Prostaglandins, Leukotrienes and Essential FattyAcids (2001) 65(5&6), 281–286 & 2001 Harcourt Publishers Ltd doi:10.1054/plef.2001.0326, available online at http://www.idealibrary.com on
Flavonoids differentially inhibit guinea pig epidermal cytosolic phospholipase A2 H. P. Kim,1 H. T. Pham,2 V. A. Ziboh2 1
College of Pharmacy, Kangweon National University, Korea Department of Dermatology, School of Medicine, University of California, USA
2
Summary Cytosolic phospholipase A2 (cPLA2) is believed to involve the regulation of essential cellular processes. Like other cell types, epidermal cPLA2 may participate in various metabolic processes including eicosanoid generation. In this investigation, we demonstrated the presence of cPLA2 in guinea pig epidermis.The epidermal cPLA2 is Ca2+-dependent, active at micromolar concentration of Ca2+ and resistant to disulfide-reducing agents. Furthermore, it is inhibited by methyl arachidonyl fluorophosphonate (MAFP), a selective inhibitor of cPLA2, while12-epi-scalardial (a sPLA2 inhibitor) did not cause inhibition. A test of several flavonoids revealed that quercetin (flavonol) weakly inhibited cPLA2, while flavanone had negligible inhibitory activity. In contrast, amentoflavone and ginkgetin (biflavones) markedly inhibited cPLA2 activity in the epidermis. These results underscore that different flavonoids do vary in their capability to exert differential effects on arachidonate metabolism in the skin via modulation of epidermal cPLA2 activity. & 2001Harcourt Publishers Ltd
INTRODUCTION Phospholipase A2 (PLA2) is involved in many cellular processes such as normal cell growth, differentiation, immunoregulation, as well as diseased state (inflammation). Until recently, several forms of PLA2 have been reported: secretory PLA2 (sPLA2), cytosolic PLA2 (cPLA2) and intracellular Ca2+-independent PLA2 (iPLA2).1 Among these three isotypes, cPLA2 is the enzyme specific for the hydrolysis of sn-2 arachidonyl phospholipid and is active at micromolar concentrations of Ca2+ instead of the millimolar optimal concentrations of Ca2+ required by sPLA2. In addition, it has a large molecular weight (85 kD) and is resistant to disulfide reducing agents, in contrast to sPLA2. Like the other cell types, epidermal cPLA2 is thought to regulate essential cellular processes such as eicosanoid generation. For instance, cPLA2 was demonstrated to be involved in the activation of keratinocytes in cell culture and prostaglandin synthesis by UV-irradia-
Received 9 August 2001 Accepted 18 September 2001 Correspondence to: V. A. Ziboh, Department of Dermatology TB192, School of Medicine, University of California–Davis, Davis, CA 95616 USA. ;Tel.: 530-7529765; Fax: 530-752-9766; E-mail:
[email protected]
& 2001Harcourt Publishers Ltd
tion.2 This activation, in a follow-up study, was associated with cPLA2 expression using Western blotting technique.3 Recently, several forms of PLA2 including the cytosolic form were found in mouse epidermal keratinocytes4 and subsequently, regulation of cPLA2 level was demonstrated to affect growth and apoptosis in a keratinocyte cell line (HEL-30).5 Despite these reports on the cell culture systems, there are only a few reports that have shown the existence of cPLA2 in intact epidermis. For instance, a previous report demonstrated phospholipase activity in cow epidermis.6 The presence of PLA activity from rat epidermis has also been reported, but the report failed to establish whether or not the activity was due to PLA2 activity or a high lysophospholipase activity.7 The characteristics of PLA2 activity from human and rat epidermal homogenate have also been previously demonstrated, but that study was focused on the microsomal particulate enzyme.8 Recently, it was reported that cPLA2 was induced in an intact human skin by UV irradiation.9 Thus, in our present investigation, we clearly demonstrate cPLA2 activity in a guinea pig epidermal homogenate. The enzyme is Ca2+-dependent and resistant to disulfide reducing agents. Using this system, we tested the effects of five flavonoids on epidermal cPLA2 activity, since plant
Prostaglandins, Leukotrienes and Essential FattyAcids (2001) 65(5&6), 281–286
282
Kim et al.
extracts containing various flavonoids have been used in cosmetic preparations as well as in pharmaceutical preparations. The flavonoids selected in this study include: flavanone, apigenin (flavone), quercetin (flavonol), amentoflavone (biflavone) and ginkgetin (biflavone). Our data revealed that flavonoids, particularly amentoflavone and ginkgetin (biflavone), inhibited cPLA2 activity in guinea pig epidermal homogenate.
filtrate was centrifuged at 600 g for 20 minutes at 41C. The supernatant was centrifuged at 105 000 g for 60 minutes at 41C to obtain the cytosolic fraction, which was used throughout this study. Protein concentration was determined using Bradford reagent (Bio-Rad Laboratories, Hercules, CA). The usual protein concentration in our cytosolic fraction ranged from 1.6–2.0 mg/ml.
Synthesis of radioactive phospholipid substrate MATERIALS AND METHODS
Flavonoids Flavanone, apigenin and quercetin were purchased from Sigma-Aldrich (Milwaukee, MI). Amentoflavone was isolated from Sellaginella tamariscina. Ginkgetin was obtained from Ginkgo biloba leaves. Isolation and structural identification of these biflavones were carried out according to previous reports.10,11 The chemical structures of flavonoids are shown in Figure 1.
Preparation of guinea pig epidermal cytosolic fraction Guinea pig epidermis was obtained from male Hartly guinea pigs (Chales River Co.) following a previous report.12 Epidermis was cut into small pieces and homogenized in 100 mM Tris-HCl buffer, pH 7.5, with 0.8 mM EDTA, 0.8 mM EGTA, 330 mM sucrose, 2 mM phenylmethylsulfonyl fluoride (PMSF), 40 mg leupeptin/ml and 40 mg aprotinin/ml using Polytron homogenizer (Brinkmann Instruments, Westbury, NY). The homogenate (6 ml/g tissue) was filtered on cheese cloth and the
8
2′ O
7 A
C
6 5
2 3
4 O
3′
OH 4′
B
HO
O OH
6′ H
O OH OH
H
Flavanone
O
OH
Apigenin
CH3O
OH
O
HO
O Quercetin
O
OH
HO
5′
OH
HO
1-Stearyl-2-[1-14C]-arachidonyl-sn-glycero-3-phosphatidylcholine (14C-PC) was synthesized from [1-14C]-arachidonic acid (AA) and lysolecithin using guinea pig liver microsomal acyl transferase as previously reported.8 Briefly, guinea pig liver was homogenized in 20 mM Tris-HCl, pH 7.4 buffer containing 250 mM sucrose and 1 mM EDTA (1 gram wet tissue weight/5 ml buffer). The microsomal pellet was collected and resuspended in 1 ml of 50 mM Tris-HCl, pH 7.4 buffer containing 330 mM sucrose. The homogenate was centrifuged at 600 g for 20 minutes at 41C. The supernatant was collected and centrifuged at 105 000 g for 60 minutes at 41C. Protein estimation was determined using the Bradford microassay procedure (Bio-Rad Laboratories, Hercules, CA). The substrates 1-14C-arachidonic acid (50 nmol of 50 mCi/ mmol 1-14C-AA from Sigma, St Louis, MO) and 75 nmol of lysolecithin (Sigma, St Louis, MO) were dissolved in 1000 ml of ethyl ether and 500 ml of the reaction buffer containing 50 mM Tris-HCl, pH 7.4 buffer with 330 mM sucrose, 0.1 mM coenzyme A (coA), 0.1 mM dithiothreitol (DTT), 2.5 mM adenosine triphosphate (ATP), and 10 mM MgCl2. The ether layer containing the substrate was
O OCH3
OH
O
O
HO
O
OH
OH
O
Amentoflavone
OH
OH
O
Ginkgetin
Fig. 1 Chemical structures of flavonoids used in this study. Prostaglandins, Leukotrienes and Essential FattyAcids (2001) 65(5&6), 281–286
& 2001Harcourt Publishers Ltd
Differential effects of flavonoids on cPLA2 283
evaporated completely under N2 gas prior to sonication for several second bursts in an ice bucket. The reaction was initiated with the addition of 1 mg of the enzyme preparation with 1 ml of the reaction buffer and allowed to proceed for 30 minutes immersed in a 371C water bath. After 30 minutes, 1 mg of the enzyme preparation and 2 ml of the reaction buffer were added and the reaction was allowed to proceed for 120 minutes under the same conditions. The reaction was terminated with 20 ml of chloroform/methanol (2:1, v/v) and transferred to a separatory funnel. After vigorous shaking, the mixture was allowed to form two layers and the organic phase was collected. The organic extract was filtered through sodium sulfate and dried in a rotavapor (Brinkmann Instruments, Westbury, NY). The extract was resuspended in 100 ml of chloroform/methanol (2:1, v/v) and subjected to thin layer chromatography (TLC, silica gel 60, Merck) using a solvent system of chloroform/methanol/acetic acid/water (90:8:1:0.8, v/v/v/v) and authentic phospholipid and fatty acid standards as a first step in purification. The phospholipid band was scraped and eluted through a sintered funnel with chloroform/methanol (2:1, v/v). The extract was dried in a rotavapor and resuspended in 100 ml of chloroform/methanol (2:1, v/v). The synthetic 14 C-PC was further purified by a second thin layer chromatography (TLC, silica gel 60, Merck, Darmstadt, Germany) using a solvent system of chloroform/methanol/water (130:50:8, v/v/v) and authentic phospholipid and fatty acid standards. The purity and the position of 14 C-AA on the 14C-PC was determined using western diamondback rattlesnake (Crotalus atrox) venom (Sigma, St Louis, MO) as a commercial PLA2 source.
Identification of Radioactive Products The products were separated by thin layer chromatography (TLC) using the solvent mixture of ethyl acetate/isooctane/acetic acid/water (75:55:0.5:50, by volume) as a mobile phase. TLC plates were autoradiographed for 5 days, visualized by iodine vapors, and compared against reference AA. Typical separation by TLC after autoradiography is shown in Figure 2. The bands co-migrating with the authentic AA were scraped and counted in the liquid scintillation counter (LSC). The data are expressed as means dpm7 SD of 14C-AA released in at least three separate experiments. To ascertain whether the preparation contained PLC activity, the reaction products were developed on a second TLC system using chloroform/ methanol/water (65:30:4, by vol) as a mobile phase. In this sytem, 14C-AA-PC and 14C-AA-LysoPC were
PLA2 assay and identification of metabolites PLA2 activity was measured by the hydrolytic release of 14 C-AA from 1-stearyl-2-[1-14C]-arachidonyl-sn-glycero3-phosphatidylcholine (14C-PC). The reaction mixture contained 0.02 mCi 14C-PC (2 ml in DMSO), 0–40 mg cytosolic protein in HBSS buffer (36 ml) at pH 7.5 (to mimic physiological conditions), 0.8 mM EDTA, 0.8 mM EGTA, with or without 5.0 mM dithiothreitol (DTT). Incubation was started with the addition of 2 ml of varying amounts of CaCl2 solution. Total reaction volume was 40 ml/reaction tube. Reaction mixture was incubated at 371C for 0–30 min. To terminate the reaction, 40 ml of 2 % acetic acid in ethanol containing 20 mg of authentic AA was added. Extraction was carried out using 185 ml chloroform/methanol (30:7, by vol) by vigorous vortexing for 30 sec followed by centrifugation at 12 000 rpm in an Eppendorf microcentrifuge 5415C (Eppendorf). The organic phase was removed, dried under N2, and redissolved in 25 ml chloroform/methanol (2:1, by vol) solution. & 2001Harcourt Publishers Ltd
Fig. 2 TLC autoradiogram. PLA2 activity was assayed at 0.1 mM of Ca2+ concentration using 30 mg protein/incubation tube for15 min as described in Legend 3.The PLA2 reaction products were developed onTLC using ethyl acetate:iso-octane:acetic acid:water (75:55:0.5:50) as a mobile phase.TheTLC revealed 14C-AA (the product of cPLA2 catalysis from 14C-PC).
Prostaglandins, Leukotrienes and Essential FattyAcids (2001) 65(5&6), 281–286
284
Kim et al.
separated. TLC plates were autoradiographed and spots were counted on the LSC using the same procedure described above (data not shown). No PLC catalyzed radioactive products were revealed.
Characterization of epidermal PLA2 activity To characterize epidermal PLA2, we used a cell-free system of cytosolic fraction obtained from guinea pig epidermis to measure PLA2 activity at micromolar Ca2+ concentrations (100–400 mM) except in the Ca2+-dependent study. PLA2 activity was assayed as described above at different time intervals and at varying Ca2+ concentrations to establish whether or not cPLA2 is a constituent of guinea pig epidermis. To establish specificity for the epidermal cytosolic PLA2 (cPLA2), the incubations contained either methyl arachidonyl fluorophosphonate (MAFP), specific inhibitor of cPLA2, or 12-epi-scalaradial (inhibitor of soluble PLA2).
Effects of flavonoids on cPLA2 activity To determine the effects of the flavonoids on epidermal PLA2 activity, 30 mg of cytosolic protein from epidermal homogenate as previously described,12 was incubated with CaCl2 (final concentration of 100 mM free Ca2+) in a reaction buffer (modified HBSS buffer without Ca2+ or Mg2+, containing 10 mM HEPES, pH 7.5, with 0.8 mM EDTA, 0.8 mM EGTA, and 5 mM dithiothreitol) for 15 min. The flavonoids were dissolved in a mixture of DMSO/ reaction buffer (1:9, by vol) and aliquots added to the incubation mixture at varying concentrations of 0.1, 1.0,
Fig. 3 Effects of Ca2+ concentration and dithiothreitol (DTT) on PLA2 activity.PLA2 activity was measured using 30 mg protein/tube as shown in Figure 2 with (*) or without (*) 5 mM DTT for15 min. Each point represents the mean 7 SE of duplicate incubations from three experiments.
10 and 100 mM, respectively. The same amount of reaction buffer containing DMSO was added in a control experiment. The flavonoids (4 ml each) were initially preincubated with the cytosolic protein fraction (32 ml) for 10 min at 371C prior to initiation of the assay as described above. RESULTS Based on a previous study in our laboratory,12 we determined that using 30 mg of cytoplasmic protein and incubating with 14C-AA-PC were adequate for the assays of cPLA2 activity. This amount of cytoplasmic extract was used in these experiments.
Effect of Ca2+ concentration Since cellular calcium (Ca2+) is pivotal in the activity of cellular PLA2, we tested the effect of Ca2+ concentration on the PLA2 activity of the epidermal cytosolic extract. Our data revealed that the effects of Ca2+ concentration and dithiothreitol (DTT) depended on the presence of Ca2+ at micromolar concentrations (Fig. 3). These results correlated with previous reports describing cPLA2 activity in keratinocytes.13,14 However, it is not clear at present whether or not the marked increase in PLA2 activity observed above 1 mM Ca2+ concentration was due to cPLA2 alone or to other PLA2 isoforms (such as sPLA2) since other forms of PLA2 have previously been described in murine epidermal keratinocytes.4
Effect of selective inhibitors of PLA2 To strengthen our premise that our epidermal cytosolic preparation contained cPLA2, we incubated 30 mg of cytosolic protein in the presence of 100 mM Ca2+ concentration for 15 min with methyl arachinonyl fluorophosphonate (MAFP), a potent selective inhibitor of cPLA2. Our data revealed a marked suppression of the cPLA2 activity at an IC50 of 1 mM. A similar experiment with 12-epi-scalaradial (a potent inhibitor of sPLA2) revealed negligible inhibitory effect of cPLA2 even at 100 mM (Fig. 4). Under this condition, 14C-AA conversion to HETEs by 12- and 15-lipoxygenases was not detected (data not shown). Indeed, 14C-AA and 14C-AA-PC were the only radioactive compounds detected (Fig. 2). To discern whether or not our cytosolic preparation contained phospholipase C (PLC) activity which could hydrolyze 14C-PC to generate 14C-diacylglycerol (DAG), we applied a portion of the extracted hydrolysate to TLC plates and developed it in the solvent system of chloroform/methanol/water (65:30:4, by vol). No products of PLC catalysis were detected. Taken together, these results, including the Ca2+ dependency and selective inhibitor
Prostaglandins, Leukotrienes and Essential FattyAcids (2001) 65(5&6), 281–286
& 2001Harcourt Publishers Ltd
Differential effects of flavonoids on cPLA2 285
(MAFP) studies, strongly indicate that the enzyme in the crude epidermal cytosolic fraction that catalyzes the release of 14C-AA from 14C-PC is seemingly cPLA2.
Structure/activity of the flavonoids The concentration-dependent effects of the five flavonoids tested are summarized in Figure 5. Flavanone
Fig. 4 Inhibition of PLA2 activity by methyl arachidonyl fluorophosphonate (MAFP). PLA2 activity was assayed in medium containing 0.1 mM Ca2+ and using 30 mg of cytosolic protein/tube for 15 min as shown in Legend 3 with variable amounts of inhibitors: MAFP (*) and12-epi-scalardial (*), respectively. Each point represents the mean 7SE of duplicate incubations from three experiments.
(Fig. 1) lacking a C-2, 3-double bond did not inhibit cPLA2 activity even at 100 mM. Similarly, apigenin (a flavone) showed weak inhibitory activity (less than 20%) at 100 mM. Quercetin (a flavonol) exerted stronger inhibitory activity at 0.1–100 mM (25%). Interestingly, quercetin has previously been reported to inhibit group II (sPLA2),15–17 but not group I PLA2.15,16 Considering the IC50 value of quercetin on sPLA2 (2–4 mM),15 it appears that quercetin in situ may not significantly reduce arachidonic acid release via the inhibition of cPLA2 (Fig. 5A). In contrast, the biflavones (biflavonoids) tested: amentoflavone (at concentrations of 10 and 100 mM) and ginkgetin (at concentrations of 1 and 10 mM), markedly inhibited cPLA2 activity, respectively (Fig. 5B). The order of inhibitory potencies of the flavonoids on epidermal cPLA2 is: ginkgetin 4 amentoflavone quercetin apigenin. Ginkgetin, in particular, potently inhibited cPLA2 activity 50% at 10 mM, a pharmacologically achievable concentration by topical application. Because biflavones were previously reported to inhibit group II sPLA2,18 it is likely that some biflavones such as amentoflavone and ginkgetin may reduce arachidonic acid release by the inhibition of the two phospholipases (sPLA2 and cPLA2), thereby affecting several physiological and pathological conditions in the skin. DISCUSSION Flavonoids are known to be nature’s tender drug possessing antioxidant and anti-inflammatory properties.19 Because of these properties, flavonoid preparations are widely used in cosmetics as well as in crude
Fig. 5 Effects of flavonoids on cPLA2 activity. Cytosolic PLA2 activity was assayed as in Legend 3, containing 0.1 mM Ca2+, 30 mg cytosolic protein/ incubation tube for15 min at 371C.The cytosolic fraction was pre-incubated respectively with each flavonoid prior to initiation of incubation.The effects of flavonone (*), apigenin (*) and quercetin (~) are shown in 5A.The effects of biflavones, amentoflavone (*) and ginkgetin (*) are shown in 5B. Each point represents the mean 7SE of duplicate incubations from three experiments.
& 2001Harcourt Publishers Ltd
Prostaglandins, Leukotrienes and Essential FattyAcids (2001) 65(5&6), 281–286
286
Kim et al.
pharmaceutical drugs. The biological importance of the flavonoids in plant extracts has recently been greatly emphasized. To delineate their biological activity in the skin, five representative flavonoids were evaluated for their modulatory role on arachidonic acid metabolism in the skin. For instance, in our previous study, we showed that various flavonoids such as quercetin inhibit the 5-lipoxygenase pathway, amentoflavone potently inhibited the cyclooxygenase activity from guinea pig epidermis.12 Our present study does reveal that biflavones, especially ginkgetin, potently inhibited cPLA2 activity in the epidermis. Since a pathological condition such as UVB-induced PGE2 generation involves cPLA2 in keratinocytes3,9 it is reasonable to speculate that biflavones (which potently inhibit PGE2 generation) if topically applied to a lesional skin may exert a beneficial effect. Thus, data from our study indicate a dichotomy in the effects of the naturally-occurring flavonoids on the activity of enzymes involved in eicosanoid generation. For instance, biflavones have been reported to inhibit lymphocyte proliferation (T- and B-cell proliferation and mixed lymphocyte reaction) when compared to flavones/ flavonols which mainly affect the T-cell.20 Furthermore, the biflavones have been reported to exert their irreversible inhibitory effects on T-cell proliferation.21 The epidermal cPLA2 therefore provides a model for evaluating the cutaneous biological activity for the variety of flavonoids.
ACKNOWLEDGEMENTS Special thanks are given to Drs S. S. Kang, H. W. Chang and K. H. Son for the isolation of biflavones in this study.
REFERENCES 1. Dennis E. A. The growing phospholipase A2 superfamily of signal transduction enzymes. TIBS 1997; 22: 1–2. 2. Kang-Rotondo C. H., Miller C. C., Morrison A. K., Pentland A. P. Enhanced keratinocyte prostaglandin synthesis after UV injury is due to increased phospholipase activity. Am J Physiol 1993; 264: C396–C401. 3. Chen X., Gresham A., Morrison A., Pentland, A. P. Oxidative stress mediates synthesis of cytosolic phospholipase A2 after UVB injury. Biochim Biophys Acta 1996; 1299: 23–33.
4. Li-Stiles B., Lo H-H. Fisher S. M. Identification and characterization of several forms of phospholipase A2 in mouse epidermal keratinocytes. J Lipid Res 1998; 39: 569–582. 5. Lo H-H., Teichman P., Furstenberger G., Gimenez-Conti I., Fisher S. M. Suppression or elevation of cytosolic phospholipase A2 alters keratinocyte prostaglandin synthesis, growth, and apoptosis. Cancer Res 1998; 58: 4624–4631. 6. Long V. J. W., Yardley H. J. Phospholipase A activity in the epidermis. J Invest Dermatol 1972; 58: 148–154. 7. Freinkel R. K., Traczyk T. N. The phospholipase A of epidermis. J Invest Dermatol 1972; 74: 169–173. 8. Ziboh V. A., Lord J. T. Phospholipase A activity in the skin. Biochem J 1979; 184: 283–290. 9. Gresham A., Masferrer J., Chen X., Leal-Khouri S., Pentland A. P. Increased synthesis of high-molecular weight cPLA2 mediates early UV-induced PGE2 in human skin. Am J Physiol 1996; 270: C1037–1050. 10. Kang S. S., Kim J. S., Kawk W. J., Kim K. M. Flavonoids from the leaves of Ginkgo biloba. Kor J Pharmacogn 1990; 21: 111–120. 11. Shin D. I., Kim J. Flavonoid constituents of Sellaginella tamariscina. Kor J Pharmacogn 1991; 22: 207–210. 12. Kim H. P., Mani I., Iversen L., Ziboh V. A. Effects of naturallyoccurring flavonoids on epidermal cyclooxygenase and lipoxygenase from guinea-pig. Prostag Leukot Essent Fatty Acids 1998; 58: 17–24. 13. Kast R., Funstenberger G., Marks F. Activation of cytosolic phospholipase A2 by transforming growth factor-a in HEL-30 keratinocytes. J Biol Chem 1993; 268: 16795–16802. 14. McCord M. Chabof-Fletcher M., Breton J., Marshell L. A. Human keratinocytes possess an sn-2 acylhydrolase that is biochemically similar to the U937-derived 85 kDa phospholipase A2. J Invest Dermatol 1994; 102: 980–986. 15. Lindahl M., Tagesson C. Selective inhibition of Group II phospholipase A2 by quercetin. Inflammation 1993; 17: 573–582. 16. Lindahl M., Tagesson C. Flavonoids as phospholipase A2 inhibitors: importance of their structure for selective inhibition of group II phospholipase A2. Inflammation 1997; 21: 347–356. 17. Fawzy A. A., Vishwanath B. S., Franson R. C. Inhibition of human non-pancreatic phospholipase A2 by retinoids and flavonoids: mechanism of action. Agents and Actions 1988; 25: 394–400. 18. Chang H. W., Baek S. H., Chung K. W., Son K. H., Kim H. P., Kang S. S. Inactivation of phospholipase A2 by naturally occurring biflavonoid, ochnaflavone. Biochem Biophys Res Comm 1994; 205: 843–849. 19. Middleton E., Kandaswami C. Effects of flavonoids on immune and inflammatory cell functions. Biochem Pharmacol 1992; 43: 1167–1179. 20. Namgoong S. Y., Son K. H., Chang H. W., Kang S. S., Kim H. P. Effects of naturally occuring flavonoids on mitogen-induced lymphocyte proliferation and mixed lymphocyte culture. Life Sciences 1994; 54: 313–320. 21. Lee S. J., Choi J. H., Son K. H., Chang H. W., Kang S. S., Kim H. P. Suppression of mouse lymphocyte proliferation in vitro by naturally-occurring biflavonoids. Life Science 1993; 57: 551–558.
Prostaglandins, Leukotrienes and Essential FattyAcids (2001) 65(5&6), 281–286
& 2001Harcourt Publishers Ltd
