Available online at www.sciencedirect.com

www.elsevier.com/locate/yexpr

Experimental Parasitology 118 (2008) 324–330

Trypanosoma cruzi: Antiprotozoal activity of parthenolideobtained from Tanacetum parthenium (L.) Schultz Bip. (Asteraceae,

Compositae) against epimastigote and amastigote forms

Erika Izumi a, Luis Gustavo Morello b, Tania Ueda-Nakamura b,Sueli Fumie Yamada-Ogatta a, Benedito Prado Dias Filho a,b,

Diogenes Aparıcio Garcia Cortez c, Izabel Cristina Piloto Ferreira c,Jose Andres Morgado-Dıaz d, Celso Vataru Nakamura a,b,*

a Programa de Pos-graduacao em Microbiologia, Universidade Estadual de Londrina, Rodovia Celso Garcia Cid, PR 445, Km 380,

CEP 86051-990, Campus Universitario, Londrina, Parana, Brazilb Departamento de Analises Clınicas, Laboratorio de Microbiologia Aplicada aos Produtos Naturais e Sinteticos, Bloco I-90 Sala 123 CCS,

Universidade Estadual de Maringa, Av. Colombo 5790, CEP 87020-900, Maringa, Parana, Brazilc Departamento de Farmacia e Farmacologia, Universidade Estadual de Maringa, Brazild Divisao de Biologia Celular, Instituto Nacional do Cancer, Rio de Janeiro, RJ, Brazil

Received 21 June 2007; received in revised form 13 August 2007; accepted 29 August 2007Available online 7 September 2007

Abstract

This study reports the activity of crude extracts, fractions and parthenolide (pure compound) obtained from Tanacetum parthenium

against two forms of the parasite Trypanosoma cruzi. Feverfew is a traditional herbal medicine that has been used for the treatment ofmigraine, fever and arthritis. Activity against epimastigote forms was observed for crude extracts, fractions and parthenolide, and a pro-gressive increase in the antitrypanosomal effect was observed in the course of the purification process. The pure compound showedIC50/96h and IC90/96h of 0.5 lg/ml and 1.25 lg/ml, respectively. The cytotoxic effect of parthenolide in LLMCK2 cells was 3.2 lg/ml(CC50/96h) and the selectivity index was 6.4. No hemolysis was detected for the pure compound. The internalization index of T. cruzi

in LLMCK2 cells was reduced almost 51% at the concentration of 2 lg/ml of parthenolide, and 96.6% at 4 lg/ml. Scanning and trans-mission electron microscopy permitted observation of morphological modifications and ultrastructural alterations.� 2007 Elsevier Inc. All rights reserved.

Index Descriptors and Abbreviations: Trypanosoma cruzi; Epimastigotes; Tanacetum parthenium; In vitro study

1. Introduction

Tanacetum parthenium (Asteraceae), popularly knownas feverfew, is a traditional herbal medicine that has been

0014-4894/$ - see front matter � 2007 Elsevier Inc. All rights reserved.

doi:10.1016/j.exppara.2007.08.015

* Corresponding author. Address: Departamento de Analises Clınicas,Laboratorio de Microbiologia Aplicada aos Produtos Naturais e Sintet-icos, Bloco I-90 Sala 123 CCS, Universidade Estadual de Maringa, Av.Colombo 5790, CEP 87020-900, Maringa, Parana, Brazil. Fax: +55 443261 4860.

E-mail address: cvnakamura@uem.br (C.V. Nakamura).

used for the treatment of fever, migraine and arthritis inAsia and Europe for centuries (Won et al., 2004). In Brazil,this plant is available commercially in capsules (Tanaceto)for prophylactic treatment of migraine. Its effects havebeen attributed to the content of sesquiterpene lactones,notably parthenolide. Crude extracts of this herb areknown for their antimicrobial and anti-inflammatory prop-erties (Parfitt, 1999; Tiuman et al., 2005a). Parthenolide hasbeen studied for its ability to inhibit IjB kinase, which isimportant in cytokine-mediated signaling (Kwok et al.,2001). The compound is also known for inducing apoptosis

E. Izumi et al. / Experimental Parasitology 118 (2008) 324–330 325

and metastasis reduction in carcinoma cells (Ralstin et al.,2006) and shows antiparasitic activity against Leishmania

amazonensis (Tiuman et al., 2005b).The protozoan parasite Trypanosoma cruzi is the etio-

logic agent of Chagas’ disease. It has a high prevalence inLatin America, affecting about 18 million people, and 100million people live in high-risk areas (WHO, 2000). Itstreatment is still a challenge, because the only two availabledrugs, nifurtimox and benznidazole, have severe side effects(De Castro, 1993). As a consequence, few people completethe long course of treatment, and most of them give it upbecause of side effects. Treatment options for T. cruzi infec-tions are suboptimal due to the toxicity and limited effec-tiveness of the available drugs. From the beginning of the1990s and with the revival of phytotherapy, biologicalresearch on the antiprotozoal effects of plants has beenongoing. Natural compounds of medicinal plants haveshown lower toxicity together with higher specificity, creat-ing an optimistic view of new treatments for diseases(Cowan, 1999; Luize et al., 2006).

Several studies of crude plant extracts have already iden-tified potential compounds to treat Chagas’ disease. Crudeextracts and fractions from some Colombian plants of thefamily Annonaceae (Osorio et al., 2007) were effectiveagainst epimastigote forms of T. cruzi as well as medicinalplants from Argentina (Sulsen et al., 2006), China and theMediterranean (Schinella et al., 2002). In addition to impor-tant in vitro studies, in vivo experiments also have been donewith different plant extracts and isolated compounds forevaluating their effect in host organisms. Bizimana et al.(2006) found that the aqueous extract of Terminalia avicen-

nioides was able to reduce parasitemia and showed low cyto-toxicity to mammal cells. Ferreira et al. (2007) studied theeffect of the alkaloidal extract and purified compound can-thin-6-one from Zanthoxylum chiloperone in acutely andchronically infected mice. The treatments led to 80–100%animal survival compared to untreated controls.

Our objective was to determine the antiprotozoal activ-ity of the crude extract, fractions and an isolated com-pound (parthenolide) obtained from T. parthenium

against T. cruzi epimastigote and amastigote forms. Thecytotoxic activity against sheep blood cells and Rhesusmonkey kidney cells was also determined.

2. Materials and methods

2.1. Plant material

Dried aerial parts of T. parthenium (Lot 166871) werekindly furnished by the Herbarium Laboratorio BotanicoLtda (Colombo, Parana, Brazil).

2.2. Extraction, fractionation and identification of the

compound

One thousand grams of plant material was sequen-tially extracted by exhaustive maceration in 10 l etha-

nol:water (9:1) at room temperature in the dark. Thesupernatants were filtered, concentrated by rotary evapo-rator at approximately 40 �C and lyophilized. The pow-der resulting from lyophilization, soluble in water, wastermed the aqueous crude extract (WCE). The residuewas dissolved in ethanol or ethyl-acetate and was namedaccording to the solvent used. The aqueous (WCE), eth-anolic (ACE) and ethyl-acetate (ECE) crude extractswere assayed for their activity against epimastigote formsof T. cruzi. Subsequently, the ethanolic crude extract waschromatographed on a silica-gel 60 column (70–230mesh—Merck�) using hexane, dichloromethane, ethyl-acetate and methanol. Each fraction (hexane, HF;dichloromethane, DF; ethyl-acetate, EF; methanol, MF)was tested for antiprotozoal activity. The hexane anddichloromethane fractions were subfractioned, resultingin isolation of a pure compound that was identified assesquiterpene lactone, 4,5-epoxy-germacra-1-(10),11(13)-dien-12,6-olide (parthenolide) by chromatography-massspectrometry (Micromass Quattro LC); nuclear magneticresonance (NMR; Gemini 2000 BB; Varian), 1H NMR(300 MHz), and 13C NMR (75.5 MHz) analyses inCDCl3; infrared analysis (Bomem-MV 100; Hartmann& Braun-Michelson); and UV analysis (CARY 1E UV–vis; Varian).

2.3. Parasites and culture procedure

Epimastigote forms of T. cruzi Y strain were main-tained at 28 �C by weekly transfers in liver infusion tryp-tose medium-LIT (Camargo, 1964) supplemented with10% heat-inactivated fetal bovine serum (FBS) (GibcoInvitrogen Corporation, NY, USA). Amastigote formswere grown in monolayers of LLMCK2 cells in DMEM(Gibco Invitrogen) with 10% fetal bovine serum in 5%CO2 at 37 �C.

2.4. Antitrypanosomal activity

Stock solutions of the crude plant extracts, fractions,and parthenolide at the indicated concentrations were pre-pared in dimethylsulfoxide (DMSO–Sigma Chemical Co.,St. Louis, MO, USA). The DMSO final concentration inthe assay did not exceed 1.0%. For the assay, parasites wereharvested during the exponential phase of growth, resus-pended in LIT and plated on 24-well plates at1 · 106 cells/ml. In each well were included different con-centrations of crude plant extracts (1 lg/ml up to1000 lg/ml), fractions (1 lg/ml up to 100 lg/ml) or parthe-nolide (1 lg/ml up to 10 lg/ml) and incubated for 96 h at28 �C. Cell density was performed by counting of parasitesin a hemocytometer (Improved Double Neubauer) with anOlympus CX 31 light microscope. Benznidazole (N-benzyl-2-nitro-1-imidazolacetamide—Roche Pharmaceuticals, Riode Janeiro, Brazil) was used as the reference drug. Allassays were carried out twice in duplicate on differentoccasions.

326 E. Izumi et al. / Experimental Parasitology 118 (2008) 324–330

2.5. Interaction with host cells

In a 24-well plate, 2.5 · 105 cells/ml (LLMCK2) inDMEM medium, supplemented with 10% FBS were platedon round cover slips. The plate with the cells was incubatedfor 24 h at 37 �C, 5% CO2 until confluence was achieved.Trypomastigotes were added to the wells at a concentrationof 20 parasites per cell and incubated for 24 h. After thisperiod, the wells were washed three times with PBS forremoving the non-internalized parasites. Cells were thenincubated for a maximum period of 96 h with parthenolideat 2 or 4 lg/ml. Control and treated cells were fixed after48, 72 and 96 h with methanol, stained with Giemsa andthe cover slips were permanently prepared with Entellan�

(Merck). The internalization index (number of infectedcells · mean of amastigotes per cell) was determined bycounting 200 cells in a light microscope.

2.6. Cytotoxicity assay

The cytotoxic effect was evaluated against LLMCK2 linecells in 96-well plates. A suspension of 2.5 · 104 cells wasadded to each well, and then grown as a monolayer for24 h at 37 �C in 5%-CO2 air mixture. After this period, dif-ferent concentrations of crude extracts, fractions, or par-thenolide were added to the wells and the plate wasincubated for 96 h under the same conditions as above.The cells were fixed in 10% trichloroacetic acid at 4 �Cfor 1 h, washed five times with distilled water, and allowedto dry at room temperature. A solution of 4% sulforhoda-mine B (in 1% acetic acid) was added to each well and theplate was kept protected from light for 30 min at 4 �C. Thewells were washed four times with 1% acetic acid and analiquot of 150 ll, 10 mM Tris-base was added and homog-enized for 15 min. The absorbance was read at 530 nm in amicroplate spectrophotometer (Bio Tek—Power WaveXS). Data were calculated as the percentage of inhibition.The concentration of 50% cellular toxicity (CC50) wasdefined as the concentration which reduced the OD530 oftreated cells to 50% of that of untreated cells.

2.7. Hemolytic assay

This experiment also evaluated the cytotoxicity of T.

parthenium. A 4% suspension of freshly defibrinized sheepblood was prepared by adding blood to sterile 5% glucosesolution. One milliliter of red blood cell suspension wasadded to each test tube, and different concentrations ofcrude plant extracts, fractions and Amphotericin B(AMPB), as the reference drug, which may cause anemia,were added, gently mixed and incubated at 37 �C for120 min. Samples were centrifuged at 1000g for 10 min.Absorbance of the supernatant was determined at540 nm. Hemolysis was calculated according to the equa-tion: Hemolysis (%) = 100 – [(Ap�As)/(Ap�Ac) · 100];where Ap, As and Ac are the absorbance of the positive con-trol, test sample and negative control, respectively. The

negative control was the red blood cell suspension aloneor with 1.0% DMSO, and the positive control was redblood cells lyzed with Triton X-100. These tests were per-formed in duplicate on separate occasions.

2.8. Scanning electron microscopy

After treatment with parthenolide (IC50/96h = 0.5 lg/mland IC90/96h = 1.25 lg/ml), epimastigote forms werewashed with 0.01 M PBS and fixed in 2.5% glutaraldehydein 0.1 M sodium cacodylate buffer for 2 h. The parasiteswere washed three times in 0.1 M sodium cacodylate buffer,placed on the poly-L-lysine-coated coverslip, dehydrated indifferent concentrations of ethanol, critical point-dried inCO2, sputter-coated with gold and observed in a ShimadzuSS-550 scanning electron microscope.

2.9. Transmission electron microscopy

After the treatments with parthenolide at IC50/96h andIC90/96h concentrations, epimastigote forms were washedin 0.01 M phosphate-buffered saline and fixed in 2.5% glu-taraldehyde in 0.1 M sodium cacodylate buffer. The cellswere postfixed in a solution containing 1% osmium tetrox-ide and 0.8% potassium ferrocyanide in 0.1 M cacodylatebuffer, washed in the same buffer, dehydrated in differentconcentrations of acetone and embedded in Epon resin.Thin sections were stained with uranyl acetate and lead cit-rate, and examined in a Zeiss 900 transmission electronmicroscope.

2.10. Statistical analyses

Statistical analysis was performed with the programGraphPad Prism 4 (GraphPad Software, San Diego, CA,USA). Student’s t-test was applied and a p-value less than0.05 was regarded as significant.

3. Results

The compound isolated from the hexane and dichloro-methane fractions of T. parthenium was identified as 4,5-epoxy-germacra-1-(10),11(13)-dien-12,6-olide (partheno-lide). The 1H, 13C NMR spectral data analyses in CDCl3;infrared and UV analysis were identical to the valuesreported by Tiuman et al. (2005b).

3.1. Antitrypanosomal activity

The effects of the crude plant extracts on epimastigoteforms of T. cruzi Y strain are shown in Fig. 1a. At1000 lg/ml all the crude extracts showed similar effects,inhibiting more than 90% of the parasites’ growth. Atlow concentrations, ACE was the most effective extract,with 32 and 68.5% growth inhibition of parasites at 1 lg/ml and 10 lg/ml, respectively. At the same concentrations,no significant activity was obtained with the polar crude

Fig. 1. Effects of Tanacetum parthenium on epimastigote forms ofTrypanosoma cruzi cultured for 96 h in the presence of the indicatedconcentrations. (a) Crude extracts (WCE, ACE and ECE). (b) Fractions(HF, DF, EF and MF). (c) Parthenolide and Benznidazole. The results arefrom two experiments in duplicate and are shown as percentages ±standard deviations of growth inhibition in relation to untreated proto-zoans. *Statistical difference (p < 0.05).

Table 1Cytotoxicity, growth inhibition, and selectivity index of the crude extracts,fractions and the isolated compound parthenolide of Tanacetum

parthenium

Drugs CC50/96h (lg/ml) IC50/96h (lg/ml) SI*

ACE 38.0 ± 13.0 3.0 ± 1.0 12.6 ± 0.1ECE 265.0 ± 15.0 50.0 ± 12.0 5.3 ± 1.0WCE 432.5 ± 32.5 280.0 ± 5.0 1.5 ± 0.09DF 14.5 ± 0.5 2.1 ± 1.1 6.9 ± 1.8HF 170.0 ± 20.0 2.1 ± 0.4 80.9 ± 6.1EF 74.0 ± 6.0 79.0 ± 4.0 0.9 ± 0.03MF 265.0 ± 5.0 40.0 ± 8.0 6.6 ± 1.2Parthenolide 3.2 ± 0.08 0.5 ± 0.2 6.4 ± 2.8

* Selectivity index (SI) = CC50/IC50.

Fig. 2. Hemolytic properties of ethyl-acetate crude extract (ACE),dichloromethane fraction (DF) and pathenolide obtained from aerialparts of Tanacetum parthenium. Amphotericin B (AMPB) was included inthe assay as a reference drug.

E. Izumi et al. / Experimental Parasitology 118 (2008) 324–330 327

plant extracts (ECE and WCE). The IC50/96h (50% inhibi-tory concentration after 96 h of incubation) values forACE, ECE and WCE crude plant extracts were 3.0, 50.0and 280.0 lg/ml, respectively. The results obtained withfractions are presented in Fig. 1b. The apolar fractionsHF and DF showed very strong activity against parasiteproliferation. On the other hand, the polar fractions (EFand MF) showed no significant effect on epimastigoteforms of T. cruzi. The IC50/96h calculated for both HFand DF fractions were 2.1 lg/ml, and for EF and MF were79.0 lg/ml and 40.0 lg/ml, respectively. Fig. 1c shows theeffect of the purified compound parthenolide and benzni-dazole (BZ), a drug used as reference, on the growth of epi-

mastigote forms. The IC50/96h concentration forparthenolide was 0.5 lg/m, while that for BZ was 1.8 lg/ml. Student’s t-test (p < 0.05) indicated significant differ-ences between these two compounds, confirming the effi-cacy of parthenolide.

3.2. Cytotoxicity and hemolytic assays

The 50% cytotoxic concentrations after 96 h of incuba-tion (CC50/96h) of crude extracts, fractions and partheno-lide on LLMCK2 cells are shown in Table 1. The lowestcytotoxicity was observed for WCE, with CC50/96h of432.5 lg/ml, whereas ECE was similar to the MF fraction,with a CC50/96h of 265.0 lg/ml. The CC50/96h determinedfor parthenolide was 3.2 lg/ml, and the selectivity indexwas 6.4.

The ACE crude extract and DF fraction at 500 lg/mlcaused 93% and 11% hemolysis, respectively. The EC50/2h

(50% effective concentration after 2 h of incubation) calcu-lated for ACE and DF were 240 lg/ml and 700 lg/ml,respectively. Parthenolide and benznidazole did not causeany lysis of erythrocytes at concentrations below 100 lg/ml. Cells treated with Amphotericin B (AMPB) at 8.4 lg/ml showed 50% lysis, while at 100 lg/ml they showed90.5% lysis (Fig. 2). The red blood cell control with or with-

328 E. Izumi et al. / Experimental Parasitology 118 (2008) 324–330

out 1.0% DMSO did not show hemolysis, whereas the Tri-ton X-100 positive control showed 100% hemolysis.

3.3. Interaction with host cells

The effect of the drug on the intracellular amastigoteswas observed during 96 h of incubation. For the first per-iod analyzed at 48 h, the interaction index was calculatedas 394 for 2 lg/ml and 40 for 4 lg/ml of parthenolide,which represents a decrease of 36 and 93% compared tothe control (IS = 621). At 72 h, the interaction index was465 and 47, for 2 lg/ml and 4 lg/ml, respectively, whilethe control was determined as 1164. For the maximum time

Fig. 3. Interaction of Trypanosoma cruzi with host cells LLMCK2.Interaction index after 48–96 h of treatment with parthenolide.

Fig. 4. Scanning electron microscopy of epimastigote forms of Trypanosoma

Bars = 2 lm in (A–C); Bars = 1 lm in (D) and (E).

studied, 96 h, the interaction index was 1081 for the con-trol, and 528.5 and 36.5 for 2 and 4 lg/ml, correspondingto 51% and 96.6% decrease, respectively (Fig. 3).

3.4. Scanning and transmission electron microscopy

Morphological alterations of T. cruzi treated with par-thenolide were observed by scanning electron microscopy.In epimastigote forms, the IC50 (0.5 lg/ml) of the com-pound mainly caused distortion of the parasite body, swell-ing and flattening (Fig. 4B and C). The concentration of1.25 lg/ml (IC90) caused a decrease in cell length, and cellshape was severely affected (Fig. 4D and E). Fig. 4A showsthe characteristic elongated shape of an untreated proto-zoan with a terminal flagellum and flagellar pocket.

Ultrastructural alterations of epimastigote forms wereobserved by transmission electron microscopy. Partheno-lide IC50/96h and IC90/96h concentrations caused alterationsin organelles such as the nucleus, reservosomes, mitochon-drion and vacuoles, as well as the arrangement of internalmembranes (Fig. 5). Many cells treated with 0.5 lg/mlshowed two or three nuclei (Fig. 5C), also an increasednumber of reservosomes (Fig. 5D) and formation of mem-branes around them (Fig. 5B). At 1.25 lg/ml, normal cellswere not observed, and an effect of swelling on the mito-chondrion and distortion of internal membranes, togetherwith an excess of vacuoles, were seen (Fig. 5E and F). Someconcentric structures, myelin-like figures, appeared in theanterior part of the cell in both treatments. Fig. 5A showscontrol cells, which have a terminal flagellum, nucleus andthe single giant mitochondrion with a branched structure

cruzi. (A) Control; (B) and (C) IC50 treated; (D) and (E) IC90 treated.

Fig. 5. Transmission electron microscopy of Trypanosoma cruzi epim-astigotes. (A) Control; (B–D), Treated IC50; (E) and (F), IC90; N, Nucleus;K, Kinetoplast; R, Reservosomes; arrowhead, Membranes envelopingreservosomes; arrow, Concentric membranes; asterisks, vacuoles.Bar = 1 lm.

E. Izumi et al. / Experimental Parasitology 118 (2008) 324–330 329

containing a disk-shaped aggregate of DNA termed thekinetoplast.

4. Discussion and conclusion

The data from our study indicated that parthenolide iseffective against proliferative epimastigote and amastigoteforms of T. cruzi Y strain. A progressive increase in theantitrypanosomal effect was observed in the course of thepurification process of the compound. Bioassay-guidedfractionation led us to hexane and dichloromethane frac-tions, which showed the best activity against epimastigoteforms of T. cruzi Y strain. The sesquiterpene lactone, par-thenolide, was almost four times more effective against epi-mastigote forms than the reference drug, benznidazole.This is in accordance with previous studies in which sesqui-terpenes have shown interesting antiparasitic properties(Kayser et al., 2003). For instance, sesquiterpene lactonesisolated from different plant species have shown in vitro

activity against T. cruzi (Brengio et al., 2000; Jimenez-Ortizet al., 2005). When tested against Leishmania, this sesqui-terpene lactone had a strong activity against the infectivepromastigote form, with an IC50 (0.37 lg/ml) similar to

that obtained for epimastigote forms of T. cruzi (Tiumanet al., 2005b).

The aqueous crude extract had the lowest cytotoxicityagainst LLMCK2 cells but showed a very low selectiveindex. The best selectivity was obtained with the hexanefraction, and the hemolytic effect of dichloromethane wassmaller than those of crude plant extracts. Dichlorometh-ane, methanol and hydroalcoholic crude extracts, togetherwith purified compounds obtained from Lychnophora poh-

lii were tested against T. cruzi. The sesquiterpene lactonesshowed the best activity, and goyazensolide showed thelowest IC50, 15.8 lg/ml (Grael et al., 2005). Contrary toour study, these sesquiterpene lactones showed stronghemolytic activity, which was not observed forparthenolide.

The experiments of interaction of T. cruzi withLLMCK2 cells showed the activity of parthenolide onamastigotes inside mammalian cells. Fifty-one percentreduction of the interaction index was observed at 2 lg/ml after 96 h, whereas at a concentration of 4 lg/ml thereduction was even higher, but partial cytotoxicity wasobserved. It does not influence the internalization of theparasites, because the drug incubation starts after the inter-nalization process. Taking into account that parthenolidewas able to keep the interaction index constant (around400–500 and 40 for treatments with 2 lg/ml and 4 lg/ml,respectively) during the treatment of cells infected with T.

cruzi, this may indicate an effect on the internal parasitereplication.

Scanning electron microscopy revealed severe altera-tions in the morphology of the epimastigote forms of theparasite caused by parthenolide. Distortion of the cell bodyand decrease of the body and flagellum length wereobserved. The structure and shape of the plasma mem-brane is determined by submembrane microtubules. Itwas previously observed that parthenolide exerts in vitro

stimulatory activity on tubulin assembly, inducing the for-mation of well-organized microtubule polymers (Migliettaet al., 2004). Considering the effect of the drug on microtu-bule polymerization, we cannot discard the possibility thatthis process may have caused the alterations seen by exter-nal observation of the cells. The ultrastructure analyses bytransmission electron microscopy showed multinuclearcells, an increase of reservosomes, swelling of the mito-chondrion, which could be better seen in the kinetoplastmembrane and internal membrane organization, with theformation of concentric structures, myelin-like figures.The increased number of reservosomes may indicate adirect relationship to increased activity of cysteine protein-ase caused by parthenolide, as previously reported in Leish-

mania (Tiuman et al., 2005b). The effect of parthenolide onthe mitochondrion may suggest an effect on structural com-ponents in this organelle, possibly proteins in the mem-branes or enzymes. Glutathione (GSH) in cells functionsto prevent the oxidation of molecules, especially in themitochondrion and nucleus. Studies with mammalian cellsshowed that the redox state is involved in regulation of the

330 E. Izumi et al. / Experimental Parasitology 118 (2008) 324–330

cell cycle and cell death. GSH, the main intracellular anti-oxidant, is important in these processes: its depletion leadsto cell death, and its increase inhibits cell proliferation. Par-thenolide is reported to have strong antioxidant activity,increasing the levels of glutathione in mammal cells (Herre-ra et al., 2005). In the present study, whether parthenolideis causing its antiproliferative effect through GSH regula-tion, remains to be determined. Disruption of internalmembranes was detected, mainly close to the region wherethe Golgi apparatus is typically found, but it was not pos-sible to conclude that the concentric structures were due tobreakdown of this organelle.

Natural products show good potential for the discoveryof new agents to combat tropical diseases such as Chagas’disease. In this study the biological effect of T. parthenium

on T. cruzi Y strain was reported. Extracts of this plantmay result in potential sources for the development ofmore effective compounds for treatment of Chagas’ dis-ease. As shown in the present study, bioguided fraction-ation of T. parthenium resulted in the isolation of pureparthenolide with trypanocidal activity. This activity repre-sents an exciting advance in the search for new antiproto-zoal agents from natural sources. However, furtherin vitro and in vivo studies are necessary to confirm theseresults.

Acknowledgments

This study was supported through grants from DECIT/SCTIE/MS and MCT by Conselho Nacional de Desen-volvimento Cientıfico e Tecnologico (CNPq), Financiadorade Estudos e Projetos (FINEP), Programa de Nucleos deExcelencia (PRONEX/Fundacao Araucaria) and Progra-ma de Pos-graduacao em Microbiologia da UniversidadeEstadual de Londrina.

References

Bizimana, N., Tietjen, U., Zessin, K.H., Diallo, D., Djibril, C., Melzig,M.F., Clausen, P.H., 2006. Evaluation of medicinal plants from Malifor their in vitro and in vivo trypanocidal activity. J. Ethnopharmacol.103, 350–356.

Brengio, S.D., Belmonte, S.A., Guerreiro, E., Giordano, O.S., Pietrobon,E.O., Sosa, M.A., 2000. The sesquiterpene lactone dehydroleucodine(DhL) affects the growth of cultured epimastigotes of Trypanosoma

cruzi. J. Parasitol. 86, 407–412.Camargo, E.P., 1964. Growth and differentiation in Trypanosoma cruzi.

Origin of metacyclic trypanosomes in liquid media. Revista InstitutoMedicina Tropical Sao Paulo 6, 93–100.

Cowan, M.M., 1999. Plant products as antimicrobial agents. Clin.Microbiol. Rev. 12, 564–582.

De Castro, S.L., 1993. The challenge of Chagas’ disease chemotherapy: anupdate of drugs assayed against Trypanosoma cruzi. Acta Tropica. 53,83–98.

Ferreira, M.E., Nakayama, H., Arias, A.R., Schinini, A., Bilbao, N.V.,Serna, E., Lagoutte, D., Soriano-Agaton, F., Poupon, E., Hocquem-iller, R., Fournet, A., 2007. Effects of canthin-6-one alkaloids fromZanthoxylum chiloperone on Trypanosoma cruzi-infected mice. J.Ethnopharmacol. 109, 258–263.

Grael, C.F.F., Albuquerque, S., Lopes, J.L.C., 2005. Chemical constitu-ents of Lycnophora pohlii and trypanocidal activity of plant crudeextracts and of isolated compounds. Fitoterapia 76, 73–82.

Herrera, F., Martin, V., Rodriguez-Blanco, J., Garcia-Santos, G.,Antolin, I., Rodriguez, C., 2005. Intracellular redox state regula-tion by parthenolide. Biochem. Biophys. Res. Commun. 332, 321–325.

Jimenez-Ortiz, V., Brengio, S.D., Giordano, O., Tonn, C., Sanchez, M.,Burgos, M.H., Sosa, M.A., 2005. The trypanocidal effect of sesquiter-pene lactones helenalin and mexicanin on cultured epimastigotes. J.Parasitol. 91, 170–174.

Kayser, O., Kiderlen, A., Croft, S., 2003. Natural products as antiparasiticdrugs. Parasitol. Res. 90, 55–62.

Kwok, B.H.B., Koh, B., Ndubuisi, M.I., Elofsson, M., Crews, C.M., 2001.The anti-inflammatory natural product parthenolide from the medic-inal herb feverfew directly binds to and inhibits IjB kinase. Chem.Biol. 8, 759–766.

Luize, P.S., Ueda-Nakamura, T., Dias-Filho, B.P., Cortez, D.A.G.,Nakamura, C.V., 2006. Activity of neolignans isolated from Piper

regnellii (Miq.) C.DC. var. pallescens (C.DC.) Yunck against Trypan-

osoma cruzi. Biol. Pharm. Bull. 29, 2126–2130.Miglietta, A., Bozzo, F., Gabriel, L., Bocca, C., 2004. Microtubule-

interfering activity of parthenolide. Chemico-Biological Interact. 149,165–173.

Osorio, E., Arango, G.J., Jimenez, N., Alzate, F., Ruiz, G., Gutierrez, D.,Paco, M.A., Gimenez, A., Robledo, S., 2007. Antiprotozoal andcytotoxic activities in vitro of Colombian Annonaceae. J. Ethnophar-macol. 111, 630–635.

Parfitt, K., 1999. Martindale: The Complete Drug Reference, 32nd ed. ThePharmaceutical Press, p. 447.

Ralstin, M.C., Gage, E.A., Yip-Schneider, M.T., Klein, P.J., Wiebke,E.A., Schmidt, C.M., 2006. Parthenolide cooperates with NS398 toinhibit growth of human hepatocellular carcinoma cells througheffects on apoptosis and G0–G1 cell cycle arrest. Mol. Cancer Res. 4,387–399.

Schinella, G.R., Tournier, H.A., Prieto, J.M., Rıos, J.L., Buschiazzo, H.,Zaidenberg, A., 2002. Inhibition of Trypanosoma cruzi growth bymedical plant extracts. Fitoterapia 73, 569–575.

Sulsen, V., Guida, C., Coussio, J., Paveto, C., Muschietti, L.,Martino, V., 2006. In vitro evaluation of trypanocidal activity inplants used in Argentine traditional medicine. Parasitol. Res. 98,370–374.

Tiuman, T.S., Ueda-Nakamura, T., Dias Filho, B.P., Cortez, D.A.G.,Nakamura, C.V., 2005a. Studies on the effectiveness of Tanacetum

parthenium against Leishmania amazonensis. Acta Protozool. 44, 245–251.

Tiuman, T.S., Ueda-Nakamura, T., Cortez, D.A.G., Dias Filho, B.P.,Morgado-Dıaz, J.A., De Souza, W., Nakamura, C.V., 2005b. Antile-ishmanial activity of parthenolide, a sesquiterpene lactone isolatedfrom Tanacetum parthenium. Antimicrob. Agents Chemother. 49, 176–182.

WHO. 2000. Special programme for research and training in tropicaldisease (TDR). Natural Product Parasitology Diseases TDR News 62,4.

Won, Y.K., Ong, C.N., Shi, X., Shen, H.M., 2004. Chemopreventiveactivity of parthenolide against UVB-induced skin cancer and itsmechanisms. Carcinogenesis 25, 1449–1458.