- Email: [email protected]
Ligustrazine is a vasodilator of human pulmonary and bronchial arteries
European Journal of Pharmacology, 191 (1990) 345-350 Elsevier
345
EIP 51634
onary
an
Shufang Liu, Yingnan Cai ‘, Timothy W. Evans, David G. McCormack, Gwenda R. Barer and Peter J. Barnes Department
i
of Thoracic Medicine. National Heart and Luitg InstitutL. London and ’ Department of Medicine, Royal Haliamshire Hospital, Shejfield, U.K. Received 11 September
1990, accepted 25 September
1990
We have investigated the dilator effect of ligustrazine, the semisynthetic principle of a traditional Chinese herbal remedy, on human pulmonary and bronchial arteries in vitro. Ligustrazine caused a concentration-dependent relaxation of human small pulmonary arteries, which was independent of endothelium. Although ligustrazine was equally potent in inducing dilatation of pulmonary and bronchial arteries, it was about 10 times more potent in relaxing small pulmonary arteries (300-500 pm i.d.) compared with lobar pulmonary arteries (7-8 mm id.). By contrast, the relaxant responses of small and lobar pulmonary arteries to sodium nitroprusside was not significantly different. Ligustrazine was equally potent in relaxing prostaglandin F,,- or 5-hydroxytryptamine-precontracted pulmonary arteries, suggesting that it is not a prostaglandin F,, or 5-hydroxytryptamine antagonist. Preincubating the vessels with propranolol (1 pM) or indomethacin (10 PM) had no significant effect on the ligustrazine-induced vasodilatation. However, ligustrazine caused concentration-dependent inhibition of calcium-evoked contraction when applied to rat aorta in calcium-free K”-depolarizing medium. We conclude that ligustrazine is a dilator of human pulmonary and bronchial arteries, which is endothelium-independent and that ligustrazine preferentially relaxes pulmonary resistance vessels rather than large conduit pulmonary arteries. Ligustrazine; Pulmonary artery; Bronchial artery; Vasodilators; (Human)
1. Introduction
Vasodilators have been used to treat pulmonary hypertension for several years, but adverse side-effects have restricted their clinical use (Hyman and Kadowitz, 1984). One of the main complications is systemic hypotension (Packer et al., 1982), such that a selective pulmonary vasodilator is clinically desirable (Hyman and Kadowitz, 1984). Evidence from a number of experimental studies suggests that most of the increase in pulmonary vascular resistance observed during hypoxia is
Correspondence to: P.J. Barnes, Department of Thoracic Medicine, National Heart and Lung Institute, Dovehouse Street, London SW3 6LY, U.K. 0014-2999/90/$03.50
localized to precapillary arterioles (Grover et al., 1983; Fishman, 1985). Consequently, a drug that predominantly relaxes small resistance rather than large conduit pulmonary arteries would be beneficial in patients ‘with pulmonary hypertension. Ligustrazine (for the chemical structure see Cai and Barer, 1989) is a chemically identified and semisynthesized herbal extract (Beijing Institute of Pharmaceutical Industries, 1977) of a traditional Chinese herb, Ligustraci chuanxiong, which has been used as a mixture with other medical herbs to treat various cardiovascular diseases for more than a thousand years. Ligustrazine reverses hypoxic pulmonary vasoconstriction in isolated blood perfused rat lungs (Cai and Suggett, 1988) and the isolated right lower lobe of ferrets (Cai et al., 1988) in a dose-dependent manner and reduces
0 1990 Elsevier Science Publishers B.V. (Biomedical
Division)
ary artery pressure, right ventricular hy~pby and museularization of arterioles when n chronically to rats exposed to hypoxia (10% _) for 2 weeks (Cai and Barer, 1989). The purpose of this study was to study the effects of ~g~s~~ne on human p~rnon~ vessel and to compare its effects on human targe and small pulmonary arteries and on bronchial arteries.
2.1. Tissue preparation Human small (300-500 gm i.d.), segmental (5-6 mm id.) and lobar (7-8 mm id.) pulmonary arteries were obtained from patients undergoing lobectomy or pneumonectomy for the removal of bronchial carcinoma. Human bronchial arteries (100-200 pm i.d.) were removed from patients undergoing heart-lung or lung ~~spl~tation for the treatment of cystic fibrosis. Rat thoracic aorta was dissected from male Wistar rat (250 g) killed by a blow to the head. All human vascular tissue was used on the day of surgery- The vessels were cleaned of surroundmg tissues and sectioned into rings of 3-4 mm in length. In the rings of some human pulmonary arteries and rat aorta, the endothelium was removed by gently rotating a steel probe (for small vessels. fine probe was used) in the lumen for 20-30 s. Removal of the endotherm was confirmed by the loss of relaxant response in precontracted vessels to acetylchohue (0.1-10 PM). 2.2. Organ bath studies Vessel rings (except human segmental and lobar pulmonary arteries) were suspended in 2 ml organ baths by a pair of stainless steel wires (for small vessels, a pair of fine wires was used) inserted into the lumen of the vessels, one wire was fixed and the other attached to a force transducer (FT.03 Grass, Quincy, MA, USA). Rings of human segmental and lobar pulmonary arteries were fixed vertically between two hooks in 10 ml organ baths, one hook was fixed and the other attached to the
force transducer. The organ baths contained Krebs-Henseleit (K-H) solution of the following composition (in mM): NaCl 118, KC1 5.9, MgSO, .7H,O 1.2, CaCI, .6H,O 2.5, NaH,PO, 1.2, glucose 5.6 and NaHC03 25.5. The organ baths were rn~n~ed at 37O C and aerated with a gas mixture of 95% 0, and 5% CO,. The rings were stretched gradually to a resting tension of 600 mg, 1.5 and 2 g for human small pulmonary and bronchial arteries, human segmental and lobar pulmonary arteries, and rat aorta respectively, which were found to be optimal for measuring changes in tension in these vessels. The rings were then allowed to equilibrate in normal K-H solution for at least 1 h before starting studies and were washed with fresh K-H solution every 15-20 min. 2.3. Relaxation studies The rings of human puImona~ and bronchial arteries were precontracted with 1 FM prostaglandin Fzo (PGF,,) or 1 FM serotonin (S-HT). After a stable contraction was obtained, cumulative concentration-response curves to ligustrazine and sodium nitroprusside were performed. In some of the experiments, human small pulmonary arteries were preincubated with 1 PM propranolol or 10 FM indomethacin for 20 min before studying relaxation. 2.4. experiments in calcium free solution After a 1 h equilibration period in normal K-H solution and a 5-10 min incubation in calcium-free K-H solution (prepared by o~t~g calcium from normal K-H solution) containing 0.2 mM EGTA (ethyleneglycol-bis-( &aminoethyl ether) N,N,N’, N’-tetraacetic acid), cumulative concentration-response curves of rat aortic rings to CaCl, (0.01-20 mM) were obtained in a calcium-free K+-depolarizing medium (composition in mM: NaCll7, KC1 100, KH,PO, 1.0, MgSO, 1.2, NaHCO, 25, glucose 11.5). The preparations were washed, equilibrated in normal K-H solution again and incubated with EGTA, before being incubated with
347
various concentrations of ligustrazine in calciumfree K-H solution for 30 min. Concentration-response curves to CaCl, in the presence of ligustrazine were constructed in a calcium-free depolarizing medium. Control experiments (no ligustrazine) were carried out on matched rings from the same animal.
2.5. Drugs The following drugs were used: prostaglandin F,, (Upjohn Ltd, Crawley, Sussex, UK), ligustrazine hydrochloride (The Fourth Drug Manufacturer, Beijing, China), propranolol (I.C.I., Macclesfield, Cheshire, UK) and calcium chloride, acetylcholine, serotonin, sodium nitroprusside, EGTA and indomethacin (Sigma, Poole, U.K.). Drugs (except indomethacin which was in an alkaline buffer) were dissolved in distilled water and diluted with fresh K-H solution before use.
2.6. Analysis of results Relaxation was expressed as the percentage of PGF,, or 5-HT-induced contraction. Contraction was expressed as the percentage of the maximal contraction induced by the same agent in the same ring. -Log EC,, values and the mean values at each concentration point were used to compare the relaxant response of different vessels and the same vessels in the presence and absence of an intact endothelium or an antagonist to ligustrazine, where EC,, being the concentration causing 50% reduction of the PGF,,-induced tone. For analysing the contractile respsones to CaCl, in the presence and absence ligustrazine, - log EC,, values and the mean values at each concentration point were compared. E& value was calculated using a computer programe. All values in the text were expressed as means +S.E.M. Results were compared by using paired and unpaired Student’s t-test, or one way analysis of variance and Bonferroni corected t-test when multiple comparison was made. P values < 0.05 were considered to be significant.
3. Results 3.1. Effect of endothelial denudation on the relaxant response of human small pulmonary artety to ligustrazine Figure 1 shows the effect of endothelial removal on the relaxant response of human small pulmonary artery to ligustrazine. The relaxant response of the vessels with and withoat an intact endothelium to ligustrazine was not significantly different at any concentration, indicating that removal of endothelium had no significant influence on ligustrazine-induced relaxation. 3.2. Comparison of human bronchial and small pulmonary arteries Ligustrazine caused a concentration-dependent relaxation of both human small pulmonary and bronchial arteries. The mean values of relaxation induced by ligustrazine at each concentration point in bronchial arteries were not significantly different from those in pulmonary arteries with and without endothelium. The -log EC,,, value was
I
7
I
6
I
5
r
4
I
3
ILigusti-ozinel, (-log r4)
Fig. 1. Comparison of concentration-response curves to ligustrazine in human small pulmonary arteries with (v) and without (v) endothelium. Tension is expressed as a percentage of 1 pM prostaglandin F,, (PGF,,) contraction. Data represent mean vahres with vertical bars indicating S.E.M.; n = g-11.
(n = 7). 4.32 + 0.10 (n = 11) and 4.11
bronchial artery and pulmonary and with endothelium, respecindicating that the potency of in relaxing bronchial arteries was not different from that of relaxing
3.3. C~lparison
of
large
and
small
human
~~i~~~~~a~arteries
Ligustrazine relaxed human small pulmonary arteries to a significantly greater extent than lobar pulmonary arteries at all concentrations used (fig. 2) and segmental pulmonary arteries at higher concentrations (fig. 2). Thus, ligustrazine caused a 938 relaxation of small pulmonary arteries, but only 65% relaxation of segmental and lobar pulmonary arteries at the concentration of 1 mM. Although sodium nitroprusside was slightly more potent in relaxing lobar pulmonary arteries than small pulmonary arteries (fig. 3), this was not significantly different.
o-
98
i
I
I
I
I
I
7 6 5 4 3 [sodium nltroprussidel. (-log M)
Fig. 3. Comparison of concentration-response curves of human small (0) and lobar ( ) pulmonary arteries to sodium nitroprusside. Tension is expressed as a percentage of 1 pM PGF,,-induced contraction. Data represent mean values with vertical bars indicating S.E.M.; n = 5-7.
3.4. Effects of antagonistic agents on the vasodilator action of ligustrazine
7
6
;
4
;
[Liyustrozinel.f-log MI
Fig. 2. Comparison of concentration-response curves of human small (0). segmental (0) and lobar ( ) pulmonary arteries to ligustrazine. Tension is expressed as a percentage of 1 pM PGF,, contraction. Data represent mean values with vertical bars indicating S.E.M.; n = 3-11. * P -z 0.05, compared with small pulmonary arteries.
The ligustrazine-induced relaxations of 5-HTand PGF,,-precontracted human small pulmonary arteries were not significantly different at any concentration. The -log EC,, value were 3.75 + 0.10 (n = 4) and 4.0 + 0.21 (n = 4) for 5-HT and PGF,,-precontracted vessels, respectively (P > 0.05), indicating that ligustrazine is not a 5-HT or PGF,, antagonist. Preincubating human small pulmonary arteries with propranolol (1 PM) or indomethacin (10 PM) similarly had no significant effect on the vasodilator action of ligustrazine. The -log EC,, values were 3.96 kO.16 (n = 9) and 3.90 + 0.10 (n = 8), for control and propranolol pretreated vessels, and 3.85 _+0.13 (n = 4) and 3.72 f. 0.18 (n = 4) for control and indomethacin-pretreated vessels, suggesting that neither /?-adrenoceptors nor vasodilator cyclooxygenase products were involved in the ligustrazine-induced vasodilatation.
11 “I
5.0
405
‘I.0
3,s
3.0
2.5
2,o
1.7
1cOc121, t-log MI
Fig. 4. Effects of ligustrazine on concentration-response curves of rat aortic rings bathed in Ca 2+-free K +-depolarizing medium to CaCl,. Cumulative concentration-response curves to Ca2+ were constructed after incubating the rings with 0.3 (0). 1 (A) and 3 (e) mM ligustrazine (n = 6 each group). ( Tension is expressed as the percentage of CaCI, maximal contraction. Data represent mean values with vertical bars indicating S.E.M. * P -z 0.05 compared with cotrol group.
3.5. Effects of ligustrazine on CaCI,-induced traction
con-
In rat aortic rings bathed in Ca”+-free K+-depolarizing medium, ligustrazine shifted the CaCl, concentration-response curve significantly to right and reduced the maximal response to CaCl, in concentration-dependent manner (fig. 4). Thus, the -log EC,, values were 3.71, 3.47, 2.96 and 2.17 for control group and other groups pretreated with 0.3, 1 and 3 mM of ligustrazine.
4. Discussion
Our data demonstrate that ligustrazine is a dilator of both human pulmonary and bronchial arteries, which is endothelium-independent and that ligustrazine preferentially relaxes human small pulmonary resistance arteries in comparison with large conduit arteries. Ligustrazine appears to have a calcium antagonistic property.
Our conclusion of ligustrazine preferentially relaxing small pulmonary arteries is supported by our sodium nitroprusside studies in which we showed that the potency of sodium nitroprusside in relaxing small and lobar human pulmonary arteries was not significantly different using the same method. By contrast, nitrate vasodilators have been demonstrated to act preferentially on large coronary arteries (Winbmy et al., 1969; Schnaar and Sparks, 1972). Although several vasodilators have been demonstrated to have different effects on large and small systemic arteries (Winbury et al., 1969; Norton and Detar, 1970; Schnaar and Sparks, 1972), our data show that ligustrazine relaxes human small pulmonary arteries to a significantly greater extent than lobar pulmonary arteries, in what we believe to be the first demonstration of this phenomenon in human pulmonary vessels. These results are consistant with our previous data showing that ligustrazine is significantly more potent in relaxing intrapuhnonary arteries than extrapulmonary arteries in the rat (Liu et al., 1989). Hypoxic pulmonary vasoconstriction occurs predominantly in small pulmonary resistance arteries (Grover et al., 1983; Fishman, 1985) and may contribute to pulmonary hypertension (Voelkel and Weir, 1989). The characteristic action of ligustrazine in predominantly relaxing small pulmonary resistance arteries raises the possibility that ligustrazine may become a useful drug in the treatment of pulmonary hypertension when reduction of the pulmonary blood pressure is needed. Indeed, it has been reported that ligustrazine causes dose-dependent inhibition of hypoxic vasoconstriction in isolated blood perfused rat lungs and the right lower lobe of ferret lungs (Cai and Suggett, 1988; Cai et al., 1988). Furthermore, when given chronically (80 mg/kg i.p.) to rats exposed to 10% 0, for 2 weeks, ligustrazine attenuates the muscularization of pulmonary arterioles and right ventricular hypertrophy (Cai and Barer, 1989) which are the typical pathological features of pulmonary hypertension. It has been reported that ligustrazine reduces pulmonary blood pressure more than systemic blood pressure in ferrets (Cai et al., 1988) but our results indicate that ligustrazine is equally potent
in causing vasodilatation of human pulmonary and bronchial arteries. The difference between e.~ results could be explained by a species difference. ~4d~tion~Iy. the bronchial arteries may be functionally different from the systemic arteries of other regions and the response to ligustrazine may therefore be different from that of systemic arteries of other regions. In studying the possible mechanisms by which ligustrazine causes vasodilatation, we investigated whether ligustrazine may interact with a specialcaused by ised receptor. The vasodilatation ligustrazine was not significantly different in vessels precontracted with FGF,, from those preconn-acted with S-HT. In addition, preincubation of the pulmonary artery rings with propranolol and indomethacin had no significant effect on the reIaxant responses to ligustrazine. These results indieates that ligustrazine is not a PGF,, or 5-HT antagonist and that neither &adrenoceptors nor the release of vasodilator cyclooxygenase products were involved in its vasodilator action. The concentration-dependent inhibition of Ca’ ‘-induced contraction in calcium-free K+-depolarizing medium suggests that ligustrazine has a calcium antagonistic property. Although ligustrazinc itself is not potent, its differential effect on small pulmonary arteries is of potential clinical value and it may provide the basis for the development of new pulmonary vasodilator in the future.
Acknowledgements We thank the British Heart Foundation and Chest Heart and Stroke Foundation (U.K.) for financial support.
References Beijing institute of Pharmaceutical Industries. 1977. Extraction, isolation and structural identification of ligustrazine, Chung Hua I Hsueh Tsa Chih 57.420 (Chinese). Cai, Y. and G.R. Barer, 1989, Effects of ligustrazine on pulmonary vascular changes induced by chronic hypoxia in rats, Clin. Sci. 77. 515. Cai. Y.. D. Bee and G.R. Barer, 1988, Pulmonary vascular action of ligustrazine. principle of a traditional Chinese remedy, Clin. Sci. 75 (Suppl. 19). 55P. Cai. Y. and A.J. Suggett. 1988. Pulmonary vasodilator action of ligustrazine, active principle of a traditional Chinese remedy. J. Physiol. 399, 45P. Fishman. A.P., 1985, Pulmonary circulation, in: Handbook of Physiology, the Respiratory System, Section 3, Volume 1, eds. A.P. Fishman. A.B. Fisher and S.R. Geiger (Waverly Press, Baltimore) p. 93. Grover, B.F.. W.W. Wagner, Jr., LF. McMurtry and J.T. Reeves, 1983, Pulmonary circulation, in: Handbook of Physiology, the Cardiovascular System, Volume 3. eds. J.T. Shepherd, F.M. Abboud and S.R. Geiger (Waverly Press, Baltimore) p. 65. Hyman, A-L. and P.J. Kadowitz, 1984. Vasodilator therapy for pulmonary hypertensive disorders, Chest 85, 145. Liu. S.F.. L.Y. Rui, Y.N. Cai. P.J. Barnes. T.W. Evans and G.R. Barer, 1989, Dilator action of ligustrazine on human and rat pulmonary arteries, Clin. Sci. 77 (Suppl.), 27P. Norton, J.M. and R. Detar, 1970, Adenosine and isolated coronary vascular smooth muscle, Physiologist 13, 273. Packer, M.. B. Greenberg, B. Massie and H. Dash, 1982, Deleterious effects of hydralazine in patients with pulmonary hypertension, N. Engl. J. Med. 306, 1326. Schnaar, R.L. and H.V. Sparks. 1972, Response of large and small coronary arteries to nitroglycerin, NaNOr. and adenosine, Am. J. Physiol. 223, 223. Voelkel, N.F. and E.K. Weir, 1989. Etiologic mechanism in primary pulmonary hypertension, in: Pulmonary Vascular Physiology and Pathophysiology, eds. E.K. Weir and J.T. Reeves (Marcel Dekker, New York) p. 513. Winbury, M.M., B.B. Howe and M.A. Hefner, 1969, Effect of nitrates and other coronary dilators on large and small coronary vessls, J. Pharmacol. Exp. Ther. 168, 70.
345
EIP 51634
onary
an
Shufang Liu, Yingnan Cai ‘, Timothy W. Evans, David G. McCormack, Gwenda R. Barer and Peter J. Barnes Department
i
of Thoracic Medicine. National Heart and Luitg InstitutL. London and ’ Department of Medicine, Royal Haliamshire Hospital, Shejfield, U.K. Received 11 September
1990, accepted 25 September
1990
We have investigated the dilator effect of ligustrazine, the semisynthetic principle of a traditional Chinese herbal remedy, on human pulmonary and bronchial arteries in vitro. Ligustrazine caused a concentration-dependent relaxation of human small pulmonary arteries, which was independent of endothelium. Although ligustrazine was equally potent in inducing dilatation of pulmonary and bronchial arteries, it was about 10 times more potent in relaxing small pulmonary arteries (300-500 pm i.d.) compared with lobar pulmonary arteries (7-8 mm id.). By contrast, the relaxant responses of small and lobar pulmonary arteries to sodium nitroprusside was not significantly different. Ligustrazine was equally potent in relaxing prostaglandin F,,- or 5-hydroxytryptamine-precontracted pulmonary arteries, suggesting that it is not a prostaglandin F,, or 5-hydroxytryptamine antagonist. Preincubating the vessels with propranolol (1 pM) or indomethacin (10 PM) had no significant effect on the ligustrazine-induced vasodilatation. However, ligustrazine caused concentration-dependent inhibition of calcium-evoked contraction when applied to rat aorta in calcium-free K”-depolarizing medium. We conclude that ligustrazine is a dilator of human pulmonary and bronchial arteries, which is endothelium-independent and that ligustrazine preferentially relaxes pulmonary resistance vessels rather than large conduit pulmonary arteries. Ligustrazine; Pulmonary artery; Bronchial artery; Vasodilators; (Human)
1. Introduction
Vasodilators have been used to treat pulmonary hypertension for several years, but adverse side-effects have restricted their clinical use (Hyman and Kadowitz, 1984). One of the main complications is systemic hypotension (Packer et al., 1982), such that a selective pulmonary vasodilator is clinically desirable (Hyman and Kadowitz, 1984). Evidence from a number of experimental studies suggests that most of the increase in pulmonary vascular resistance observed during hypoxia is
Correspondence to: P.J. Barnes, Department of Thoracic Medicine, National Heart and Lung Institute, Dovehouse Street, London SW3 6LY, U.K. 0014-2999/90/$03.50
localized to precapillary arterioles (Grover et al., 1983; Fishman, 1985). Consequently, a drug that predominantly relaxes small resistance rather than large conduit pulmonary arteries would be beneficial in patients ‘with pulmonary hypertension. Ligustrazine (for the chemical structure see Cai and Barer, 1989) is a chemically identified and semisynthesized herbal extract (Beijing Institute of Pharmaceutical Industries, 1977) of a traditional Chinese herb, Ligustraci chuanxiong, which has been used as a mixture with other medical herbs to treat various cardiovascular diseases for more than a thousand years. Ligustrazine reverses hypoxic pulmonary vasoconstriction in isolated blood perfused rat lungs (Cai and Suggett, 1988) and the isolated right lower lobe of ferrets (Cai et al., 1988) in a dose-dependent manner and reduces
0 1990 Elsevier Science Publishers B.V. (Biomedical
Division)
ary artery pressure, right ventricular hy~pby and museularization of arterioles when n chronically to rats exposed to hypoxia (10% _) for 2 weeks (Cai and Barer, 1989). The purpose of this study was to study the effects of ~g~s~~ne on human p~rnon~ vessel and to compare its effects on human targe and small pulmonary arteries and on bronchial arteries.
2.1. Tissue preparation Human small (300-500 gm i.d.), segmental (5-6 mm id.) and lobar (7-8 mm id.) pulmonary arteries were obtained from patients undergoing lobectomy or pneumonectomy for the removal of bronchial carcinoma. Human bronchial arteries (100-200 pm i.d.) were removed from patients undergoing heart-lung or lung ~~spl~tation for the treatment of cystic fibrosis. Rat thoracic aorta was dissected from male Wistar rat (250 g) killed by a blow to the head. All human vascular tissue was used on the day of surgery- The vessels were cleaned of surroundmg tissues and sectioned into rings of 3-4 mm in length. In the rings of some human pulmonary arteries and rat aorta, the endothelium was removed by gently rotating a steel probe (for small vessels. fine probe was used) in the lumen for 20-30 s. Removal of the endotherm was confirmed by the loss of relaxant response in precontracted vessels to acetylchohue (0.1-10 PM). 2.2. Organ bath studies Vessel rings (except human segmental and lobar pulmonary arteries) were suspended in 2 ml organ baths by a pair of stainless steel wires (for small vessels, a pair of fine wires was used) inserted into the lumen of the vessels, one wire was fixed and the other attached to a force transducer (FT.03 Grass, Quincy, MA, USA). Rings of human segmental and lobar pulmonary arteries were fixed vertically between two hooks in 10 ml organ baths, one hook was fixed and the other attached to the
force transducer. The organ baths contained Krebs-Henseleit (K-H) solution of the following composition (in mM): NaCl 118, KC1 5.9, MgSO, .7H,O 1.2, CaCI, .6H,O 2.5, NaH,PO, 1.2, glucose 5.6 and NaHC03 25.5. The organ baths were rn~n~ed at 37O C and aerated with a gas mixture of 95% 0, and 5% CO,. The rings were stretched gradually to a resting tension of 600 mg, 1.5 and 2 g for human small pulmonary and bronchial arteries, human segmental and lobar pulmonary arteries, and rat aorta respectively, which were found to be optimal for measuring changes in tension in these vessels. The rings were then allowed to equilibrate in normal K-H solution for at least 1 h before starting studies and were washed with fresh K-H solution every 15-20 min. 2.3. Relaxation studies The rings of human puImona~ and bronchial arteries were precontracted with 1 FM prostaglandin Fzo (PGF,,) or 1 FM serotonin (S-HT). After a stable contraction was obtained, cumulative concentration-response curves to ligustrazine and sodium nitroprusside were performed. In some of the experiments, human small pulmonary arteries were preincubated with 1 PM propranolol or 10 FM indomethacin for 20 min before studying relaxation. 2.4. experiments in calcium free solution After a 1 h equilibration period in normal K-H solution and a 5-10 min incubation in calcium-free K-H solution (prepared by o~t~g calcium from normal K-H solution) containing 0.2 mM EGTA (ethyleneglycol-bis-( &aminoethyl ether) N,N,N’, N’-tetraacetic acid), cumulative concentration-response curves of rat aortic rings to CaCl, (0.01-20 mM) were obtained in a calcium-free K+-depolarizing medium (composition in mM: NaCll7, KC1 100, KH,PO, 1.0, MgSO, 1.2, NaHCO, 25, glucose 11.5). The preparations were washed, equilibrated in normal K-H solution again and incubated with EGTA, before being incubated with
347
various concentrations of ligustrazine in calciumfree K-H solution for 30 min. Concentration-response curves to CaCl, in the presence of ligustrazine were constructed in a calcium-free depolarizing medium. Control experiments (no ligustrazine) were carried out on matched rings from the same animal.
2.5. Drugs The following drugs were used: prostaglandin F,, (Upjohn Ltd, Crawley, Sussex, UK), ligustrazine hydrochloride (The Fourth Drug Manufacturer, Beijing, China), propranolol (I.C.I., Macclesfield, Cheshire, UK) and calcium chloride, acetylcholine, serotonin, sodium nitroprusside, EGTA and indomethacin (Sigma, Poole, U.K.). Drugs (except indomethacin which was in an alkaline buffer) were dissolved in distilled water and diluted with fresh K-H solution before use.
2.6. Analysis of results Relaxation was expressed as the percentage of PGF,, or 5-HT-induced contraction. Contraction was expressed as the percentage of the maximal contraction induced by the same agent in the same ring. -Log EC,, values and the mean values at each concentration point were used to compare the relaxant response of different vessels and the same vessels in the presence and absence of an intact endothelium or an antagonist to ligustrazine, where EC,, being the concentration causing 50% reduction of the PGF,,-induced tone. For analysing the contractile respsones to CaCl, in the presence and absence ligustrazine, - log EC,, values and the mean values at each concentration point were compared. E& value was calculated using a computer programe. All values in the text were expressed as means +S.E.M. Results were compared by using paired and unpaired Student’s t-test, or one way analysis of variance and Bonferroni corected t-test when multiple comparison was made. P values < 0.05 were considered to be significant.
3. Results 3.1. Effect of endothelial denudation on the relaxant response of human small pulmonary artety to ligustrazine Figure 1 shows the effect of endothelial removal on the relaxant response of human small pulmonary artery to ligustrazine. The relaxant response of the vessels with and withoat an intact endothelium to ligustrazine was not significantly different at any concentration, indicating that removal of endothelium had no significant influence on ligustrazine-induced relaxation. 3.2. Comparison of human bronchial and small pulmonary arteries Ligustrazine caused a concentration-dependent relaxation of both human small pulmonary and bronchial arteries. The mean values of relaxation induced by ligustrazine at each concentration point in bronchial arteries were not significantly different from those in pulmonary arteries with and without endothelium. The -log EC,,, value was
I
7
I
6
I
5
r
4
I
3
ILigusti-ozinel, (-log r4)
Fig. 1. Comparison of concentration-response curves to ligustrazine in human small pulmonary arteries with (v) and without (v) endothelium. Tension is expressed as a percentage of 1 pM prostaglandin F,, (PGF,,) contraction. Data represent mean vahres with vertical bars indicating S.E.M.; n = g-11.
(n = 7). 4.32 + 0.10 (n = 11) and 4.11
bronchial artery and pulmonary and with endothelium, respecindicating that the potency of in relaxing bronchial arteries was not different from that of relaxing
3.3. C~lparison
of
large
and
small
human
~~i~~~~~a~arteries
Ligustrazine relaxed human small pulmonary arteries to a significantly greater extent than lobar pulmonary arteries at all concentrations used (fig. 2) and segmental pulmonary arteries at higher concentrations (fig. 2). Thus, ligustrazine caused a 938 relaxation of small pulmonary arteries, but only 65% relaxation of segmental and lobar pulmonary arteries at the concentration of 1 mM. Although sodium nitroprusside was slightly more potent in relaxing lobar pulmonary arteries than small pulmonary arteries (fig. 3), this was not significantly different.
o-
98
i
I
I
I
I
I
7 6 5 4 3 [sodium nltroprussidel. (-log M)
Fig. 3. Comparison of concentration-response curves of human small (0) and lobar ( ) pulmonary arteries to sodium nitroprusside. Tension is expressed as a percentage of 1 pM PGF,,-induced contraction. Data represent mean values with vertical bars indicating S.E.M.; n = 5-7.
3.4. Effects of antagonistic agents on the vasodilator action of ligustrazine
7
6
;
4
;
[Liyustrozinel.f-log MI
Fig. 2. Comparison of concentration-response curves of human small (0). segmental (0) and lobar ( ) pulmonary arteries to ligustrazine. Tension is expressed as a percentage of 1 pM PGF,, contraction. Data represent mean values with vertical bars indicating S.E.M.; n = 3-11. * P -z 0.05, compared with small pulmonary arteries.
The ligustrazine-induced relaxations of 5-HTand PGF,,-precontracted human small pulmonary arteries were not significantly different at any concentration. The -log EC,, value were 3.75 + 0.10 (n = 4) and 4.0 + 0.21 (n = 4) for 5-HT and PGF,,-precontracted vessels, respectively (P > 0.05), indicating that ligustrazine is not a 5-HT or PGF,, antagonist. Preincubating human small pulmonary arteries with propranolol (1 PM) or indomethacin (10 PM) similarly had no significant effect on the vasodilator action of ligustrazine. The -log EC,, values were 3.96 kO.16 (n = 9) and 3.90 + 0.10 (n = 8), for control and propranolol pretreated vessels, and 3.85 _+0.13 (n = 4) and 3.72 f. 0.18 (n = 4) for control and indomethacin-pretreated vessels, suggesting that neither /?-adrenoceptors nor vasodilator cyclooxygenase products were involved in the ligustrazine-induced vasodilatation.
11 “I
5.0
405
‘I.0
3,s
3.0
2.5
2,o
1.7
1cOc121, t-log MI
Fig. 4. Effects of ligustrazine on concentration-response curves of rat aortic rings bathed in Ca 2+-free K +-depolarizing medium to CaCl,. Cumulative concentration-response curves to Ca2+ were constructed after incubating the rings with 0.3 (0). 1 (A) and 3 (e) mM ligustrazine (n = 6 each group). ( Tension is expressed as the percentage of CaCI, maximal contraction. Data represent mean values with vertical bars indicating S.E.M. * P -z 0.05 compared with cotrol group.
3.5. Effects of ligustrazine on CaCI,-induced traction
con-
In rat aortic rings bathed in Ca”+-free K+-depolarizing medium, ligustrazine shifted the CaCl, concentration-response curve significantly to right and reduced the maximal response to CaCl, in concentration-dependent manner (fig. 4). Thus, the -log EC,, values were 3.71, 3.47, 2.96 and 2.17 for control group and other groups pretreated with 0.3, 1 and 3 mM of ligustrazine.
4. Discussion
Our data demonstrate that ligustrazine is a dilator of both human pulmonary and bronchial arteries, which is endothelium-independent and that ligustrazine preferentially relaxes human small pulmonary resistance arteries in comparison with large conduit arteries. Ligustrazine appears to have a calcium antagonistic property.
Our conclusion of ligustrazine preferentially relaxing small pulmonary arteries is supported by our sodium nitroprusside studies in which we showed that the potency of sodium nitroprusside in relaxing small and lobar human pulmonary arteries was not significantly different using the same method. By contrast, nitrate vasodilators have been demonstrated to act preferentially on large coronary arteries (Winbmy et al., 1969; Schnaar and Sparks, 1972). Although several vasodilators have been demonstrated to have different effects on large and small systemic arteries (Winbury et al., 1969; Norton and Detar, 1970; Schnaar and Sparks, 1972), our data show that ligustrazine relaxes human small pulmonary arteries to a significantly greater extent than lobar pulmonary arteries, in what we believe to be the first demonstration of this phenomenon in human pulmonary vessels. These results are consistant with our previous data showing that ligustrazine is significantly more potent in relaxing intrapuhnonary arteries than extrapulmonary arteries in the rat (Liu et al., 1989). Hypoxic pulmonary vasoconstriction occurs predominantly in small pulmonary resistance arteries (Grover et al., 1983; Fishman, 1985) and may contribute to pulmonary hypertension (Voelkel and Weir, 1989). The characteristic action of ligustrazine in predominantly relaxing small pulmonary resistance arteries raises the possibility that ligustrazine may become a useful drug in the treatment of pulmonary hypertension when reduction of the pulmonary blood pressure is needed. Indeed, it has been reported that ligustrazine causes dose-dependent inhibition of hypoxic vasoconstriction in isolated blood perfused rat lungs and the right lower lobe of ferret lungs (Cai and Suggett, 1988; Cai et al., 1988). Furthermore, when given chronically (80 mg/kg i.p.) to rats exposed to 10% 0, for 2 weeks, ligustrazine attenuates the muscularization of pulmonary arterioles and right ventricular hypertrophy (Cai and Barer, 1989) which are the typical pathological features of pulmonary hypertension. It has been reported that ligustrazine reduces pulmonary blood pressure more than systemic blood pressure in ferrets (Cai et al., 1988) but our results indicate that ligustrazine is equally potent
in causing vasodilatation of human pulmonary and bronchial arteries. The difference between e.~ results could be explained by a species difference. ~4d~tion~Iy. the bronchial arteries may be functionally different from the systemic arteries of other regions and the response to ligustrazine may therefore be different from that of systemic arteries of other regions. In studying the possible mechanisms by which ligustrazine causes vasodilatation, we investigated whether ligustrazine may interact with a specialcaused by ised receptor. The vasodilatation ligustrazine was not significantly different in vessels precontracted with FGF,, from those preconn-acted with S-HT. In addition, preincubation of the pulmonary artery rings with propranolol and indomethacin had no significant effect on the reIaxant responses to ligustrazine. These results indieates that ligustrazine is not a PGF,, or 5-HT antagonist and that neither &adrenoceptors nor the release of vasodilator cyclooxygenase products were involved in its vasodilator action. The concentration-dependent inhibition of Ca’ ‘-induced contraction in calcium-free K+-depolarizing medium suggests that ligustrazine has a calcium antagonistic property. Although ligustrazinc itself is not potent, its differential effect on small pulmonary arteries is of potential clinical value and it may provide the basis for the development of new pulmonary vasodilator in the future.
Acknowledgements We thank the British Heart Foundation and Chest Heart and Stroke Foundation (U.K.) for financial support.
References Beijing institute of Pharmaceutical Industries. 1977. Extraction, isolation and structural identification of ligustrazine, Chung Hua I Hsueh Tsa Chih 57.420 (Chinese). Cai, Y. and G.R. Barer, 1989, Effects of ligustrazine on pulmonary vascular changes induced by chronic hypoxia in rats, Clin. Sci. 77. 515. Cai. Y.. D. Bee and G.R. Barer, 1988, Pulmonary vascular action of ligustrazine. principle of a traditional Chinese remedy, Clin. Sci. 75 (Suppl. 19). 55P. Cai. Y. and A.J. Suggett. 1988. Pulmonary vasodilator action of ligustrazine, active principle of a traditional Chinese remedy. J. Physiol. 399, 45P. Fishman. A.P., 1985, Pulmonary circulation, in: Handbook of Physiology, the Respiratory System, Section 3, Volume 1, eds. A.P. Fishman. A.B. Fisher and S.R. Geiger (Waverly Press, Baltimore) p. 93. Grover, B.F.. W.W. Wagner, Jr., LF. McMurtry and J.T. Reeves, 1983, Pulmonary circulation, in: Handbook of Physiology, the Cardiovascular System, Volume 3. eds. J.T. Shepherd, F.M. Abboud and S.R. Geiger (Waverly Press, Baltimore) p. 65. Hyman, A-L. and P.J. Kadowitz, 1984. Vasodilator therapy for pulmonary hypertensive disorders, Chest 85, 145. Liu. S.F.. L.Y. Rui, Y.N. Cai. P.J. Barnes. T.W. Evans and G.R. Barer, 1989, Dilator action of ligustrazine on human and rat pulmonary arteries, Clin. Sci. 77 (Suppl.), 27P. Norton, J.M. and R. Detar, 1970, Adenosine and isolated coronary vascular smooth muscle, Physiologist 13, 273. Packer, M.. B. Greenberg, B. Massie and H. Dash, 1982, Deleterious effects of hydralazine in patients with pulmonary hypertension, N. Engl. J. Med. 306, 1326. Schnaar, R.L. and H.V. Sparks. 1972, Response of large and small coronary arteries to nitroglycerin, NaNOr. and adenosine, Am. J. Physiol. 223, 223. Voelkel, N.F. and E.K. Weir, 1989. Etiologic mechanism in primary pulmonary hypertension, in: Pulmonary Vascular Physiology and Pathophysiology, eds. E.K. Weir and J.T. Reeves (Marcel Dekker, New York) p. 513. Winbury, M.M., B.B. Howe and M.A. Hefner, 1969, Effect of nitrates and other coronary dilators on large and small coronary vessls, J. Pharmacol. Exp. Ther. 168, 70.