Respiration Physiology, 87 (1992) 105-114
105
© 1992 Elsevier Science Publishers B.V. All rights reserved. 0034-5687/92/$03.50 RESP 01859
Hypoxic contraction of pre-stretched human pulmonary artery Masatoshi Ohe, Masahiko Ogata, Dai Katayose and Tamotsu Takishima 7he First Department of Internal Medicine. Tohoku University School of Medicine. Sendal. Japan (Accepted 18 September 1991) Abstract. To clarify the mechanism ofhypoxic pulmonary vasoconstriction in man, human pulmonary artery segments (2 mm O.D.) were suspended and changes in isometric force were measured. The arteries were contracted by hypoxia (Pc), 43 + 2 Tort) developing a tension of 127 + 36 mg over the course of 15 rain. This contraction was completely blocked by 10 -6 M L-isoproterenol, 10-6M nitroglycerin, partially blocked by 10- ~-I0- 6 M verapamil, unchanged by 10- 6 M phentolamine, ! 0- 6 M L-propranolol, 10- 6 M diphenhydramine, 10- 6 M guanethidine, 10- 7 M FPL 55712 and enhanced by 10- 6 M BAY K 8644, 10- 3 M procaine, 3 × 10- 6 M quinacrine, 10 - 6 M indomethacin or 10- 6 M methylene blue. Removal of the endothelium significantly enhanced the magnitude ofhypoxia.inducedcontraction. These results suggest that the human pulmonary artery constricts in response to hypoxia, at least in part, through activation of the voltage-dependent Ca 2 ÷ channels and that neither ~,/t, Hz receptors, the lipoxygenase pathway nor neural reflexes are involved. They also show that the endothelium is not required for hypoxic contraction and that its presence reduces sensitivity to hypoxia.
Endothelium, hypoxic pulmonary vasoconstriction; Hypoxic pulmonary vasoconstriction, dependenceon voltage-gated Ca 2 + channels; Ion channels, Ca 2 +, hypoxic pulmonary vasoconstriction; Mammals, humans
Unlike systemic arteries, the pulmonary artery constricts in response to hypoxia, an event which is known as hypoxic pulmonary vasoconstriction (HPV) (Euler and Liljestrand, 1946). However, it has been reported that the isolated pulmonary artery is usually insensitive to hypoxia and constricts only under specific conditions. For example, Lloyd (1968) observed hypoxic contraction of isolated rabbit lobar pulmonary artery only when a layer of lung remained. Detar and Gellai (1971) found that rabbit lobar pulmonary artery contracted in response to hypoxia after prolonged exposure to hypoxia. Moreover, Souhrada and Dickey (1976) reported that the isolated lobar pulmonary artery of rabbit contracts in response to hypoxia if glucose is absent in the perfusate. However, it has recently been reported that pulmonary arteries with an outer diameter Correspondence address: T. Takishima, The First Department of' Internal Medicine, Tohoku University School of Medicine, Seiryo-machi !-1, Aoba-ku, Sendal 980, Japan.
106
M. OHE et al.
of 300 #m isolated from cats show depolarization and constriction with hypoxia (Po, 50 Tort) in a phys.iological solution and that this phenomenon is rarely seen in arteries of more than 500 #m in outer diameter (Madden et aL, 1985). Moreover, Rodman et al. (1989) reported that even the main extrapulmonary artery without pre-constriction contracts in response to hypoxia. This contraction is more reproducible in rings preconstricted with either phenylephrine, norepinephrine, KCi, angiotensin II or the thromboxane A 2 mimetic u 46619. It is generally accepted that pulmonary vessel strips frequently require an 'initial tone' with some agonist before hypoxic vasoconstriction is found. It has recently been reported that isolated human pulmonary artery constricts with hypoxia under histamine-mediated contraction (Hoshino et al., 1988). Histamine was used as an agonist, since hypoxia alone did not alter tension in pulmonary arterial strips and since histamine had a more potent contractile effect on the isolated human pulmonary artery than did serotonin or prostaglandin F2~. However, no studies on the hypoxic contraction of isolated human pulmonary artery without pre-constriction have been reported thus far. The purpose of this study was to investigate whether human pulmonary artery without pre-constriction contracts in response to hypoxia and, if so, whether the contraction is similar to the known characteristics of the hypoxic presser response seen in isolated pulmonary arteries of other species.
Methods
Preparation of human intrapulmonary arteo, rings. Lung segments were obtained with the donor's consent from 38 patients who underwent surgery for removal of a malignant tumor or lung abscess. The mean age of the patients was 60 (range 26-79 years). None of the patients exhibited clinical evidence of pulmonary hypertension. Immediately after removal, the tissue was placed in a cold preoxygenated salt solution for transport to the laboratory. The salt solution contained the following (in raM): Tris-HCl (pH 7.4), 23.8; NaCI, 125; KCI, 2.7; CaCI,, 1.8, glucose. I I.0. intrapulmonary arterial branches that were located far from tumors or other pathologically altered tissues, were rapidly isolated, and gently cleaned of lung parenchyma, fat and adhering connective tissue under a dissection microscope. Care was taken to avoid stretching. Ring segments 2 mm in outer diameter and 3 mm in length were cut from these vessels. Experiments were started within 5 h of lung removal. Recording ¢~fmuscle telrsion. Human pulmonary artery prepared in this manner was suspended vertically between a rigid support and the arm of a strain gauge by slipping it over two pins, and placed in a chamber (50 ml) filled with modified Krebs-Ringer bicarbonate solution (37 + 0.5 °C). The solution contained the following (in raM): NaCi, 118.0; KCI, 4.7; KH2PO4, 1.2; MgSO4"7H20, 1.2; NaHCO3, 25.0; CaCl:. 2H20, 2.5; glucose, 5.5. The upper stainless steel-wire supporting the ring was attached to a force-displacement transducer (UL- 10-120, Shinko Kohgyo). We selected
HYPOXIC CONTRACTION OF THE HUMAN PULMONARY ARTERY
107
a resting tension of 2-3 g on the basis of preliminary experiments showing that rings equilibrated in this manner responded to 40 mM KCI with maximal active tension development. A gas mixture containing 15% O2, 80% N2 and 5~ CO2 or 95~o Ne and 5~o CO2 was allowed to bubble vigorously at a constant flow rate through a ball f'dter located at the bottom of the chamber. With the mixture containing 15% oxygen, the Po2 of the bath solution was 104 + 3 (SE) Torr (normoxia) and with that containing no oxygen, Po, was 43 _+2 Torr (hypoxia). The resulting pH and Poe2 of the experimental solution were the same for either gas mixture (pH 7.44 + 0.0022, Pco, 34 _+ 1 Torr). The muscle chamber was filled with the solution from the bottom of the chamber and washing was accomplished by allowing the solution to drain from the top of the chamber. Periodically, throughout the experiment, Po.~,pH and Poe., were monitored with a gas analyzer (Model 813, Instrumentation Laboratory Inc. Lexington, MA).
Experimentaiprotocol. Rings were equilibrated for 120 min or until a stable baseline passive tension of 2-3 g was maintained at which time the chamber was washed out with fresh solution. No arteries showed spontaneous development of force during equilibration. Tissue viability was assessed by noting stable and consistent increases in isometric force after depolarizations with 40 mM KCI which was subsequently washed off; the tissue was allowed to equilibrate for 60 min. After re-equilibration, three consecutive 15 min hypoxic challenges were induced and each acute hypoxic challenge was followed by a 15 rain normoxic recovery period. Drugs were first added at 30 min intervals just after the second challenge, and one additional challenge was made except in the case of verapamil, for which, in order to obtain the cumulative concentrationresponse curves, three additional chaU0nges were made. There were no significant changes in the pH, Po.~ or Pco.~ of the solution after addition of these drugs. in experiments designed to test the effect of endothelium removal on the hypoxic response, rings were taken from the chamber after the second hypoxic challenge, the entire intimal surfaces of which were rubbed for 30-40 sec, using a short length of stainless-steel tubing, without stretching the vessel walls. After re.equilibration for 90 min, the destruction of endothelial cells was verified by means of a pharmacologic criterion; the absence of relaxation in response to 10-s-10-6 M acetylcholine (ACh) after precontraction with 10 -s M phenylephrine. Additional hypoxic challenge was then attempted.
Drugs and chemicals. The following pharmacological agents were used: verapamil (Eisai); EGTA; BAY K 8644 (B ayer); procaine; phentolamine (Ciba); L-isoproterenol; L-propranolol (ICI); diphenhydramine; guanethidine (Ciba); nitroglycerin; indomethacin; sodium 7-{3.(4-acetyl-3-hydroxy-2-propylphenoxy)-2-hydroxypropoxy~-4oxo-8-propyl-4H- l-benzopyran-2-carboxylate (FPL 55712, Fisons); quinacrine (Sigma); methylene blue (Sigma) and L-NG-monomethyl arginine (L-NMMA). L-NMMA was a gift from Dr. Sakuma (Hokkaido University, Sapporo, Japan). All drugs were prepared in distilled water, except for BAY K 8644 and indomethacin,
108
M. OHE el al.
which were dissolved in dimethyl suifoxide (DMSO) and ethanol, respectively. Drugs were added to the muscle chamber in aliquots of up to 500 #1. Concentrations are expressed as final concentrations in the bath solution. Because the number of samples was limited, we could not fully evaluate dose-response curves to determine the appropriate doses of these drugs. However, we did establish the drug doses in this experiment after confirming their effectiveness on agonist-induced contractions.
Calculations and statistical analysis. Muscle contraction was measured as the increase in tension above the baseline. All values are expressed as the means + SE. n represents the number of rings tested. Comparisons were made using Student's t-test. The level of statistically significant difference was taken to be P < 0.05.
Results
Control hypoxic challenge. As the Po.. in the bath solution was decreased, the resting tension increased and the mean value of the tension after 15 rain of hypoxia was 127 + 36 mg (n = 12). Maximal contraction was usually sustained when hypoxia lasted for more than 15 rain. Reoxygenation caused an abrupt relaxation and the tension declined to the original baseline values. Repeated hypoxic challenges did not alter the amplitude of the response to as many as five challenges.
Drug.pretreatmem hypoxic challenge. To determine the possible role of a voltagedependent Ca 2 * influx in the hypoxic contraction of human pulmonary artery, we examined the effects of the Ca 2 + antagonist, verapamil and a voltage-dependent Ca 2 + agonist, BAY K 8644 and DMSO (vehicle control of BAY K 8644) on the hypoxic response. Verapamil at 10-s-10 =~' M did not change baseline tone in normoxia, Compared with the preceding control value, cumulative administration of 10°s-10 °° M verapamil resulted in decreased hypoxic contraction (Fig. !, n = 5), Similarly, Ca a + -free solution containing I mM EGTA decreased hypoxic contraction by 14 + 7% (n = 3). BAY K 8644 at 10 -6 M slightly increased has©line tone in normoxia and markedly enhanced hypoxic contraction by 396 _+ 106% (P < 0.05, n = 7). However, the hypoxic response was unchanged in the vessel treated with 3 x 10-4 M DMSO (n = 4). Procaine at 10 - ~ M, a dose which markedly decreased the norepinephrine-induced ( 10 - s M) elevated tension of the artery, slightly increased the baseline tone in normoxia and enhanced the hypoxic response by 295 + 45% (P < 0.05, n = 5). Table i summarizes the effects of the drugs (in each case at 10- ~' M concentration) guanethidine, diphenhydramine, phentolamine, L-propranolol, L-isoproterenol and nitroglycerin on the hypoxic response. Only the latter two drugs altered resting tension or hypoxic contraction. Figure 2 shows a typical example of the effect of 10 -6 M L-propranolol on hypoxic contraction of a ring. However, L-isoproterenol and nitroglycerin (both at 10 - o M) completely inhibited the hypoxic contraction. Figure 3 illustrates a typical example of the effect of 10- 6 M nitroglycerin on the hypoxic response.
HYPOXIC CONTRACTION OF THE HUMAN PULMONARY ARTERY
109
(% Control ) 10(]
m
4O 2(]
Control # 11)~ 1~' 104 Verapamil ooneentration ( M ) Fig. !. Effect of increasing concentrations of verapamil on the hypoxic response. Verapamil was added to muscle chambers cumulatively for 30 min prior to the next hypoxic response. Verapamil at 10- ~-I0- 6 M partially inhibited hypoxic contraction compared with preceding control values. Each point represents the mean + SE. n = 5; * P < 0.05; ** P < 0.01.
TABLE ! Effects of drugs on the hypoxic response.
I0 ~ ~ M I0 ° * M I0 ° ~ M I 0 ~' M IO ~ M
guanethidine diphenhydraminc phentolamine l.-propranolol L-isoprotercnol
10- f' M nitroglycerin
(5) (5) (6) (8) (9)
(9)
Hypoxic response (% of control)
P value
104 -4- 8 117+9 97+4 106 + 5 NR NR
NS NS NS NS
Values are shown as means + SE, Numbers in parentheses denote the number of rings tested, NS, not significant; NR, no measurable response.
PO:
150I'1
(To-, oLt V-I
I
t I
t
L-Propranolol 10+M
,
.
15 rain
Fig. 2. Typical example of the effect of L-propranolol on the hypoxic response. L-.Propranolol at 10- 6 M did not alter resting tension or hypoxic contraction.
l lO
M. OHE et al.
PO:
(Torr)
'"[1 I
I
I
I
I
OL
! Nitroglycerin 10"M
.°
iS rain
Fig. 3. Typical example of the effect of nitroglycerin on the hypoxie response. Nitroglycerin at 10-6 M did not change baseline tone in normoxia, but completely inhibited hypoxic contraction.
A summary of the effects of arachidonic acid metabolite blockers on the hypoxic contraction of rings is given in Table 2. Indomethacin at 10- 6 M, an inhibitor of the cyclooxygenase pathway, slightly increased baseline tone in normoxia and augmented the hypoxic response to 185 + 37~o of its preceding values (P < 0.05). However, the hypoxic response remained unchanged in the ethanol-treated vessels. FPL 55712 at 10- 7 M, a competitive antagonist of slow-reacting substance of anaphylaxis (Augstein et eL, 1973), affected neither the baseline tone nor the hypoxic response. Quinacrine at 3 x 10- 6 M, a phospholipase A2 inhibitor, increased baseline tone slightly in normoxia and markedly enhanced the hypoxic response. Methylene blue at 10-¢'M, an inhibitor of soluble guanylate cyclase, increased baseline tone slightly in normoxia and markedly enhanced the hypoxic response by 386 _+ 110~, of its preceding values (P < 0.05, n -- 5). L-NMMA at 10 =4 M increased baseline tone in normoxia and enhanced the hypoxie response by 262 + 48% of its preceding values (n = 3). To examine the role of the endothelium in hypoxic contraction of isolated human pulmonary artery, the hypoxic response was compared in the same arterial ring before and after the endothelium was rubbed. Figure 4 summarizes the effect of rubbing the endothelium on the hypoxic response. Hypoxic contraction after such rubbing was enhanced as compared with the values determined beforehand (P < 0.01, n = 6).
TABLE 2 Effects of arachidonic acid metabolite blockers on the hypoxic response.
10- ~' M indomethacin (8) 10 7 M FPL 55712 (4) 3 x I 0 * M quinacrine (6)
Hypoxic response (% of control)
P value
185 + 37 97 _+ 10 274 + 16
P < 0.05 NS P < 0.05
Values are given as means + SE, Numbers in parentheses represent the number of rings tested, NS, not significant,
HYPOXIC CONTRACTION OF THE HUMAN PULMONARY ARTERY
11 1
3OO
loo
-I-
0
E (÷)
E (-)
Fig. 4. Summary of the effect on hypoxic contraction of isolated human pulmonary artery caused by rubbing the endothelium. Data are means + SE. E( + ), endothelium not rubbed; E( - ), endothelium rubbed; n ffi 6; * P < 0.01.
Discussion
Pulmonary vascular smooth muscle frequently requires an initial tone produced by an agonist, such as a prostaglandin or histamine, before hypoxic vasoconstriction is observed. If initial tone is blocked with a specific receptor blocker, the HPV is also lost. However, the hypoxic contractile response produced in this study was elicited by the direct action of hypoxia alone and required no pre-constriction. We have termed thi~ contraction 'pre-stretched contraction', in the present study, the reason why isolated human pulmonary artery, even in the case of an outer diameter of 2 mm, constricted in response to hypoxia without pre-constriction is unclear. However, when an experiment was begun more than 10 h after lung removal and the handling of the pulmonary vascular smooth muscle was inappropriate, the response to hypoxia was also lost. This preparation is the first to demonstrate hypoxic contraction of human pulmonary artery without pre-constriction in vitro and has the advantage of allowing observation of the direct effect of hypoxia on human pulmonary vascular smooth muscle. Currently, transmembrane Ca 2 ÷ influx through the voltage-dependent Ca 2 ÷ channel is thought to be the most important factor in control of HPV (McMurtry, 1985). In the present study, verapamil had no effect on the normoxic force and decreased the maximum hypoxic response, whereas BAY K 8644 increased normoxic force and the subsequent maximum hypoxic response. These results show that voltage-dependent Ca 2 + influx also plays an important role in the hypoxic contraction of pulmonary artery in man. However, because the hypoxic contraction was still present in the absence of extracellular Ca 2 ÷, we suppose that hypoxic contraction is not dependent only on extracellular Ca 2 + as reported in hypoxic histamine-mediated contraction in man
112
M, OHE etal.
(Hoshino et al., 1988). Moreover, the data obtained with phentolamine, L-propranolol, diphenhydramine, guanethidine, indomethacin and FPL 55712 also show that neither the 0g ]/, H~ receptors, lipoxygenase pathway nor neural reflexes are involved in this response, and that vasodilatory prostaglandins prevent the hypoxic pulmonary response in man. There are several differences between the hypoxic response in our study and that recently observed in a human histamine-HPV model (Hoshino et al., 1988): For example, (1)The hypoxic response observed in the current work was not inhibited by diphenhydramine or procaine. In contrast, hypoxic contraction was enhanced by 10-3 M procaine. (2)With reoxygenation, tension returned to the original baseline values which was not the case in hypoxic histamine-mediated contraction. These discrepancies may be due to differences between the unstimulated and H,-receptor stimulated conditions as described above. That procaine at 5 x I0-3 M inhibited the hypoxic histamine-mediated contraction may derive from the tacts that histamine partially promotes Ca 2 + release from intraceUular storage sites (William, 1985) and that procaine acts to inhibit Ca2 + release. In our study, procaine at 10-3 M enhanced the hypoxic pulmonary response. This may be due to some other actions of procaine, such as blockage of K + channels (Kurihara, 1973). A similar result has been reported previously using intrapulmonary rabbit artery (Lloyd, 1966). The hypoxia used in our study was mild with a Po, of 43 Torr. Therefore, no significant differences exist between the Po: employed in this study and that in mixed venous blood. However, the step changes in Po: in cat small pulmonary arteries produced stimulus-response curves that decreased in a single step from 400 to 30 Tort and a Po: of 20 Torr developed a peak tension equal to that seen at 50 Torr, although this tension was transient and dropped rapidly from the peak (Madden et ai., 1985). Similarly, Sylvester et al. (1980) reported that the maximum response in isolated pig lungs occurred at Po~ between 30 and 60 Torr and diminished at Po: below 30 Ton. Therefore, the Po~,of 43 Torr used in the present study was considered as appropriate for hypoxic stimulation of the isolated pulmonary artery. Hypoxic contraction of isolated human pulmonary artery was enhanced by removal of endothelial cells. Moreover, this contraction was also enhanced by administration of 3 × 10 - 6 M quinacrine, 10- ~ M methylene blue or 10- 4 M L-NMMA. These doses of quinacrine and methylene blue have been reported to inhibit the action of endothelium-dependent relaxing factor (EDRF) on ACh in human pulmonary artery (Greenberg etal., 1987). The effects of methylene blue include alteration in smooth muscle cell redox state, damage to the endothelium through superoxide production or change in catecholamine levels. The effects of quinacrine include inhibition of the modulatory effect of platelet-activating factor (Otamiri and Tagesson, 1989) as well as inhibition of phospholipase A2. However, the major effect of methylene blue and quinacrine is to block guanylate cyclase, and EDRF relaxes smooth muscle by activating guanylate cyclase (Ignarro, 1989). Hence, blocking guanylate cyclase with methylene blue and quinacrine would be expected to block the action of EDRF. Nitric oxide is a major component of EDRF. L-NMMA at 10- 4 M, an inhibitor of the synthesis of
HYPOXIC CONTRACTION OF THE HUMAN PULMONARY ARTERY
113
nitric oxide from L-arginine (Sakuma et aL, 1988), enhanced the hypoxic response. Moreover, removal of endothelium enhances the hypoxic vasoconstrictor response. These results are consistent with the observations of Robertson et al. (1990) in which hypoxic pulmonary vasoconstriction was more than doubled by blocking nitric oxide production with L-NMMA in isolated perfused rat lung. It has been reported that the in vitro antagonists of EDRF, eicosatetraynoic acid, nordihydroguaiaretic acid and hydroquinone (Brashers etai., 1988) and methylene blue (Mazmanian etal., 1989), significantly augment the hypoxic pressor response in the isolated perfused rat lung preparation. However, these observations conflict with those of other studies in which hypoxic contraction in porcine main pulmonary artery was reported to be completely endothelium dependent (Holden and McCall, 1984) and in which chemical inhibition of EDRF by methylene blue or hemoglobin was found to reduce the hypoxic contraction of isolated rat pulmonary artery (Rodman et al., 1987). The reason for this difference is unclear. However, apart from species differences, the latter studies were performed using ring preparations of the main pulmonary artery, where reactivity of the vascular bed may differ from that of the isolated perfused lung preparation and intrapuimonary artery used in our study. In summary, the pre-stretched human pulmonary artery constricted in response to hypoxia, at least in part, through activation of the voltage-dependent Ca 2 + channel, a, //, H~ receptors, lipoxygenase pathways and neural reflexes are not involved in this response. In addition, our results suggest that the endothelium does not contribute at all to hypoxic contraction of human pulmonary artery.
Acknowledgements.We are grateful to Hoken Kagaku Institution for their support and to R, Scott for his critical reading of tlte manuscript. A part of this work was presented at the 60th Scientific Session of the American Heart Association in Anaheim, CA, 15-17 November 1987.
References Augstein, J., J.B. Farmer, T.B. Lee, P. Sheard and M,L. Tattersall (1973). Selective inhibitor of slow reacting substance of anaphylaxis. NatI~re New Biol, 245: 215-217, Brashers, V. L., M.J. Peach and C. E. Rose, Jr. (1988). Augmentation of hypoxicpulmonary vasoconstriction in the isolated perfused rat lung in vitro antagonists ofendothelium-dependent relaxation. J. Clin, Invest. 82: 1495-1502. Detar, R. and M. Gellai (1971). Oxygen and isolated vascular smooth muscle from the main pulmonary artery of the rabbit. Am. J. Physiol. 221: 1791-1794. Greenberg, B., K. Rhoden and P.J. Barnes (1987). Endothelium-dependent relaxation of human pulmonary arteries. Am. J. Physiol. 252: H434-H438. Holden, W.E. and E. McCall (1984). Hypoxia-induced contractions of porcine pulmonary artery strips depend on intact endothelium. Exp. Lung Res. 7: I01-I 12. Hoshino, Y., H. Obara, M. Kusunoki, Y. Fujii and S. lwai (1988). Hypoxic contractile response in isolated human pulmonary artery: role of calcium ion. J. Jppl. Physiol. 65: 2468-2474. Ignarro, L.J. (1989), Endothelium-derived nitric oxide: actions and properties. FASEB J. 3: 31-36. Kurihara, S, (1973). The effect of procaine on the urinary bladder smooth muscle of bullfrogs.Jap. J. Physiol. 23: 309-324.
114
M. OHE et al.
Lloyd, T. C., Jr. (! 966). Po,-dependent pulmonary vasoconstriction caused by procaine. J. Appl. Physiol. 21: 1439-1442. Lloyd, T.C., Jr. (1968). Hypoxic pulmonary vasoconstriction: role of perivascular tissue. J. Appl. Physiol. 25: 560-565. Madden, J.A., C.A. Dawson and D.R. Harder (1985). Hypoxia-induced activation in small isolated pulmonary arteries from the cat. 3. Appl. Physiol. 59: !13-118. Mazmanian, G.M., B. Baudet, C. Brink, J. Cerrina, S. Kirkiacharian and M. Weiss (I~89). Methylene blue potentiates vascular reactivity in isolated rat lungs. J. Appl. Physiol. 66: 1040-1045. McMurtry, !. F. (! 985). BAY K 8644 potentiates and A23187 inhibits hypoxic vasoconstriction in rat lungs. Am. J. Physiol. 249: H741-H746. Otamiri, T. and C. Tagesson (1989). Role of phospholipase Ae and oxygenated free radicals in mucosal damage after small intestinal ischemia and reperfusion. Am. J. Surg. 157: 562-566. Robertson, B.E., J.B. Warren and P.C.G. Nye (1990). Inhibition of nitric oxide synthesis potentiates hypoxic vasoconstriction in isolated rat lungs. Exp. Physiol. 75: 255-25?. Rodman, D.M., T. Yamaguchi, R.F. O'Brien and I.F. McMurtry (1987). Hypoxic contraction of isolated pulmonary artery is reduced by endothelial denudation, hemoglobin, and methylene blue (Abstract). Am. Ray. Respir. Dis. 135: A131. Rodman, D.M., T. Yamaguchi, R.F. O'Brien and I.F. McMurtry (1989). Hypoxic contraction of isolated rat pulmonary artery. J. Pharmacol. Exp. That. 248: 952-959. Sakuma, l., D. 3. Stueher, S. S. Gross, C. Nathan and R. Levi (I 988) Identification of arglnine as a precursor of endothelium-derived relaxing factor. Proc. Natl. Acad. $ci. USA 85: 8664-8667. Souhrada, J. F. and D.W. Dickey (1976). Effectof substrata on hypoxic response of small pulmonary artery. J. Appl. Physiol. 40: 533-538. Sylvester, J.T., A.L. Harabin, M.D. Peake and R.S. Frank (1980). Vasodilator and constrictor responses to hypoxia in isolated pig lungs. J. Appl. Physiol. 49: 820-825. Von Euler, U.S. and G. Liljestrand (1946). Observations on the pulmonary arterial blood pressure in the cat, Acts Physlol, $cand, 12: 301-320. William, W, D. (1985). Histamine and 5-hydroxytryptamine (serotonin) and their antagonists. In: The Pharmacological Basis of Therapeutics; edited by A.G. Gilman, L.S. Goodman, T.W, Rail and F. Murad. New York: Macmillan, p. 610,
105
© 1992 Elsevier Science Publishers B.V. All rights reserved. 0034-5687/92/$03.50 RESP 01859
Hypoxic contraction of pre-stretched human pulmonary artery Masatoshi Ohe, Masahiko Ogata, Dai Katayose and Tamotsu Takishima 7he First Department of Internal Medicine. Tohoku University School of Medicine. Sendal. Japan (Accepted 18 September 1991) Abstract. To clarify the mechanism ofhypoxic pulmonary vasoconstriction in man, human pulmonary artery segments (2 mm O.D.) were suspended and changes in isometric force were measured. The arteries were contracted by hypoxia (Pc), 43 + 2 Tort) developing a tension of 127 + 36 mg over the course of 15 rain. This contraction was completely blocked by 10 -6 M L-isoproterenol, 10-6M nitroglycerin, partially blocked by 10- ~-I0- 6 M verapamil, unchanged by 10- 6 M phentolamine, ! 0- 6 M L-propranolol, 10- 6 M diphenhydramine, 10- 6 M guanethidine, 10- 7 M FPL 55712 and enhanced by 10- 6 M BAY K 8644, 10- 3 M procaine, 3 × 10- 6 M quinacrine, 10 - 6 M indomethacin or 10- 6 M methylene blue. Removal of the endothelium significantly enhanced the magnitude ofhypoxia.inducedcontraction. These results suggest that the human pulmonary artery constricts in response to hypoxia, at least in part, through activation of the voltage-dependent Ca 2 ÷ channels and that neither ~,/t, Hz receptors, the lipoxygenase pathway nor neural reflexes are involved. They also show that the endothelium is not required for hypoxic contraction and that its presence reduces sensitivity to hypoxia.
Endothelium, hypoxic pulmonary vasoconstriction; Hypoxic pulmonary vasoconstriction, dependenceon voltage-gated Ca 2 + channels; Ion channels, Ca 2 +, hypoxic pulmonary vasoconstriction; Mammals, humans
Unlike systemic arteries, the pulmonary artery constricts in response to hypoxia, an event which is known as hypoxic pulmonary vasoconstriction (HPV) (Euler and Liljestrand, 1946). However, it has been reported that the isolated pulmonary artery is usually insensitive to hypoxia and constricts only under specific conditions. For example, Lloyd (1968) observed hypoxic contraction of isolated rabbit lobar pulmonary artery only when a layer of lung remained. Detar and Gellai (1971) found that rabbit lobar pulmonary artery contracted in response to hypoxia after prolonged exposure to hypoxia. Moreover, Souhrada and Dickey (1976) reported that the isolated lobar pulmonary artery of rabbit contracts in response to hypoxia if glucose is absent in the perfusate. However, it has recently been reported that pulmonary arteries with an outer diameter Correspondence address: T. Takishima, The First Department of' Internal Medicine, Tohoku University School of Medicine, Seiryo-machi !-1, Aoba-ku, Sendal 980, Japan.
106
M. OHE et al.
of 300 #m isolated from cats show depolarization and constriction with hypoxia (Po, 50 Tort) in a phys.iological solution and that this phenomenon is rarely seen in arteries of more than 500 #m in outer diameter (Madden et aL, 1985). Moreover, Rodman et al. (1989) reported that even the main extrapulmonary artery without pre-constriction contracts in response to hypoxia. This contraction is more reproducible in rings preconstricted with either phenylephrine, norepinephrine, KCi, angiotensin II or the thromboxane A 2 mimetic u 46619. It is generally accepted that pulmonary vessel strips frequently require an 'initial tone' with some agonist before hypoxic vasoconstriction is found. It has recently been reported that isolated human pulmonary artery constricts with hypoxia under histamine-mediated contraction (Hoshino et al., 1988). Histamine was used as an agonist, since hypoxia alone did not alter tension in pulmonary arterial strips and since histamine had a more potent contractile effect on the isolated human pulmonary artery than did serotonin or prostaglandin F2~. However, no studies on the hypoxic contraction of isolated human pulmonary artery without pre-constriction have been reported thus far. The purpose of this study was to investigate whether human pulmonary artery without pre-constriction contracts in response to hypoxia and, if so, whether the contraction is similar to the known characteristics of the hypoxic presser response seen in isolated pulmonary arteries of other species.
Methods
Preparation of human intrapulmonary arteo, rings. Lung segments were obtained with the donor's consent from 38 patients who underwent surgery for removal of a malignant tumor or lung abscess. The mean age of the patients was 60 (range 26-79 years). None of the patients exhibited clinical evidence of pulmonary hypertension. Immediately after removal, the tissue was placed in a cold preoxygenated salt solution for transport to the laboratory. The salt solution contained the following (in raM): Tris-HCl (pH 7.4), 23.8; NaCI, 125; KCI, 2.7; CaCI,, 1.8, glucose. I I.0. intrapulmonary arterial branches that were located far from tumors or other pathologically altered tissues, were rapidly isolated, and gently cleaned of lung parenchyma, fat and adhering connective tissue under a dissection microscope. Care was taken to avoid stretching. Ring segments 2 mm in outer diameter and 3 mm in length were cut from these vessels. Experiments were started within 5 h of lung removal. Recording ¢~fmuscle telrsion. Human pulmonary artery prepared in this manner was suspended vertically between a rigid support and the arm of a strain gauge by slipping it over two pins, and placed in a chamber (50 ml) filled with modified Krebs-Ringer bicarbonate solution (37 + 0.5 °C). The solution contained the following (in raM): NaCi, 118.0; KCI, 4.7; KH2PO4, 1.2; MgSO4"7H20, 1.2; NaHCO3, 25.0; CaCl:. 2H20, 2.5; glucose, 5.5. The upper stainless steel-wire supporting the ring was attached to a force-displacement transducer (UL- 10-120, Shinko Kohgyo). We selected
HYPOXIC CONTRACTION OF THE HUMAN PULMONARY ARTERY
107
a resting tension of 2-3 g on the basis of preliminary experiments showing that rings equilibrated in this manner responded to 40 mM KCI with maximal active tension development. A gas mixture containing 15% O2, 80% N2 and 5~ CO2 or 95~o Ne and 5~o CO2 was allowed to bubble vigorously at a constant flow rate through a ball f'dter located at the bottom of the chamber. With the mixture containing 15% oxygen, the Po2 of the bath solution was 104 + 3 (SE) Torr (normoxia) and with that containing no oxygen, Po, was 43 _+2 Torr (hypoxia). The resulting pH and Poe2 of the experimental solution were the same for either gas mixture (pH 7.44 + 0.0022, Pco, 34 _+ 1 Torr). The muscle chamber was filled with the solution from the bottom of the chamber and washing was accomplished by allowing the solution to drain from the top of the chamber. Periodically, throughout the experiment, Po.~,pH and Poe., were monitored with a gas analyzer (Model 813, Instrumentation Laboratory Inc. Lexington, MA).
Experimentaiprotocol. Rings were equilibrated for 120 min or until a stable baseline passive tension of 2-3 g was maintained at which time the chamber was washed out with fresh solution. No arteries showed spontaneous development of force during equilibration. Tissue viability was assessed by noting stable and consistent increases in isometric force after depolarizations with 40 mM KCI which was subsequently washed off; the tissue was allowed to equilibrate for 60 min. After re-equilibration, three consecutive 15 min hypoxic challenges were induced and each acute hypoxic challenge was followed by a 15 rain normoxic recovery period. Drugs were first added at 30 min intervals just after the second challenge, and one additional challenge was made except in the case of verapamil, for which, in order to obtain the cumulative concentrationresponse curves, three additional chaU0nges were made. There were no significant changes in the pH, Po.~ or Pco.~ of the solution after addition of these drugs. in experiments designed to test the effect of endothelium removal on the hypoxic response, rings were taken from the chamber after the second hypoxic challenge, the entire intimal surfaces of which were rubbed for 30-40 sec, using a short length of stainless-steel tubing, without stretching the vessel walls. After re.equilibration for 90 min, the destruction of endothelial cells was verified by means of a pharmacologic criterion; the absence of relaxation in response to 10-s-10-6 M acetylcholine (ACh) after precontraction with 10 -s M phenylephrine. Additional hypoxic challenge was then attempted.
Drugs and chemicals. The following pharmacological agents were used: verapamil (Eisai); EGTA; BAY K 8644 (B ayer); procaine; phentolamine (Ciba); L-isoproterenol; L-propranolol (ICI); diphenhydramine; guanethidine (Ciba); nitroglycerin; indomethacin; sodium 7-{3.(4-acetyl-3-hydroxy-2-propylphenoxy)-2-hydroxypropoxy~-4oxo-8-propyl-4H- l-benzopyran-2-carboxylate (FPL 55712, Fisons); quinacrine (Sigma); methylene blue (Sigma) and L-NG-monomethyl arginine (L-NMMA). L-NMMA was a gift from Dr. Sakuma (Hokkaido University, Sapporo, Japan). All drugs were prepared in distilled water, except for BAY K 8644 and indomethacin,
108
M. OHE el al.
which were dissolved in dimethyl suifoxide (DMSO) and ethanol, respectively. Drugs were added to the muscle chamber in aliquots of up to 500 #1. Concentrations are expressed as final concentrations in the bath solution. Because the number of samples was limited, we could not fully evaluate dose-response curves to determine the appropriate doses of these drugs. However, we did establish the drug doses in this experiment after confirming their effectiveness on agonist-induced contractions.
Calculations and statistical analysis. Muscle contraction was measured as the increase in tension above the baseline. All values are expressed as the means + SE. n represents the number of rings tested. Comparisons were made using Student's t-test. The level of statistically significant difference was taken to be P < 0.05.
Results
Control hypoxic challenge. As the Po.. in the bath solution was decreased, the resting tension increased and the mean value of the tension after 15 rain of hypoxia was 127 + 36 mg (n = 12). Maximal contraction was usually sustained when hypoxia lasted for more than 15 rain. Reoxygenation caused an abrupt relaxation and the tension declined to the original baseline values. Repeated hypoxic challenges did not alter the amplitude of the response to as many as five challenges.
Drug.pretreatmem hypoxic challenge. To determine the possible role of a voltagedependent Ca 2 * influx in the hypoxic contraction of human pulmonary artery, we examined the effects of the Ca 2 + antagonist, verapamil and a voltage-dependent Ca 2 + agonist, BAY K 8644 and DMSO (vehicle control of BAY K 8644) on the hypoxic response. Verapamil at 10-s-10 =~' M did not change baseline tone in normoxia, Compared with the preceding control value, cumulative administration of 10°s-10 °° M verapamil resulted in decreased hypoxic contraction (Fig. !, n = 5), Similarly, Ca a + -free solution containing I mM EGTA decreased hypoxic contraction by 14 + 7% (n = 3). BAY K 8644 at 10 -6 M slightly increased has©line tone in normoxia and markedly enhanced hypoxic contraction by 396 _+ 106% (P < 0.05, n = 7). However, the hypoxic response was unchanged in the vessel treated with 3 x 10-4 M DMSO (n = 4). Procaine at 10 - ~ M, a dose which markedly decreased the norepinephrine-induced ( 10 - s M) elevated tension of the artery, slightly increased the baseline tone in normoxia and enhanced the hypoxic response by 295 + 45% (P < 0.05, n = 5). Table i summarizes the effects of the drugs (in each case at 10- ~' M concentration) guanethidine, diphenhydramine, phentolamine, L-propranolol, L-isoproterenol and nitroglycerin on the hypoxic response. Only the latter two drugs altered resting tension or hypoxic contraction. Figure 2 shows a typical example of the effect of 10 -6 M L-propranolol on hypoxic contraction of a ring. However, L-isoproterenol and nitroglycerin (both at 10 - o M) completely inhibited the hypoxic contraction. Figure 3 illustrates a typical example of the effect of 10- 6 M nitroglycerin on the hypoxic response.
HYPOXIC CONTRACTION OF THE HUMAN PULMONARY ARTERY
109
(% Control ) 10(]
m
4O 2(]
Control # 11)~ 1~' 104 Verapamil ooneentration ( M ) Fig. !. Effect of increasing concentrations of verapamil on the hypoxic response. Verapamil was added to muscle chambers cumulatively for 30 min prior to the next hypoxic response. Verapamil at 10- ~-I0- 6 M partially inhibited hypoxic contraction compared with preceding control values. Each point represents the mean + SE. n = 5; * P < 0.05; ** P < 0.01.
TABLE ! Effects of drugs on the hypoxic response.
I0 ~ ~ M I0 ° * M I0 ° ~ M I 0 ~' M IO ~ M
guanethidine diphenhydraminc phentolamine l.-propranolol L-isoprotercnol
10- f' M nitroglycerin
(5) (5) (6) (8) (9)
(9)
Hypoxic response (% of control)
P value
104 -4- 8 117+9 97+4 106 + 5 NR NR
NS NS NS NS
Values are shown as means + SE, Numbers in parentheses denote the number of rings tested, NS, not significant; NR, no measurable response.
PO:
150I'1
(To-, oLt V-I
I
t I
t
L-Propranolol 10+M
,
.
15 rain
Fig. 2. Typical example of the effect of L-propranolol on the hypoxic response. L-.Propranolol at 10- 6 M did not alter resting tension or hypoxic contraction.
l lO
M. OHE et al.
PO:
(Torr)
'"[1 I
I
I
I
I
OL
! Nitroglycerin 10"M
.°
iS rain
Fig. 3. Typical example of the effect of nitroglycerin on the hypoxie response. Nitroglycerin at 10-6 M did not change baseline tone in normoxia, but completely inhibited hypoxic contraction.
A summary of the effects of arachidonic acid metabolite blockers on the hypoxic contraction of rings is given in Table 2. Indomethacin at 10- 6 M, an inhibitor of the cyclooxygenase pathway, slightly increased baseline tone in normoxia and augmented the hypoxic response to 185 + 37~o of its preceding values (P < 0.05). However, the hypoxic response remained unchanged in the ethanol-treated vessels. FPL 55712 at 10- 7 M, a competitive antagonist of slow-reacting substance of anaphylaxis (Augstein et eL, 1973), affected neither the baseline tone nor the hypoxic response. Quinacrine at 3 x 10- 6 M, a phospholipase A2 inhibitor, increased baseline tone slightly in normoxia and markedly enhanced the hypoxic response. Methylene blue at 10-¢'M, an inhibitor of soluble guanylate cyclase, increased baseline tone slightly in normoxia and markedly enhanced the hypoxic response by 386 _+ 110~, of its preceding values (P < 0.05, n -- 5). L-NMMA at 10 =4 M increased baseline tone in normoxia and enhanced the hypoxie response by 262 + 48% of its preceding values (n = 3). To examine the role of the endothelium in hypoxic contraction of isolated human pulmonary artery, the hypoxic response was compared in the same arterial ring before and after the endothelium was rubbed. Figure 4 summarizes the effect of rubbing the endothelium on the hypoxic response. Hypoxic contraction after such rubbing was enhanced as compared with the values determined beforehand (P < 0.01, n = 6).
TABLE 2 Effects of arachidonic acid metabolite blockers on the hypoxic response.
10- ~' M indomethacin (8) 10 7 M FPL 55712 (4) 3 x I 0 * M quinacrine (6)
Hypoxic response (% of control)
P value
185 + 37 97 _+ 10 274 + 16
P < 0.05 NS P < 0.05
Values are given as means + SE, Numbers in parentheses represent the number of rings tested, NS, not significant,
HYPOXIC CONTRACTION OF THE HUMAN PULMONARY ARTERY
11 1
3OO
loo
-I-
0
E (÷)
E (-)
Fig. 4. Summary of the effect on hypoxic contraction of isolated human pulmonary artery caused by rubbing the endothelium. Data are means + SE. E( + ), endothelium not rubbed; E( - ), endothelium rubbed; n ffi 6; * P < 0.01.
Discussion
Pulmonary vascular smooth muscle frequently requires an initial tone produced by an agonist, such as a prostaglandin or histamine, before hypoxic vasoconstriction is observed. If initial tone is blocked with a specific receptor blocker, the HPV is also lost. However, the hypoxic contractile response produced in this study was elicited by the direct action of hypoxia alone and required no pre-constriction. We have termed thi~ contraction 'pre-stretched contraction', in the present study, the reason why isolated human pulmonary artery, even in the case of an outer diameter of 2 mm, constricted in response to hypoxia without pre-constriction is unclear. However, when an experiment was begun more than 10 h after lung removal and the handling of the pulmonary vascular smooth muscle was inappropriate, the response to hypoxia was also lost. This preparation is the first to demonstrate hypoxic contraction of human pulmonary artery without pre-constriction in vitro and has the advantage of allowing observation of the direct effect of hypoxia on human pulmonary vascular smooth muscle. Currently, transmembrane Ca 2 ÷ influx through the voltage-dependent Ca 2 ÷ channel is thought to be the most important factor in control of HPV (McMurtry, 1985). In the present study, verapamil had no effect on the normoxic force and decreased the maximum hypoxic response, whereas BAY K 8644 increased normoxic force and the subsequent maximum hypoxic response. These results show that voltage-dependent Ca 2 + influx also plays an important role in the hypoxic contraction of pulmonary artery in man. However, because the hypoxic contraction was still present in the absence of extracellular Ca 2 ÷, we suppose that hypoxic contraction is not dependent only on extracellular Ca 2 + as reported in hypoxic histamine-mediated contraction in man
112
M, OHE etal.
(Hoshino et al., 1988). Moreover, the data obtained with phentolamine, L-propranolol, diphenhydramine, guanethidine, indomethacin and FPL 55712 also show that neither the 0g ]/, H~ receptors, lipoxygenase pathway nor neural reflexes are involved in this response, and that vasodilatory prostaglandins prevent the hypoxic pulmonary response in man. There are several differences between the hypoxic response in our study and that recently observed in a human histamine-HPV model (Hoshino et al., 1988): For example, (1)The hypoxic response observed in the current work was not inhibited by diphenhydramine or procaine. In contrast, hypoxic contraction was enhanced by 10-3 M procaine. (2)With reoxygenation, tension returned to the original baseline values which was not the case in hypoxic histamine-mediated contraction. These discrepancies may be due to differences between the unstimulated and H,-receptor stimulated conditions as described above. That procaine at 5 x I0-3 M inhibited the hypoxic histamine-mediated contraction may derive from the tacts that histamine partially promotes Ca 2 + release from intraceUular storage sites (William, 1985) and that procaine acts to inhibit Ca2 + release. In our study, procaine at 10-3 M enhanced the hypoxic pulmonary response. This may be due to some other actions of procaine, such as blockage of K + channels (Kurihara, 1973). A similar result has been reported previously using intrapulmonary rabbit artery (Lloyd, 1966). The hypoxia used in our study was mild with a Po, of 43 Torr. Therefore, no significant differences exist between the Po: employed in this study and that in mixed venous blood. However, the step changes in Po: in cat small pulmonary arteries produced stimulus-response curves that decreased in a single step from 400 to 30 Tort and a Po: of 20 Torr developed a peak tension equal to that seen at 50 Torr, although this tension was transient and dropped rapidly from the peak (Madden et ai., 1985). Similarly, Sylvester et al. (1980) reported that the maximum response in isolated pig lungs occurred at Po~ between 30 and 60 Torr and diminished at Po: below 30 Ton. Therefore, the Po~,of 43 Torr used in the present study was considered as appropriate for hypoxic stimulation of the isolated pulmonary artery. Hypoxic contraction of isolated human pulmonary artery was enhanced by removal of endothelial cells. Moreover, this contraction was also enhanced by administration of 3 × 10 - 6 M quinacrine, 10- ~ M methylene blue or 10- 4 M L-NMMA. These doses of quinacrine and methylene blue have been reported to inhibit the action of endothelium-dependent relaxing factor (EDRF) on ACh in human pulmonary artery (Greenberg etal., 1987). The effects of methylene blue include alteration in smooth muscle cell redox state, damage to the endothelium through superoxide production or change in catecholamine levels. The effects of quinacrine include inhibition of the modulatory effect of platelet-activating factor (Otamiri and Tagesson, 1989) as well as inhibition of phospholipase A2. However, the major effect of methylene blue and quinacrine is to block guanylate cyclase, and EDRF relaxes smooth muscle by activating guanylate cyclase (Ignarro, 1989). Hence, blocking guanylate cyclase with methylene blue and quinacrine would be expected to block the action of EDRF. Nitric oxide is a major component of EDRF. L-NMMA at 10- 4 M, an inhibitor of the synthesis of
HYPOXIC CONTRACTION OF THE HUMAN PULMONARY ARTERY
113
nitric oxide from L-arginine (Sakuma et aL, 1988), enhanced the hypoxic response. Moreover, removal of endothelium enhances the hypoxic vasoconstrictor response. These results are consistent with the observations of Robertson et al. (1990) in which hypoxic pulmonary vasoconstriction was more than doubled by blocking nitric oxide production with L-NMMA in isolated perfused rat lung. It has been reported that the in vitro antagonists of EDRF, eicosatetraynoic acid, nordihydroguaiaretic acid and hydroquinone (Brashers etai., 1988) and methylene blue (Mazmanian etal., 1989), significantly augment the hypoxic pressor response in the isolated perfused rat lung preparation. However, these observations conflict with those of other studies in which hypoxic contraction in porcine main pulmonary artery was reported to be completely endothelium dependent (Holden and McCall, 1984) and in which chemical inhibition of EDRF by methylene blue or hemoglobin was found to reduce the hypoxic contraction of isolated rat pulmonary artery (Rodman et al., 1987). The reason for this difference is unclear. However, apart from species differences, the latter studies were performed using ring preparations of the main pulmonary artery, where reactivity of the vascular bed may differ from that of the isolated perfused lung preparation and intrapuimonary artery used in our study. In summary, the pre-stretched human pulmonary artery constricted in response to hypoxia, at least in part, through activation of the voltage-dependent Ca 2 + channel, a, //, H~ receptors, lipoxygenase pathways and neural reflexes are not involved in this response. In addition, our results suggest that the endothelium does not contribute at all to hypoxic contraction of human pulmonary artery.
Acknowledgements.We are grateful to Hoken Kagaku Institution for their support and to R, Scott for his critical reading of tlte manuscript. A part of this work was presented at the 60th Scientific Session of the American Heart Association in Anaheim, CA, 15-17 November 1987.
References Augstein, J., J.B. Farmer, T.B. Lee, P. Sheard and M,L. Tattersall (1973). Selective inhibitor of slow reacting substance of anaphylaxis. NatI~re New Biol, 245: 215-217, Brashers, V. L., M.J. Peach and C. E. Rose, Jr. (1988). Augmentation of hypoxicpulmonary vasoconstriction in the isolated perfused rat lung in vitro antagonists ofendothelium-dependent relaxation. J. Clin, Invest. 82: 1495-1502. Detar, R. and M. Gellai (1971). Oxygen and isolated vascular smooth muscle from the main pulmonary artery of the rabbit. Am. J. Physiol. 221: 1791-1794. Greenberg, B., K. Rhoden and P.J. Barnes (1987). Endothelium-dependent relaxation of human pulmonary arteries. Am. J. Physiol. 252: H434-H438. Holden, W.E. and E. McCall (1984). Hypoxia-induced contractions of porcine pulmonary artery strips depend on intact endothelium. Exp. Lung Res. 7: I01-I 12. Hoshino, Y., H. Obara, M. Kusunoki, Y. Fujii and S. lwai (1988). Hypoxic contractile response in isolated human pulmonary artery: role of calcium ion. J. Jppl. Physiol. 65: 2468-2474. Ignarro, L.J. (1989), Endothelium-derived nitric oxide: actions and properties. FASEB J. 3: 31-36. Kurihara, S, (1973). The effect of procaine on the urinary bladder smooth muscle of bullfrogs.Jap. J. Physiol. 23: 309-324.
114
M. OHE et al.
Lloyd, T. C., Jr. (! 966). Po,-dependent pulmonary vasoconstriction caused by procaine. J. Appl. Physiol. 21: 1439-1442. Lloyd, T.C., Jr. (1968). Hypoxic pulmonary vasoconstriction: role of perivascular tissue. J. Appl. Physiol. 25: 560-565. Madden, J.A., C.A. Dawson and D.R. Harder (1985). Hypoxia-induced activation in small isolated pulmonary arteries from the cat. 3. Appl. Physiol. 59: !13-118. Mazmanian, G.M., B. Baudet, C. Brink, J. Cerrina, S. Kirkiacharian and M. Weiss (I~89). Methylene blue potentiates vascular reactivity in isolated rat lungs. J. Appl. Physiol. 66: 1040-1045. McMurtry, !. F. (! 985). BAY K 8644 potentiates and A23187 inhibits hypoxic vasoconstriction in rat lungs. Am. J. Physiol. 249: H741-H746. Otamiri, T. and C. Tagesson (1989). Role of phospholipase Ae and oxygenated free radicals in mucosal damage after small intestinal ischemia and reperfusion. Am. J. Surg. 157: 562-566. Robertson, B.E., J.B. Warren and P.C.G. Nye (1990). Inhibition of nitric oxide synthesis potentiates hypoxic vasoconstriction in isolated rat lungs. Exp. Physiol. 75: 255-25?. Rodman, D.M., T. Yamaguchi, R.F. O'Brien and I.F. McMurtry (1987). Hypoxic contraction of isolated pulmonary artery is reduced by endothelial denudation, hemoglobin, and methylene blue (Abstract). Am. Ray. Respir. Dis. 135: A131. Rodman, D.M., T. Yamaguchi, R.F. O'Brien and I.F. McMurtry (1989). Hypoxic contraction of isolated rat pulmonary artery. J. Pharmacol. Exp. That. 248: 952-959. Sakuma, l., D. 3. Stueher, S. S. Gross, C. Nathan and R. Levi (I 988) Identification of arglnine as a precursor of endothelium-derived relaxing factor. Proc. Natl. Acad. $ci. USA 85: 8664-8667. Souhrada, J. F. and D.W. Dickey (1976). Effectof substrata on hypoxic response of small pulmonary artery. J. Appl. Physiol. 40: 533-538. Sylvester, J.T., A.L. Harabin, M.D. Peake and R.S. Frank (1980). Vasodilator and constrictor responses to hypoxia in isolated pig lungs. J. Appl. Physiol. 49: 820-825. Von Euler, U.S. and G. Liljestrand (1946). Observations on the pulmonary arterial blood pressure in the cat, Acts Physlol, $cand, 12: 301-320. William, W, D. (1985). Histamine and 5-hydroxytryptamine (serotonin) and their antagonists. In: The Pharmacological Basis of Therapeutics; edited by A.G. Gilman, L.S. Goodman, T.W, Rail and F. Murad. New York: Macmillan, p. 610,
