Journal of Ethnopharmacology 93 (2004) 319–323

In vitro study of the antioxidant and immunomodulatory activity of

aqueous infusion of Bidens pilosa夽
Celia Abajo a , Marı́a Ángeles Boffill b , Jaime del Campo a , Marı́a Alexandra Méndez a ,
Yisel González b , Montserrat Mitjans a , Marı́a Pilar Vinardell a,∗
a Department of Fisiologia-Divisió, IV, Facultat de Farmàcia, Av. Joan XXIII s/n, E-08028 Barcelona, Spain
b Unidad de Toxicologı́a, Villa Clara, Cuba

Received 15 October 2003; received in revised form 5 March 2004; accepted 31 March 2004

Available online 28 May 2004

Abstract

Bidens pilosa is an annual plant from tropical America with anti-inflammatory properties in hepatitis, laryngitis, headache and digestive dis-
orders, among others. Its wide pharmacological applications can be attributed to its chemical composition, with inhibitory effects on pathogenic
microorganisms and flavonoids, which show strong antioxidant capacities. We investigated the antioxidant activity of an aqueous infusion
of Bidens pilosa by studying its protective effect on the hemolysis induced by an initiator of radicals such as 2,2 -azobis(2-amidinopropane)
dihydrochloride (AAPH).
The immunomodulatory activity of the infusion was tested using whole blood cells. Cytokine production increased in whole blood stimulated
or not by lipopolysaccharides (LPSs).
The infusion is also characterized by its capacity to protect erythrocytes from the phototoxic effect of chlorpromazine, which allows its use
as a potential photoprotector. Finally, it did not show ocular irritation, as demonstrated by the effect on hemoglobin denaturation.
This study supports the health benefits of the ingestion of the infusion.
© 2004 Elsevier Ireland Ltd. All rights reserved.

Keywords: Antioxidant; Free radicals; Bidens pilosa; Photoprotection; Cytokine

1. Introduction pilosa have revealed the presence of alkaloids, saponins,

Bidens alba L. var. radiata Schultz-Bip, family Aster-
aceae, synonymous Bidens pilosa, is an annual plant from
tropical America, 30–100 cm in height with yellow flowers.
It is widely applied as an anti-inflammatory agent in hep-
atitis, laryngitis, headache and digestive disorders, among
others. In Cuba, it is widely used in infusions for several
disorders (Lastra et al., 2001). Its high pharmacological ap-
plications can be attributed to the chemical compounds, es-
pecially polyacetylenes, which inhibit pathogenic microor-
ganisms and flavonoids, which are active anti-inflammatory
agents (Pereira et al., 1999). Phytochemical studies of Bidens
Abbreviations: AAPH, 2,2 -azobis-(2-amidinopropane) dihydrochlo-
ride; IC50 , inhibitory concentration 50; IL-1␤, interleukin-1␤; LPS, lipo-
polysaccharide; ROS, reactive oxygen species; TMB, 3,3 ,5,5 -tetra-
methylbenzidine; TNF␣, tumor necrosis factor ␣
夽 A part of this paper was presented as a poster in the V Symposium
flavonoids, polyacetylenes and triterpens (Hoffmann and
Hölzl, 1988), acyl chalcones and glucosides and 1-phenyl
hepta 1,3,5-triyne (Zollo et al., 1995). Bidens pilosa is used
in traditional medicine to treat malaria (Brandäo et al.,
1997) and shows high antibacterial activity (Rabe and van
Staden, 1997). Increasing evidence suggests that oxidative
damage to cell components has a relevant pathophysiolog-
ical role in several types of human diseases (Ames et al.,
1993). Reactive oxygen species (ROS) have been reported
to damage erythrocytes in patients with blood pathologies
(Vives-Corrons et al., 1995; Rice-Evans et al., 1986). Ery-
throcytes are highly susceptible to oxidative damage owing
to the high polyunsaturated fatty acid content of their mem-
branes and the high cellular concentrations of oxygen and
hemoglobin, a powerful potential promoter of oxidative pro-
cesses (Clemens et al., 1987). These are the most common
sensitive configurations damaged by the action of oxygen

of Oxidative Stress, La Havana, Cuba, 2003.

free radical species, which may exert their damaging action
∗ Corresponding author. Tel.: +34 934024505; fax: +34 934035901. by membrane lipid oxidation, leading to the lysis of red
E-mail address: mpvinardellmh@ub.edu (M. Pilar Vinardell). blood cells. Lipid peroxidation is one of the consequences

0378-8741/$ – see front matter © 2004 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.jep.2004.03.050
320 C. Abajo et al. / Journal of Ethnopharmacology 93 (2004) 319–323
of oxidative damage and has been suggested as a general
mechanism of cell injury and death (i.e. hemolysis).
In intact isolated erythrocytes, 2,2 -azobis-(2-amidino-
propane) dihydrochloride (AAPH) induces the extensive
oxidation of membrane components and promotes struc-
tural alterations that lead to the rupture of the erythrocyte
membrane and to hemoglobin leakage (Sato et al., 1995).
In this paper, we have studied the antioxidant and im-
munomodulatory activity of an infusion of Bidens pilosa,
the most common preparation of this plant, with wide ap-
plications in popular medicine in Cuba and other countries
around the world.
2. Material and methods
2.1. Plant material
Aerial parts of fresh Bidens pilosa were collected from the
Medicinal Plants Farm of Santa Clara, Villa Clara, Cuba, in
June 2002. Plant material was authenticated by J. Matos, a
specialist of the Flora and Fauna Entity. A voucher specimen
is kept in the Herbarium of Villa Clara Medical University.
This plant was dried at 50 ◦ C for 3 days.
2.2. Preparation of infusion
An aqueous infusion of 8 g of the dried plant was prepared
by addition of 250 ml of water. After 2 min of ebullition and
10 min at room temperature, it was filtered and prepared for
use.
2.3. Chemicals
The following reagents were obtained from Sigma Chemi-
cal Co. (St. Louis, MO): sodium chloride, sodium phosphate
dibasic, chlorpromazine, 2,2 -azobis(2-amidinopropane) di-
hydrochloride, lipopolysaccharide (LPS) from Escherichia
coli, LPS from Salmonella abortus equi.
2.4. Antioxidant effect
Blood samples were obtained from healthy donors by
venipuncture (Blood Bank, Hospital Clinic, Barcelona,
Spain). Citrated blood was centrifuged at 1000 × g for
10 min and washed three times with phosphate-buffered
saline (PBS). Supernatant and buffy coat were carefully re-
moved by aspiration after each wash. Washed erythrocytes
were finally suspended in buffered saline.
We determined the hemolysis of erythrocytes mediated by
AAPH (a peroxyl radical initiator) using a modification of a
method described elsewhere (Miki et al., 1987). Twenty-five
microlitre aliquots of erythrocyte suspension were incubated
in the presence of 100 mM AAPH to induce hemolysis at
37 ◦ C for 2.5 h. The inhibitory concentration 50 (IC50 ) of the
hemolysis induced by AAPH was determined. Hemolysis
was monitored by spectrophotometry at 540 nm.
2.5. Photoprotective activity
The method used was an adaptation of the method
described by Pape et al. (2001). Erythrocyte suspension
was incubated for 30 min with the infusion equivalent to
2 mg/ml. Thereafter, the erythrocyte suspension was added
to a 24-well plate with various concentrations of chlor-
promazine, a photohemolytic compound. The cells were
exposed to UVA (1000 ␮W/cm2 ) and UVB (220 ␮W/cm2 )
delivered through an OSRAM lamp for 120 min. The ery-
throcyte suspension was then centrifuged and hemolysis
was monitored in the supernatant.
2.6. Cytokine production by whole blood cells stimulated
by LPS
We studied the effect of the extracts on the production of
tumor necrosis factor ␣ (TNF␣) and interleukin-1 ␤ (IL-1␤)
by whole blood cells stimulated with lipopolysaccharide
from Escherichia coli and Salmonella abortus equi. Hun-
dred microlitre of endotoxin solution (10 ␮g/ml) was added
to 200 ␮l of human blood, 100 ␮l of the infusion and 800 ␮l
of sterile saline solution (Sigma) in sterile polypropylene
tubes. The assay mixture was incubated at 37 ◦ C for 18 h in
a thermoblock. Cell-free supernatants obtained by centrifu-
gation at 3000 × g for 1 min were stored at −80 ◦ C until
cytokine measurement.
The viability of the cells after exposure to the extracts was
evaluated by the trypan blue exclusion test (Mascotti et al.,
2000).
Cytokines were determined in 96-well plates (Diaclone
Research, France). Samples or standard were added at
100 ␮l/well, and 50 ␮l/well biotylinated anti-TNF␣ or
anti-IL-1␤ was added to all wells and incubated for 3 h
at room temperature. The plate was then washed three
times, after which 100 ␮l of streptavidin–HRP was added
to all wells. The plate was incubated for 30 min at room
temperature and washed three times. The chromogen
3,3 ,5,5 -tetramethylbenzidine (TMB) solution was added
at 100 ␮l/well and the plate was incubated at room tem-
perature in the dark for 15 min. The chromogenic reaction
was stopped by the addition of 100 ␮l of 0.2 M H2 SO4 per
well. The optical densities (OD) were read on a plate reader
(Bio-Rad) at 450 nm.
A linear standard curve was generated by plotting the av-
erage absorbance on the vertical axis versus the correspond-
ing TNF␣ or IL-1␤ standard concentration on the horizontal
axis. The amount of TNF␣ and IL-1␤ in each sample was
determined by extrapolating OD values to TNF␣ or IL-1␤
concentrations in the standard curves. Human blood without
stimulation was used as a negative control and the level of
cytokines was used as a blank to calculate cytokine produc-
tion.
 C. Abajo et al. / Journal of Ethnopharmacology 93 (2004) 319–323
2.7. Study of potential eye irritant action
The method is based on the hemoglobin denaturation in-
duced by irritant products (Hayashi et al., 1995; Vinardell
et al., 1999) and was modified using hemoglobin from hu-
man erythrocytes (Mitjans et al., 2003). Hemoglobin was
exposed to various concentration of the infusion in 96-well
plates and its absorbance was recorded. Increasing concen-
trations of irritant products progressively reduce hemoglobin
absorbance. Non-irritant products do not alter hemoglobin
absorbance.
2.8. Statistical analysis
The differences between control and treatment tests were
analyzed using the Student’s t-test. They were considered
significant at P < 0.05 (represented by an asterisk). All
experiments were repeated at least three times. Data are
expressed as means ± S.E.M.
3. Results and discussion
In this study, we first tested the efficacy of the aqueous
infusion of Bidens pilosa as an inhibitor of AAPH-induced
erythrocyte hemolysis. AAPH, a water-soluble free radical
generator, was used to stimulate the in vivo conditions of ox-
idative stress and peroxyl radicals were generated by thermal
decomposition of an azo compound in oxygen. The advan-
tages of this method are that the AAPH decomposes ther-
mally to generate radicals without biotransformation or en-
zymes and the rate of radical generation is easily controlled
by adjusting the concentration of initiator (Ko et al., 1997).
The antioxidant effect of the infusion of Bidens pilosa on
the hemolysis induced by AAPH is shown in Fig. 1, which
represents the percentage of hemolysis inhibition at several
concentrations. The amount of Bidens pilosa infusion that
halved the hemolysis induced by AAPH was 6 ␮l, which
corresponds to an IC50 of 1.19 mg of dry weight per millilitre
of infusion. Thus, the oxidative hemolysis of erythrocytes
120
% Hemolysis Inhibition
100
80
60
40
20
0
0 1
Concentration (mg/mL)
Fig. 1. Antioxidant action of infusion of Bidens pilosa as determined by
the inhibition of hemolysis in human whole blood induced by AAPH
(Mean ± S.E.M.).
321
induced by AAPH was suppressed by an aqueous infusion of
Bidens pilosa, which is a very active antioxidant and exerts
its protective effect at low amounts.
The activity of the infusion of Bidens pilosa may be at-
tributed to the compounds found by HPLC (data not shown).
We detected that the infusion contains gallic acid and poly-
meric polyphenolic material. A small part of this material
is of proanthocyanidin type, since it was depolymerised by
thiolysis. The rest are hydrolysable tannins, as ascertained
by their elution time in HPLC and their UV absorption at
280 and 320 nm. The antioxidant activity of polyphenolic
material has been demonstrated in biological models and
extensively reviewed (Rice-Evans et al., 1996).
The photoprotective effect of Bidens pilosa has been
described after incubation of erythrocytes treated with the
infusion in presence of chlorpromazine, a phototoxic sub-
stance (Torinuki and Tagami, 1986; Eberlein-Konig et al.,
1997). Fig. 2 shows the photohemolysis induced by chlor-
promazine in erythrocytes pre-incubated or not with the
infusion of Bidens pilosa. The hemolysis induced by chlor-
promazine in erythrocytes incubated with the infusion and
in non-incubated erythrocytes differs significantly. The pho-
tohemolysis induced by chlorpromazine revealed a HC50 of
65.44 mg/ml. This value increases in blood cells incubated
in presence of the infusion, with a HC50 of 112.10 mg/ml,
which represents about 40% of the photoprotection induced
by the infusion.
It is known that the LPS from Escherichia coli induces
the production of cytokines by human whole blood, which
allows the study in vitro of the immunomodulatory action
of several compounds (Hermann et al., 2003). The aqueous
infusion of Bidens pilosa showed an increase in cytokine
production after stimulation with LPS from Escherichia coli
and Salmonella abortus equi. Fig. 3 shows the effect of the
infusion of Bidens pilosa on the production of TNF␣ by
10
Fig. 2. Photohemolysis induced by chlorpromazine in presence and ab-
sence of the infusion of Bidens pilosa (2 mg/ml). The symbol ( ) indicates
significant differences.
322 C. Abajo et al. / Journal of Ethnopharmacology 93 (2004) 319–323
Fig. 3. Individual results of TNF␣ production by human whole blood
exposed to LPS from Salmonella abortus equi and the effect of the
incubation with 100 ␮l of the aqueous infusion of Bidens pilosa. The
symbol ( ) indicates significant differences.
Fig. 4. Effect of the aqueous infusion of Bidens pilosa (100 ␮l) on the
TNF␣ and IL-1␤ production by human whole blood stimulated by LPS
from Escherichia coli. The symbol ( ) indicates significant differences.
whole blood stimulated by LPS from Salmonella abortus
equi in various subjects. The addition of Bidens pilosa to
whole blood significantly increases (P < 0.01) TNF␣ pro-
duction (between four and eight times, depending on the
subject). Fig. 4 shows the production of TNF␣ and IL-1␤ by
whole blood stimulated by LPS from Escherichia coli and
the effect of Bidens pilosa. TNF␣ and IL-1␤ production in
2.5
2
Hemoglobin absorbance
1.5
0.5
0
0 10
Fig. 6. Potential ocular irritation as determined by the measurement of human hemoglobin absorbance after incubation with various amounts of the
aqueous infusion of Bidens pilosa (error bars smaller than symbols).
Fig. 5. Effect of the aqueous infusion of Bidens pilosa (100 ␮l) on TNF␣
and IL-1␤ production by human whole blood without previous stimulation.
The symbol ( ) indicates significant differences.
whole blood stimulated by LPS in the presence of Bidens
pilosa significantly increases. To test whether the effect of
Bidens pilosa depends on the presence of LPS, we incu-
bated whole blood with only the infusion of Bidens pilosa
and determined cytokine production. The levels of cytokine
in whole blood from healthy donors were very low and were
used as blank. Fig. 5 shows the effect of Bidens pilosa on
the production of TNF␣ and IL-1␤ in non-stimulated whole
blood. TNF and IL-1␤ production was doubled, which may
be due to the action of Bidens pilosa alone.
IL-1␤ and TNF␣ are currently regarded as central regula-
tory cytokines in the induction of macrophage antimicrobial
activities (Wang et al., 1994). The antimicrobial activity of
the infusion may increase the amount of cytokine (Rabe
and van Staden, 1997) in a similar way to other plants
(Gomez-Flores et al., 2000; Rimbach et al., 2000).
Fig. 6 shows the hemoglobin denaturation induced by
Bidens pilosa. Increased amounts of the infusion did not
enhance the denaturation of hemoglobin, which points to the
non-irritant action of the infusion, thus allowing its use in
ophthalmic preparations (Mitjans et al., 2003).
20 30 40 50 60 70 80
Bidens pilosa aqueous infusion (µL)
 C. Abajo et al. / Journal of Ethnopharmacology 93 (2004) 319–323
In this paper, we have demonstrated in vitro some prop-
erties of the aqueous infusion of Bidens pilosa that are in
accordance with the general use of the infusion in traditional
medicine.
However, its mechanism of action associated with boost-
ing the immune response remains to be elucidated. The aque-
ous infusion of Bidens pilosa was found to enhance cytokine
production by whole blood. Monokines such as interleukin-1
and TNF␣ are essential mediators of host inflammatory re-
sponses to natural immunity.
These results shed some light on the mechanism of ac-
tion of Bidens pilosa. The regulation of immune parameters
induced by the infusion of Bidens pilosa may be clinically
relevant in numerous disease processes like chronic viral in-
fection, tuberculosis, AIDS and cancer.
In summary, although the data presented in this paper
yielded an incomplete picture of the effects of the infusion of
Bidens pilosa on the immune system, we have demonstrated
that the infusion enhances cytokine production by whole
blood. Further studies with animal models are necessary to
clarify how and to what extent this activation occurs in vivo.
Acknowledgements
We thank Robin Rycroft for language assistance and Ari-
adna Selga for HPLC analysis. This study was supported by
the project 501 from the Fundació Bosch i Gimpera of the
Universitat de Barcelona.
References
Ames, B.N., Shigenaga, M.K., Hagen, T.M., 1993. Oxidants, antioxidants,
and the degenerative diseases of aging. Proceedings of the Natural
Academy of Sciences USA 90, 7915–7922.
Brandäo, 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 (Asteraceae) correlated with the pres-
ence of acetylene and flavonoid compounds. Journal of Ethnopharma-
cology 17, 131–138.
Clemens, M.R., Ruess, M., Bursa, Z., Waller, H.D., 1987. The relationship
between lipid composition of red blood cells and their susceptibility
to lipid peroxidation. Free Radical Research Communication 3, 265–
271.
Eberlein-Konig, B., Bindl, A., Przybilla, B., 1997. Phototoxic properties
of neuroleptic drugs. Dermatology 194, 131–135.
Gomez-Flores, R., Calderon, C.L., Scheibel, L.W., Tamez-Guerra, P.,
Rodrı́guez-Padilla, C., Tamez-Guerra, R., Weber, R.J., 2000. Immu-
noenhancing properties of Plantago major leaf extract. Phytotherapy
Research 14, 617–622.
Hayashi, T., Itagaki, H., Fokuda, T., Tamura, U., Sato, Y., Suzuki, Y.,
1995. Hemoglobin denaturation caused by surfactants. Biological and
Pharmaceutical Bulletin 18, 540–545.
323
Hermann, C., von Aulock, S., Graf, K., Hartung, T., 2003. A model of
human whole blood lymphokine release for in vitro and ex vivo use.
Journal of Immunological Methods 275, 69–79.
Hoffmann, B., Hölzl, J., 1988. New chalcones from Bidens pilosa. Planta
Medica 54, 52–54.
Ko, F.N., Hsiao, G., Kuo, Y.H., 1997. Protection of oxidative hemolysis
by demethyldiioseugenol in normal and beta-thalassemic red blood
cells. Free Radical Biology and Medicine 22, 215–222.
Lastra, H., Valdés, I., Ponce de León, R.H., 2001. Bidens pilosa Linné.
Revista Cubana de Planta Medica 1, 232–233.
Mascotti, K., Mc Cullough, J., Burger, S.R., 2000. HPC viability mea-
surement: trypan blue versus acridine orange and propidiym iodide.
Transfusion 40, 693–696.
Miki, M., Ramai, H., Mino, M., Yamamoto, Y., Niki, E., 1987. Free radical
chain oxidation of rat red blood cells by molecular oxygen and its
inhibition by ␣-toxopherol. Archives in Biochemistry and Biophysics
258, 373–380.
Mitjans, M., Abajo, C., del Campo, J., Vinardell, M.P., 2003. Estudio
comparativo de dos métodos alternativos para evaluar la irritación
ocular utilizando hemoglobina. Revista de Toxicologia 22, 131.
Pape, W.J., Maurer, T., Pfannenbecker, U., Steling, W., 2001. The red
bood cell phototoxicity test (photohaemolysis and haemoglobin oxida-
tion): EU/COLIPA validation programme on phototoxicity (phase II).
Atla-Alternatives to Laboratory Animals 29, 145–162.
Pereira, R.L., Ibrahin, T., Lucchetti L da Silva, A.J., Goncalves de
Morales, V.L., 1999. Immunosuppressive and anti-inflammatory effects
of methanolic extract and the polyacetylene isolated from Bidens pi-
losa L. Immunopharmacology 43, 31–37.
Rabe, R., van Staden, J., 1997. Antibacterial activity of South African
plants used for medicinal purposes. Journal of Ethnopharmacology 56,
81–87.
Rice-Evans, C., Omorphos, S.C., Baysal, E., 1986. Sickle cell membranes
and oxidative damage. Biochemical Journal 237, 265–269.
Rice-Evans, C., Miller, N.J., Paganga, G., 1996. Structure-antioxidant
activity relationship of flavonoids and phenolic acids. Free Radical
Biology and Medicine 20, 933–956.
Rimbach, G., Guo, Q., Moini, H., Saliou, C., Takayama, K., Virgili, F.,
Packer, L., 2000. Nitric oxide synthesis and TNF-␣ secretion in RAW
2647 macrophages. Mode of action of a fermented papaya preparation.
Life Sciences 67, 679–694.
Sato, Y., Kamo, S., Rakahasi, T., Suzuki, Y., 1995. Mechanism of free
radical-induced hemolysis of human erythrocytes: hemolysis by a
water-soluble radical initiator. Biochemistry 34, 8940–8949.
Torinuki, W., Tagami, H., 1986. Role of complement in chlorpromazine-
induced phototoxicity. Journal of Investigative Dermatology 86, 142–
144.
Vinardell, M.P., Gonzalez, S., Infante, M.R., 1999. Adaptation of
hemoglobin denaturation for assessment of ocular irritation of surfac-
tants and manufactured products. Journal of Toxicology – Cutaneous
and Ocular Toxicology 18, 375–384.
Vives-Corrons, J.L., Miguel-Garcı́a, A., Pujades, M.A., Miguel-Sosa, A.,
Cambiazzo, S., Linares, M., Dibarrarts, M.T., Calvo, M.A., 1995. In-
creased susceptibility of microcytic red blood cells to in vitro oxidative
stress. European Journal of Haematology 55, 327–331.
Wang, Y., Huang, D.S., Watson, R.R., 1994. In vivo and in vitro cocaine
modulation on production of cytokines in C57BL/6 mice. Life Sciences
54, 401–411.
Zollo, A., Kuiate, J.R., Menut, C., Lamaty, G., Bessiere, J.M., Chalcat,
J.C., Garry, R.P., 1995. Aromatic plants of tropical central Africa. XX:
the occurrence of 1-phenyl-1,3,5-triyne in the essential oil of Bidens
pilosa L. from Cameroon. Flavour and Fragrance Journal 10, 97–100.