268 Planta Med. 56(1990)
Inhibition of Lenkotriene and Platelet Activating Factor Synthesis in Leukocytes by the Sesquiterpene Lactone Scandenolide MafelC. Ysrael'andKevinD. Croft2'3 2
Research Centre for Natural Sciences, University of Santo Tomas, Espana, Manila, Philippines Department of Medicine, University of Western Australia, Royal Perth Hospital, 35 Victoria Square, Perth, Western Australia 6000 Address for correspondence
Abstract The sesquiterpene lactone scandenolide, isolated from the Philippines medicinal plant Mi/can/a cordata, at a dose of 100 RM completely inhibited whole blood chemiluminescence in response to the activators PMA and zymosan. In isolated inflammatory rat leukocytes this compound inhibited both leukotriene B4 and 5-
The prostaglandins and leukotrienes are known to be mediators of inflammation (4) and more recently the ether linked phospholipid, platelet activating factor (PAP), has also been implicated as a potent inflammatory mediator (5). We have found that scandenolide inhibits in a dose-dependent manner the formation of the lipoxygenase products leukotriene B4 (LTB4) and 5-hydroxyeico-
satetraenoic acid (5-HETE) as well as PAP in calcium
HETE production with IC50 of 15 RM and 30 pM, respec-
ionophore stimulated rat peritoneal leukocytes. In contrast the formation of the cyclooxygenase product thromboxane
tively. The formation of the cyclooxygenase product thromboxane B2 was not inhibited in the concentration
B2 (TXB2) was not inhibited by scandenolide.
range 10 to 200 sM of scandenolide. The formation of the
Materials and Methods
potent inflammatory mediator platelet activating factor (PAP) was suppressed by i.tM concentrations of scan-
Test compounds
denolide with an IC50 of< 20 RM. The presence of a com-
pound in M. cordata which inhibits some of the inflam-
matory mediators such as leukotrienes and PAP may explain at least in part some of its medicinal properties.
Key words
Scandenolide, isolated fromM. cordafa, purified by recrystallisation and identified hy spectroscopic methods (6), was provided by Mrs. A. Aguinaldo, Research Center, University of Santo Tomas. Coronopilin, isotated from Ambrosia psilosfachya, purified by recrystallisation (m.p. 179—180°C), was a gift from Dr.
E. Ghisalberti, Department of Organic Chemistry, University of Western Australia.
Mi/can/a cordata, sesquiterpene lactone, scandenolide, anti-inflammatory action.
Luminol-dependent chemi/uminescence Luminol-amplified chemiluminescence of whole
blood was measured using the luminescence channel of a Chronolog whole blood aggregometer according to the methods of
Introduction
The leaf of the plant Mi/can/a cordata (Burm.) B. L. Robinson, is used for medicinal purposes in
the Philippines (1). During a routine screening of compounds isolated from Philippine medicinal plants, a ses-
Kovacs et al. (7) and Bird and Giroud (2). Chemiluminescence (measured in mV) was induced by standard additions of either opsonized zymosan (0.6mg/ml) or phorhol myristate acetate (PMA, 10 pg/ml).
Ce//isolation and incubation
quiterpene lactone, scandenolide from M. cord ata showed potent inhibition ofluminol-dependent cherniluminescence of whole blood exposed to opsonized zymosan, a technique used to detect compounds with potential anti-inflammatory activity (2). Since a number of plants known to contain ses-
Rat peritoneal leukocytes were prepared according to the method ofTerano et al. (8) and more than 85% of exudate cells were neutrophils as assessed by size and differential staining using a Technicon Hi instrument. Pooled leukocytes were washed
quiterpene lactones have historically been used as an-
per ml. Cells were pre-incubated with test compounds in doses ranging from 10 to 200 sM for 15mm at 37°C prior to stimulation
tiinflammatory and antipyretic treatments (3), we decided to investigate the effect of scandenolide on lipid mediators released by inflammatory rat polymorphonuclear leukocytes.
and resuspended in Hepes-buffered Hank's solution containing 0.1 % bovine serum albumin (pH 7.4) at a concentration of io cells with the calcium ionophore A23187 (2.5 g/ml) for 10 mm. Control incubations were carried out with vehicle only (10 il of dimethyl
sulphoxide per I ml incubation). In cells preincubated with test compounds or vehicle there was no detectable release of LTB4 or PAP prior to stimulation with A23 187.
Downloaded by: Universite Laval. Copyrighted material.
Received: April 10, 1989
Planta Med. 56(1990) 269
Inhibition ofLeulcotriene and Platelet Activating Factor Synthesis in Leakocytes 100
Leukotriene and throinboxane analysis Prostaglandin B2 (PGB2, 50 ng) was added to cell-
free supernatant from the cell incubations as internal standard. 3 with 2N formic acid and the mixture extracted with ethyl acetate (2 ml). The organic solvent was evaporated under vacuum and the residue reconstituted in
The solution was acidified to p11 =
100 p1 of HPLC mobile phase. Analysis was carried out using a Wat-
ers 5 p "Nova Pak" C18 column (4 x 150 mm) with methanol: water: acetic acid (75 : 25 : 0.01 v/v), pH adjusted to 6.2 with NH4OH,
-J
C a: Iz C
0
0 40
as mobile phase at a flow rate ofo.8ml/min. UV ahsorbance was
20
monitored at 270 nm (LTB4 and PUB2) and 234nm (5-HETE) using
a Waters Lambda-Max Model 481 variable wavelength detector. LTB4 and 5-HETE were quantitated by comparison of peak areas with the internal standard and response factors were calculated by injection of a range of known amounts of PUB2, synthetic LTB4
20
00
100 140 00 120 SCANDENOLIDE CONC. (1iM)
40
100
100
200
(Miles Laboratories, Slough, UK), and 5-HETE (Cayman Chemical,
TXB2 antisera from Seragen. Samples were assayed neat and after serial dilution with phosphate buffered saline according to previously described methods (9).
Fig. 1 The effect of scandenolide an the formation of leukotriene B4 j), 5-HETE (U) and TXB2 (0) by ionophore-stimulated rat peritoneal leukocytes. Results are expressed as percent of control and represent the mean standard error of six experiments.
Platelet activating factor analysis The cell pellet obtained following centrifugation at 2000 g for 5 mm was extracted with 2 ml chloroform : methanol (2: 1). After evaporation of solvent, the lipid residue was dissolved in ethanol (50 il) and reconstituted in Hepes buffered Tyrodes solution containing 0.25% bovine serum albumin to a total volume of 1 ml. The extracted PAP was quanitated by bioassay using 14Cserotonin labelled rabbit platelets as previously described (10, 11).
Standard curves for the assay were constructed using authentic PAP (Sigma Chemicals), and results were obtained as ng per io
-J 0 a:
H z
0 0 0
cells.
Phospholipase A2 activity Leukocytes were prelabelled with [5,6,8,9,11,12,
14,15-3H]-arachidonic acid (4Onmol, specific activity 2lOCi/ mmol, Arnersham) by 60 mm incubation at 37°C in Hepes-buffered
Hanks' solution containing 0.1% bovine serum albumin. Leuko-
20
60
40
80
100
120
140
160
SCANOENOLIDE CONC. (vM)
cytes were washed three times and resuspended in fresh buffer before incubation with test compounds (10—100 pM) or vehicle con-
Fig. 2 The effect of scandenolide on the formation of platelet activating factor (PAF) by ionophore-stimulated rat peritoneal leukocytes. Results are expressed as percent of control and represent the mean standard error of
trol as described above. Following incubation with calcium
four duplicate experiments.
ionophore A23187 (2.5 pg) for 10mm, samples were centrifuged and the amount of label released into the supernatant was assessed by bquid scintillation counting.
Results Scandenolide at a concentration of 100 pM completely inhibited the luminol dependent chemilumines-
cence of whole blood exposed to opsonized zymosan or PMA.
In concentrations ranging from 10 to
Since most PAP formed in stimulated leuko-
cytes remains cell bound (12), measurements of this substance were made on organic extracts of the cell pellet. Formation of PAP in control incubations was 30 6 ng/107 cells. Scandenolide showed strong inhibition of PAP formation with an IC50 of c 20 pM and nearly complete inhibition at doses up to 100 pM (Fig. 2).
Similar experiments conducted with the
200 pM scandenolide showed a dose-dependent inhibition
sesquiterpene lactone coronopihn at the same doses
of the lipoxygenase products LTB4 and 5-HETE (Fig. 1) with an IC50 of 15 pM for LTB4 and 30 pM for 5-HETE. In contrast
showed no inhibition of either LTB4 or PAP production, with no significant difference between treatment and con-
inhibitory effects of scandenolide on the cyclooxygenase
trols at each concentration tested (up to 200 1.tM).
metabolite TXB2 were only slight over the entire concentration range tested (Fig. 1). Control incubations of leukocytes 3 ng/107 cells of in these experiments produced 62.7 LTB4, 50.0 11 ng/107 cells of 5-HETE and 43.0 5 ng/107 cells of TXB2. Results are quoted as % of the control incubation. There was only slight inhibition of phospholipase A2
activity with doses of scandenolide up to 100 pM. At this dose phospholipase activity remained at 60% of control values.
Downloaded by: Universite Laval. Copyrighted material.
Ann Arbor, USA). TXB2 was measured in the supernatant by a specific radioimmunoassay developed in this laboratory using
270 PlantaMed. 56 (1990)
The sesquiterpene lactones are a group of compounds known to have a wide range of biological activities (13). Plants containing these compounds have been used as anti-inflammatory remedies (3) and the mode of ac-
tion of a number of individual sesquiterpenes has been studied (14, 15). Sesquiterpene lactones such as helenalin have been shown to uncouple oxidative phosphorylation of neutrophils, inhibit lysosomal enzyme activity and neutrophil chemotactic migration. At relatively high concentrations (in the mM range) cyclooxygenase activity was sup-
pressed by about 30% (14). The a-methylene-y-lactone moiety was necessary for anti-inflammatory activity as well
as the /3-unsaturated cyclopentenone ring as found in helenalin. Our studies with scandenolide, which possesses some of these structural features, indicate that it is a potent inhibitor of lipoxygenase but not cyclooxygenase activity at RM concentrations. We presume that suppression of LTB4 and 5-HETE is due to direct effects on the 5-lipoxygenase enzyme rather than on phospholipase A2 since effects on the latter enzyme would cause a decrease in all arachidonic acid metaholites including TXB2. Inhibition of LTB4 maybe an important anti-inflammatory effect considering the potent chemotactic activity of this compound.
Scandenolide also suppressed the formation of the potent inflammatory mediator PAP, the first activity of this type attributed to a sesquiterpene lactone. Although the point at which PAP biosynthesis is inhibited remains to be elucidated, this activity of scandenolide maybe of biological relevance for anti-inflammatory effects. It is difficult for us to postulate on the struc-
tural requirements that may be necessary for leukotriene and PAP inhibition since only two sesquiterpene lactones have been tested in this assay system. However, sesquiterpene lactones with the reactive a-methylene-y-lactone moiety maybe expected to interact with enzymes and structural proteins giving rise to a number of biological and possibly toxic effects (13).
The hiochemical activity described here for one of the constituents of M. cordata may explain at least in part some of the reputed medicinal properties of this plant.
Acknowledgements We would like to thank Alicia Aguinaldo, Research Center, University of Santo Tomas, Manila, for the sample of scandenolide. Support of the Development Assistance Committee of the University of Western Australia and the Network for the Chemistry of Biologically-important Natural Products (Australian Development Assistance Bureau) is gratefully acknowledged.
References 1 Quisumbing, E. (1978) Medicinal Plants of the Philippines, Katha Publishing Company, Quezon City, Philippines. 2 Bird, J., Giroud, J. P. (1985)J. Pharmacol. Methods 14, 305. Hall, I. H.. Lee, K. H., Starnes, C. 0., Sumida, Y., Wu, R. Y., Waddell, 1. G., Cochran, J. W., Gerhart, K. G. (1979) J. Pharm. Sd. 68, 537. " Salmon, J. A., Higgs, G. A. (1987) Brit. Med. Bull. 43, 285.
Braquet, P., Touqul, L., Shen, T. Y., Vargaftig, B. B. (1987) Pharmacol. Rev. 39, 97. 6 Herz, W., Subramaniam, P. S., Santhanam, P. S., Aota, K., Hall, A. L. (1970) J. Org. Chem. 35, 1453. Kovacs, I. B., Meyrick Thomas, B. H., Mackay, A. H., Rustin, M. H. A., Kirby, J. D. T. (1986) Clin. Sd. 70, 257.
Terano, T., Salmon, J. A., Moncada, 5. (1984) Biochem. Pharmacol. 33, 3071. Mahoney, D. P., Barden, A., Beilin, L. J., Vandongen, It. (1983) Prost. Leuk. Med. 12,11. 10 Croft, K. 0., Codde, J. P., Barden, A., Vandongen, H., Beilin, L. J. (1988) Clin. Exp. Pharm. Physiol. 15, 517. Croft, K. 0., Sturm, M. J.. Codde, J. P., Vandongen, R., Beilin, L. J. (1986) Life Sciences 38, 1875. 12 Lynch, J. M., Henson, P.M. (1986)1 Immunol. 137, 2653.
13 lyle, G. W., Witzel, D. A. (1983) In: Encyclopedic Handbook of Natural Toxins, (Keeler, R., Tu, A., eds.), pp. 543—584, Marcel Dekker, New York. 14 Hall, I. H., Starnes, C. 0., Lee, K. H., Waddell, T. 6. (1980) J. Pharm. Sci. 69, 537. ' Hall, I. H., Lee, K. H., Sykes, H. C. (1987) Planta Med. 53, 153.
Downloaded by: Universite Laval. Copyrighted material.
Discussion
MafeiC. Ysraeletaf
Inhibition of Lenkotriene and Platelet Activating Factor Synthesis in Leukocytes by the Sesquiterpene Lactone Scandenolide MafelC. Ysrael'andKevinD. Croft2'3 2
Research Centre for Natural Sciences, University of Santo Tomas, Espana, Manila, Philippines Department of Medicine, University of Western Australia, Royal Perth Hospital, 35 Victoria Square, Perth, Western Australia 6000 Address for correspondence
Abstract The sesquiterpene lactone scandenolide, isolated from the Philippines medicinal plant Mi/can/a cordata, at a dose of 100 RM completely inhibited whole blood chemiluminescence in response to the activators PMA and zymosan. In isolated inflammatory rat leukocytes this compound inhibited both leukotriene B4 and 5-
The prostaglandins and leukotrienes are known to be mediators of inflammation (4) and more recently the ether linked phospholipid, platelet activating factor (PAP), has also been implicated as a potent inflammatory mediator (5). We have found that scandenolide inhibits in a dose-dependent manner the formation of the lipoxygenase products leukotriene B4 (LTB4) and 5-hydroxyeico-
satetraenoic acid (5-HETE) as well as PAP in calcium
HETE production with IC50 of 15 RM and 30 pM, respec-
ionophore stimulated rat peritoneal leukocytes. In contrast the formation of the cyclooxygenase product thromboxane
tively. The formation of the cyclooxygenase product thromboxane B2 was not inhibited in the concentration
B2 (TXB2) was not inhibited by scandenolide.
range 10 to 200 sM of scandenolide. The formation of the
Materials and Methods
potent inflammatory mediator platelet activating factor (PAP) was suppressed by i.tM concentrations of scan-
Test compounds
denolide with an IC50 of< 20 RM. The presence of a com-
pound in M. cordata which inhibits some of the inflam-
matory mediators such as leukotrienes and PAP may explain at least in part some of its medicinal properties.
Key words
Scandenolide, isolated fromM. cordafa, purified by recrystallisation and identified hy spectroscopic methods (6), was provided by Mrs. A. Aguinaldo, Research Center, University of Santo Tomas. Coronopilin, isotated from Ambrosia psilosfachya, purified by recrystallisation (m.p. 179—180°C), was a gift from Dr.
E. Ghisalberti, Department of Organic Chemistry, University of Western Australia.
Mi/can/a cordata, sesquiterpene lactone, scandenolide, anti-inflammatory action.
Luminol-dependent chemi/uminescence Luminol-amplified chemiluminescence of whole
blood was measured using the luminescence channel of a Chronolog whole blood aggregometer according to the methods of
Introduction
The leaf of the plant Mi/can/a cordata (Burm.) B. L. Robinson, is used for medicinal purposes in
the Philippines (1). During a routine screening of compounds isolated from Philippine medicinal plants, a ses-
Kovacs et al. (7) and Bird and Giroud (2). Chemiluminescence (measured in mV) was induced by standard additions of either opsonized zymosan (0.6mg/ml) or phorhol myristate acetate (PMA, 10 pg/ml).
Ce//isolation and incubation
quiterpene lactone, scandenolide from M. cord ata showed potent inhibition ofluminol-dependent cherniluminescence of whole blood exposed to opsonized zymosan, a technique used to detect compounds with potential anti-inflammatory activity (2). Since a number of plants known to contain ses-
Rat peritoneal leukocytes were prepared according to the method ofTerano et al. (8) and more than 85% of exudate cells were neutrophils as assessed by size and differential staining using a Technicon Hi instrument. Pooled leukocytes were washed
quiterpene lactones have historically been used as an-
per ml. Cells were pre-incubated with test compounds in doses ranging from 10 to 200 sM for 15mm at 37°C prior to stimulation
tiinflammatory and antipyretic treatments (3), we decided to investigate the effect of scandenolide on lipid mediators released by inflammatory rat polymorphonuclear leukocytes.
and resuspended in Hepes-buffered Hank's solution containing 0.1 % bovine serum albumin (pH 7.4) at a concentration of io cells with the calcium ionophore A23187 (2.5 g/ml) for 10 mm. Control incubations were carried out with vehicle only (10 il of dimethyl
sulphoxide per I ml incubation). In cells preincubated with test compounds or vehicle there was no detectable release of LTB4 or PAP prior to stimulation with A23 187.
Downloaded by: Universite Laval. Copyrighted material.
Received: April 10, 1989
Planta Med. 56(1990) 269
Inhibition ofLeulcotriene and Platelet Activating Factor Synthesis in Leakocytes 100
Leukotriene and throinboxane analysis Prostaglandin B2 (PGB2, 50 ng) was added to cell-
free supernatant from the cell incubations as internal standard. 3 with 2N formic acid and the mixture extracted with ethyl acetate (2 ml). The organic solvent was evaporated under vacuum and the residue reconstituted in
The solution was acidified to p11 =
100 p1 of HPLC mobile phase. Analysis was carried out using a Wat-
ers 5 p "Nova Pak" C18 column (4 x 150 mm) with methanol: water: acetic acid (75 : 25 : 0.01 v/v), pH adjusted to 6.2 with NH4OH,
-J
C a: Iz C
0
0 40
as mobile phase at a flow rate ofo.8ml/min. UV ahsorbance was
20
monitored at 270 nm (LTB4 and PUB2) and 234nm (5-HETE) using
a Waters Lambda-Max Model 481 variable wavelength detector. LTB4 and 5-HETE were quantitated by comparison of peak areas with the internal standard and response factors were calculated by injection of a range of known amounts of PUB2, synthetic LTB4
20
00
100 140 00 120 SCANDENOLIDE CONC. (1iM)
40
100
100
200
(Miles Laboratories, Slough, UK), and 5-HETE (Cayman Chemical,
TXB2 antisera from Seragen. Samples were assayed neat and after serial dilution with phosphate buffered saline according to previously described methods (9).
Fig. 1 The effect of scandenolide an the formation of leukotriene B4 j), 5-HETE (U) and TXB2 (0) by ionophore-stimulated rat peritoneal leukocytes. Results are expressed as percent of control and represent the mean standard error of six experiments.
Platelet activating factor analysis The cell pellet obtained following centrifugation at 2000 g for 5 mm was extracted with 2 ml chloroform : methanol (2: 1). After evaporation of solvent, the lipid residue was dissolved in ethanol (50 il) and reconstituted in Hepes buffered Tyrodes solution containing 0.25% bovine serum albumin to a total volume of 1 ml. The extracted PAP was quanitated by bioassay using 14Cserotonin labelled rabbit platelets as previously described (10, 11).
Standard curves for the assay were constructed using authentic PAP (Sigma Chemicals), and results were obtained as ng per io
-J 0 a:
H z
0 0 0
cells.
Phospholipase A2 activity Leukocytes were prelabelled with [5,6,8,9,11,12,
14,15-3H]-arachidonic acid (4Onmol, specific activity 2lOCi/ mmol, Arnersham) by 60 mm incubation at 37°C in Hepes-buffered
Hanks' solution containing 0.1% bovine serum albumin. Leuko-
20
60
40
80
100
120
140
160
SCANOENOLIDE CONC. (vM)
cytes were washed three times and resuspended in fresh buffer before incubation with test compounds (10—100 pM) or vehicle con-
Fig. 2 The effect of scandenolide on the formation of platelet activating factor (PAF) by ionophore-stimulated rat peritoneal leukocytes. Results are expressed as percent of control and represent the mean standard error of
trol as described above. Following incubation with calcium
four duplicate experiments.
ionophore A23187 (2.5 pg) for 10mm, samples were centrifuged and the amount of label released into the supernatant was assessed by bquid scintillation counting.
Results Scandenolide at a concentration of 100 pM completely inhibited the luminol dependent chemilumines-
cence of whole blood exposed to opsonized zymosan or PMA.
In concentrations ranging from 10 to
Since most PAP formed in stimulated leuko-
cytes remains cell bound (12), measurements of this substance were made on organic extracts of the cell pellet. Formation of PAP in control incubations was 30 6 ng/107 cells. Scandenolide showed strong inhibition of PAP formation with an IC50 of c 20 pM and nearly complete inhibition at doses up to 100 pM (Fig. 2).
Similar experiments conducted with the
200 pM scandenolide showed a dose-dependent inhibition
sesquiterpene lactone coronopihn at the same doses
of the lipoxygenase products LTB4 and 5-HETE (Fig. 1) with an IC50 of 15 pM for LTB4 and 30 pM for 5-HETE. In contrast
showed no inhibition of either LTB4 or PAP production, with no significant difference between treatment and con-
inhibitory effects of scandenolide on the cyclooxygenase
trols at each concentration tested (up to 200 1.tM).
metabolite TXB2 were only slight over the entire concentration range tested (Fig. 1). Control incubations of leukocytes 3 ng/107 cells of in these experiments produced 62.7 LTB4, 50.0 11 ng/107 cells of 5-HETE and 43.0 5 ng/107 cells of TXB2. Results are quoted as % of the control incubation. There was only slight inhibition of phospholipase A2
activity with doses of scandenolide up to 100 pM. At this dose phospholipase activity remained at 60% of control values.
Downloaded by: Universite Laval. Copyrighted material.
Ann Arbor, USA). TXB2 was measured in the supernatant by a specific radioimmunoassay developed in this laboratory using
270 PlantaMed. 56 (1990)
The sesquiterpene lactones are a group of compounds known to have a wide range of biological activities (13). Plants containing these compounds have been used as anti-inflammatory remedies (3) and the mode of ac-
tion of a number of individual sesquiterpenes has been studied (14, 15). Sesquiterpene lactones such as helenalin have been shown to uncouple oxidative phosphorylation of neutrophils, inhibit lysosomal enzyme activity and neutrophil chemotactic migration. At relatively high concentrations (in the mM range) cyclooxygenase activity was sup-
pressed by about 30% (14). The a-methylene-y-lactone moiety was necessary for anti-inflammatory activity as well
as the /3-unsaturated cyclopentenone ring as found in helenalin. Our studies with scandenolide, which possesses some of these structural features, indicate that it is a potent inhibitor of lipoxygenase but not cyclooxygenase activity at RM concentrations. We presume that suppression of LTB4 and 5-HETE is due to direct effects on the 5-lipoxygenase enzyme rather than on phospholipase A2 since effects on the latter enzyme would cause a decrease in all arachidonic acid metaholites including TXB2. Inhibition of LTB4 maybe an important anti-inflammatory effect considering the potent chemotactic activity of this compound.
Scandenolide also suppressed the formation of the potent inflammatory mediator PAP, the first activity of this type attributed to a sesquiterpene lactone. Although the point at which PAP biosynthesis is inhibited remains to be elucidated, this activity of scandenolide maybe of biological relevance for anti-inflammatory effects. It is difficult for us to postulate on the struc-
tural requirements that may be necessary for leukotriene and PAP inhibition since only two sesquiterpene lactones have been tested in this assay system. However, sesquiterpene lactones with the reactive a-methylene-y-lactone moiety maybe expected to interact with enzymes and structural proteins giving rise to a number of biological and possibly toxic effects (13).
The hiochemical activity described here for one of the constituents of M. cordata may explain at least in part some of the reputed medicinal properties of this plant.
Acknowledgements We would like to thank Alicia Aguinaldo, Research Center, University of Santo Tomas, Manila, for the sample of scandenolide. Support of the Development Assistance Committee of the University of Western Australia and the Network for the Chemistry of Biologically-important Natural Products (Australian Development Assistance Bureau) is gratefully acknowledged.
References 1 Quisumbing, E. (1978) Medicinal Plants of the Philippines, Katha Publishing Company, Quezon City, Philippines. 2 Bird, J., Giroud, J. P. (1985)J. Pharmacol. Methods 14, 305. Hall, I. H.. Lee, K. H., Starnes, C. 0., Sumida, Y., Wu, R. Y., Waddell, 1. G., Cochran, J. W., Gerhart, K. G. (1979) J. Pharm. Sd. 68, 537. " Salmon, J. A., Higgs, G. A. (1987) Brit. Med. Bull. 43, 285.
Braquet, P., Touqul, L., Shen, T. Y., Vargaftig, B. B. (1987) Pharmacol. Rev. 39, 97. 6 Herz, W., Subramaniam, P. S., Santhanam, P. S., Aota, K., Hall, A. L. (1970) J. Org. Chem. 35, 1453. Kovacs, I. B., Meyrick Thomas, B. H., Mackay, A. H., Rustin, M. H. A., Kirby, J. D. T. (1986) Clin. Sd. 70, 257.
Terano, T., Salmon, J. A., Moncada, 5. (1984) Biochem. Pharmacol. 33, 3071. Mahoney, D. P., Barden, A., Beilin, L. J., Vandongen, It. (1983) Prost. Leuk. Med. 12,11. 10 Croft, K. 0., Codde, J. P., Barden, A., Vandongen, H., Beilin, L. J. (1988) Clin. Exp. Pharm. Physiol. 15, 517. Croft, K. 0., Sturm, M. J.. Codde, J. P., Vandongen, R., Beilin, L. J. (1986) Life Sciences 38, 1875. 12 Lynch, J. M., Henson, P.M. (1986)1 Immunol. 137, 2653.
13 lyle, G. W., Witzel, D. A. (1983) In: Encyclopedic Handbook of Natural Toxins, (Keeler, R., Tu, A., eds.), pp. 543—584, Marcel Dekker, New York. 14 Hall, I. H., Starnes, C. 0., Lee, K. H., Waddell, T. 6. (1980) J. Pharm. Sci. 69, 537. ' Hall, I. H., Lee, K. H., Sykes, H. C. (1987) Planta Med. 53, 153.
Downloaded by: Universite Laval. Copyrighted material.
Discussion
MafeiC. Ysraeletaf
