.0
15.0
23.0
23.0
4
Acetylsalicylic acid
11.6
10.7
12.4
12.3
13.7
13.7
5
Indomethacin 20
10
16.0
8.3
3.0
16.0
10.0
13.3
6
Meclofenamate
11.6
4.7
6.0
12.0
7.3
8.7
7
Indomethacin 22
24
11.3
19.7
20.3
11.3
22.0
22.0
8
Meclofenamate
16
15.0
16.0
16.0
15.0
16.0
16.0
20
25
20
* Pulmonary art_erialpressure and pressure-changes are given in mm Hg. * * Peak rise in PpA during 3 min of ventilation hypoxia.
this panel represents baseline level of ppAbefore each individual hypoxic test. With the exception of exp. no. 3, Table I, the baseline level of PPAdid not change. The right hand panel in Fig. 2 demonstrates the change in APpz\(A(APpA)) from I to 11, from I1 to I11 and from I11 to IV. Thus, a continuous increase in the response to hypoxia would give positive values of A(APpA), and if the increase was linear the values would have been identical. A sudden increase in the response, in excess of a steady linear rise, would have increased A(APpA) compared with the preceding one. Following drug administration this is exactly what happened in the majority of preparations. The response to hypoxia was lifted to a higher plateau. When all the values of A(APpA) from I to I1 were regarded as one group and compared to the group of A(APPA) from I1 to 111, this last group was significantly increased (P i0.04, Wilcoxon two-sided test). In no case did inhibitors of PG synthesis cause a reduction in the pressor response to acute hypoxia. Undiminished responses could be elicited as long as 48 min after administration of indomethacin, which is in agreement with the normal response pattern in this preparation. Table I gives a general overview of preparation parameters such as volume of perfusate, flow and baseline inflow pressure (PpA).In addition the table shows which inhibitor drug that was given to each individual lung preparation and also the numeric values of pressor responses. None of the inhibitors of PG-synthesis used had any effect on the vasoconstrictor response to kallidin (100 pg) injected intra-arterially. Neither hypoxic tests nor the administration of inhibitors of PG biosynthesis caused consistent bronchomotor responses, judged from the observation of tracheal pressure.
ALVEOLAR HYPOWA AND PROSTAGLANDINS
99
Discussion Since inhibition of P G biosynthesis was not experimentally verified in the present study, we have relied upon the works of other investigators when deciding the dosage of inhibitors and the length of the observation periods. The use of nonsteroid anti-inflammatory drugs is a tool to test the involvement of PGs in biological systems (Ferreira and Vane 1974 b). Piper and Vane (1969) found that 0.1-0.5 yg/ml of indomethacin and 1-5 pg/ml of aspirin blocked PG synthesis in isolated guinea-pig lungs perfused with Krebs-Ringer solution. I n our experiments 2 mg of each agent tested was diluted in about 20 ml of blood, i.e. approximately 1 0 0 ,ug/ml. Since about 90 per cent of indomethacin is bound to plasma-proteins (Vane 1973), this is equivalent to an effective dose of this particular drug of some 10 pg/ml, i.e. about 100 times the minimum dose recommended by the above workers. Herbazynska-Cedro and Vane (1973) have also shown that renal PG-synthesis in vivo was inhibited by indomethacin 2 mg/kg. Plasma protein binding of acetyl salicylic acid is 5&80 per cent (Vane 1973). The fraction of meclofenamate which is bound to plasma protein is not known (Vane 1973). However, when evaluated on a weight basis as well as on a molar basis, this drug is the most potent inhibitor of PG-synthesis of the three used in the present work (Ferreira and Vane 1974 a). Perfused rat lungs have a considerable capacity to release PGs secondary to a wide variety of stimuli (Bakhle and Vane 1974). Whether intrapulmonary synthesis and release of PGs will cause a constrictor or dilator response depends on the balance between the different subgroups of PGs which act on vascular smooth muscle cells. PGF,, is a potent pulmonary vasoconstrictor agent (Anggird and Bergstrom 1963), whereas PGE, is able to dilate pulmonary resistance vessels, provided there is a certain degree of smooth muscle tone (Hauge et al. 1967). The release of PGs from lungs is dependent on a rapidly induced biosynthesis, since only negligible amounts of PGs are stored in lung tissue (Piper and Vane 1971). Accordingly, the present observation that pharmacological blockade of PG synthesis did not diminish the vasoconstrictor response to hypoxia, indicates that vasoconstrictor PGs are not mediators of this response. An unexpected finding was that nonsteroid, anti-inflammatory drugs sometimes induced a moderate potentiation of the vasoconstrictor effect caused by hypoxia. This observation is supported, however, by a short recent report from Weir et al. (1974), in which it is stated that meclofenamate and indomethacin augment the pulmonary pressor response to hypoxia in intact dogs. As an explanation it is tempting to speculate whether pulmonary arterial hypertension due to hypoxia might induce synthesis and release of some vasodilatory PGs. PG release has been induced by various chemical and mechanical stimuli with distortion of cell membranes (Piper and Vane 1971). The release of some vasodilatory PGs from lungs during severe pulmonary hypertension with subsequent moderation of vascular transmural pressure can be regarded as a functional and advantageous phenomenon modulating the response to hypoxia. Such a postulated mechanism might then be prevented by administration of PG synthesis inhibitors. A secondary release of this type may be the reason why other investigators have observed PG-like biologically active substances in the effluent perfusate from lungs during hypoxia (Said and Yoshida 1974).
100
J. VAAGE, L. BJERTNAZS AND A. HAUGE
Recently, Said and coworkers (1974) reported that the hypoxic pressor response of cat lungs in situ is reduced by aspirin. There are, however, several reasons why this particular work is difficult to assess. In these experiments aspirin by itself elevated pulmonary vascular resistance (PVR), whereas the PVR-level obtained during hypoxia was unaltered. When a reduction in the PVR-response to hypoxia was claimed, it was based on calculations in per cent of the new PVR base-line levels, assuming a linear system. Furthermore, no tests with a pulmonary vasoconstrictor agent was carried out in order to detect unspecific damping of vascular smooth muscle reactivity. Such general and non-specific depression of vascular smooth muscles has been described for rabbit lungs following the administration of other nonsteroid anti-inflammatory drugs, such as sodium salicylate and phenylbutazone (Hauge et al. 1966). Finally, we would like to point out that also Said and coworkers found that indomethacin (1 to 10 mg/kg) “sometimes even enhanced“ the pulmonary arterial pressor response in cats. L. B. and J. V. are Research Fellows of the Norwegian Council for Science an d the Humanities. This support and additional support, through the Institute of Physiology, from the Nansen Foundation and the Anders Jahre Foundation for the Promotion of Science together with the generous gifts of indomethacin (Dumex, Copenhagen, Denmark), acetylsalicylic acid (Nyegaard & Co., Oslo, Norway) and sodium meclofenamate (Parke-Davis, Pontypool, Mon., England) are all gratefully acknowledged.
References A N G G ~ R E. D , and S. BERGSTROM, Biological effects of an unsaturated trihydroxy acid (PGFaa) from normal swine lung. Acta physiol. scand. 1963. 58. 1-12. BAKHLE,Y. S. and J. R. VANE,Pharmacokinetic function of the pulmonary circulation. Physiol. Reo. 1974. 54. 1007-1045. BERRY,P. A., .I. F. EDMONDS and J. M. WYLLIE, Release of prostaglandin E, and unidentified factors from ventilated lungs. Brit. J . Surg. 1971. 58. 189-192. EULER,U. S. VON and G. LILJESTRAND, Observations on the pulmonary arterial blood pressure in the cat. Acta physiol. seand. 1946. 12. 301-320. FERREIRA, S. H. and J. R. VANE,New aspects of the mode of action of nonsteroid anti-inflammatory drugs. Ann. Rev. Pharmacol. 1974 a. 14. 57-73. FERREIRA, S. H. and J. R. VANE,Aspirin and prostaglandins, in “The prostaglandins”, edited by Peter W. Ramwell, Plenum Press, New York, 1974 b. Vol. 2, pp. 1-47. HAAS,F. and E. H. BERGOFSKY, Role of the mast cell in the pulmonary pressor response to hypoxia. J. clin. Invest. 1972. 51. 3154-3162. HAUGE,A., Conditions governing the pressor response to ventilation hypoxia in isolated, perfused rat lungs. Acta physiol. scand. 1968 a. 72. 33-44. HAUGE,A., Role of histamine in hypoxic pulmonary hypertension in the rat. I. Blockade and potentiation of endogenous amines, kinins and ATP. Circulat. Res. 1968 b. 22. 371-383. HAUGE, A., P. K. M. LUNDEand B. A. WAALER, The effect of bradykinin, kallidin and eledoisin upon the pulmonary vascular bed of an isolated blood-perfused rabbit lung preparation. Acra physiol. scand. 1966. 66. 269-277. HAUGE,A., P. K. M. LUNDEand B. A. WAALER, Effects of prostaglandin E, and adrenaline on the pulmonary vascular resistance (PVR) in isolated rabbit lungs. Life Sci. 1967. 6. 673-680. HAUGE,A. and N. C. STAUB,Prevention of hypoxic vasoconstriction in cat lung by histamine releasing agent 48/80. J. appl. Physiol. 1969. 26. 693-699. HERBAZYNSKA-CEDRO, K. and J. R. VANE,Contribution of intrarenal generation of prostaglandin to autoregulation of renal blood flow in the dog. Circulat. Res. 1973. 33. 428-36. KAY,J. M., J. C. WAYMIRE and R. T. GROVER, Lung mast cell hyperplasia and pulmonary histamine-forming capacity in hypoxic rats. Amer. J. Phjwiol. 1974. 226. 178-184.
ALVEOLAR HYPOXIA AND PROSTAGLANDINS
101
LILJESTRAND, G . , Discussion remarks to Anggard and Samuelsson: “The metabolism of prostaglandins in lung tissue’’ in Proc. 2nd Nobel Symp. Prostaglandins, edited by S . Bergstrom and B. Samuelsson. Almqvist & Wiksell, Stockholm 1967, pp. 107-108. PIPER,P. J., Release and metabolism of prostaglandins in lung tissue. Pol. J. Pharmacol. Pharmacy 1974. 26. 61-72. PIPER,P. J. and J. VAAGE,Unpublished observations. 1973. PIPER, P. J. and J. R. VANE,Release of additional factors in anaphylaxis and its antagonism by anti-inflammatory drugs. Nature (Lond.) 1969. 223. 29-37. PIPER, P. J. and J. R. VANE,The release of prostaglandin from lung and other tissues. Ann. N . Y. Acad. Sci. 1971. 180. 363-385. SAID,S . I., S. KITAMURA and C. VREIM,Prostaglandin release from the lung during mechanical ventilation at large tidal volumes. J. clin. Invest. 1972. 51. 83a. Release of prostaglandins and other humoral mediators during hypoxic breathSAID, S. I. and T. YOSHIDA, ing anli pulmonary edema. Chest 1974. 66. Suppl. (part 2), 12 A-I3 A. and C. VREIM,Pulmonary alveolar hypoxia: Release of prostaglanSAID,S. I., T. YOSHIDA,S. KITAMURA dins and other humoral mediators. Science 1974. 185. 1181-1183. SILOVE,E. D. and A. J. SIMCHA,Histamine-induced pulmonary vasodilatation in the calf: relationship to hypoxia. J. appl. Physiol. 1973. 35. 830-836. SMITH,A. P., “The prostaglandins”, edited by P. W. Ramsvell. Plenum Press, New York-London, 1973, vol. 1, pp. 203-218. SUSMANO, A. and R. A. CARLETON, Prevention of hypoxic pulmonary hypertension by chlorpheniramine. J. appl. Physiol. 1971. 31. 531-535. VANE,J. R., Prostaglandins and Aspirin-like drugs. Pharmacology and the future of man. Proc. 5th Int. congr. Pharmacology, San Francisco 1972, vol. 5. pp. 352-378 (Karger, Basel). 1973. WEIR,E. K., J. T. REEVES and R. F. GROVER,Meclofenamate and indomethacin augment the pulmonary pressor response to hypoxia and exogenous prostaglandin FZa.The Physiologist 1974. 17. 355.
FYSaGLQGlSKA INSTITUTIONEW I IiAROLIMSXA INSTITUTET .-.- .- STOCKHOLM
