Immunosuppressive and anti-inflammatory effects of methanolic extract and the polyacetylene isolated from Bidens pilosa L.

Immunopharmacology 43 Ž1999. 31–37

Immunosuppressive and anti-inflammatory effects of methanolic extract and the polyacetylene isolated from Bidens pilosa L. Rachel L.C. Pereira a , Tereza Ibrahim a , Leonardo Lucchetti b, Antonio Jorge R. da Silva b, Vera Lucia Gonc¸alves de Moraes a,) a

b

Depto. de Bioquımica Medica, ICB, UniÕersidade Federal do Rio de Janeiro, 21941-590, Rio de Janeiro, Brazil ´ ´ Nucleo de Pesquisa de Produtos Naturais, UniÕersidade Federal do Rio de Janeiro, 21941-590, Rio de Janeiro, Brazil ´ Accepted 3 March 1999

Abstract The immunomodulatory effect of the methanolic extract obtained from dried leaves of Bidens pilosa L. ŽAsteraceae. and the polyacetylene 2-O-b-D-glucosyltrideca-11E-en-3,5,7,9-tetrayn-1,2-diol ŽPA-1. isolated from it was investigated. The extract inhibited the proliferative response in two in vitro models: human lymphocytes stimulated by 5 mg mly1 phytohemagglutinin ŽPHA. or to 100 nM 12-O-tetradecanoyl phorbol-13-acetate ŽTPA. plus 0.15 mM ionomycin and murine lymphocytes stimulated by 5 mg mly1 concanavalin A ŽCon A. or in the mixed leukocyte reaction ŽIC 50 s 12.5 to 25 mg mly1 .. PA-1 was 10-fold more potent than the original extract in blocking both human and murine lymphocyte proliferation ŽIC 50 s 1.25 to 2.5 mg mly1 .. In mice, the intraperitoneal Ži.p.. administration of methanolic extract of B. pilosa significantly reduced the size of the popliteal lymph node ŽPLN. after the inflammation induced by zymosan. One week after the injection of zymosan Ž150 mg. in the foot pad, PLN weighed 4.6 " 0.6 mg in comparison with 0.5 " 0.07 mg of the contralateral non-inflamed foot pad. The i.p. treatment with 10 mg extract from day 2 to day 6 after zymosan injection reduced the PLN weight to 1.8 " 0.3 mg. The data suggest an immunosuppressive activity of components of B. pilosa that may explain its popularly perceived anti-inflammatory effect. q 1999 Elsevier Science B.V. All rights reserved. Keywords: Bidens pilosa; Lymphocyte; Immunomodulation; PKC; Phorbol ester TPA; Polyacetylene

1. Introduction In recent years, plants have become increasingly important as a source of biologically active natural products. It is estimated that 25% of all medicines contain plant derivatives ŽFarnsworth, 1984.. In many cases, plant components have been used as starting material for semi-synthetic drugs: an example is taxol, obtained from Taxus baccata. )

Corresponding author. Tel.: q55-21-560-8895; Fax: q55-21270-8647; E-mail: [email protected]

Among the medicinal plants are species of Asteraceae ŽCompositae., the largest flowering plant family in the world. This family is characterized by the presence of polyacetylenes in all parts of the plant. Several compounds with biological actions have been isolated from different species of Asteraceae ŽBohlmann, 1988; Geissberger and Sequin, 1991; Hudson et al., 1993.. Among these is Bidens pilosa L. widely distributed in tropical regions. Although it is considered to be a weed of no economic importance, in Mexican traditional medicine it is used for treating stomach disorders, in South Africa and Taiwan as

0162-3109r99r$ - see front matter q 1999 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 2 - 3 1 0 9 Ž 9 9 . 0 0 0 3 9 - 9

32

R.L.C. Pereira et al.r Immunopharmacology 43 (1999) 31–37

food and to treat infections and in Brazil for treatment of malaria and liver disorders ŽChih et al., 1995; Brandao ˜ et al., 1997; Rabe and van Staden, 1997.. Bioactive compounds isolated from B. pilosa have been reported to have antibiotic, anti-inflammatory and anti-malarial activities, as well as to inhibit prostaglandin synthesis ŽGeissberger and Sequin, 1991; Alvarez et al., 1996; Jager et al., 1996; Rabe and van Staden, 1997.. The habitual use of plants such as B. pilosa might confer beneficial effects by increasing nonspecific defense mechanisms ŽWagner and Proksch, 1985.. Compounds that stimulate the immune system are useful as adjuvants in the treatment of certain fungal, bacterial and parasitic diseases, in which the infective agent depresses the host defenses as a strategy for survival. On the other hand, immunosuppressive compounds are useful in situations in which immune responses are undesirable, such as transplant rejection, allergies or autoimmune diseases. The possibility of finding an immunosuppressive mechanism underlying the perceived efficacy of B. pilosa led us to examine in vitro and in vivo its effects on lymphocyte activation. The results reported here show that the methanolic extract has an immunosuppressive effect on lymphocyte activation that may be associated with the presence of the polyacetylene 2-O-b-D -glucosyltrideca-11 E-en3,5,7,9-tetrayn-1,2-diol, previously shown to be present in Bidens camphyloteca ŽRedl et al., 1994. and now isolated for the first time from B. pilosa.

2.2. Cell proliferation PBMC Ž10 6 cells mly1 . were cultivated in the presence of 5 mg mly1 phytohemagglutinin ŽPHA. or 100 ng mly1 12-O-tetradecanoyl phorbol-13acetate ŽTPA. plus 0.15 mM ionomycin for 96 h in flat-bottom 96-well microplates at 378C in a 5% CO 2 atmosphere. Murine lymphocytes Ž10 6 cells mly1 . were cultivated in the presence of 5 mg mly1 Con A. Proliferation was measured by w3 Hxthymidine incorporation into cellular DNA: 0.5 mCi welly1 was added 6 h before the end of the culture period, which was 96 h for human and 72 h for murine lymphocytes. Cells were harvested and the radioactivity was assessed using a liquid scintillation counter. In the mixed leukocyte reaction ŽMLR., C57 Bl 6 lymphocytes Ž10 6 cells mly1 . gamma-irradiated with 2000 rads during 30 min were co-cultivated with BALBrc lymphocytes Ž4 = 10 6 cells mly1 .. After 120 h at 378C in a 5% CO 2 atmosphere, proliferation was measured as above. 2.3. Zymosan-induced arthritis Groups of 5 B10.ArSg SnJ mice ŽInstituto Nacional de Cancer, Rio de Janeiro, Brasil. received an injection Ž10 ml. of 15 mg mly1 zymosan prepared in saline, in the left hind footpad. The control group received saline. The popliteal lymph node ŽPLN. weight was used to follow the development of arthritis. After different intervals, animals were sacrificed and left and right Žcontrol. PLNs were dissected out, freed of attached adipose tissue and weighed.

2. Materials and methods 2.4. Crude extract of B. pilosa L. 2.1. Cells Peripheral blood mononuclear cells ŽPBMC. were obtained by separating heparinized blood from healthy volunteers by Ficoll-Hypaque density gradient centrifugation. Murine lymphocytes were obtained from spleens of normal BALBrc and C57 Bl 6 mice ŽInstituto Nacional de Cancer, RJ, Brazil.. Spleens were lysed and forced through a steel mesh. After haemolysis with ammonium chloride buffer, cells were extensively washed by centrifugation and resuspended in RPMI medium containing 5% fetal bovine serum ŽFBS..

2.4.1. Origin B. pilosa was collected between June and July in the Botanical Gardens of Rio de Janeiro and identified by Dr. Roberto Loureiro Steves ŽMuseu Nacional Rio de Janeiro.. 2.4.2. Preparation Leaves were dried at 408C during 3 days to yield 250 g, which were ground in 2 l methanol at room temperature for 7 days. The extract obtained was filtered, concentrated and lyophilized, affording 45 g of an amorphous dark powder that was resuspended

R.L.C. Pereira et al.r Immunopharmacology 43 (1999) 31–37

33

in methanol:water Ž9:1, vrv. and stored in the dark at 48C. 2.5. Isolation of the polyacetylene 2-O-b-D-glucosyltrideca-11E-en-3,5,7,9-tetrayn-1,2-diol (PA-1) The methanol:water solution was washed with petrol Ž60–808C., concentrated and fractionated under vacuum liquid chromatography on silica gel H, using solvents of increasing polarity Ž n-hexane, ethyl acetate and methanol.. The fractions displaying characteristic polyacetylene ultraviolet absorption spectra were collected and rechromatographed on RP18 eluted with a mixture of methanolrwater Ž7:3, vrv.. The polyacetylene-containing fractions were subjected to isocratic preparative HPLC Žmobile phase: methanol:water 7:3, vrv.. PA-1 was collected as the fraction corresponding to a sharp and symmetrical signal at tr s 7 min. The spectroscopic data for PA-1 wUV, IR Žfilm., 1 H NMR Ž200 MHz., 13 C NMR Ž50 MHz., and MSx identified the structure of the polyacetylene as 2-O-b-D-glucosyltrideca-11 E-en-3,5, 7,9-tetrayn-1,2-diol. Fig. 1 shows the chemical structure of the polyacetylene, identical to the compound isolated from B. camphylotheca ŽRedl et al., 1994. 2.6. Treatment with methanolic extract and PA-1 In vitro lyophilized samples of methanolic extract or PA-1 were resuspended before use in RPMI containing 5% FBS. They were added at the beginning of the proliferation assay or after different time intervals. The viability of lymphocytes was evaluated by the Trypan blue exclusion method.

Fig. 2. The inhibitory effects of methanolic extract and PA-1 on human lymphocyte proliferation. Human peripheral blood mononuclear cells Ž10 6 cells mly1 . were incubated with 5 mg mly1 PHA ŽA. or 100 nM TPAq0.15 mM ionomycin ŽB. for 96 h. Proliferation was measured by w3 Hxthymidine incorporation into cellular DNA, added 6 h before the end of the culture period. Values are means"S.E.M. of five experiments.

In vivo—from day 2 to 6 after zymosan, mice received a daily intraperitoneal Ži.p.. injection of 0.2 ml of water Žcontrol. or the solution containing 1, 5 or 10 mg methanolic extract dissolved in water. After different time intervals, mice were sacrificed and PLN weights were evaluated as described in Section 2.3. 2.7. Chemicals and reagents Tissue culture media ŽRPMI 1640., PHA, FicollHypaque, concanavalin A ŽCon A., TPA, ionomycin,

Fig. 1. Chemical structure of the polyacetylene 2-O-b-D-glucosyltrideca-11E-en-3,5,7,9-tetrayn-1,2-diol ŽPA-1..

R.L.C. Pereira et al.r Immunopharmacology 43 (1999) 31–37

34

liquid chomatography, RP18 was purchased from Aldrich Chemical ŽMilwaukee, WI, USA. and silica gel H from Merck ŽDarmstadt, Germany.. Solvents of analytical and spectrometric grade were purchased from Vetec ŽRio de Janeiro, Brazil.. 2.8. Statistical analysis Results were expressed as means " S.E.M. Data were analyzed statistically using a Mann–Whitney U-test, with differences considered to be significant when p - 0.05. Fig. 3. The inhibitory effects of methanolic extract and PA-1 on murine lymphocyte proliferation. BALBrc mouse spleen lymphocytes Ž10 6 cells mly1 . were incubated with ŽA. 5 mg mly1 Con A for 72 h or ŽB. co-cultivated Ž4=10 6 cells mly1 . with C57 Bl 6 gamma-irradiated spleen lymphocytes Ž10 6 cells mly1 . for 120 h. Proliferation was measured by w3 Hxthymidine incorporation into cellular DNA, added 6 h before the end of the culture period. In ŽB., the maximal value taken as reference was 3500"458 cpm. Values are means"S.E.M. of five experiments.

zymosan and tritiated thymidine Ž59 mCi mmoly1 . were purchased from Sigma ŽSt. Louis, MO.. Fetal bovine serum ŽFBS. was purchased from Fazenda Pigue ŽRJ, Brazil.. HPLC columns were purchased from Supelco ŽBellefonte, PA, USA.. HPLC apparatus, UV spectrophotometer Žmodel 1601. and refractive index detector model RID-6A from Shimadzu ŽTokyo, Japan.. The Gemini 200 NMR spectrophotometer was from Varian ŽPalo Alto, CA, USA.; AutoSpec Q mass spectrometer from VG ŽManchester, England.; IR model 783 spectrophotometer from Perkin-Elmer ŽNorwalk, CO, USA.; In the vacuum

3. Results 3.1. In Õitro effect of methanolic extract of B. pilosa and the polyacetylene PA-1 Both the methanolic extract of B. pilosa and PA-1 suppressed human lymphocyte proliferation ŽFig. 2.. Addition of increasing concentrations of extract resulted in a dose-dependent inhibition of the proliferation induced by 5 mg mly1 PHA ŽFig. 2A. or by 100 nM TPA plus 0.15 mM ionomycin ŽFig. 2B.. In both cases, the calculated IC 50 values were between 12.5 and 25 mg mly1 extract. Fig. 2 also shows that the isolated polyacetylene PA-1 was 10fold more potent to inhibit lymphocyte proliferation than the original crude extract, with IC 50 values around 1.5 mg mly1 . To investigate whether the inhibition caused by the methanolic extract and PA-1 was specific for human lymphocytes, we tested for a possible effect

Table 1 Effect of methanolic extract and PA-1 on different stages of murine lymphocyte proliferation Treatment

Control

w3 HxThymidine incorporation Žcpm. Time of addition Žh. a Extract Ž100 mg mly1 .

None 5 mg mly1 Con A

840 39 240

PA-1 Ž10 mg mly1 .

0

24

48

66

0

24

48

66

867 1485

678 3258

750 17 564

840 42 450

856 733

910 669

983 714

799 38 634

a Extract or PA-1 was added at the beginning of the culture period together with Con A or after the time indicated. All cell cultures were analyzed after 72 h; viability of lymphocytes was assessed by the trypan blue exclusion method. w3 HxThymidine was present during the last 6 h. Results are averages of three determinations, with S.E. 10% of mean.

R.L.C. Pereira et al.r Immunopharmacology 43 (1999) 31–37

on mouse lymphocytes. Fig. 3A shows that murine lymphocyte proliferation stimulated by 5 mg mly1 Con A was also blocked by both extract and PA-1, with estimated IC 50 values of 30 and 2.5 mg mly1 , respectively. To examine whether alloantigenspecific responses were also affected by the plant material, lymphocytes from BALBrc mice were cocultivated in the presence of gamma-irradiated lymphocytes derived from C57 Bl 6 mice. Proliferation in this mixed leukocyte reaction was also blocked in a dose-dependent manner by the methanolic extract and PA-1 ŽFig. 3B.. The inhibitory effect of methanolic extract and PA-1 on the proliferative response of lymphocytes of different origins could be due to an impairment of w3 Hxthymidine uptake. To check this possibility, methanolic extract Ž100 mg mly1 . or PA-1 Ž10 mg mly1 . was added at different intervals to murine lymphocyte cultures stimulated or not with the mitogen ŽCon A.. Table 1 shows that when plant material were added 6 h before the end of the culture period together with the radionuclide there was no inhibition of lymphocyte proliferation. 3.2. In ÕiÕo effect of methanolic extract of B. pilosa To address the question as to whether the direct suppressive effect of B. pilosa observed in vitro also

35

Fig. 5. The inhibitory effect of methanolic extract on PLN weight after inflammation with zymosan. The protocol was the same as in Fig. 4, except that mice were treated with B. pilosa from day 2 to day 6. PLN weight was evaluated one week after the injection of zymosan. Values are means"S.E.M. of five experiments.

occurs in vivo, we tested the B. pilosa preparations on the inflammation induced by zymosan. Mice received an injection of 10 ml containing 150 mg zymosan suspension in the left hind footpad. The development of the inflammatory process was followed by comparing the weight of the adjacent popliteal lymph node weight ŽPLN. to the 0.5 " 0.07 mg normally found for the non-inflamed foot pad Žright side.. Fig. 4 shows that one week after injection, PLN had increased in weight almost 10-fold, reaching a maximum in 3 weeks and decreasing spontaneously after 1 month. To examine the effect of plant material, mice received intraperitoneally different doses of the methanolic extract dissolved in water from day 2 to day 6 after the zymosan injection. Fig. 5 shows that a statistically significant prevention of inflammation was achieved by treating mice with 10 mg extract, daily for 5 days.

4. Discussion Fig. 4. Left popliteal lymph node ŽPLN. weight after inflammation with zymosan. Mice received a single injection of 150 mg zymosan in the left hind footpad. After different intervals they were sacrificed and the right and left PLN were dissected out and weighed. Right PLN taken as control ŽC. weighed 0.5 mg throughout the experiment. Values are means"S.E.M. of five experiments.

The results presented here show for the first time that the extract of B. pilosa L. has an immunosuppressive effect. They suggest that this effect may be due to the presence of the polyactylene shown for the first time to be present in this species.

36

R.L.C. Pereira et al.r Immunopharmacology 43 (1999) 31–37

B. pilosa L. widely used in tropical regions to treat inflammatory processes triggered by bacteria, fungi of helminths ŽWatt and Breyer-Brandwijk, 1962; Morton, 1981; Alvarez et al., 1996; Brandao ˜ et al., 1997.. To assess its immunomodulatory activity, we first tested the in vitro effect of the crude extract of B. pilosa on lymphocyte cultures. The proliferative response of lymphocytes from different species to various stimuli were completely inhibited by the methanolic extract of B. pilosa. After stimulation, T lymphocytes undergo intracellular events that involve phosphorylation and lead to an autocrine growth in which the stimulated naıve ¨ lymphocytes proliferate in response to their own production of interleukin-2 ŽIL-2. and the receptor for IL-2 ŽSchwartz, 1990.. Among the substances that induce lymphocyte proliferation are the lectins that bind to carbohydrate residues. The activation of the T lymphocyte upon binding of a lectin triggers intracellular reactions that generate second messengers such as diacylglicerol; this molecule stimulates PKC activity. The same occurs in the mixed leukocyte reaction. Diterpene phorbol esters, due to their lipophilicity, cross the cell membrane and, like the endogenous diacylglycerol, also activate PKC ŽCastagna et al., 1992.. Thus, the PKC activity is a common step in the proliferative pathway, triggered by the different stimuli used in this work. This means that addition of protein kinase C inhibitors to the culture medium would block lymphocyte proliferation ŽMey and Revillard, 1998.. The present data suggest a blockade of a common step in the proliferative response such as the PKC pathway. This possibility is currently under investigation. The results obtained in vitro led us to examine a possible in vivo effect of B. pilosa. We observed that a single injection of zymosan in the mouse foot pad led a significant increase in the adjacent PLN weight. The peripheral lymphoid organs, mainly the lymph nodes are drain sites where the antigens are concentrated and also the place into which naıve ¨ lymphocytes preferentially migrate. In addition, accessory cells, which are required for lymphocyte activation, are abundant in these tissues ŽAbbas et al., 1997.. The treatment of mice with B. pilosa for 5 days significantly reduced the increase in PLN weight, probably by inhibiting lymphocyte proliferation, as observed in vitro. These results are in line

with those that demonstrate an anti-inflammatory action of extracts of B. pilosa ŽGeissberger and Sequin, 1991; Jager et al., 1996.. Our results also indicate that the polyacetylene PA-1, for the first time isolated from B. pilosa may be involved in the immunosuppressive effect found in the crude extract. The same polyacetylene had previously been isolated from B. camphylotheca ŽBauer et al., 1992., a species traditionally used in Hawaiian folk medicine ŽAkana, 1922.. The polyacetylenes isolated from B. camphylotheca inhibited prostaglandin biosynthesis ŽRedl et al., 1994.. Taken together, the data indicate a potent immunosuppressive action of components present in the species B. pilosa, suggesting a promising application as an anti-inflammatory drug.

Acknowledgements We would like to thank Dr. Martha M. Sorenson for the careful revision of the manuscript and Dr. Vivian M. Rumjanek for the relevant criticisms. This work was supported by CNPq, FINEP, FAPERJ, FUJB, Brazil. T. Ibrahim was recipient of a graduate fellowship from CAPES and R.L.C. Pereira from CNPq.

References Abbas, A.K., Lichtman, A.H., Pober, J.S., 1997. Cellular and Molecular Immunology. W.B. Saunders, USA. Akana, A., 1922. Hawaiian Herbs of Medicinal Value. Pacific Book House, Honolulu. Alvarez, L., Marquina, S., Villarreal, M.L., Alonso, D., Aranda, E., Delgado, G., 1996. Bioactive polyacetylenes from Bidens pilosa. Planta Med. 62, 355–357. Bauer, R., Redl, K., Davis, B., 1992. Four polyacetylenes glucosides from Bidens camphylotheca. Phytochemistry 31, 2035– 2037. Bohlmann, F., 1988. In: Lam, J. ŽEd.., Chemistry and Biology of Naturally Occurring Acetylenes and Related Compounds. Elsevier, Amsterdam. Brandao, ˜ M.G.L., Krettli, A.U., Soares, L.S.R., Nery, C.G.C., Marinuzzi, H.C., 1997. Antimalarial activity of extracts and fractions from Bidens pilosa and other Bidens species ŽAsteraceae. correlated with the presence of acetylene and flavonoid compounds. J. Ethnopharmacol. 57, 131–138. Castagna, M., Takaz, Y., Akaz, K., Kaibuchi, K., Sano, K., Kikkawa, U., Nishisuka, Y., 1992. Direct activation of cal-

R.L.C. Pereira et al.r Immunopharmacology 43 (1999) 31–37 cium activated phospholipid dependent protein kinase by tumor promoting phorbol esters. J. Biol. Chem. 257, 7847–7851. Chih, H.W., Lin, C.C., Tang, K.S., 1995. Anti-inflammatory activity of Taiwan folk medicine ‘ham-hong-chho’ in rats. A. J. Chin. Med. 23, 273–278. Farnsworth, N.R., 1984. In: Krogsgaard-Larsen, P., Christensen, S.G., Kofod, H. ŽEds.., Natural Products and Drug Development. Munksgaard, Copenhagen, pp. 17–30. Geissberger, P., Sequin, U., 1991. Constituents of Bidens pilosa L.: do the components found so far explain the use of this plant in traditional medicine?. Acta Trop. 48, 251–261. Hudson, J.B., Graham, E.A., Rossi, R., Carpita, A., Neri, D., Towers, G.H.N., 1993. Biological activities of terthiophenes and polynes from Asteraceae. Planta Med. 59, 447–450. Jager, A.K., Hutchings, A., van Staden, J., 1996. Screening of Zulu medicinal plants for prostaglandin-synthesis inhibitors. J. Ethnopharmacol. 52, 95–100. Mey, A., Revillard, J.P., 1998. Mitogenic response of murine B lymphocytes to Salmonella typhimurium lipopolysaccharide

37

requires protein kinase C-dependent late tyrosine phosphorylations. Infect. Immun. 66, 2547–2552. Morton, J.F., 1981. In: Thomas, C.C. ŽEd.., Atlas of Medicinal Plants of Middle America-Bahamas to Yucatan. Springfield. Rabe, T., van Staden, J., 1997. Antibacterial activity of South African plants used for medicinal purposes. J. Ethnopharmacol. 56, 81–87. Redl, K., Breu, W., Davis, B., Bauer, R., 1994. Anti-inflammatory active polyacetylenes from Bidens camphylotheca. Planta Med. 60, 58–62. Schwartz, R.H., 1990. A cell culture model for T-lymphocyte clonal anergy. Science 248, 1349–1356. Wagner, H., Proksch, A., 1985. Economic and Medicinal Plant Research. Immunostimulatory Drugs of Fungi and Higher Plants. Academic Press, London, pp. 113–153. Watt, J.M., Breyer-Brandwijk, M.G., 1962. The Medicinal and Poisonous Plants of Southern and Eastern Africa. E&S Livingstone, London, pp. 205–206.