Mechanisms of Relaxant Action of a Pyranocoumarin from Peucedanum japonicum in Isolated Rat Thoracic Aorta

• Abstract
• Introduction
• Materials and Methods
• Extraction and isolation
• Preparation of ring segments from rat thoracic aorta
• Vasorelaxation
• Assessment of the role of the endothelium
• Pharmacological characterization of 1-evoked relaxation
• Assessment of the involvement of Ca²⁺ channel blockade
• Chemicals
• Data analysis
• Results
• Discussion
• Acknowledgements
• References

Abstract

The CHCl₃-soluble fraction obtained from the MeOH extract of Peucedanum japonicum Thunb. inhibited phenylephrine-induced vasoconstriction in isolated rat thoracic aorta. We isolated a vasorelaxing compound, as one of the bioactive components, which was identified as (+)-cis-4′-O-acetyl-3′-O-angeloylkhellactone (1), a pyranocoumarin, and examined the mechanisms of vasorelaxant effect caused by 1. This compound (1) (10^-6 - 10^-4 M) concentration-dependently relaxed the isolated rat thoracic aorta pre-contracted with phenylephrine (PE). This vasorelaxant potency was diminished by endothelial removal (by 20 %), L-N ^G-nitro-arginine or methylene blue (MB), but not indomethacin treatment. These findings indicate that the vasorelaxant effect of 1 was partially endothelium dependent and mediated by nitric oxide and cyclic GMP pathway. To determine if the effect of 1 was mediated through the activation of some of the receptors known to lead to vascular relaxation, the effects of atropine, triprolidine and propranolol were determined. 1-induced vasorelaxation was not affected by atropine, triprolidine and propranolol. Compound 1 inhibited high potassium (80 mM)-induced, calcium-dependent contractions in a concentration-dependent manner. But it slightly relaxed the rat aorta precontracted with PE in the presence of nifedipine, a blocker of voltage-operated calcium channels. Tetraethylammonium (TEA, a non-specific K⁺ channel blocker) did not affect the vasodilatory effect of 1 against PE-induced contraction. Mechanisms of the vasorelaxant effect of 1 were multiple, including endothelium dependence and Ca²⁺ channel blockade.

Introduction

Increase in cytosolic Ca²⁺ is essential for almost fundamental cellular processes. These Ca²⁺ increases are the consequence of Ca²⁺ release from intracellular Ca²⁺ stores and of Ca²⁺ influx through more or less specific plasmalemmal ion channels [1]. These complex processes are under continuous debate. Ca²⁺ is released from endoplasmic reticulum, mitochondria, nucleus [2], and Golgi apparatus [3] in response to inositol (1,4,5)-trisphosphate or to Ca²⁺ inophores, etc.

In addition to Ca²⁺ release from intracellular stores, Ca²⁺ entry is also essential for most sustained cellular responses. In excitable tissue, such as smooth muscle, Ca²⁺ influx may be operated by (a) membrane receptors; (b) second messengers; (c) voltage (voltage-dependent Ca²⁺ channels); (d) any conjectural association [4].

Vascular endothelium plays an important role in the response of arterial muscles [5]. Endothelium-dependent mechanisms are involved in the actions of many vasodilating compounds [6]. These compounds mostly interact, through their respective receptors, with the endothelium to cause the release of either endothelium-derived relaxing factor (EDRF) which was identified as nitric oxide by Furchgott et al. [4] and Palmer et al. [7] or prostacyclin, which then exerts an inhibitory effect on vascular smooth muscle tone through the activation guanylyl cyclase or adenylyl cyclase, respectively [4], [5].

Peucedanum japonicum Thunb. (Umbelliferae) is widely distributed in Korea, Japan, and China. The root of this plant has been used as a folk medicine in the treatment of cold, cough, and headache. Several coumarins have been isolated from the root of this plant [8], [9], [10]. In our continuing search for bioactive natural products, we isolated a pyranocoumarin, (+)-cis-4′-acetyl-3′-angeloylkhellactone (1). In this paper, we examined the vasorelaxant effect of 1 on rat thoracic aorta and tried to elucidate its mode of action.

Materials and Methods

Extraction and isolation

The roots of Peucedanum japonicum were purchased from herbal store in Taejon, Korea. The authenticity of the material was confirmed by K. H. Bae, College of Pharmacy, Chungnam National University. The voucher specimen (PB013-006) has been deposited in the Korea Plant Extract Bank, Korea Research Institute of Bioscience and Biotechnology.

The dried, crushed roots (1.2 kg) of P. japonicum were extracted with MeOH (10 L × 2 times) at room temperature. The combined MeOH extract were concentrated in vacuo to give viscous extract (111 g). A portion (100 g) of the MeOH extract was suspended in distilled H₂O, extracted with CHCl₃. The CHCl₃-soluble fractions were concentrated in vacuo to give 50 g, which chromatographed on silica gel column (1,000 g, 230 - 400 mesh, Merck) eluted with the increasing concentrations of EtOAc in hexane (hexane/EtOAc, 10 : 1→ 3 : 1). The elution gave 16 fractions (each of 1,000 ml) and each fraction was monitored for a vasorelaxing effect on phenylephrine-induced vasoconstriction in isolated rat thoracic aorta. The two major vasorelaxing fractions (9.8 g, Fr. 9 - Fr. 10) were rechromatographed on Sephadex LH-20 eluting mixture of solvents (CHCl₃/ MeOH = 1 : 1) and LiChroprep RP-18 Lobar column (25 × 310 mm, 40 - 63 μm) eluting in a stepwise manner with MeOH-H₂O (40 : 60, 50 : 50, 60 : 40, 70 : 30). Finally, the active fraction was purified by semi-preparative HPLC (column: J’sphere ODS-H80, 4 μm, 150 × 20 mm I.D., flow rate; 7 ml/min, UV detector; 254 nm, solvent; 70 % MeOH) to give pure a compound (1, 34 mg) that inhibited phenylephrine-induced vasoconstriction.

(+)-cis-4′-Acetyl-3′-angeloylkhellactone (1): White crystals; [α]_D ²⁰: + 54.4 ° (c = 1.0, CHCl₃); (FAB-MS, m/z = 387, [M+H]⁺, C₂₁H₂₂O₇. ¹H-NMR, ¹³C-NMR, and MS data, consistent with literature values reported previously [9], [10].

Preparation of ring segments from rat thoracic aorta

Male Sprague-Dawley rats weighing 200 - 250 g, purchased from Jeil animal Laboratories (Anseong, Korea) were housed in an air-conditioned room with a 12 h reverse light-dark cycle and fed Purina mouse chow and tap water ad libitum. To preserve the functional integrity of the endothelium, we used ring segments rather than helical strips. Male Sprague-Dawley rats (200 - 250 g) were killed by stunning and exsanguination. The thoracic aorta was then excised rapidly. The tissue was cleaned of connective tissues and cut into rings of ∼ 4 mm long. The rings were maintained in Krebs solution (in mM): NaCl 118, KCl 4.6, CaCl₂ 2.5, MgSO₄ 1.2, KH₂PO₄ 1.2, NaHCO₃ 24, and glucose 11 aerated with 95 % O₂/5 % CO₂ and kept at 37 °C for 60 - 90 min for equilibration. Contractions were measured by suspending the tissue rings between two stainless-steel hooks in an organ bath with one hook attached to the end of a fixed support rod and the other to a isometric transducer (Havard App LTD, 50 - 7905) connected to a Havard Oscilloscophy. The tissues were left to equilibrate further under 1.5 - 2.0 g resting tension for 30 - 60 min before the experiments were started.

Vasorelaxation

For the assessment of vasorelaxation property and potency, the thoracic aorta was first contracted to 80 % of maximal contraction by phenylephrine (PE, 3.0 × 10^-6 M for the arteries with endothelium and 10^-6 M for preparations without endothelium). In each case, the initial tensions were adjusted to similar magnitudes so that different levels of vascular muscle tones would not complicate the effects of these conditions or drug magnitudes on the effects of 1. After the contractile response was stabilized at the plateau level (∼10 min), varying concentrations of 1 were cumulatively added to the preparation for the construction of concentration-response curves (CRC). Responses were recorded on a physiograph (Havard Oscilloscophy). Compound 1-evoked inhibitory responses were expressed as percentages of relaxation. The effect of corresponding concentrations of the dissolution vehicle dimethyl sulfoxide (DMSO) alone was examined in the various environments. Each preparation was used for only one set of experiments in which a reference compound (1) dose-response curve was first established, the arterial ring was washed with Krebs’ solution at intervals of 20 min for 1.5 - 2 h and treated with a drug, and the effects of 1 were examined again. The contractile forces evoked by PE before and after drug treatments were adjusted to similar magnitudes.

Assessment of the role of the endothelium

We studied the involvement of the endothelium by comparing the effects of 1 in endothelium-intact and endothelium-denuded preparations. The endothelium was removed by carefully threading a cotton ball through the lumen of the thoracic aorta ring and gently passing it back and forth a few times. Loss of functional integrity of the endothelium was confirmed by the loss of relaxant response to 10^-6 M acetylcholine (ACh).

In separate experiments, the relaxation pattern of 1 on PE-induced pre-constriction was first established. To define the particular pathways and mediators involved, the tissue was then treated with L-N ^G-nitro-arginine, a nitric oxide (NO) synthase inhibitor [11], methylene blue (MB), a guanylyl cyclase inhibitor, and indomethacin, a cyclooxygenase inhibitor. A second contraction of similar magnitude was then produced by PE before treatment again with 1 to ensure that inhibitor-induced changes in vasorelaxant effect of 1 would not be complicated by different levels of VSM tone.

Pharmacological characterization of 1-evoked relaxation

 To determine if the effect of 1 was mediated through the activation of some of the receptors known to lead to vascular relaxation, the effects of atropine (muscarinic-receptor antagonist), triprolidine (histamine H₁ receptor antagonist), or propranolol (β-adrenoceptor antagonist), were determined. The ability of 1 to produce relaxation before and after 10 min receptor antagonist treatment was compared.

Assessment of the involvement of Ca²⁺ channel blockade

To assess the involvement of receptor-operated Ca²⁺ channel and voltage-operated Ca²⁺ channel blockade, we compared the effects of 1 on PE-induced contraction (in the presence of nifedipine, 2.0 × 10^-6 M) and high extracellular K⁺ (80 mM KCl) on endothelium-denuded preparations, and the relaxation effects of 1 were observed and compared. High K⁺ solution without Ca²⁺ was prepared by substituting NaCl in the solution with 80 mM KCl.

In another experiment, we studied the possible involvement of K⁺ channel activation. The effect of tetraethylammonium (TEA, K⁺ channel blocker) on the relaxant potency of 1 was studied. The tissue was pretreated with tetraethyl ammonium (TEA, 10^-2 M) for 1 h, and then observed the effects of 1 on PE-induced contraction.

Chemicals

Phenylephrine HCl (PE), acetylcholine HCl, L-N ^G-nitro-arginine, methylene blue (MB), indomethcin, atropine, triprolidine, DL-propranolol hydrochloride, tetraethyl ammonium (TEA), nifedipine and DMSO were purchased from Sigma Chemical. (+)-cis-4′-acetyl-3′-angeloylkhellactone (1) was prepared from the root of P. japonicum in our laboratory.

PE was dissolved in 0.9 % saline containing 0.1 % ascorbic acid and stored in a freezer. On the day of experiment, a final dilution of PE was made with Krebs solution. Indomethacin was dissolved in 0.1 % NaHCO₃. Compound 1 was dissolved in DMSO to a stock solution of 10^-1 M, and further dilution was made with Krebs’ solution. All other drugs were dissolved in Krebs’ solution.

Data analysis

The concentrations of drugs were expressed as final bath concentration. Results are expressed or plotted as the mean ± S.E.M. Student’s t-test was used for statistical analyses; P values of less than 0.05 were considered to be significant.

Results

The CHCl₃-soluble fraction obtained from the MeOH extract inhibited phenylephrine-induced vasoconstriction at 100 μg/ml. The CHCl₃-soluble fraction was chromatographed on silica gel eluted with increasing concentrations of EtOAc in hexane. The active fraction was further rechromatographed on Sephadex LH-20, Lichroprep RP-18 Lobar column and finally semi-preparative HPLC to yield 1. This compound was identified as (+)-cis-4′-acetyl-3′-angeloylkhellactone based on the comparison of its spectral data (Fig. [1]).

In the endothelium-intact thoracic aorta ring pre-constricted by PE, additions of 1 (10^-6 - 10^-4 M) to the incubation medium resulted in concentration-dependent relaxation. Difficulty with dissolution of 1 had precluded testing at higher concentrations. The onset of relaxation began 30 - 60 s after the addition of 1, reached a stable plateau at 10 - 12 min. Based on relaxation obtained by 1 (10^-4 M) as maximal response, the medium effective concentration (EC₅₀) was 1.78 × 10^-5 M (Fig. [2]). Recovery of the response to PE took 50 - 60 min after the solution was changed several times. No desensitization was observed with repeated exposure to 1 (data not shown).

We studied the possible effects of DMSO under the various experimental conditions. DMSO at the concentrations used (as much as 0.1 %) had no effect.

The relaxation induced by 1 was attenuated by endothelium removal. In endothelium-denuded rings, the 1 (3.0 × 10^-5 M)-induced relaxation of 63.7 % was significantly less than the 100 % inhibition induced by the same concentration of 1 in the endothelium-intact preparations (Fig. [2]).

As shown in Fig. [3]a, L-N ^G-nitro-arginine pretreatment shifted the concentration response curve (CRC) of 1 to the right, and the 1 (3.0 × 10^-5 M)-induced response was decreased to 54 %. 1-induced relaxation was also inhibited by treatment with MB (3 × 10^-5 M), a guanylyl cyclase inhibitor, for 10 min. The inhibitory effect of MB was higher than that of L-N ^G-nitro-arginine (L-NOARG). On the other hand, indomethacin treatment did not affect the effect of 1.

Compound 1-induced vasorelaxation was not affected by atropine (2 × 10^-7 M, 10 min), triprolidine (5 × 10^-6 M, 10 min) or propranolol (3 × 10^-6 M, 10 min) (Fig. [3]b). The relaxant potency of 1 in the ring pre-contracted by PE was significantly diminished by pre-treating with 2 μM of nifedipine [2 μM nifedipine completely blocked the high K⁺ (80 mM)-induced contraction] for 15 min (Table [1]). 2 μM of nifedipine attenuated the contractile responsibility to PE by half. In Ca²⁺-free Krebs’ solution containing 80 mM K⁺, cumulative addition of Ca²⁺ (0.03 to 3 mM) caused a stepwise increase of the contraction in rat aorta with endothelium and without endothelium. 1 (3.0 × 10^-6 - 3.0 × 10^-5 M) inhibited these contractions in a concentration-dependent manner (Fig. [4]). TEA (10^-2 M), a nonspecific K⁺ channel blocker, did not affect the vasodilatory effect of 1 against PE-induced contraction (data not shown).

Discussion

In order to evaluate the possible mechanisms of action underlying the vasorelaxant effect of 1, we explored several commonly recognized vasorelaxant mechanisms, including endothelium-related factors, calcium channels, and muscarinic, histaminergic and β-adrenoceptor activation and K⁺ channel activation, in the isolated thoracic aorta rings preparations.

Compound 1 relaxed PE-pre-contracted aortic rings concentration-dependently. Endothelium removal attenuated the response of 1 by ∼20 %, shifting the CRC to the right in an approximately parallel manner. This result suggests that the vasorelaxant effect of 1 is partially dependent on endothelium-related factors. Vasorelaxant substances generally believed to be released by the endothelium including EDRF, prostacyclin, endothelium-derived hyperpolarizing factor (EDHF), and possible other substances [12], [13]. In our study, L-N ^G-nitro-arginine (an effective NO synthase inhibitor) and MB (a guanylyl cyclase inhibitor) diminished the vasorelaxant potency of 1. Therefore, the vasorelaxant effect of 1 apparently was at least partially dependent on a functionally intact endothelium with NO and guanylyl cyclase. On the other hand, indomethacin, a cyclooxygenase inhibitor, had no effects on the action of 1, thus excluding the involvement of prostacyclin.

In an attempt to identify a specific receptor with which 1 might react to produce its action, we tested the effects of atropine, a muscarinic receptor antagonist; tripolidine, an H₁ histaminergic receptor antagonist, and propranolol, a β-adrenoceptor antagonist. None of these antagonists had any significant effects on the actions of 1. Therefore, the vasorelaxant effect apparently was not mediated by any of these receptors.

The present study demonstrated that 1 inhibited both the contractile responses to PE and high K⁺. High K⁺-induced and Ca²⁺-dependent contractions were suppressed by 1 in a concentration-dependent manner. It has been reported that high K⁺-induced contraction in smooth muscle is mediated by an increase in Ca²⁺ influx through voltage-operated Ca²⁺ channels [14]. Since 1 inhibited Ca²⁺-dependent contraction in high K⁺ medium, it may be a blocker of voltage-operated Ca²⁺ channels. On the other hand, the vasorelaxant effect of 1 in PE-pre-contracted aorta rings was significantly attenuated by the treatment of nifedipine (2 μM), which completely blocked high K⁺ (80 mM)-induced contraction. These findings indicate that 1 may be a voltage-operated Ca²⁺ channel blocker rather than a receptor-operated Ca²⁺ channel blocker. However, it has been reported that these two types of Ca²⁺ channels are functionally not completely separated in rat aorta [in case of rabbit aorta, voltage- and receptor-operated Ca²⁺ channels are independent] [15]. Thus, further experiments are needed to determine if 1 is a selective inhibitor of voltage- and receptor-operated Ca²⁺ channels in rabbit aorta.

 The possibility of K⁺ activation by 1 was investigated. TEA, a non-selective potassium channel blocker, had no significant effect on the vasorelaxant effect of 1, thus the involvement of K⁺ channels might be excluded. Meisheri and colleagues [16] described several properties for the identification of K⁺ channel activation, including attenuation by increased extracellular K⁺ and susceptibility to K⁺ channel blockers. Therefore, further experiments are needed to clarify the involvement of K⁺ activation.

We demonstrated that 1 induced relaxation in pre-contracted thoracic aorta of rat in a concentration-dependent manner. The underlying mechanisms were complex, however, involving interaction with the endothelium through the NO-guanylyl cyclase pathway, and voltage-operated Ca²⁺ channel blockade.

Acknowledgements

This research was supported by a grant (code #PF002101-00) from Plant Diversity Research Center of 21^st Century Frontier Research Program funded by Ministry of Science and Technology of Korean government.

References

• 1 Berridge M J. Capacitative calcium entry. Biochem.  J.. 1995;  312 1-11
  PubMedSearch in Google Scholar
• 2 Himpens B, De Smedt H, Cstells R. Kinetics of nucleo-cytoplasmic transients in DDT1MF-2 cells. Am. J.  Physiol.. 1992;  260 C978-85
  PubMedSearch in Google Scholar
• 3 Chandra S, Kable E PW, Morrison G H, Webb W W. Calcium sequestration in the Golgi apparatus of cultured mammalian cells revealed by laser scanning confocal microscopy and ion microscopy. J.  Cell Sci.. 1990;  100 742-52
  PubMedSearch in Google Scholar
• 4 Felder C C, Singer-Lahat D, Mathes C. Voltage-independent calcium channels. Regulation by receptor and intracellular calcium stores. Biochem.  Pharmacol.. 1994;  48 1997-2004
  PubMedSearch in Google Scholar
• 5 Vanhoutte P M. Endothelium and control of vascular function.  Hypertension.. 1989;  13 658-67
  PubMedSearch in Google Scholar
• 6 De Mey J G, Claeys M, Vonhoutte P M. Endothelium dependent inhibitory effects of acetylcholine, adenosine triphosphate, thrombin and arachidonic acid in the canine femoral artery. J. Pharmacol. Exp.  Ther.. 1982;  222 166-73
  PubMedSearch in Google Scholar
• 7 Palmer R MJ, Ferrige A G, Moneada S. Nitric oxide release accounts for the biological activity of endothelium-derived relaxing factor.  Nature. 1987;  327 524-6
  CrossrefPubMedSearch in Google Scholar
• 8 Duh C Y, Wang S K, Wu Y C. Cytotoxic pyranocoumarins from the aerial parts of Peucedanum japonicum .  Phytochemistry. 1991;  30 2812-4
  CrossrefPubMedSearch in Google Scholar
• 9 Ikeshiro Y, Mase I, Tomita Y. Dihydropyranocoumarins from roots of Peucedanum japonicum .  Phytochemistry.. 1992;  31 4303-6
  CrossrefPubMedSearch in Google Scholar
• 10 Ikeshiro Y, Mase I, Tomita Y. Coumarins glycosides from Peucedanum japonicum .  Phytochemistry. 1994;  35 1339-41
  CrossrefPubMedSearch in Google Scholar
• 11 Moore P K, Al-Swateh O A, Chong N WS, Evans R, Mirza-Zadeh S, Gibson A. L-N ^G-nitroarginine (L-NOARG) inhibits endothelium dependent vasodilatation in the rabbit aorta and perfused rat mesentery. Br. J.  Pharmacol.. 1989;  98 905-14
  PubMedSearch in Google Scholar
• 12 Beny J L, Brunet P C. Neither nitric oxide nor nitroglycerin accounts for all the characteristics of endothelially mediated vasodilatation of pig coronary arteries.  Blood Vessels. 1988;  25 3-9
  PubMedSearch in Google Scholar
• 13 Feletou M, Vanhoutte P M. Endothelium-dependent hyperpolarization of canine coronary smooth muscle. Br. J.  Pharmacol.. 1988;  93 515-4
  PubMedSearch in Google Scholar
• 14 Karaki H, Nakagawa H, Urakawa N. Comparative effects of verapamil and sodium nitroprusside on contraction and ⁴⁵Ca²⁺ uptake in the smooth muscle of rabbit aorta, rat aorta and guinea pig taenia coli. Br. J.  Pharmacol.. 1984;  81 393-4
  PubMedSearch in Google Scholar
• 15 Bolton T B. Mechanisms of action of transmitters and other substance on smooth muscle. Physiol.  Rev.. 1979;  59 607-18
  PubMedSearch in Google Scholar
• 16 Meisheri K D, Cipkus-Dubray L A, Oleynek A J. A sensitive in vitro functional assay to detect K⁺ channel-dependent vasodilators. J. Pharmacol. Exp.  Ther.. 1990;  24 251-61
  PubMedSearch in Google Scholar

Dr. Hyun Sun Lee

Cardiovascular Research Laboratory

Korea Research Institute of Bioscience and Biotechnology

P.O. Box 115, Yusong, Taejon 305-333

Korea

Email: leehs@mail.kribb.re.kr

Fax: +82-42-861-2675

Phone: +82-42-860-4314

References

• 1 Berridge M J. Capacitative calcium entry. Biochem.  J.. 1995;  312 1-11
  PubMedSearch in Google Scholar
• 2 Himpens B, De Smedt H, Cstells R. Kinetics of nucleo-cytoplasmic transients in DDT1MF-2 cells. Am. J.  Physiol.. 1992;  260 C978-85
  PubMedSearch in Google Scholar
• 3 Chandra S, Kable E PW, Morrison G H, Webb W W. Calcium sequestration in the Golgi apparatus of cultured mammalian cells revealed by laser scanning confocal microscopy and ion microscopy. J.  Cell Sci.. 1990;  100 742-52
  PubMedSearch in Google Scholar
• 4 Felder C C, Singer-Lahat D, Mathes C. Voltage-independent calcium channels. Regulation by receptor and intracellular calcium stores. Biochem.  Pharmacol.. 1994;  48 1997-2004
  PubMedSearch in Google Scholar
• 5 Vanhoutte P M. Endothelium and control of vascular function.  Hypertension.. 1989;  13 658-67
  PubMedSearch in Google Scholar
• 6 De Mey J G, Claeys M, Vonhoutte P M. Endothelium dependent inhibitory effects of acetylcholine, adenosine triphosphate, thrombin and arachidonic acid in the canine femoral artery. J. Pharmacol. Exp.  Ther.. 1982;  222 166-73
  PubMedSearch in Google Scholar
• 7 Palmer R MJ, Ferrige A G, Moneada S. Nitric oxide release accounts for the biological activity of endothelium-derived relaxing factor.  Nature. 1987;  327 524-6
  CrossrefPubMedSearch in Google Scholar
• 8 Duh C Y, Wang S K, Wu Y C. Cytotoxic pyranocoumarins from the aerial parts of Peucedanum japonicum .  Phytochemistry. 1991;  30 2812-4
  CrossrefPubMedSearch in Google Scholar
• 9 Ikeshiro Y, Mase I, Tomita Y. Dihydropyranocoumarins from roots of Peucedanum japonicum .  Phytochemistry.. 1992;  31 4303-6
  CrossrefPubMedSearch in Google Scholar
• 10 Ikeshiro Y, Mase I, Tomita Y. Coumarins glycosides from Peucedanum japonicum .  Phytochemistry. 1994;  35 1339-41
  CrossrefPubMedSearch in Google Scholar
• 11 Moore P K, Al-Swateh O A, Chong N WS, Evans R, Mirza-Zadeh S, Gibson A. L-N ^G-nitroarginine (L-NOARG) inhibits endothelium dependent vasodilatation in the rabbit aorta and perfused rat mesentery. Br. J.  Pharmacol.. 1989;  98 905-14
  PubMedSearch in Google Scholar
• 12 Beny J L, Brunet P C. Neither nitric oxide nor nitroglycerin accounts for all the characteristics of endothelially mediated vasodilatation of pig coronary arteries.  Blood Vessels. 1988;  25 3-9
  PubMedSearch in Google Scholar
• 13 Feletou M, Vanhoutte P M. Endothelium-dependent hyperpolarization of canine coronary smooth muscle. Br. J.  Pharmacol.. 1988;  93 515-4
  PubMedSearch in Google Scholar
• 14 Karaki H, Nakagawa H, Urakawa N. Comparative effects of verapamil and sodium nitroprusside on contraction and ⁴⁵Ca²⁺ uptake in the smooth muscle of rabbit aorta, rat aorta and guinea pig taenia coli. Br. J.  Pharmacol.. 1984;  81 393-4
  PubMedSearch in Google Scholar
• 15 Bolton T B. Mechanisms of action of transmitters and other substance on smooth muscle. Physiol.  Rev.. 1979;  59 607-18
  PubMedSearch in Google Scholar
• 16 Meisheri K D, Cipkus-Dubray L A, Oleynek A J. A sensitive in vitro functional assay to detect K⁺ channel-dependent vasodilators. J. Pharmacol. Exp.  Ther.. 1990;  24 251-61
  PubMedSearch in Google Scholar

Dr. Hyun Sun Lee

Cardiovascular Research Laboratory

Korea Research Institute of Bioscience and Biotechnology

P.O. Box 115, Yusong, Taejon 305-333

Korea

Email: leehs@mail.kribb.re.kr

Fax: +82-42-861-2675

Phone: +82-42-860-4314