UNIVERSITÉ D’ANTANANARIVO ------FACULTE DES SCIENCES ------ECOLE DOCTORALE Valorisation des Ressources Naturelles Renouvelables

MÉMOIRE

pour l’obtention de l’

HABILITATION À DIRIGER DES RECHERCHES

Etudes chimique et biologique pour l’assurance qualité et la valorisation des plantes

Volume 1 : Publications et communications

par

RAOELISON Emmanuel Guy Maître de Conférences au Département Pharmacie de la Faculté de Médecine d’Antananarivo

présenté le 26 août 2016 devant le jury composé de :

Président : Madame RAKOTOVAO Marcelle, Professeur Titulaire Directeur : Monsieur RANDRIANTSOA Adolphe, Professeur Titulaire Rapporteurs : Madame RAZAFIMAHEFA RAMILISON Dorothée, Professeur Titulaire Monsieur RAMANOELINA Panja Armand, Professeur Titulaire Examinateurs : Madame RAZANAMAHEFA Bakonirina, Professeur Titulaire Monsieur RASAMINDRAKOTROKA Andry, Professeur Titulaire

Malgré les progrès réalisés en médecine, en biologie, en chimie durant ces dernières décennies, l’arsenal thérapeutique dont nous disposons s’avère encore insuffisant pour faire face aux nombreuses maladies telles le cancer, la malaria, les infections fongiques, bactériennes, virales, le diabète, etc. S’ajoutent à cette liste l’apparition de nouvelles maladies comme le SIDA, la résurgence de souches microorganiques résistantes aux traitements classiques. Le développement de nouveaux médicaments par la découverte de nouveaux composés actifs, nouvelles têtes de séries, reste toujours d’actualité dans la recherche. Plusieurs sources sont déjà explorées à ce jour dont les plantes. En effet, les plantes fournissent de nombreux principes actifs de médicaments dont la quinine, la vinblastine, les cardiotoniques, l’acide acétylsalicylique entre autres qui sont encore utilisés de nos jours. Dans la médecine traditionnelle, ces plantes sont utilisées sous diverses formes de préparations pour traiter de nombreuses affections. L’intégration de la médecine traditionnelle dans le système de santé selon la recommandation de l’OMS confirme les potentialités indéniables de cette forme de médecine et témoigne sa reconnaissance officielle. Pourtant, la plante n’est pas dénuée d’effet toxique et le contrôle qualité doit être exigé à toute préparation végétale à des fins thérapeutiques.

L’objectif de nos travaux de recherche est donc de valoriser ces potentialités des plantes par la recherche de nouveaux composés d’intérêt thérapeutique d’une part, et par la mise au point et validation de méthode analytiques pour assurer la qualité de ces plantes et/ou préparations à base de plantes, d’autre part.

Les publications relatives à ces travaux de recherche sont compilées dans ce Volume 1, publications et communications de mémoire HDR tandis que la synthèse de ces travaux de recherche est donnée dans le Volume 2. Au total, ces travaux ont fait l’objet de 15 publications parues dans des revues scientifiques de niveau international et 1 article dans un livre récemment sorti en 2016 intitulé Essential Oils in Food Preservation, Flavor and Safety. Parmi les revues scientifiques, on cite Natural Product Communication, Ethnopharmacologia, Phytochemical Analysis, Food and Chemical Toxicology, Journal of Chromatography B, Journal of Pharmaceutical and Biomedical Analysis, BMC Complementary and Alternative Medicine, Molecules, Journal of Essential Oil Bearing Plants, Analytical Chemistry Letters, Canadian Journal of Microbiology, Tropical Journal of Pharmaceutical Research, Journal of Ethnopharmacology, et Journal of Food Science.

Ces résultats de travaux ont été également présentés sous forme de communications scientifiques orale ou affichée.

PUBLICATIONS PUBLICATIONS Les 16 articles ci-après composant ce Volume 1 sont présentés selon leur ordre de citation dans le Volume 2, synthèse des travaux ; ils sont notés P.01 à P.16 :

P.01 : GE. Raoelison, M. Rafamantanana, R. Razafindrazaka, A. Randriantsoa, S. Urverg- Ratsimamanga, N. Morel, J. Quetin-Leclercq. Vasorelaxant alkaloids from Spirospermum penduliflorum (MENISPERMACEAE) a plant used to treat hypertension in malagasy traditional medicine. Natural Product Communications 2013, 8(5), 575-578.

P.02 : H. Rakotoarimanana, H. Rafatro, V. Jeannoda, G. Raoelison, RB. Robijaona, S. Ratsimamanga-Urverg. Composé antiplasmodial isolé du Tsilaitra (Noronhia divaricata Perr., Oleaceae). Ethnopharmarcologia 2008, 41, 71-73.

P.03 : E.F. Queiroz, J-L. Wolfender, GE. Raoelison, K. Hostettmann. Determination of the absolute configuration of 6-alkylated -pyrones from Ravensara crassifolia by LC-NMR. Phytochemical Analysis 2003, 14, 34-39.

P.04 : J-R. Ioset, G.E. Raoelison, K. Hostettmann. Detection of aristolochic acid in Chinese phytomedicines and dietary supplements used as slimming regimens. Food and Chemical Toxicology 2003, 41, 29-36.

P.05 : M.H. Rafamantanana, E. Rozet, G.E. Raoelison, K. Cheuk, S. Urverg Ratsimamanga, Ph. Hubert, J. Quetin-Leclercq. An improved HPLC-UV method for the simultaneous quantification of triterpenic glycosides and aglycones in leaves of Centella asiatica (L.) Urb (APIACEAE). Journal of Chromatography B 2009, 877, 2396-2402.

P.06 : MH. Rafamantanana, B. Debrus, EG. Raoelison, E. Rozet, P. Lebrun, S. Ratsimamanga- Urverg, P. Hubert, J. Leclercq-Quetin. Application of design experiments and design space methodology for the HPLC-UV separation optimization of aporphine alkaloids from leaves of Spirospermum penduliflorum Thouars. Journal of Pharmaceuticals and Biomedical Analysis 2012, 62, 23 – 32.

P.07 : S. Afoulous, H. Ferhout, EG. Raoelison, A. Valentin, B. Moukarzel, F. Couderc, J. Bouajila. Chemical composition and anticancer, antiinflammatory, antioxidant and antimalarial activities of leaves essential oil of Cedrelopsis grevei. Food and Chemical Toxicology 2013, 56, 352- 362.

P.08 : AT H. Mossa, TM. Heikal, M. Belaiba, EG. Raoelison, H. Ferhout, J. Bouajila. Antioxidant activity and hepatoprotective potential of Cedrelopsis grevei on cypermethrin oxidative stress in liver damage in male mice. BMC Alternative and Complementary Medicine 2015, 15, 251- 260.

P.09 : S. Afoulous, H. Ferhout, EG. Raoelison, A. Valentin, B. Moukarzel, F. Couderc and J. Bouajila. Helichrysum gymnocephalum essential oil : chemical composition and cytotoxic, antimalarial and antioxidant activities, attribution of the activity origin by correlations. Molecules 2011, 16 (10), 8273- 8291.

P.10 : D. Ben Hassine, D. Khlifi, H. Ferhout, G. Raoelison, J. Bouajila. Chapter 44 : Curry Plant (Helichrysum sp.) Oils dans Essential Oils in Food Preservation, Flavor and Safety. Academic Press, First Edition Elsevier, 2016, 395-403, ISBN : 9780124166417.

P.11 : DJR. Rabehaja, G. Raoelison, H. Ihandriharison, PAR. Ramanoelina, J. Casanova, F. Tomi. Volatile Components from Cymbopogon giganteus (Hochst) Chiov var. madagascariensis (A. Camus). Journal of Essential Oil Bearing Plants 2010, 13 (5), 522-527. P.12 : DJR. Rabehaja, EG. Raoelison, H. Ihandriharison, PA. Ramanoelina, A. Bighelli, J. Cassanova, F. Tomi. Combined analysis by chromatographic and spectroscopic techniques : Composition of the essential oil of ‘Andriambolamena’, a wild aromatic plant from Madagascar. TACL Analytical Chemistry Letters 2014, 4(1), 57-64.

P.13 : EF. Rakotonirina, G. Chataigné, EG. Raoelison, C. Rabemanantsoa, F. Munhaut, EJ Mondher, S. Urverg-Ratsimamanga, J. Marchand-Brynaert, AM. Corbisier, S. Deklerck, J. Quetin- Leclercq. Characterization of an endophytic whorlforming Streptomyces from Catharanthus roseus stems producing polyene macrolide antibiotic. Canadian Journal of Microbiology 2012, 58(5), 617-627.

P.14: EF. Rakotoniriana, JF. Rajaonarison, GE. Raoelison, JP. Rajaonarivelo, N. Manga, M. Solofoniaina, B. Rakotonirina, D. Randriamampionona, C. Rabemanantsoa, K. Cheuk, S. Urverg Ratsimamanga, J. Quetin Leclercq. Antimicrobial activity of 23 endemic plants in Madagascar. Tropical Journal of Pharmaceutical Research 2010, 9 (2), 165-171.

P.15: O. Rangasamy, G. Raoelison, E.F. Rakotoniriana, K. Cheuk, S. Urverg-Ratsimamanga, J. Quetin-Leclercq, A. Gurib-Fakim and A.H. Subratty. Screening for anti-infective properties of several plants of the Mauritian flora. Journal of Ethnopharmacology 2007, 109, 331-337.

P.16: Y. Yvon, EG. Raoelison, R. Razafindrazaka, A. Randriantsoa, M. Romdhane, N. Chabir, M.G. Mkaddem, J. Bouajila. Relation between chemical composition or antioxidant activity and antihypertensive activity of six essential oils. Journal of Food Science 2012, 77(8), H184-H191. P.01 : GE. Raoelison, M. Rafamantanana, R. Razafindrazaka, A. Randriantsoa, S. Urverg- Ratsimamanga, N. Morel, J. Quetin-Leclercq. Vasorelaxant alkaloids from Spirospermum penduliflorum (MENISPERMACEAE) a plant used to treat hypertension in malagasy traditional medicine. Natural Product Communications 2013, 8(5), 575-578.

NPC Natural Product Communications

EDITOR-IN-CHIEF HONORARY EDITOR DR. PAWAN K AGRAWAL PROFESSOR GERALD BLUNDEN Natural Product Inc. The School of Pharmacy & Biomedical Sciences, 7963, Anderson Park Lane, University of Portsmouth, Westerville, Ohio 43081, USA Portsmouth, PO1 2DT U.K. [email protected] [email protected]

EDITORS ADVISORY BOARD PROFESSOR ALEJANDRO F. BARRERO Department of Organic Chemistry, Prof. Berhanu M. Abegaz Prof. Karsten Krohn University of Granada, Gaborone, Botswana Paderborn, Germany Campus de Fuente Nueva, s/n, 18071, Granada, Spain [email protected] Prof. Viqar Uddin Ahmad Prof. Chiaki Kuroda

Karachi, Pakistan Tokyo, Japan PROFESSOR ALESSANDRA BRACA Dipartimento di Chimica Bioorganicae Biofarmacia, Prof. Øyvind M. Andersen Prof. Hartmut Laatsch Universita di Pisa, Bergen, Norway Gottingen, Germany via Bonanno 33, 56126 Pisa, Italy Prof. Giovanni Appendino Prof. Marie Lacaille-Dubois [email protected] Novara, Italy Dijon, France PROFESSOR DEAN GUO State Key Laboratory of Natural and Biomimetic Drugs, Prof. Yoshinori Asakawa Prof. Shoei-Sheng Lee School of Pharmaceutical Sciences, Tokushima, Japan Taipei, Taiwan

Peking University, Prof. Lee Banting Prof. Francisco Macias Beijing 100083, China Portsmouth, U.K. Cadiz, Spain [email protected]

Prof. Julie Banerji Prof. Imre Mathe PROFESSOR YOSHIHIRO MIMAKI Kolkata, India Szeged, Hungary

School of Pharmacy, Prof. Anna R. Bilia Tokyo University of Pharmacy and Life Sciences, Prof. Ermino Murano Florence, Italy Trieste, Italy Horinouchi 1432-1, Hachioji, Tokyo 192-0392, Japan [email protected] Prof. Maurizio Bruno Prof. M. Soledade C. Pedras

Palermo, Italy Saskatoon, Canada PROFESSOR STEPHEN G. PYNE

Department of Chemistry Prof. César A. N. Catalán Prof. Luc Pieters University of Wollongong Tucumán, Argentina Antwerp, Belgium

Wollongong, New South Wales, 2522, Australia Prof. Josep Coll [email protected] Prof. Peter Proksch Barcelona, Spain Düsseldorf, Germany

PROFESSOR MANFRED G. REINECKE Department of Chemistry, Prof. Geoffrey Cordell Prof. Phila Raharivelomanana Texas Christian University, Chicago, IL, USA Tahiti, French Polynesia

Forts Worth, TX 76129, USA Prof. Ana Cristina Figueiredo Prof. Luca Rastrelli [email protected] Lisbon, Portugal Fisciano, Italy

PROFESSOR WILLIAM N. SETZER Prof. Cristina Gracia-Viguera Prof. Monique Simmonds Department of Chemistry Murcia, Spain Richmond, UK

The University of Alabama in Huntsville Huntsville, AL 35809, USA Prof. Duvvuru Gunasekar Dr. Bikram Singh [email protected] Tirupati, India Palampur, India

PROFESSOR YASUHIRO TEZUKA Prof. Kurt Hostettmann Prof. John L. Sorensen Institute of Natural Medicine Lausanne, Switzerland Manitoba, Canada

Institute of Natural Medicine, University of Toyama, Prof. Martin A. Iglesias Arteaga 2630-Sugitani, Toyama 930-0194, Japan Prof. Valentin Stonik Mexico, D. F, Mexico Vladivostok, Russia [email protected]

Prof. Leopold Jirovetz PROFESSOR DAVID E. THURSTON Prof. Winston F. Tinto Vienna, Austria Department of Pharmaceutical and Biological Chemistry, Barbados, West Indies

The School of Pharmacy, Prof. Vladimir I Kalinin Prof. Sylvia Urban University of London, 29-39 Brunswick Square, Vladivostok, Russia Melbourne, Australia

London WC1N 1AX, UK Prof. Niel A. Koorbanally david.thurston @pharmacy.ac.uk Prof. Karen Valant-Vetschera Durban, South Africa Vienna, Austria

INFORMATION FOR AUTHORS

Full details of how to submit a manuscript for publication in Natural Product Communications are given in Information for Authors on our Web site http://www.naturalproduct.us.

Authors may reproduce/republish portions of their published contribution without seeking permission from NPC, provided that any such republication is accompanied by an acknowledgment (original citation)-Reproduced by permission of Natural Product Communications. Any unauthorized reproduction, transmission or storage may result in either civil or criminal liability.

The publication of each of the articles contained herein is protected by copyright. Except as allowed under national “fair use” laws, copying is not permitted by any means or for any purpose, such as for distribution to any third party (whether by sale, loan, gift, or otherwise); as agent (express or implied) of any third party; for purposes of advertising or promotion; or to create collective or derivative works. Such permission requests, or other inquiries, should be addressed to the Natural Product Inc. (NPI). A photocopy license is available from the NPI for institutional subscribers that need to make multiple copies of single articles for internal study or research purposes.

To Subscribe: Natural Product Communications is a journal published monthly. 2013 subscription price: US$2,395 (Print, ISSN# 1934-578X); US$2,395 (Web edition, ISSN# 1555-9475); US$2,795 (Print + single site online); US$595 (Personal online). Orders should be addressed to Subscription Department, Natural Product Communications, Natural Product Inc., 7963 Anderson Park Lane, Westerville, Ohio 43081, USA. Subscriptions are renewed on an annual basis. Claims for nonreceipt of issues will be honored if made within three months of publication of the issue. All issues are dispatched by airmail throughout the world, excluding the USA and Canada. 2013 NPC Natural Product Communications Vol. 8

No. 5 Vasorelaxant Alkaloids from Spirospermum penduliflorum 575 - 578 (Menispermaceae), a Plant Used to Treat Hypertension in Malagasy Traditional Medicine

Guy E. Raoelisona, Mamy H. Rafamantananaa,d, René Razafindrazakab, Adolphe Randriantsoab, Suzanne Urverg-Ratsimamangaa,b, Nicole Morelc and Joëlle Quetin-Leclercqd,* aLaboratoire de Phytochimie, Institut Malgache de Recherches Appliquées, BP: 3833, Antananarivo 101, Madagascar bLaboratoire de Pharmacologie expérimentale, Institut Malgache de Recherches Appliquées (IMRA), BP: 3833, Antananarivo 101, Madagascar cLaboratoire de physiologie, Institute of Neuroscience, Université catholique de Louvain, Av. Hippocrate 55, Bruxelles 1200, Belgium dGroupe de Recherche en Pharmacognosie, Louvain Drug Research Institute, UCL Bte B1 7203, Université catholique de Louvain, Av. E Mounier 72, Bruxelles 1200, Belgium [email protected]

Received: October 31st, 2012; Accepted: March 21st, 2013

Spirospermum penduliflorum Thouars (Menispermaceae) is widely used on the eastern coast of Madagascar to treat hypertension. The aim of the present study was to analyse the vasorelaxant properties of different leaf extracts. The activity of the n-hexane, dichloromethane and methanolic extracts was tested on phenylephrine-contracted aorta. The dichloromethane extract was shown to be the most effective. Further fractionation of this extract led to the isolation of an active fraction relaxing phenylephrine-contracted aorta with an IC50 of 0.18 µg/mL {log IC50 (µg/mL) -0.74 ± 0.03} but was much less effective on KCl induced contractions. Bioassay-guided fractionation of this fraction led to the isolation of two aporphinoid alkaloids, neolitsine and dicentrine, which at concentrations of 0.1 µM and 1 µM displaced to the right the phenylephrine concentration-contraction curve. Our results show that Spirospermum penduliflorum extracts possess vasorelaxant activity in vitro that could be related to the presence of dicentrine in the extracts having an 1 antagonist activity. This finding is not in accord with the previous studies by Rasoanaivo et al. where no alkaloids were detected in the leaves of Spirospermum penduliflorum.

Keywords: Spirospermum penduliflorum, Menispermaceae, Aporphine alkaloids, Antihypertensive activity.

Madagascar is one of the lands where traditional medicine based on the use of plants has an important place in society. Several plants in the Malagasy flora are alleged to possess therapeutic virtues and are largely used by the local population. Many of these plants are used without pharmacological or phytochemical data or clinical evaluation. Our studies lead us to investigate plants used for the treatment of hypertension. Among these, Spirospermum penduliflorum (Menispermaceae) is mentioned to be effective against arterial hypertension. The aim of the present study was to verify the traditional use of the plant and to determine the nature of the bioactive compounds by a bio-guided fractionation approach combining chromatographic methods with vasorelaxant activity Figure 1: Relaxing effect of different extracts of Spirospermum penduliflorum leaves tests. in rat aorta contracted by phenylephrine (1 µM). (n = 6).

When tested on phenylephrine-contracted aorta (Figure 1), the n- Table 1: Vasorelaxing activities of the four fractions of the DCM extract on phenylpephrine-induced contraction of rat aorta at 1µg/mL. hexane, dichloromethane (DCM) and methanol (MeOH) extracts showed significant vasorelaxant activity characterized by log IC50 Fractions RR RR1 RR2 RR3 (µg/mL) values of 1.4 ± 0.02, < -0.5 and 0.29 ± 0.01, respectively Relaxation (%) (n = 4) 12.5 ± 0.09 100 ± 8.50 7.2 ± 0.6 37.5 ± 0.4 (respective IC50 values were 25 µg/mL, < 0.3 µg/mL and 1.9 µg/mL). Isoprenaline, used as a reference standard, exhibited The relaxation of phenylephrine-induced contraction by RR1 was vasorelaxant activity with an IC50 of 1.7 µg/mL. tested in rat aorta with (E+) and without endothelium (E-). Figure 2A shows that the activity of RR1 was not significantly different in Fractionation of the dichloromethane extract was then undertaken either the presence or absence of functional endothelium: the log by preparative TLC giving 4 fractions named RR, RR1, RR2 and IC50 values (µg/mL) were -0.74 ± 0.03 and - 0.61 ± 0.02 in the RR3. Each fraction was tested at 1 µg/mL for its vasorelaxant presence and absence of endothelium, respectively (IC50 values activity on rat aorta contracted by phenylephrine (Table 1). RR1 were 0.18 and 0.24 µg/mL) (p > 0.05, n = 6). To determine the showed the highest vasorelaxant activity, which was further potential role of β-adrenergic receptors in the vasorelaxant effect of investigated. RR1, aorta rings were pre-incubated with propranolol (1 µM) before the contractile response to phenylephrine. 576 Natural Product Communications Vol. 8 (5) 2013 Raoelison et al.

Figure 2: (A) Relaxation evoked by the RR1 fraction of the phenylephrine-induced contraction of rat aorta with (E+) and without endothelium (E-), and with endothelium in the presence of propranolol (1 µM) (n = 6); (B) Comparison of the relaxant effect of Figure3: Effect of dicentrine on phenylephrine evoked contraction. RR in the rat aorta (E-) contracted either by phenylephrine (1 µM) or by KCl 1 (A) Effect of dicentrine on the concentration-effect curves of phenylephrine in (100 mM) solution (n = 6). endothelium-denuded rat aorta; (B) Relaxation of phenylephrine-precontracted rat aorta by increasing concentrations of dicentrine, compared with the effect of prazosin. Figure 2A shows that propranolol did not affect the vasorelaxant activity of RR1. Log IC50 (µg/mL) of RR1 in the presence of dicentrine, respectively, p < 0.05 dicentrine vs control). Dicentrine propranolol was -0.54 ± 0.02 (0.29 µg/mL). also relaxed aorta pre-contracted with phenylephrine, with a logIC50 (M) value of- 5.53 ± 0.05 (IC50 value was 2.95 µM) (Figure 3B). The potential involvement of voltage-dependent Ca2+ channels (VDCs) in the relaxing activity of RR1 was tested by measuring the High blood pressure represents the main cause of death in the effect of RR1 on KCl-induced contraction. When KCl (100 mM) world. A recent study in 2009 on the prevalence of hypertension in was used to evoke the contraction of the aorta, RR1 produced a Antananarivo, Madagascar revealed that 28.0 % of the adult concentration-dependent relaxation. However, RR1 was less potent population suffer from this disease with a mean age of 49 years. in inhibiting KCl-contraction than phenylephrine-contraction High blood pressure prevalence increased from 19.1 % in 2000 to (Figure 2B). The IC50 value for KCl-contraction was higher than 28.0 % in 2009 in Antananarivo [3]. 30 µg/mL (n = 6). Traditional phytomedical practice is common in Madagascar and The developed TLC chromatogram of RR1 showed two well Spirospermum penduliflorum is used as a traditional medicine to separated spots at Rf 0.62 and 0.81, which gave positive reactions treat hypertension. We showed here that crude extracts of its leaves with Dragendorff’s reagent. Fractionation by successive open exhibited significant concentration-dependent relaxation on rat aorta column chromatography, followed by purification on preparative pre-contracted by phenylephrine, the dichloromethane extract being TLC led to the isolation of two active compounds, 1 and 2. the most effective (IC50 < 0.3 µg/mL). We further studied the mechanism of action of the most active fraction (RR1) after a first Comparison of our spectroscopic and mass spectrometric data with fractionation of this extract. We found that the vasorelaxant activity those in the literature, and with a reference sample, led us to the was not inhibited by the β-adrenergic antagonist propranolol identification of compound 1 as dicentrine [1,2], and compound 2 as indicating that the active compound(s) relaxed rat aorta by a neolitsine [2]. mechanism other than the stimulation of the β-adrenergic receptor.

When vasoconstriction was evoked by KCl (100 mM), RR1 was markedly less active than on phenylephrine pre-contracted aorta. This reduced relaxant activity of RR1 on KCl-induced contraction indicates that its activity is not related to inhibition of VDCs but to an  adrenergic receptor inhibition. Vasorelaxant activity can also be mediated by an effect of endothelium-released nitric oxide (NO). We observed that the activity on RR1 was not affected by removing the endothelium, indicating that endothelial NO does not contribute to the effect of RR1.

Bioguided fractionation of the RR1 fraction led to the isolation of two aporphine alkaloids identified as dicentrine and neolitsine. Indeed, the Menispermaceae family is known to be a rich source of

aporphine alkaloids [4-6]. Previous studies of the root bark of Because of the small amount of neolitsine that could be isolated, the Spirospermum penduliflorum have shown the presence of the determination of its pharmacological profile was not possible. bisbenzylisoquinoline alkaloids limacine and palmitine, and Dicentrine was tested for its relaxing activity on phenylephrine- aporphine alkaloids [7,8]. The authors mentioned that they did not induced contraction in endothelium-denuded artery rings. As shown detect alkaloids in the leaves. That was not corroborated by our in Figure 3A, preincubation of the aorta with dicentrine shifted the study. It is well known that production of secondary metabolites concentration-contraction curve of phenylephrine to the right, in a such as alkaloids is influenced by numerous factors such as the age concentration-dependent manner. The EC50 value of phenylephrine of the plant or the leaves, and the nutrients present in the soil [9]. was significantly increased from 12 nM to 52 nM and 341 nM in The influence of factors on alkaloid production could be the cause the absence and in the presence of 0.1 µM and 1 µM of dicentrine, of this discordance, probably linked to the sensitivity of the respectively (logEC50 values were - 7.9 ± 0.02 in control aorta, detection method used. - 7.14 ± 0.05 and - 6.47 ± 0.03, in the presence of 0.1 µM and 1 µM Vasorelaxant alkaloids from Spirospermum penduliflorum Natural Product Communications Vol. 8 (5) 2013 577

In agreement with the interaction of the RR1 fraction with the contraction, acetylcholine (1 µM) was injected into the bath adrenergic pathway, dicentrine induced a concentration-dependent solution to relax the aorta in order to verify the presence of parallel shift of the phenylephrine concentration-contraction curve, endothelium and its integrity. After 60 min of recuperation time, the suggesting that it could act as an antagonist of the α1-adrenergic artery rings were stimulated with phenylephrine (1 µM) and receptor. This result is in accordance with those reported in the cumulative concentrations of the extracts (from 0.3 to 50 µg/mL), literature. Indeed, a previous study on dicentrine isolated from from a stock solution in DMSO/H2O (10:90: v/v) were added into Lindera megaphylla showed that it is an α1-adrenoreceptor the organ bath (20 mL) at the steady-state of contraction to evaluate competitive antagonist [10]. Other studies on dicentrine corroborated the vasorelaxant activity. When required, aorta rings were incubated this allegation [11,12]. Literature data also report that neolitsine is a in the presence of propranolol (1 µM) for 30 min before inducing potent vasorelaxing agent on precontracted rat aorta preparations the contraction with phenylephrine. [13]. The presence of dicentrine and neolitsine may support the use of the plant as an antihypertensive, but standardisation is necessary Relaxing activity on KCl induced contractions: After 2 h as it seems that the alkaloid concentration may be highly variable; equilibration with Krebs solution, the medium was removed and such a quality control method has recently been described [14]. replaced with depolarizing solution containing: NaCl 27 mM, KCl However, the total safety of the use of the plant is uncertain because 100 mM, NaHCO3 15 mM, MgCl2 1.25 mM, CaCl2 1.25 mM, and dicentrine and neolitsine were shown to be cytotoxic [15-17], and glucose 11 mM to induce contraction. After contraction had reached so further toxicological studies are needed. a maximum, the organ was rinsed 3 times with normal Krebs solution. After 30 min equilibration, the medium was replaced by Experimental depolarizing Krebs solution. Cumulative concentrations of the

Plant material: Fresh leaves of Spirospermum penduliflorum extracts (from 0.03 to 10 µg/mL) from a stock solution in were collected from the east coast of Madagascar. Botanical DMSO/H2O (10:90, v/v) were added to the organ bath (20 mL) at identification was made by the botanist Benja Rakotonirina. the steady state of contraction. A voucher specimen was deposited at the herbarium of IMRA Phenylephrine concentration-response curves: Concentration- (n° AML13). -9 -5 response curves of phenylephrine (from 10 M to 3x10 M) were Chemical compounds: The following reagents were purchased realized in endothelium-denuded aorta rings either without or in the from Sigma (St Louis, MO): isoprenaline, acetylcholine, D-glucose, presence of different concentrations of dicentrine (0.1 µM and 1 µM phenylephrine, propranolol, prazosin, dimethylsulfoxyde and N- from stock solution in DMSO/H2O (10:90, v/v)). Contraction was nitro-L-arginine. L-Dicentrine was purchased from Sequoia expressed as percent of the maximum response obtained in the Research Products (Pangbourne, UK). Methanol, ethylacetate, and absence of inhibitor in the same artery ring. The concentration- dichloromethane were purchased from Scharlab (Barcelona, Spain). effect curves were built and the concentrations of phenylephrine producing 50% of the maximum aorta contraction (EC50) were All the salts (KCl, NaCl, NaHCO3, MgCl2, and CaCl2) were purchased from Prolabo (VWR international, Haasrode, Belgium). calculated and compared.

Extraction: The plant material used was in the form of finely Chromatographic methods: TLC was realized on silica gel 60F254 ground, dried leaves. Extraction was performed by successive plates (Merck) using n-hexane/ethyl acetate/ethanol/NH4OH maceration of the plant material (750 g) in increasing polarity (6/6/2.5/0.01) as mobile phase. The separated components were solvents, namely n-hexane, dichloromethane, and methanol (2.5 L visualized under visible and ultraviolet light (254 and 365 nm) or of each). The extracts were filtered and dried under reduced using Dragendorff’s spray reagent. Fractionation was made by pressure. successive open CC using either silica gel 60 (0.063 – 0.200 mm, Merck Darmstadt, Germany) as stationary phase or Sephadex Gel Rat aorta: Wistar rats of either sex weighing between 200 to 300 g LH-20 (Pharmacia Biotech Healthcare, Diegem, Belgium) using as were used in the study. Animals were sacrificed, the thoracic aorta eluant a mixture of dichlomethane-methanol (1:1, v/v). Final isolated and cut into rings of 2 mm in length. The rings were purification was achieved by preparative TLC on Silica Gel 60 mounted under a tension of 2g in organ baths containing Krebs GF254 plates (Merck Darmstadt, Germany). The mobile phase was a mixture of ethyl acetate/methanol (9:1). solution (NaCl 122 mM, KCl 5.9 mM, NaHCO3 15 mM, MgCl2

1.25 mM, CaCl2 1.25 mM, and glucose 11 mM). When required, the endothelium was removed from the rat aorta rings by gently Identification method: Identification of the isolated compounds rubbing the luminal surface with a cotton rod before mounting the was performed with a LCQ Advantage Thermo Finnigan (Waltham, aorta ring in the organ bath. The bath solution was maintained at MA, USA) mass spectrometer piloted by X-Calibur software. Mass spectra were determined using an APCI source in positive mode, a 37°C and gassed with 95% O2 and 5% CO2. Aorta were equilibrated in the medium for 2 h and the bath solution was changed every 30 vaporizer temperature of 455°C, a sheath gas flow rate of 45 (a.u.) 13 1 min. After 1 h of equilibration, the tension was adjusted to 2 g. and a capillary voltage of 2V. C and H NMR spectra were Contractions and relaxations were recorded with an Ugobasile 7003 recorded on a Bruker Avance DRX-400 spectrometer in CDCl3, 1 13 isometric force transducer. All experiments were performed in CD3OD or DMSO at 400 MHz ( H) and 100 MHz ( C), at 30°C. A accordance with international guidelines and the local ethics combination of COSY, HMQC, HMBC and ROESY experiments 1 13 committee. were used when necessary for the assignment of H and C chemical shifts. Pharmacological experiments Statistical analysis: The relaxant effect of the tested products was Relaxing activity on the contraction induced by phenylephrine: expressed as percent of the steady-state contraction induced by the The relaxing activity was tested on phenylephrine pre-contracted agonist or the KCl-depolarizing solution. The log values of EC , aorta either in the absence or presence of endothelium. In both 50 which is defined as the concentration producing 50% of the cases, after the equilibration, phenylephrine (1 µM) was added to maximum response, or of IC (concentration inhibiting the the organ bath (20 mL) to induce contraction. At the steady-state 50 contraction by 50%) were determined from the non-linear 578 Natural Product Communications Vol. 8 (5) 2013 Raoelison et al.

regression of the experimental data (Prism, GraphPad) and used for Acknowledgements- Financial support of the Commission the statistical analysis. Each test was repeated in 3 inter-day Universitaire pour le Devéloppement (CUD), Belgique is gratefully experiments. The results are presented as the mean ± S.E.M. of n acknowledged. Authors wish to thank A. Randrianirina for technical observations. Values were analyzed using Student’s t-test and were help. The authors gratefully thank the Belgian National Fund for considered to be significantly different when p< 0.05. Scientific Research (FNRS) (FRFC 2.4555.08 and credit for researcher 1.5128.11), the Special Fund for Research (FSR) and the faculty of medicine of UCL.

References [1] Stévigny C, Wautier MC, Habib Jiwan JL, Chiap P, Hubert P, Quetin-Leclercq J.(2004) Development and validation of a high performance liquid chromatography method for quantitative determination of aporphine alkaloids from different samples of Cassytha filiformis. Planta Medica, 70, 764-770. [2] Stévigny C, Habib Jiwan JL, Rozenberg R, De Hoffmann E, Quetin-Leclercq J.(2004) Key fragmentation patterns of aporphine alkaloids by electrospray ionization with multistage mass spectrometry. Rapid Communication of Mass Spectrometry, 18, 523-528. [3] Rabarijaona LMPH, Rakotomalala DP, Rakotoniriana El-CJ, Rakotoarimanana S, Randrianasolo O. (2009) Prévalence et séverité de l’hypertension artérielle de l’adulte en milieu urbain à Antananarivo. Revue d’anesthésie-réanimation et de médecine d’urgence (Septembre- Octobre), 1, 24-27. [4] Kashiwaba N, Ono M, Toda J, Suzuki H, Sano T.(2000) Aporphine glycosides from Stephania cepharantha seeds. Journal of Natural Products, 63, 477-479. [5] Guinaudeau H, Shamma M, Tantisewie B, Pharadai K. (1982) Aporphine alkaloids oxygenated at C-7. Journal of Natural Products, 45, 355–357. [6] Blanchfield JT, Sands DPA, Kennard CHL, Byriel KA, Kitching W. (2003) Characterization of alkaloids from some Australian Stephania (Menispermaceae) species. Phytochemistry, 63, 711–720. [7] Rasoanaivo P, Ratsimamanga-Urverg S, Rakoto-Ratsimamanga A. (1995) Isoquinoline alkaloids of Strychnopsis thouarsii and Spirospermum penduliflorum. Biochemical Systematics and Ecology, 23, 679-680. [8] Ratsimamanga-Urverg S, Rasoanaivo P, Ramiaramanana L, Randrianarivelojosia M, Rafatro H, Verdier F, Le Bras J, Rakoto-Ratsimamanga A. (1992) In vitro antimalarial activity and chloroquine potentiating action of BBIQ enantiomers from Strychnopsis thouarsii and Spirospermum penduliflorum. Planta Medica, 58, 540-543. [9] Aniszewski T. (2007) In Alkaloids-Secret of Life: Alkaloid Chemistry, Biological Significance, Applications and Ecological Role. Biological Significance of Alkaloids, 143. Elsevier, Amsterdam, the Netherlands, Oxford, UK. 316 p. [10] Che-Ming T, Sheu-Meei Y, Feng-Nien K, Chien-Child C, Yu-Ling H, Tur-Fu H. (1991) Dicentrine, a natural vascular α1-adrenoreceptor antagonist isolated from Lindera megaphylla. British Journal of Pharmacology, 104, 651-656. [11] Sheu-Meei Y, Shih-Yuan H, Feng-Nien K, Chien-Chih C, Yu-Ling H, Tur-Fu H, Che-Ming T. (1992) Haemodynamic effects of dicentrine, a novel α1-adrenoceptor antagonist: comparison with prazosin in spontaneously hypertensive and normotensive Wistar-Kyoto rats. British Journal of Pharmacology, 106, 797-801. [12] Mustafa MR, Achike FI. (2000) Dicentrine is preferentially antagonistic to rat aortic than splenic alpha 1-adrenoceptor stimulation. Acta Pharmacologica Sinica, 21, 1165-1168. [13] Tsai TH, Wang GJ, Lin LC. (2008) Vasorelaxing alkaloids and flavonoids from Cassytha filiformis. Journal of Natural Products, 71, 289-291. [14] Rafamantanana MH, Debrus B, Raoelison GE, Rozet E, Lebrun P, Ratsimamanga SU, Hubert P, Quetin-Leclercq J. (2012) Application of design of experiments and design space methodology for the HPLC-UV separation optimization of aporphine alkaloids from leaves of Spirospermum penduliflorum Thouars. Journal of Pharmaceutical and Biomedical Analysis, 62, 23-32. [15] Ray-Ling H, Chien-Chih C, Yu-Lin H, Jun-Chih O, Cheng-Po H, Chieh-Fu C, Chungming C. (1998) Anti-tumor effects of D-dicentrine from the Root of Lindera megaphylla. Planta Medica, 64, 212-215. [16] Stévigny C, Block S, De Pauw-Gillet MC, De Hoffmann E, Labres G, Adjakidjé V, Quetin-Leclercq J. (2002) Cytotoxic aporphine alkaloids from Cassytha filiformis. Planta Medica, 68, 1042-1044. [17] Konkimalla BV, Efferth T. (2010) Inhibition of epidermal growth factor receptor over-expressing cancer cells by the aphorphine-type isoquinoline alkaloid, dicentrine. Biochemical Pharmacology, 79, 1092–1099. Natural Product Communications Vol. 8 (5) 2013 Published online (www.naturalproduct.us)

Eight New Alkyne and Alkene Derivatives from Four Saussurea Species Collected in China Yoshinori Saito, Yuko Iwamoto, Yasuko Okamoto, Takayuki Kawahara, Xun Gong, Chiaki Kuroda and Motoo Tori 631

The Apoptotic Activity of one VLC Fraction of the Sponge Petrosia tuberosa on Human Cervical Cells and the Subsequent Isolation of a Bioactive Polyacetylene Avin Ramanjooloo, Girish Beedessee, Deepak Arya, Rob WM. vanSoest, Thierry Cresteil and Daniel E.P. Marie 635

Chemical Characterization, Mineral Content and Radical Scavenging Activity of Sideritis scardica and S. raeseri from R. Macedonia and R. Albania Marija Karapandzova, Bujar Qazimi, Gjoshe Stefkov, Katerina Bačeva, Trajče Stafilov, Tatjana Kadifkova Panovska and Svetlana Kulevanova 639

Phytochemical and Micromorphological Traits of Geranium dalmaticum and G. macrorrhizum (Geraniaceae) Dario Kremer, Dubravka Vitali Čepo, Valerija Dunkić, Ivna Dragojević Müller, Ivan Kosalec, Nada Bezić and Edith Stabentheiner 645

GC-MS Fingerprints and Other Physico-chemical Characteristics of Rare Unifloral Prunus cerasus L. Honey Piotr Marek Kuś, Igor Jerković, Carlo Ignazio Giovanni Tuberoso, Zvonimir Marijanović and Mladenka Šarolić 651

Volatile Fraction Composition and Total Phenolic and Flavonoid Contents of Elionurus hensii—Antioxidant Activities of Essential Oils and Solvent Extracts Yin Yang, Marie-Cécile De Cian, Samuel Nsikabaka, Pierre Tomi, Thomas Silou, Jean Costa and Julien Paolini 655

Leaf Essential Oils of Six Vietnamese Species of Fissistigma (Annonaceae) Martina Höferl, Do Ngọc Dai, Tran Dinh Thang, Leopold Jirovetz and Erich Schmidt 663

Studies on the Antimicrobial and Antioxidant Activity and Chemical Composition of the Essential Oils of Kitaibelia vitifolia Pavle Mašković, Marija Radojković, Mihailo Ristić and Slavica Solujić 667

Angiotensin Converting Enzyme Inhibition Activity of Fennel and Coriander Oils from India Sushil Kumar Chaudhary, Niladri Maity, Neelesh Kumar Nema, Santanu Bhadra, Bishnu Pada Saha and Pulok Kumar Mukherjee 671

Effect of Coriander Oil (Coriandrum sativum) on Planktonic and Biofilm Cells of Acinetobacter baumannii Andreia F. Duarte, Susana Ferreira, Rosário Oliveira and Fernanda C. Domingues 673

Essential Oil from Caesalpinia peltophoroides Flowers – Chemical Composition and in vitro Cytotoxic Evaluation Bianca A. de Carvalho, Olivia S. Domingos, Murilo Massoni, Marcelo H. dos Santos, Marisa Ionta, João Henrique G. Lago, Carlos R. Figueiredo, Alisson L. Matsuo and Marisi G. Soares 679

Antimicrobial, Antioxidant, and Cytotoxic Activities of the Essential Oil of Tarchonanthus camphoratus Nasser A. Awadh Ali, Mohamed A. Al-Fatimi, Rebecca A. Crouch, Annika Denkert, William N. Setzer and Ludger Wessjohann 683 Natural Product Communications 2013 Volume 8, Number 5

Contents

Gerald Blunden Award (2012) Page

Cytotoxic Agents of the Crinane Series of Amaryllidaceae Alkaloids Jerald J. Nair, Jaume Bastida, Francesc Viladomat and Johannes van Staden 553

Original Paper

Chemosystematics of the Thai Liverwort Cheilolejeunea (Marchantiophyta, Lejeuneaceae) Phiangphak Sukkharak and Yoshinori Asakawa 565

Cytotoxic Properties of Marrubium globosum ssp. libanoticum and its Bioactive Components Mariangela Marrelli, Filomena Conforti, Daniela Rigano, Carmen Formisano, Maurizio Bruno, Felice Senatore and Francesco Menichini 567

Cytotoxic Scalarane Sesterterpenoids from a Marine Sponge Hippospongia sp. Yuh-Ming Fuh, Mei-Chin Lu, Chia-Hung Lee and Jui-Hsin Su 571

Ursane-Type Saponins from Zygophyllum cornutum Soumeya Bencharif-Betina, Tomofumi Miyamoto, Chiaki Tanaka, Zahia Kabouche, Anne-Claire Mitaine-Offer and Marie-Aleth Lacaille-Dubois 573

Vasorelaxant Alkaloids from Spirospermum penduliflorum(Menispermaceae), a Plant Used to Treat Hypertension in Malagasy Traditional Medicine Guy E. Raoelison, Mamy H. Rafamantanana, René Razafindrazaka, Adolphe Randriantsoa, Suzanne Urverg-Ratsimamanga, Nicole Morel and Joëlle Quetin-Leclercq 575

PPARα Signaling is Activated by Cocoa in Mouse Liver Marco Fidaleo and Claudia Sartori 579

Chemical Analysis of Flowers of Bombax ceiba from Nepal Khem Raj Joshi, Hari Prasad Devkota and Shoji Yahara 583

Chemical Investigation of Caragana arborescens Shoots Daniil N. Olennikov, Larisa M. Tankhaeva and Vyacheslav V. Partilkhaev 585

A New Metabolite from the Endophytic Fungus Penicillium citrinum Xinlan Li, Liang Zhang, Yanhui Liu, Zhiyong Guo, Zhangshuang Deng, Jianfeng Chen, XuanTu and Kun Zou 587

Antioxidant Activity of the Isoflavonoids from the Roots of Maackia amurensis Nadezda I. Kulesh, Sergey A. Fedoreyev, Marina V. Veselova, Natalia P. Mischenko, Vladimir A. Denisenko, Pavel S. Dmitrenok, Yakov F. Zverev and Svetlana V. Zamyatina 589

The Effect of Pyridinecarbothioamides on Isoflavonoid Production in Genista tinctoria Cultures in Vitro Lenka Tůmová, Věra Klimešová and Anna Vildová 593

A New Homoisoflavanone from the Rhizomes of Polygonatum cyrtonema Li-She Gan, Jin-Jie Chen, Man-Fei Shi, and Chang-Xin Zhou 597

Two 2-Phenylbenzofuran Derivatives from Morus atropurpurea Wen-Jing Wang, Dong-Ling Wu, Sen-Tai Liao, Chun-Lin Fan, Guo-Qiang Li, Xian-Tao Zhang, Ying Wang, Xiao-Qi Zhang and Wen-Cai Ye 599

10-Deoxy-10-hydroxyascochlorin, a New Cell Migration Inhibitor and Other Metabolites from Acremonium sp., a Fungal Endophyte in Ephedra trifurca W. M. Anoja P. Wanigesekara, E. M. Kithsiri Wijeratne, A. Elizabeth Arnold and A. A. Leslie Gunatilaka 601

Polyphenols in the Aqueous Extracts of Garden Thyme (Thymus vulgaris) Chemotypes Cultivated in Hungary Blanka Szilvássy, Gábor Rak, Szilvia Sárosi, Ildikó Novák, Zsuzsanna Pluhár and László Abrankó 605

Resveratrol Production from Hairy Root Cultures of Scutellaria baicalensis Sang-Won Lee, Young Seon Kim, Md. Romij Uddin, Do Yeon Kwon, Yeon Bok Kim, Mi Young Lee, Sun-Ju Kim and Sang Un Park 609

Anti-periodontal Pathogen and Anti-inflammatory Activities of Oxyresveratrol Waranyoo Phoolcharoen, Sireerat Sooampon, Boonchoo Sritularak, Kittisak Likhitwitayawuid, Jintakorn Kuvatanasuchati and Prasit Pavasant 613

The Triple Botanical Origin of Russian Propolis from the Perm Region, Its Phenolic Content and Antimicrobial Activity Milena Popova, Boryana Trusheva, Rail Khismatullin, Natalja Gavrilova, Galina Legotkina, Jaroslav Lyapunov and Vassya Bankova 617

Comparing Different Solvent Extracts of Rhus semialata var. roxburghiana Stem against Ferrous Ion-Induced Lipid Peroxidation in Mice Liver Mitochondria Pei-Chin Lin, Wei-Fung Bi, Che-Hsuan Lin, Fei-Peng Lee and Ling-Ling Yang 621

Antioxidant Action of Solid Preparation of Xingnaojing in SHRSP Yang Liu, Naomi Yasui, Aya Kishimoto, Jian-ning Sun and Katsumi Ikeda 627

Antioxidant and Antityrosinase Activity of Cissus quadrangularis Extract Ikuko Suzu, Hiroki Goto, Nami Hiwatashi, Shinichiro Hattori, Kanjana Rotjanapan, Wilairat Leeanansaksiri and Seiji Okada 629

Continued inside backcover P.02 : H. Rakotoarimanana, H. Rafatro, V. Jeannoda, G. Raoelison, RB. Robijaona, S. Ratsimamanga-Urverg. Composé antiplasmodial isolé du Tsilaitra (Noronhia divaricata Perr., Oleaceae). Ethnopharmarcologia 2008, 41, 71-73.

", "-,, ,

*{5fr\t?:V, ç}? I {:i tct - L** pæ r a niï,ü%*e r:r * *i*nîuz*

3 pffisé entt âææ l-m|* awæâ w æx&mwtwæ q mh W*{Y.u &wmÇwffiw

,i- i .:i 1 .::::1.,.;.y,tJiit-,,":''' 1:1.L1 I i, fi alcq:l*a rr rvl ;: r: ;lr.n . ?-1. ff. &4: ïitl * '..f 1.:i.",!:1:1'a,.:,11i:: ;-. i:r.:: ri:i...,iiiri fiat$rmary ;* *.q*.-ür ç'r,:fl{jI'li

mzelllütJ$ dliÊ lêïei»-!îtbû fillit ..11i al(:irll ?!)1,"ar.trt2t2.tlt:1ttë)

§&'

r":':1 ?: t:.;rtt'it'a''t ttvirr,,ltilitlobt'CcU\l:rlÿ!ti,.':ttt1ÿ-:,!'t:''\,: '}'*e .i':r'1". "' "i''{2r'n

'W41q tt'à?r):;1ilz2'')tkt' {: i)'cials lunt}:t;tl1{)titl, {\,i:i\l7i;3?rnt)t.l!1* î:tt:t'::l;: tzii:,r:zti') '''iÜti}i:lii?

:z"i:'n {"3 {il't 4r4Tï4 tlTi{"ïi{}i"r ? i4 I} * ; {. -1ï î"

i. câll* d,1tli.l* Ùn1 ÉtÈ Dans la rnaiantt) cie$ ârtlcles s'rientifiqus§ Publiée sui' lb ùaiJ{list}"1{:'J o:l pr,lircù,q-i Êv.f:âl'in*n1aux applirllsn dail$ l'}tl.lJil.e ûoâ !lilv;1i12 ?)ÿ*r1ert'.' y est dit que : ia rnaladio tûttch$ aclueilêinont plus dc fl{, payî dû,:.$ tàrri\!, Êti i-l,il*it ianc alnlieîi l$ môfldê (Saissy 2t101 lGeatili|ri et Dirtla' 1Q86i scin in{:i(.ler,.À â{il i[:1âi&1ir-\ €1.ai ?.tt)§i éstiniéB de 300 à S0O million§ de cas éliniquÉs Par an avÉc r-li {ùm l»-c nies ân'JÈÊiqÜû5 dF-) tlrâ do décès de {,5 à 2,7 miltions {§alss1', 2661 : Fnçr*r +l iil )\ iA t;1iile a*?, er'qLl4iâs 6'tt,rtûbr.{ii 'liqilat;, ,)i;l goilt Ia ctlî-111)/)$tliÜ1' 2001b}. La quasi'tôlalité ds décès due âu paludisrner t.:!§! trJiriirtrable luqadagasciir qili intlitltte*s Ët ûrrtrB?1l daîx {lei: rÉrncde*q â Flasmëdfurn fel1iparum (RDgiBr Bt âl , 2t)0141 {-ûr.ro esl dçnt; tte iiâ()a.l* .1e I.,-r5{)çiaî.;»t1 üa ç:l,.rcieura ,LanleJij palu(ii::î]€. c*l1ec1éëË"' cürâtôtel que lo paludiSmê rê8tê en cB dÔbul dÊ) Xrlilr.a.çiiirJâ !t lrseiiiian,',*ls uti?isés rsnlre ,e $i1 §"É pramlèro sndêmi* Ùopicâi€ n'iondlale' Ën Alrique il '$tlsaha*DrlrÉ r€ste un dÉâ prûblômes maièuls de santé putiiqtrc nva;ç $§f, r:rilli*t§ de porsonnes vi,J.lnt en z-ones impaludées (Plcltar*. 2Cit1) ,,.i/:x r4i;*t:? i . .,,'i...i;1ia,Tr t + a.:ti|\r.Ot |?1\!t'tr:.\/ 4tii,-.l:.tiè la difit)§ion et l'âuqmentâli$n de lâ. /é$i§lârl1ê *es Uérnergence, ':/ :;titirl:,|, i :i . fa1,!liu:.r''tir î çouÇhes te Plaenadium tatciparum aux médicân:c:nls àhtipâiudÈ ques convortionnels disponlbles'sur lë rnarÊhé sûnl rêsl'lorl$3lllés dë par parudisllto !'ât0(âvâtion dramatiqile de le rmrtalité le iTrïJfrë tt . i>: .3i.rj|,)i/:;,r pour à luttÉ r:É llé;{Lr ')tl

?4. ù 1 , ::| 2

SStÊÊ Ë tlÂL '.L** 1';s. cri,il,ü.!ri$ Lî {:ri}iî:iJ'lttT)

Apriè$ s(ichage et , des exlrâj!§ b,,.tt.t; r,:y't*,,,a:,i,açrtsiirlr,r*6 ûiri ér'é prépârès. €n âpÉÎiquünt fâ rrlêilûde irasetqüû *D lracliônnemenl bioguide, !e§ molécules at:\rtrâ' t"1n1 à14 iilléç* bt\ deux élapeo..loul d:êl§êrd, l'Ôr(trâit br'.rle àlà,iléci11t't,)plT/tlà Dat p"]$ 5a{8 êu ôhârhon âüti, puis 6 ët4 frârtionné nar FafI?0B ljquid-s, liqilide er)lrr? eau. d'Un9 pôIl, €l d€$ sDivarlB ;ri"ÿanruÇË Ih€x&r!Ê, d'rühlororfléTt arre. arÈ?a1à d éthyle). d'ar-rtr(.+ fiar i E,rs*rt* l*s èxtrâ.its i.g§u§ üe cE$ lraÇtistrrêmênts cnt è1é paRgÈ. $uûÜÉi.;§i,/efnênl à / tralsrs uÊ€ coloflnê shi§Eâlegrapl.lrquÊ utillsiiint leâ phaâes fiNbi6.; it SeÊhadEx LH-2û (Fhaimecia FinÈ Chêrir{ri:it;\ !)1 t(r !;lli;a i4()-ü3 ptm, Gedumn §lËû, futerck),Àcotte éteprj. ;'ûvlrr1 i ., tiit &t1üe Brr:,r: Hfl dgs Syelemes cle saivants â polarite ü!uà.i31i1t2 ü\l{tâtti flat ieÿ mhlaàng*:; 116 di6hlorornéthâne, d'acÉt,atù 4i:l?yr: q1 ir; r,tùltfitî{)t îrguÿa 2 : §lrul\1,jri, r:h\miqun

ÂprÈs étude delê parliâ cùiï:ittruÉ, ,âLr§.nô âctfil:§ Qu{a\*fi1 41éi*a- lÉÉË rl* i'ârlrâil rJe Ia*ilt'o 4e l',lot,:!iihi;'t divâicttâ {1o11i. i'un* (.1r,4i7 0F1 ljilr iî :.âirl:i1icillôt à \'itk)e tv,fr'":lariq[iô I e '{peatttj: 4l'd,f4 {:5r.,1lriâtl(l

l:|j,tlft 1 11 1 î1';lititr). i1fi1il:,t3,i1il)iù|ic. i!1 vlva 1aii tltsull*!! a;1t1§ iq|lallllâât, 1 lali\Ttt:2 l,ü§11È !3 r;hür,rtit: T;himAlllÙ

t:r:{-ri{re llnülâllirlue *rl llLri(il r1o,1r1ti alâr}si it1 trn:rlt}â v*üi}là1. it Ë."i1 's1{li}\i: ,)ir,l('ibttù r1àr\,; là ÿut::itt:}iB *i *1 .:yiaÿlÀ§n1û {.,+5 iltllJleti ,tilgé\'âte\ (tttôlü ùl Çlûa 1**7\ Ü'êâi rilr â,]ris ilrQclu;1 i$sr; ,,u 'lri,1'tlJCl:la.r?r(] riuLr:inl!âllo iLils d(i lzl lii)iijrr1f,È1i4\ ilrittt)rlii;lri irr1û1iquàfi1 la 4lÉ1Düî1È"' lî.ci*trùâl1{t ul iti .2Ûfr.'l't Cetle 1*1àî:rl+ 4!)'i.

:-ril§$i r})rl;c)11!:,Jë p^t *â DTQ\ nét(t iltli*iltlt:ë ti'à i;. tlüiç\ârvé vé:ù+lt1k? i1/oçl\ir.dvt';1 àl al.. 2ü{}21. fiàlrit lÈ îrôT}tiE a;fu':'l|, l at:iie iunulariqrr+ cïi dùLtà d't,fiË âôtivilËr fiÈliuàt)i{:iû* i'fdcrz+l fli el.. 1§â0j T#trl r&{.:(-in,ï}âîfl ii a 614 aleüJ1.i'ü$1. 1}1re. ir'.4 pü{irlâr1 +tte ufi n{-1',r!,el ât*ni

Ëlgure 1 : Spectie f*l-li'l peyl'y61ir*id&lior' slrtn:turala \1* l ar,ida raorâ.)fir2l ot\ Él\ir,iùptÊ"t*rlliün dil canûer (iil*ül al û1..2ürJ4\. A rtctsr: lunulâriquÉ atnn&.i$Ë.Ar1ce, f'eâi lir ç*emi{r* f

Les info/rnêliôns, cor!:r.riiiaûl ie

j:: fii,ri:!! r, , ,/:lr*i{: 11. :r .;:i..i:!:' tr | I 'a |:,t":tt,'";

-'1 j'|rr t. ::, ..1\ ces ieullii]n, aFlrèi! i.tpplicatiôn de la fiùJYell!2 trfri"):'r',; : I :. I i,:i'..:1.: 1 i.t,,'. r,.\i..i,-;:'..""1 ::.: : lerrainlabr;,l'âcicÉ iunu,atiqüâ â, eÈtTê airirr:;. ét"t t::rlt' i ': i i" ?p^kF Kil';:t.tî11) t>..?.ttt:t:r'l p*;udiqüz da I'ac,irat lirilulalique Étt ccilsiiléré tomnr i'l'1.i{'itl {:f St\'t,'i i.1 i:tchêf l. l1r{\at.,r';) "(ii:'il\2ilr.:l |:, a';:çJ!4j i1l1hi1i..fr ç1 itcdll\.èr+r (:el 'ÿrrtr:!t'; >'; t\»'t6l lltlr,r\iil aëJn:zt t:)'î]' I I'eÿ.Vail du lsdrjtr,r A\t''' i h'.r': .':t'r Faurrait :irstilier a't'a*eta\ian d6 rrrrt.nf.rart;t/l ,ia<,t11:t }t,:)t:lx:n Abi]'a,ÿt l'1ùË C1»,t1!-),) ) li::a i1l ?*1' l+!' rri aütrce plaàlp5 zl Ç{tl â'4 d'étr'38 mêrite d'Aîÿ p|\)E ]tf 'rr:l trn..i11:t1 I: ) i)è|t^bt.1 .i it11'':1t11 i.) it'.t"),12j b'|tt't:ft1 4/t\:.'' ir1-a4 ta.! t'!t: i'nlai;ai^t:.ttr\)l ttli-c!r-iJi;a!: i At'i:)rri: .:,.'tt L,riLz\ f..trr.t'r"1,'1. lÉÿ ? a iltri1: r':it;e Tableâu 1 : AüiivitÉ aîlipla§ft\odi:à\a tnvtÿûdê l'âcidÈ !ltttLrilf '.'t5-,1-';,5 iti :)..,i;r::t::':.rt til:. L,t,1,i)t1 t, ) lirr".r' a:,1 '.la.J )a, a1a1.!rt r11r't\\12't\'r I j' i'') tTll t1',ti'!: d@ la Èrôis§ànÇe d\t Pi,1Én"fidit1rî'ryçElii nigarienslscha2 t' '«i:r' : i<:c!{titirt(1^1 rl î\t.t,t't.l:aiiri ?l,t,l'2;l tl1 ,lit';:\i à :.i1ù,t::i)\i}î1\âit:11 t!11 '1 ij,j*1J{j Atiit1tcn){1 /\.jpr!: i.:hÉtti.\!l{:, . lrti:1 i1i.)-11*

7tê** '?"1a\ r -: : i"ta1 '1 i:;,: t trita:1t.tr a, j.:irirjér O t:{ktt':1.\i111 Li 311ô.:! ù t)f'tL1111. -i ijailê?alfl y',ïtu!ÿ;4i iri P Sllrrltlg, : t',!|.:i:-\i St.l:ti!',J'1a;i\1i,:1ttK'ii|"e!r+.\t.l||qii,.1 i1 llÉlY llrlYilkll inÉ itï,i, rl,rltiàti1rirr2 i)1.,1,.1!\l: t'i,!it:C!1t:jitisrf 52 ilt: ??t''tli 1 i\)*1',ïi11,^r1 "j:tf.riçi',iiàl.t:lj \a tiê ÿ:trüLi\i/'1tt7 ,.':ï,r'tr' lr ltta,: ii ;11:ÿ.4, l'j.j,.:i{.:it1a) i"tt}Ër}lia 4 114 ;,}t '. l'ii|vt1'v1':',t pn)l,t , \l,'l!tt,:i.. :> S'.1\ l:r ?_ !;' ?.,-t' 1',*, 1',. *" /jitî i. t.i )

t tt t1., 't'^ 13û ttl itut 4*:' ,:!:tli!r, ).1 qir:;t:tttt-:ia t\.1: ,-:, i\ i)\t;f i, l-L\i |il..i.' it! i ,,rr),i l: Âr l lnil ?i:, lt: .1,''lln1§ \r.:,..,(\ ),t t.ÿi,t 4 ra l-.:: t. :.,..it.t

-ç t 4',t;2t1q,'t: l4'tlï ;).a{:,:\!.1 lllf\'tl!firaq.rüt'.e.ti?li.î::ôtir\lÂ.!11ii4.y'..'<11 tlt):iil.tr,l,:',i!t}' !;1, t:ie zt;,4q n t ii ?"9*e ÿtil i:j,,,11i:,)\1. Eir'tii, tl: t.:\l it :-)!: tJliii ir S.,ft üliîa lat '.)l.Jr1 ,ù1t;tillrctll:\i &1,1.t'tl 'i At1tl ,.r, É'ldttltill,;r' \, )l'î::titti ilatistlî: !:ii+ t:.: 1!:'!'! i1 gâlfirnalftriàrl(Ji tt\alnl\ î1 liûtt,acî t:1. * . ?,nkîl{)7111t lÀ li , l-ÿ:tllitl\s »i»i: q ,r 1,,,y:r.;1,1'r 1r1 L ;èi a.t;J!ln',i)r\i ï:. , i1nît:li,ttl A ',2*ô7 / 'rr1 r'!l'fialt:,11r':i:l\t:),ALj r)ririr7,r!.r! ;)r:rt it!llal.ti. l)itirrr'tt\a\r" '!t1 xldfrlP ':4 .: .,-i. "\?:;.r r,:rrj.nr!., ri lê lÂ?\tlilqi)t!:>Ài '-!l ;: Lftl,l.:ar;t?'t:t:: là 1jl ,t|!à'1i: ll llitljtiilitl,t il.$ 1f:'tlSliffitîtll:a ti (;(>ll i!Ùlût :itlÈtrtnq Dt l\our rÈnerciein; i |)ltlÇËl: i l.Ji'JT:i) ,t1Â|çîld B§ûk t''/t t() i:rr:' ';tt ?a':iat:i',.lrr.tt.i';)eaitilrlt;) Ar\ti)saliftt>.1ia l\-t:ti'{it-\f^.ai. l1'\:'t?i}lÈt(:1lfffitl''silrrT i i!).r;\itlIrn!q.ii]itjrl,\lliitriitll;1ri,ln lilrl'j{4 l:lliL';ir?^i./l1t)')'t:a'.a!\*1i»:1)licil Ft{tüHmile lw f1ë§eâra!: ;Jr:.| : i.,.r'. t.J i!- lolic,3t i"''''''' ii- l; . i, 1 i. r t t)a: i :r,, 1 -1,i./, f iri,r:,:t, Li nt Épri1,(-lr l ll.f) 4 11,!:,.t 1 i -Tp.?/rlourlerirla,aêlnÇrll'i. ." :.|''.ris:jirlütit''' " ';i1i!)â't;oL.!li.' cÉiri ,i'âdle§Se!'rt )ttY l(Ad11!\'xtali,trL?é§ qt.ll Bnl Llie* !"1'rl i I ' ri.,' r: 1l :iif)Ti$r,i / r:.,::./^1': ?11.î\i..Jl!\tr1:: tl r'r(.)lllri it' 1.71'tiiri'it) t'.-1'!,aiilt!),;t 3tt:1 2 i:S)! t r;rr,l" r .:i,l.ii:l:lY l"\ !ï:i1/c patlaïb{ {eut Éia"-\t r âur i'ulili';à110il de5 trJlântÜâ it\ir-:,': ':. i: , '^ "), " " ^tt'a:'.ia: nous ofri (1''1':r:' : rl':'i {i': as§lstar']1g e1 lâcilir;i.riàns du laü0rêtoire Glli 'r t:! ti r,, i . 7.)t t'(:1 !lP,alrtJ  ii)»i")1t-i, 1'l r1a':1u-', t'i.l'i'altti ri l:ÿt');tl1i'liÏ:iia ,'xpeflm0îlâliot s rl d#5 ill.i,-j(.s s!r le lelrain. 4t1i,at.-.1. ),: !A.r:a"l )!',1 i'l:t4<)i|l. .1Q1!i\ â:lr ; t'r)êl\e i'2î r)r'1 l'Éqürp.r dlt l-afrr;l;llüire ti* Pâ:uliân1ë tln$1il1rr ''rr 'i" ,.1r:, ,, ,l l!. ::.t)i:',, r,:?li({tÉi.1.1 4rrltÉ. i}lfr:! :1in.llic 1:5:;:1 Màdagâstai), à fili;ssiâirl§ 3.:r:..' r'\Pl l: +i Frtlrrjr " '1:t :rl :'".1 t\-' a',ii, )'., l:i'ùlf1:1 F a'-(',1 i1't'\21 jt)|"'"t\t '':'r'1')'ilti't'::î:)tt:l ,9ailrfr-AvanfiÉ-f :aficè li(Jt): l ' I ":'irTlit'âiiorl ::it1 ' )l 1"; ?i. ..1-.-:;.;!i3ria;rtrt1.:ri,,,.i'.'t'':l:ie,î,11\1i.1+ !.!*()atr-t.i:t:)'::/t..ln!,';t'f:i:': cùrflpD-§és isôlês it :1.t r t 1. t)t:, 1:r'.l t'),ili,4 0t''! î,i,

'1.i t:::t t.i l?.cr.ri ii !:lrir{i I it.!l''!r':.. i\ l lllif)i U$ir;':ii::4zi çra1:e:lie* r\ta.L,j-..ti:tr-t:),!)t).fii[rjr:!4.!,nri i1:C:,al,ni)/.1t),1!r!rlt'§,1rr{i:i],)lflr.rl'r:li1!LlÉr lr ,,;, j lr,i!.ri,rci aii.rtlitl lr1rj) ,t i'j, 41at la

1.))i1tt1jlj», i.:lri'.1\',1:.:iil.a..iil:)aé (l] Iârrl,io ia aili:^lrÀ i?.t.:lii:')'ihr\;:tiit]l) i:)rr,*.iêlrfi.lrri1ja\l|.t^ri'r'1a,1il."]111'-t1r\l!-'t,.:;ltit)"'filtr]Êu-iri't^"il)[1r

'il P.03 : E.F. Queiroz, J-L. Wolfender, GE. Raoelison, K. Hostettmann. Determination of the absolute configuration of 6-alkylated -pyrones from Ravensara crassifolia by LC-NMR. Phytochemical Analysis 2003, 14, 34-39.

PHYTOCHEMICAL ANALYSIS P hytochem. Anal. 14, 34-39 (2003) Published online in Wiley InterScience (www.interscience.wiley.com). DOL 10.1002ipca.684

Determination of the Absolute Configuration of 6-Alkylated o-Pyrones from Ravensarü cr&ssilolia by LC-NMR t_-

E. F. Queïioz,J.-L. Wolfender, G. Raoelison and K. Hostettmann* Institut de Pharmacognosie et Phytochimie, Université de Lausanne, BEP, CH-1015 Lausanne, SwitzerlaDd

The absolute configuration of asymmetric centres of two a-pyrones isolated from Ravensara crassifolia was determined using ihe Mosher method. The conventional analysis of the purifred ester derivatives by 'H- NMR was replaced by a rapid and sensitive method in which the a-pyrones were analysed under isocratic reversed-phaie LC-NMR cônditions prior to and after derivatisation reactions. Comparison of ttre LC-IH- NMR spectra of the actual d-pyrones with those of the corresponding Mosher's esters recorded in the acetonitrile:deuterated water solvent system exhibited shifts comparable with those obtained using conventional deuterated solvents. Based on the shifts recorded, determination of the absolute configuration was possible by application of Mosher rules. The use of LC-NMR has permitted a direct analysis of crude reaction mixtures containing less than 50 pg of the starting material. Completion of the reaction was checked by LC-MS and the crude reaction mixture was analysed by stop-flow LC-NMR. This methodology seerns very promising for the determination at the micro-scale level of the absolute configuration of natural products which are available only in very small arnounts. Copyright @) 2003 John Wiley & Sons, Ltd. Keywords: LC-NMR; LC-MS; absolutetonfiguration; Mosher's esters; d-plrones; Ravensara crassifolia.

1H-NMR INTRODUCTION the of natural products in the microgram range has become possible; This technique also allows the chromatographic resolution of an LC-peak of interest Determination of absolute configuration remains à prior to NMR detection (Wolfender et aL.,2001). Taking challenging task in the sfucture elucidation of natural advantage ofthis new hyphenated LC technique, a means products. Only a few physical methods, such as chirality of allowing the direct analysis of the crude Mosher's ester excitation (Harada and Nakanishi, 1983) and X-ray reaction mixtures has been developed and validated with crystallography (Fiud et al., 1977) provide the required model compounds of known absolute configuration. The information, but they have some limitations. A chemical aim of the study was to establish a method for method involving the synthesis of Mosher's esters has derivatisation and subsequent rapid determination of the been frequently used for the characterisation of various absolute configuration using only a few pg of natural natural products bearing secondary alcohol functions products. (Ohtani et al., l99l; Gs et al., 1994; Jeong et a1.,2000: Rasmussen et a1., 2000). In this case, the H'-NMR spectrâ of (R)- and (^f)-2-methoxy-2-phenyl-(trifluoro- methyl) acetic acid (MTPA) ester derivatives of the EXPERIMENTAL analytes are compared. The difference in chemical shifts of the diastereoisomers determined from: Preparation and purification of Mosher's esters. 65- AdP : 6s - 6n (1) (2,S-Hydroxy -6-phenylhexyl)-5,6-dihy dr o -2H -pyran-2- one (1; 5 mg in 2mL of dichloromethane) was sequen- indicates whether the alcohol is (R) or (,S). based on tially treated with pyridine (0.2 mL) and (R)-(-)-MTPA established conformational models (Dale and Mosher, chloride (100mg). The mixture was stirred at room 1968,1913;Dale et a1.,1969; Sullivan et aI.,1973;Trost temperature under an atmosphere of nitrogen for 5 h, and et a1.,1986). Although it is a very efficient method for the progress of the reaction was monitored by HPLC. The determining absolute configuration, the synthesis of reaction mixture was concentrated and dried, and the Mosher's esters requires relatively large amounts of residue dissolved in dichloromethane and washed with samples for derivatisation. When working with natural l7o sodium bicarbonate (5mL) and water (2 x 5mL). products, pure compounds are often available in only The organic layer was purif,ed by HPL{ using a Waters very limited amounts and this hinders the determination (Milford, MA, USA) Bondapak@ Cra column of absolute conflguration by conventional methods. (100 x 25mm i.d.; 10 pm) affording (S)-Mosher's ester With the recent development of LC-NMR, analysis of derivative (1a) (5 mg; 56%o yield). The (R)-Mosher's ester derivative (1b) (6 mg; 67.l%o yield) was prepared using (^ÿ)-(*)-MTPA chloride under the conditions * Correspondence to: K. HostettmÀnn, Institut de Pha'rmacognosie et Phytochimie, Université de Lausanne, BEP, CH-1015 Lausanne, Switzerland. described above. Mosher's ester derivatives of 6R- w, Emâi[: kurt.hostettmann @ipp.unil.ch (4S,6S-dihydroxy- I O-phenyl- I -decenyl )-5,6-dihydro-

Received 5 July 2002 Copyright O 2003 John Wiley & Sons, Lrd. Revised 6 October 2002 Accepted 6 October 2002 ABSOLUTE CONFIGURATION OF a-PYRONES BY LC-NMR 35

2H-pyran-2-one (2), i.e. compounds 2a and 2b, were prepared in an exactly similar fashion.

Preparation of Mosher's ester by micro-reaction. Compound 1 (50 pg in 100 pL of dichloromethane) was transferred to a 2mL HPLC vial with a capped septum and treated with 20 pL of pyridine and 100 mg of (R)- (-)-MTPA chloride. This low amount of pyridine compared with MTPA chloride was found not to affect the reaction yield, The mixture was stirred at room temperature under an atmosphere of nitrogen for 4 h and the progress of the reaction was monitored by LC-UV- Figure 1. a-Pyrones, 6S-(2S-hydroxy-6-phenylhexyl)-5,6-dihy- (10 APCI/MS pL of the reaction mixture was analysed, dr o-2 l+py ran-2-one ( 1 ) a nd 6B-(4S,6S-di hyd roxy- 1 O-phe nyl- being5Vo of the total reaction mixture). Dichloromethane 1-decenyi)-5,6-dihydro-2tÊpyran-2-one (2), isolated from the was removed with nitrogen flux, and the total mixture aerial parts of Ravensara crassifolia. was solubilised in 100 pL of acetonitrile (HPLC grade) and analysed by stop-flow LC-NMR [the number of transients (NT) was 10241. Mosher's ester derivative (1b) recorded at 210 and 254nm and PAD specra were was prepared using (S)-(+)-MTPA chloride under the recorded between 200 and 500 nm. LC-MS, which was conditions described above. performed directly after UV-PAD analysis, employed an APCI interface with the following conditions: operation, LC-UV-APCVMS analysis. Reversed-phase HPLC positive mode; capillary temperature, l50oC; vapotizer analysis of compounds 1, la, lb, 2, 2a and 2b was temperature, 70'C; sheath gas flow, 60; corona needle performed on a Hewlett-Packard (Waldbronn, Germany) current, 5 pA; spectral range, 150-900 mu). series 1100 liquid chromatograph system with photo- diode array detector (PAD), connected in series to a LC-NMR analysis. A Varian (Palo Alto, CA, USA) Finnigan MAT (San Jose, CA, USA) LCQ ion trap mass Uniçy Inova 500 MHz NMR instrument equipped with a spectrometer. Coqrpounds (250 pg of eac[) wele injected 'H["C] pulse field gradient indirect detection microflow onto a Bondapak* Cre column (100 x 8 mm i.d.; 10 pm) LC-NMR probe (flow cell, 60 pL, 3 mm i.d.) was used. which was eluted with acetoniffile:water (80:20) for Reversed-phase HPLC analysis of the compounds was 35 min at a flow rate of 0.8 ml/min. UV traces were' carried out on a Varian modular HPLC system compris-

z o o + MeOH trace$ I àD =ct CL ùËI {

(o II rat rô I + H-AT'H-A/' I I (o T (oq!to I a E TÀIr:l 1a (q-urpa

-ocH" 8" CO IE -I li 1ÿ)9T lo I @ - Ea "l § 1b (&-urpe

Figure 2. Stop-flow LC-NMR spectrum of 1 and of the Mosher's esters la and lb (sample size 250 pg; NT = 128). (For analytical &' cohditions see Experimental section.)

Copyright O 2003 John Wiley & Sons, Ltd. Phytochem. Anal. 14:34-39 (2003) 3; E. F. QI.]EIROZ ETAL.

Supp. MoCN

Supp. MoCN

rF'' ! 19'5n)

l- 'r64Ja3Z (DZu) , =2 Figure 3. Stop-flow LC-NMR WETGCOSY spectrum of lb (sample size 250 pg; Nl = 512; NT = 36). (For analytical conditions see Experi mental section.)

ing a model 9012 pump, a model 9050 UV detector and a 210 nm. The reference of the solvent signal was set at ô Valco injection valve. The separation was performed 2.00 for acetoniftile, and for ânalysis involving samples using the same conditiens as given above for LC-MS sizes of 2501tgNT was set to 128. The LC-NMR COSY analysis except that deuterated water was employed in experiment was performed using a WETGCOSY se- the mobile phase, and UV ûaces were measured only at quence (NI=512; NT=36). For the reaction mixture

+ 0.2 + 0.12

's

Figure 4. Determination of the absolute configuration with Àôg values [obtained using equation (1)] of compounds la and lb m. using the Mosher method.

Copyright O 2003 John Wiley & Sons, Ltd. Phyrochem. Anal. 14:34-39 (2003) ABSOLUTE CONFIGURATION OF c-PYRONES BY LC-NMR 3'7

- 0.(»

+ 0.(N + - r,f.â dc-ô,

Figure 5. Determination of the absolute configuration with Âôs values [obtained using equation (1)] of compounds 2a and2b using the Mosher method.

(50 pg) a good LC-IH-NMR spectra was obtained using NMR (MeCN: D2O, 80:20) stop-flow mode: ô 603 (1H, NT= 1024. d, J =9.5H2,H-3),6.85 (1H, m, H-4), 2.43 (2H, m, H-5), 4.89 (1H, m, H-6),5.68 (1H, m, H-1'),5.85 (lH, m,H-2'), LC-MS and LC-NMR data of 1, la, lb,2,2a and 2b 2.28 (lIJ,m, H-3'), 3.91 (1H, m,H-4'),1.58 (1H, m, H- 5'), 4.00 (1H, m, H-6'),1.46-1.57 (1H, H-1'), 1.32-1.44 (lH, H-8'), 1.60 (lH, m, H-9'), 2.61 (l}{, t. 9.6 Hz,H- 65-(2S-Hydroxy-6-phenylhexyl)-5,6-dihydro-2H- = pyran-2-one (1) lO'), 1 .12-7.20 (H-Ar", H-2', 3', 4", 5" 6'). LC-MS APCl-positive ion mode lru/z(relative intensity)l: 344.8 IH-NMR (in CDCI3): see Raoelison et al. (2001).LC-IH- lM + Hl+ (100), 309.2 (22),291.2 (t2),263.2 0,206.9 (11), (28). NMR stop-flow mode: 6 5.94 (1H, d, J =9.7 Hz, H-3), 171.4 7.02 (IH, m, H-4), 2.28 (2H, m, H-5), 4.65 (lH, m, H-6), 1.65-1.83 (2H, H-1',),3.75 (1H, m,H-2',),7.47 (zE,H-3',), 4',6'-Di-(S)-MTPA ester o12 (2a) 1.38-1.48 (2H,H-4',),1.25-1.52 (2H, H-s',), 2.61 (2H,t, (250 pg 128): ô J = 9.7 Hz, H-6'), 7 .16-7 .25 (H-Ar', H-2", 3', 4", 5", 6") o LC-IH-NMR stop-flow mode with NT = LC-MS APCl-positive ion mode lrt/z (relative inten- 593 (lH, d, J =9.5H2, H-3), 6.94 (lH,m, H-4), 2.33 siry)l: 275 [M + H]+ @», 239 (s0),221.2 (92), 2tt.t (2H, m, H-5), 4.81 (1H, m, H-6), 5.57 (2H,m, H-1' and (100), 171.3 (63). H-2'), l.16 (2H,m, H-3'), 4.91 (1H, m, H-4'), 2.47 (2H,m, H-5'), 4.82 (1H, m, H-6'), 2.01 (ZH,m, H-7'), (lH, (2H, H-g',), 2.59 (2H, t, 8-(S)-MTPA ester of 1 (1a) 1.26-1.29 H-S',), 1.52-1.60 J =9.6H2, H-70'), 7.12-:7 .20 (H-Ar", H-2", 3", 4", 5", (3H, OCH3, MTPA), 3.50 (3H, s, OCH3, LC-IH-NMR stop-flow mode (250 pg with NT 128, or 6'),3.43 s, = MTPA), 1.50-7.52 (10H, H-Ar"',H-Ar",2 MTPA). LC- 50 pg with NT = 1024), (reaction mixture)): ô 5.92 (lH, d, MS APCl-positive ion mode (relative intensity)l: J =9.7 Hz, H-3),6.96 (1H, m, H-4), 2.31-2.33 (2H, m, H- lm/z 776.8 (100),543.1 (22),309 (12),291 (10), 5), 4.35 (1H, m, H-6), 1.89-2.00 (2H, H-l'),5.29 (1H, m, tM+Hl* (11). H-2'), 1.65 (2H,m, H-3'), 1.30-1.60 (2H, H-4^), 1.25- 223.t 1.52 (2H, H-5',), 2.58 (2H, t, J 9.7 Hz, H-6',), 1 .tÇ7 .25 = (2b) (H-4r", H-2',3", 4",5',6"),3.42 (3H, s, OCH3-MTPA), 4',6'-Di-(]Q)-MTPA ester of 2 7.20-7.42 (H-Ar"', phenyl MTPA). LC-MS APCI-posi- LC-IH-NMR stop-flow mode (250 pg with NT 128): ô tive ion mode fm/z (relative intensity)l: 490.8 [M + H]* = 603 (1H, d, J H-3), 7.03 (1H, m, H-4), 2.37 (2s) , 2s7 (7 8) , 239 .2 (100) , 221 .2 (es) , 211 .2 (7 6) . =9.5IJ2, (2H,m, H-5), 4.88 (1H, m, H-6), 5.71 (2H,m, H-1' and H-2'), 2.51 (2H,m, H-3'), 5.18 (lH,m, H-4'), 2.43 E-(R)-MTPA (1b) ester of 1 (2H,m, H-5'), 5.11 (lH,m, H-6'), 2.05 (zH,m, H-7'), 1.29-r.34 (lH, H-S',), 1.52-1.60 (2H, H-9), 2.59 (2H, t, LC-IH-NMR stop-flow mode (250 pg with NT = 128, or J = 9.6 Hz, H- 10), 1 .12-:7 .20 (H-Ar', H-2" , 3" , 4" , 5' , 6"), 50 pg with NT 1024), (reaction mixture)): ô 5.89 (lH, d, = 3.43 (3H, s, OCH3, MTPA),3.50 (3H, s, OCH3, MTPA), J Hz,H-3),6.84 (lH, (2H, =9.7 m, H-4), 2.11-2.15 m, H- 1.50-7.52 (10H, H-Ar"', H-1\"",2 MTPA). LC-MS 5), 3.91 (1H, m, H-6), 1.80-1.91 (2H, 5.30 (1H, m, H-l'), APCl-positive ion mode (relative intensity)l: 776.8 H-2'), 1.69 (2H, m, H-3'), 1.32-1.61 (2H, H-4'), 1.25- lnlz lM + Hl+ (100), s43 (22),309 (ts),291(10),223.3 (17). 1.57 (2H, H-5',), 2.58 (2H, t, J = 9.7 Hz, H-6'), 7 .lÇ7 .25 (H-Ar', H-2" , 3" , 4' , 5u , 6'), 3.42 (3H, s, OCH3-MTPA), 7.20-7.42 (H-Ar'', phenyl MTPA). LC-MS APCI-posi- tive ion mode [m/z (relative intensity)]: 490.8 tM + Hl+ RESULTS AND DISCUSSION (2s), 2s'7 (7 7 ), 239.1 (100), 221.2 (93), 211.2 (1 2). In order to evaluate the potential of LC-NMR for the 6R-(4S,65-Düydroxy- 10-phenyl- 1-decenyl)-5,6- determination of the absolute configuration in the dihydro-2H-pyran-2-one (2) microgtam range, two d-pyrones 6S-(2^§-hydroxy-6- phenylhexyl)-5,6-dihydro-2H-pyran-2-one (L) and 6R- rH-NMR (in CDCI3): see Raoelison et al. (2001).LC-1H- (4,§,6S-dihydroxy- 1 O-phenyl- 1 -decenyl)-5,6-dihydro-

Copyright O 2003 John rùr'iley & Sons, Ltd. Phytochem. Anal. L4:34-39 (2003) 38 E. F. QUEIROZ EZAL,

Supp. MoCN

LC/ 1H-NMB of compornd la o -ocH3 6 I o H-AT" H-AT o ri 'É =

rq) «, m m MS spectrum of 1 fl) 50 4 IM+t[' n90 to I 0 50 100 l5O Àx, 250 3æ 350 .l(n 5lx, 550 . rtz. '§0 eD.0 â?.o 2|1.2 MS spectrum of la tl2 4S E 0,flH'

qD 350 utl?

Figure 6. LC-UV-MS and LC-UV-NMR (sample size 50 pg; NT = 1024) analysis of the micro-reaction mixture of compound 1 after 2 h reaction time. ) '

2H-pyran-2-one (2) (Fig. 1), isolated from the root bark mode of operation was preferred to on-flow since it of Ravensara crassifulia (Lauraceae), were selected as allows the acquisition of a higher number of transients model compounds (Raoelison er al., 2001,2002). The and permits_ two-dimensional correlation experiments classical Mosher reaction was performed on 5 mg of I such as 'H-'H WET COSY to be conducted. Solvent and 2, which were esterified with (R)- and (.ÿ)-MTPA. suppression was achieved using the "water suppression The four different ester derivatives (La, lb,2a and 2bl enhanced through effects" technique (Small- rH- ft $fED were purified by semi-preparative HPLC and the combe et al., 1995), which is both fast and reliable. NMR spectrum of each purified derivative was recorded Good-qlrality LC-IH-NMR spectra were recorded for IH-NMR in deutérated chloroform. A comparison of the all derivatives as well as for the un-derivatised natural chemical shifts of the protons adjacent to the esterified products: a signal-noise ratio higher than 3 for the H-3 hydroxyl group between La (.ÿ)-MTPA and lb (R)- doublet was obtained for all compounds with less than MTPA allowed the determination of the absolute 128 transients. Chemical shifts recorded under LC-NMR configuration of C-2' to be S. In the case of the a-pyrone conditions differed slightly from those recorded in 2, a comparison between derivatives 2a (S)-MTPA and deuterated chloroform.- Cômparison of LC-IH-NMR 2b (R)-MTPA ind rric cenffes spectra of (À)-MTPA la and (S)-MTPA lb derivatives atC-4', C-6'were ,2002). showed significant chemical shift differences for the The LC-NMR èsters was protons near to the MTPA ester (Fig. 2). These diagnostic performed using isocratic reversed-phase HPLC condi- shifts were used for the determination of absolute tions with acetonitrile:deuterated water as mobile phase. configuration. Suppression of the acetonitrile solvent A C1s radial compression column having a large internal resonance (ô 2.00) hampered a clear assignment of the diameter (8 mm) was preferred to classical analytical signals between ô 1.60 and 2.20. For the unambiguous columns because of its high loading capacity. Passage of chemical shift attribution of each proton for both the crude reaction mixture through the HPLC column Mosher's esters la and lb, 'rt-tu wPï COSY spectra separated the Mosher's ester derivatives from the were recorded (Fig. 3). With these two-dimensional reactants and residual starting material. Isocratic rather correlation experiments, precise chemical shift assign- than gradient elution was employed since all spectra ments for H-1', H-3' and H-4' were possible. Differences needed to be recorded in the same solvent for comparison of all chemical shifts of (S)-MTPA la and of (^R)-MTPA purposes. Under these isocratic LC-NMR conditions, lb were calculated using equation (1). Positive ôH values. 250 pg of a-pyrone 1 and its purified (R)- and (S)-MTPA were recorded for the d-p)'rone moiety (C-l'), which was & derivatives were analysed by stop-flow LC-NMR. This located on the right side of the Mosher model: negative

Copyright O 2003 John Wiley & Sons, Ltd. Phytochem. Anal. 14:34-39 (2003) ABSOLUTE CONFIGURATION OF a-PYRONES BY LC-NMR 39

ôH values for C-3'lC-5'indicated that the alkyl chain had completion of the esterification reaction in both cases. A to be located on the left side of the model (Fig. 4). molécular ion [M + H]* at m/z 491 (a-pyrone 274 Finally, the proton geminal to the esterified hydroxyl Da * MTPA 216 Da) was observed for the main peak, group was located on the opposite side of the plane and confirming the esterification of 1 by MTPA chloride. The the absolute configuration at C-2' was determined using remaining reaction mixture was then submitted to stop- the priority rule of Cahn-Ingold-Prelog. This arrange- flow LC-NMR analysis. Pyridine, excess of MTPA and ment indicated a sinistrorse rotation and confirmed the S traces of starting material eluted rapidly, while the ester absolute configuration for the asymmetric cerrtre_ C-2' derivatives eluted between 8 and 12 min. Under these (Fig. a). conditions, good quality spectra were recorded (signal- The same type gf LC-NMR analyses were performed noise ratio ) than 4forH-3; NT = 1024) as shown in Fig. on the o(-pyrone 2-and its Mosher's ester derivatives 2a 6. Similar results were obtained to those measured using and 2b. As in the case of 1, useful differences in the a sample size of 2501t9, thus demonstrating the proton chemical shifts owing to the diamagnetic effect of usefulness of the method when working with restricted the benzene ring of the MTPA moiety were observed. amounts of starting material. Assignment of the 'H signals between ô 1.20 and2.20 The present study has demonstrated that the chemical were obtained via a 'H-'H WET COSY experiment. shifts of the protons of Mosher's ester derivatives Calculation of the Âôs, together with the model proposed established by itop-flow LC-IH-NMR are comparable rH-lüvIR by Mosher, permitted assignment of the configura- with those obiaineà by conventional analysis. E ^S tion for both asymmetric centres atC-4' and C-6' (Fig. 5). Analysis of Âôs gave the correct absolute configuration All results presented above were obtained with 250 pg of both fl-pyrones I and2. Micro-reaction combined with of purified MTPA derivatives. In order to minimise the LC-NMR analysis seems to be very promising for a rapid consumption of starting material and to simplify sample and efficient determination of the absolute conf,guration clean-up, direct LC-NMR analysis of the whole reaction of natural products which can only be isolated in very mixture for the determination of absolute configuration small amounts. The method can be applied to mixtures of was attempted. Two aliquots (50 pg) of a-pyrone I were compounds which are difficult to isolate on the esterified, one with (À)-MTPA and the other with (^S)- preparative scale since separation of the Mosher's esters MTPA, following the procedure described in the is performed prior to NMR detection. experimental part (micro-reaction esterification). At the end of the reaction, the excess of dichlcttomethane was removed under a gentle nitrogen flux. Aliquots (10 pL) Acknowledgements equivalent to 5Eo of the residual reaction mixture were, analysed by LC-ESÀ4S using the same isocratic condi- The Swiss National Science Foundation (grant no.063670.00, Prof. K. tions as described for LC-NMR in order to wrify the Hostettmann) is gratefully acknowledged for supporting this work.

REFERENCES

Dale JA and Mosher HS. 1968. Nuclear magnetic resonance absolute configurations of marine terpenoids. J Am non-equivalence of diastereoisomeric esters of a-sub- Chem Soc 113: 4092-4096. stituted phenylacetic acids for the determination of Rasmussen HB, Christensen SB, Kvist LP, Kharazmi A and stereochemical purity. J Am Chem Soc g0: 3732-3738. Huansi AG. 2000. Absolute configuration and anti-proto- Dale JA and Mosher HS. 1973. Nuclear magnetic resonance zoal activiÿ of minquartynoic acid. J Nat Prod 63: 1295- enantiomer reagents. Configurational correlations via 1296. nuclear magnetic resonance chemical shifts of diaster- Raoelison GE, Terreaux C, Oueiroz EF, Zsila F, Simonyi M, eomeric mandelate, Gmethylmandelate, and a-methoxy- Antus S, Randriantosona A and Hostettmann K. 2001. oc-trif I uoromethyl ph e nylacetate. J Am Che m S oc 95: 512- Absolute configuration of two new 6-alkylated-a-pyrones 519. lrom Ravensara crassifolia. Helv Chim Acta 84: 3470- Dale JA, Dull DL and Mosher HS. 1969. a-Methoxy-a- 3476. trifluoromethylphenylacetic acid, a versatile reagent for Raoelison GE, Terreaux C, Oueiroz EF, Zsila F, Simonyi M, the determination of enantiomeric composition of alco- Antus S, Randriantosona A and Hostettmann K. 2003. hols and amines. J Org Chem34:254ÿ2549. Erratum: Absolute configuration of two new 6-alkylated- Fiud JC, Horeau A and Kagan HB. 1977. Methods for the d-pyrones lrom Ravensara crassifolia. Helv Chim Acta lin determination of absolute configuration . ln Stereochem- press). istry,Vol.3, Kagan HB (ed.). George Thieme: Stuttgart: 9- Smallcombe SH, Patt SL and Keiffer PA. 1995. WET solvent 21. suppression and its application to LC-NMR and high- Gu ZM, ZengL, Fang XP, Colman-Saizarbitoria T, Huo M and resolution NMR spectroscopy. J Mag Reson A 117;295- Mclaughlin JL. 1994. Determining absolute configura- 303. tions of stereo-centres in Annonaceous acetogenins Sullivan GR, Dale JA, Dull DL and Mosher HS. 1973. through formaldehyde acetal derivatives and Mosher's Correlation of configuration and 19F chemical shifts of ester methodology. J Org Chem59:5162-5172. d-methoxy-d-trifl uoromethylphenylacetate derivatives. J Harada N and Nakanishi K. 1983. Circular Dichroic Spectro- Org Chem 38:2143-2147 scopy-Exciton Coupling in Organic Stereochemistry. Trost BM, Belletire JL, Godleski S, McDougal MC and University Science Books: Mill Valley. Balkovec JM. 1986. On the use of the @methylmandelate Jeong SJ, Miyamoto T, lnagaki M,.Kine YC and Higuchi R. ester for establishment of absolute configuration of 2000. Three novel sesquiterpene alkaloids from Cyperus secondary alcohols. J Org Chem51:2370-2374. rotundus. J Nat Prod6O:673-675. Wolfender JL, Ndjoko K and Hostettmann K. 2001. The Ohtani l. Kusumi T, Kashman Y and Kakisawa H. 199i. High- potential of LC-NMR in phytochemical analysis. Phyto- field FT NMR application of Mobher method. T'he chein Anal 11:1-22. w"

Copyright O 2003 John Wiley & Sons, Ltd. Phytochem. Arul. 14:34-39 (2003) P.04 : J-R. Ioset, G.E. Raoelison, K. Hostettmann. Detection of aristolochic acid in Chinese phytomedicines and dietary supplements used as slimming regimens. Food and Chemical Toxicology 2003, 41, 29-36.

PERGAMON Food and Chemical Toxicology 41 (2003)29-36 www.elsevier

Detection of aristolochic acid in Chinese phytomedicines and dietary supplements used as slimming regimens

J.-R. Iosetu, G.E. Raoelisonu,b, K. Hostettmannu,*

^Institut de Pharmacognosie et Phytochimie, BEP, Universitë de Lausanne, CH-1015 Lausanne, Swilzerland bLaboraloire de Pharmacodynamie, Facultê des Sciences, Université d'Antananarivo, BP 906, Antananarivo 101, Madagascar

Accepted 15 July 2002

Abstract

Over the last l0 years, numêrous cases ofintoxications, leading for the most part to end-stage renal failure, have been reported alter consumption of slimming regimens madezof Chinese herbal preparations. These intoxications were associated with species of lhe Aristolochia genus, such as Aristolochiafangchi (Aristolochiaceae), known to contain very nephrotoxic and carcinogenic meta- bolites named aristolochic acids. Several commercial dietary supplements, teas and phytomedicines used as slimming regimens were analysed for their aristolochic acid I content. A preliminary detection of this toxic compound was made by thinJayer chromato- graphy. The presence of aristolochic acid I in these prgparations was confirmed by a HPLC/UV-DAD/MS analysis. A quantitative determination of aristolochic acid I was also achieved in the incriminated preparations using both UV and MS detection. Out of 42 analysed preparations, four were found to contain aristolochic acid I and t\ryo were suspected to contain aristolochic acid deriva- tivês. Immediate removal of these products from the Swiss market was called for. €) 2003 Elsevier Science Ltd. All rights reserved

Keywords: Aristolochic acid; Intoxication; LC/UV; LCiMS; Aristolochiaceae; Aristolochiafangchi

1. Introduction tolochia genus, such as Aristoiochia fangchi Y.C. Wu (Flurer et ,a1.. 2000). The most common confusion Several cases of end-stage renal failure after con- occurs wilh Stephania tetrandra S. Moore (Capparaceae) sumption of slimming regimens involving Chinese plant (Chinese name: fangchi or hanfangji) that is replaced by preparations have recently been described in the litera- the very toxic Arislolochia fangcfri (Chinese name: ture (Vanherweghem et al.. 1993; Pena et al., 1996; guangfangchi). The Arislolochia genus contains more Tanaka el al., 7997; Stenge[ and Jones, 1998; Vanher- than 800 herbaceous or shrubby, often climbing, species weghem, 1998; Chang et al., 2001; Yang et al., 2000), growing in both temperate and tropical regions and is These intoxications were due to an accidental con- well known to contain very nephrotoxic and carcino- tamination of these preparations by species belqnging to genic compounds named aristolochic acids. These lhe Aristolochia genus (Vanhaelen et al., 1994). The nitrophenanthrene derivatives were indeed associated confusion was attributed to similar Chinese vernacular with severe renal insufficiency and urothelial carcinoma denominations between plants contained in these phy- after their DNA adducts were found in related human tomedicines'and some species belonging to the ,4rrs- tissue samples (Nortier et al., 2000; Stiborova et al., 2000). The mutagenic properties of several metabolites of aristolochic acids I and II formed by rat liver lvere also determined experimentally (Schmieser et al., 1986). The Abbreviations: APCI, pressure atmospheric chemical ionisation; presence of aristolochic acids I and II has already been SPE, solid phase extraction; TLC, thin-layer chromatography. plants as * Corresponding author. TeL.: +41-21-6924561; fax: +41-21- reported in Asian medicinal as well in slimming 6924s65. products sold on the Asian market (Zhu and Phillipson, E-mail address : [email protected] (K. Hostettmann). 1996;Haihimoto et al., 1999; Lee e|al.,2002), F 0278-à915/03/$ - see front matter O 2003 Elsevier Science Ltd. All rights reserved. PII: S0278-691 5(02)00219-3 .i -30 J,-R. Ioset et al, f Food and Chemical Toxicology 41 (2003) 29-36

For these reasons, several health institutions, includ- obtained in larger quantities from the same source or ing the US Food and Drug Administration (FDA), have withdrawn from the study, These phytomedicines were recently published safety information related to the either ground vegetable material or plant extracts in the presence of aristolochic acids in plant preparations in format of tea bags, pills, powders, teas, capsules, pellets, order to prevent further cases of intoxication (informa- granules or tablets. tion available at web address: http://www.cfsan.fda.gov/ -dms/ds-bot.html)! Several analytical methods using 2.2. Standard compounds and reagenls layer chromatography (Pharmacopoeia, of the People's Republic of China, 1988; Vanherweghem et al., 1993), Aristolochic acid I used as a reference compound for fluorometric assay (Rao et al., 1975), gas-liquid chro- the analysis was purified by preparative TLC from a matography analysis equipped with a flame-ionisation mixture of aristolochic acids purchased from Sigma detection (Rao et al., 1975), HPLC coupled to a UV (reference no. A-5512) containing 670Â of aristolochic spectrophotometric detector (HPLC/UV) (Nishida and acid L This mixture was first eluted on a silicagel 60Fzs4 Fukami, 1989;Tsai eLal.,1993; Hashirnoto et al., 1999; TLC glass-backed plate (Merck) !ÿith CHCI3.MeOH: Lee et a1., 2002) or to a quadrupole ion-trap mass spec- acetic acid (65:20:5, by vol.) before being submitted to a trometer (HPLC/MS/MS) (Kite et al., 2002) have second migration on a RP-18 WFzs+ TLC glass-backed already been employed for the detection and quantifi- plate (Merck) using MeOH:H2O (35:65). The purity of . cation of aristolochic acids. With the same aim, we aristolochic acid I was checked by LCIIJV-QAD under '' recently developed two different methods-a thinJayer the conditions described in the LC/MS section. Diphe- chromatographic (TLC) assay and a LCIUV-DAD/MS nylamine (purum) was purchased from Fluka (reference analysis-for a rapid, specific, sensitive and quantitative ro.42161). determination of aristolochic acid I, the major toxic derivative found in Arisiolochia fangchi, in complex 2.3, Preparation of the samples for TLC and HPLC herb mixtures (Fig. 1). A strategy combiding th"r. two analysis methods was applied to detect the presence of aris- tolochic acid I in Chinese phytomedicines commercia- " An amount of 5 g of plant material (pills, capsules, lised in Switzerland. : plant powder or plant extract) \vas extracted with 100 ml of boiling MeOH for 3 h. After filtration, the sample was evaporated to dryness under reduced pressure and 2. Materials and methods freeze-dried. A 20-mg/ml methanolic solution of this extract was used for the qualitative detection of aris- 2.1. Origin of the analysed samples tolochic acid I by TLC and HPLC/UV-DAD/MS ana- lysis. To perform quantitative measurements, an The analysed samples were obtained from private or amount of 50 mg of this extract was dissolved in 1 ml of industrial sources after advertisement was made in pro- solvent (10% MeOH and 90%o HzO) and eluted with 5 fessional journals and local newspapers. All the selected ml of H2O by solid phase extraction (SPE) using a 3-ml samples were declared as Chinese plant mixtures used as Chromabond@ Cl8 Hydra column (Macherey-Nagel) slimming regimens in the treatment of obesity. Samples previously stabilised with l0 ml of H2O. A quantitative .available - in amounts too small for analysis were either elution of aristolochic acid I was ensured by 20 ml of a MeOH:THF solution (80:20, v/v). This percolate was then evaporated to dryness under reduced pressure and COOH freeze-dried before being dissolved in I ml of MeOH. Samples prepared for the purpose of quantification were performed in triplicate precisely Noz and weighed, 2.4. TLC conditions

The sample was applied to a silicagel 60 F254 TLC glass-backed plate and eluted with CHCI3:MeOH:acetic acid (65:20:2, by vol.), After evaporation of the solvents, the plate was sprayed with a solution of 0.5Yo dipheny- lamine in HzSO+ 600Â and heated for 10 min at 100 "C (blow-dryer or oven). Under these conditions, aris- tolochic acid I was detected under üsible light as a dark Aristolochic I acid blue or black spot which developed a yellow fluores- Fig. l, Structure of aristolochic acid I cence when observed at 366 nm. Under these conditions F J.-R. Ioset et al. I Food and Chemical Toxicology 4t (2003) 29-36 of separaüon, an approximate HRl of 72 was calculated hRf 100 for aristolochic acid I.

2.5. HPLC I UV-DAD conditions

Reversed-phase HPLC was carried out using an HPLC system HPrl100 (Hewlett-packard, palo Alto, CA, USA) equipped with a binary pump. A Hewlett- Packard (Walbronn, Germany) Il00 series on-line pho- todiode array detector (DAD) was used for detection. This instrumentation was controlled by Hp Chemsta- tion software. The separation was performed on a C-18 Symmetry@ column (250x4.6 mm i.d., 4 pm particle size, Waters, Bedford, MA, USA) with a linear MeOH hRf o (0.5% acetic acid):H2O (0.5% acetic acid) gradient (60:40 to 100:0) in 2l min, The column temperature was ABG ' set at 30 "C with a flow of I ml/min. The DAD-UV detector was set at 254 and 224 nm. A reference mixture Plate L Detection of aristolochic acid I by TLC under visible light in -- sample 16 (declared as Han- lang ji, Stephaniae tetrairdrae radix). (A) of aristolochic acids, mostly composed of aristolochic MeOH extract of sample 16 (20 mglml, 10 pl). (B) l:l mixture of A acids I and II, was employed to optimise the LC and C. (C) Aristolochic acid I (1 mg/ml, l0 pl), separation method. Addition of acetic acid was required to avoid tailing of the compounds of interest during the LC separation. Under these conditions of separation reaction with diphenylamine. This method allowed the and during the whole period of analysis (6everât weeks), detection of quantities of aristolochic acid I as small as aristolochic acid I was eluted with a constant retention I pg in visible light and 0.2 pe at 366 nm. Aristolochic time between 13.9 and 14.0 min. A volume of 20 pl was " acid I was easily detected by TLC in samples 16 arrd l7 injected. Quantification using UV detection Was per- and was suspected in sample 22. formed at 254 nm. Samples for quantitative analysis A LC/DAD-UV/MS analysis was then employed to were injected in triplicate. Each solution was diluted assess the presence of aristolochic acid I in the studied with MeOH in order to obtain a concentration within plant preparations and if necessary, to provide quanti- the range of the calibration curve. ûcation. Results of the different analysed samples are presented in Table L Aristolochic acid was detected 2.6. MS conditions through the presence of a chromatographic peak eluting at about 13.8 min exhibiting a quite typical IfV spectrum MS analyses were performed by atmospheric pressure with maxima at 224,251 (sh), 321 and 393 nm as well as chemical ionization (APCI) on a Finnigan LCe ion trap characteristic ionised fragments at mlz 295 (M-NOr+ mass spectrometer (FinniganMAT, San Jose, CA, mf z and,324 (M-OH)+ in the APCI-MS positive mode USA). The APCI optimised parameters were: capillary (see Table 2). Under the experimental conditions, a -temporature 150 "C, vaporiser temperature 450 oC, cor- detection limit of 2 ng of aristolochic acid I was deter- ona needle current 5 pA and sheath gas flow 60%. mined for both IJV and SIM-MS measurements. A These parameters ,ÿvere optimised by Flow Injection detection limit of 15 ng was obtained in the full scan MS Analysis (FIA) using aristolochic acid I as reference mode. The presence of aristolochic acid I was unam- compound. Spectra (150-700 amu) were recorded in the biguously found in four samples, three of them being positive ion mode. Quantiûcation using MS detection declared as Fang ji or Han fang ji preparations (see performed was in the full scan mode set between 150 Table l). Only traces of the toxic product could be and 700 amu. Samples for quantitative analysis were determined in sample 15, while larger amounts were dis- injected in triplicate. covered in the three other samples. Aristolochic acid I could not be detected in sample l5 by TLC. The result of the TLC screening for aristolochic acid I was tenuous in 3. Results the case of sample 22, but the presence of aristolochic acid I could be confirmed through the LC/UV-DAD/MS The samples for analysis were first screened in a TLC analysis (Fig. 2). Samples 10 and 21 were free of aris- assay using diphenylamine as detection reagent. Under tolochic acid I but the presence of other aristolochic these conditions, aristolochic acid I was seen as a dark- acids was strongly suspected, since some compounds in blue coloured spot under visible light. (see plate 1). A these prgparations exhibited UV spectra similar to that yellow fluorescence was also observed aI 366 nm after of aristolochic acid I, associated with MS ions at mlz295, F '" 12 J.-R. Ioset et al, I Food and Chemical Toxicology 41 (2003) 29-36

Table 1 Commercial samples analysed for their aristolochic acid I content

No Declaration of content Product Aristolochic Remarks presentation acid I

I Pu-Ehr tea Tea bags 2 Pu-Ehr tea Capsules t 3 Pu-Ehr tea Tea bags 4 Black tea Tea bags 5 Pu-Ehr tea Capsules 6 _s Granules 8 Green tea, Java tea, anise, blackcurrant Tea bags 9 Black tea Tea bags 9 Powder l0 Asari herba Powder Suspicion ol the presence of aristolochic acid derivatives II Pu-Ehr tea Tea herb 12 Pu-Ehr tea 'Capsules 13 Pu-Ehr tea Pills t4 b Tea herb l5 Fang ji huang qi sang Stephania and Astragalus Powder + Traces of aristolochic acid I (not visible on TL€) l6 Han fang ji Stephaniae tetrandrae radix Powder Quantification of aristolochic acid I UV: 0.044% MS: 0.040% 17 Ba zheng san (eight herbs) Powder Quantification ol aristolochic acid I UV: 0.009% MS: 0.014% 18 Shu jing huo xue tang Powder l9 Xin yi san Magnoliae fl. ( ' Powder 20 Chuan mu tor,g Clemalidis caulis Powder 2l Xi xin Asari herba Powder Suspicion of the presence of aristolochic acid derivatives 22 Han fang ji Sinomeniwn acutum Powder + Dubious result on TLC 23 San bi tang '! Powder 24 Du huo ji sheng tang Angelica pubescens Powder z5 Long dan xie gan tang Gentiwta longdancao Powder 26 Pu-Ehr tea Capsules 27 Pills 28 Powder 29 Tea herb 30 Green tea Tea herb 3l Tea bags 32 Ginseng tea Tea bags 33 Pellets 34 Shu xin xiao zhi wan Pellets 35 Jian pi he wei wan Pellets 36 Yi shen yang yin wan Pellets 5t Yi qi yang xue wan pills 38 Xiao feng san Powder 39 Ligusticum radix Powder 40 Fang ji huang qi tang Stephania and Astragalus Powder 4l Black tea Tablets 42 Xiao feng san Powder

For reasons ofpublic health care, translated Chinese names ofthe analysed preparations are given where possible, E - Products obtained without a declaration. b Product declaration only available in Chinese,

324, 342, 354 and 358 expected for aristolochic acid of aristolochic acid L Very similar results were obtained derivatives (Fig. 3 and Table 2), Quantification of aris- using these two different techniques (see Table l). Sample tolochic acid I in samples 16 and 17 was performed using 16, declared as a single plant preparation, was found to UV and MS detection, These samples were determined as contain about four times more aristolochic acid I than 17, the commercial products containing the highest amounts said to be â mixture of eight drugs. F l.-R. Ioset et al. I Food and Chemical Toxicology 41 (2003) 29-36 33

4. Discussion total safety ofuse ofthe analysed preparations since it is not yet known whether undetectable traces of aris- Owing to its high toxicity, the development of a tolochic acids can still have toxic effects if taken reg- selective and sensitive method for the detection of ularly over a long period, Other parameters such as aristolochic acid I in complex mixtures is more impor- consumer sensitivity to aristolochic acid, quantities of tant than its accurate quantification. The presence of ingested preparations, predisposition to renal insuffi- aristolochic acid I ând of its derivatives is indeed not ciency, should also be considered in the risk evaluation acceptable in plant preparations since even traces of of intoxication. Similar limits of detection were these compounds constitute a potential health risk. obtained for IfV (Hashimoto et al,, 1999: Lee et al., However, the determination of new toxicological data in 2002) and MS (Kite et aL.,2002) analysis of aristolochic the future should be considered as a sufficient motiva- acid I but the method developed here has the advantage tion to achieve both detection and quantification of of coupling both kinds of detection in association with a aristolochic acid L The detection of aristolochic acid I prior rapid concentration of the sample by SPE. A in herbal preparations was ensured by a çombination of search strategy for aristolochic acid I in herbal pre- two different techniques, namely a TLC analysis fol- parations was also proposed using a TLC pre-screening lowed by detection of aristolochic acid I through che- followed by a LC/W-DAD/MS confirmative analysis. mical derivatisation and an HPLC separation coupled Both lfV and MS detection allowed the quantification of to IfV diode array and mass spectrometry detection. aristolochic acid I in the samples undergoing Snalysis, very Details concerning the optimisation and the validation similar amounts of this toxic compound bein§ determined of these methods have already been discussed elsewhere by these two techniques. The presence of lower amounts of (Ioset et al., 2002). The combination of these two tech- aristolochic acid I in sample 16 compared with those niques proved to be a very efficient strategy in the search determined for Aristolochia fangchi in the literature for aristolochic acid I in plant preparations. The two (Hashimoto et al., 1999) could be explained by the fact samples containing the highest ôonôerrtr6tloni of aris- l}nat Aristolochia fangchi-probably mistaken for the tolochic acid I-samples 16 and l7-could easily be declared Stephania tetrqndra-had either a lower con- detected by the TLC screening. This method allowed the ' tent of aristolochic acid I or was mixed with other herbs. detection of quantities of aristolochic acid I as gmall as Sample l7-found to contain only one-quarter of the 1 pg in visible light and 0.2 pe at 366 nm, through the amount of aristolochic acid I calculated for 16-is prob- presence ol its strongly oxidative NOz group. This ably a mixture of several herbs, as declared by the supplier. redction can be considered as very specific in the context Mass spectrometry was not only useful in the quali- of plant chemistry since only very few natural products tative bnd quantitative determination of aristolochic have been characterised to contain this type of func- acid I, but also suggested the presence of other aris- tional group (see Fig. 2). Two additional samples- tolochic acids in samples l0 and 21 due to characteristic samples 15 and 22-were also found to contain aris- UV spectra associated with MS ions at mf z 295, 324, tolochic acid I after LC/UV-DAD/MS analysis. The 342,354 and 358, speciûc for aristolochic acid deriva- advantages of this technique are the specificity and sen- tives (see Fig. 3 and Table 2). Such results are not sur- sitiüty of both UV and MS detection. Indeed, aris- prising since these two preparations were declared as tolochic acid I exhibited a typical IfV spectrum as well Asari herba, a genus belonging to the Aristolochiaceae as a strongly ionised characteristic fragment al mlz 295 family and well known to contain aristolochic acid (M-NO2)+ in the APCI-MS mode due to the loss of the derivatives (Hashimoto et al., 1999). labile NO2 functional group. The sensitivity of the Finally, four out the 42 analysed preparations were detection was in the ng range for both UV and MS found to contain aristolochic acid I and the presence of detection. These results oannot of course guarantee the aristolochic acid derivatives was suspected in two other herbal mixtures, Five out of the six incriminated pro- ducts were provided by the same Chinese herb importer. Immediate removal of the products from the Swiss Table 2 market was called for. Because of an increased interest Some important aristolochic acids and their expected ions for natural alternative medicine, consumers have to face Aristolochic acids Molecular weight Expected ions* a huge variety of new drug treatments and herbal several I, III 341 342, 324,295 health supplements to combat overweight. For, Ia, IIIa 327 328,310,281 reasons, including confusion of close vernacular Chi- II 311 312,294,265 nese names, lack of quality control, misuse in the drug ry, v, u, vlr 371 372,354,325 preparation, dosage or indication and also toxicity IVa, Va, VIa, VIIa 357 358,341,311 related to some Chinese traditional herbs, the use of ' Through protonation [M+H]*, loss ol Ofl [M-17]+ and NO2 such preparations is not without risk. The results of this [M-46]+ groups. work shôw the importance for a better survey of herbal F -'34 J.- R. Ioset et al, I Food and Chemical Toxicology 4 I ( 2003 ) 29-36

.I À E ,:.

N j

èô

o

àt àe o E c8 '6 H o (! È E t â .9 â o o o !Ë E o É o gE É o o rO rf) r !6 o t § È § o ôa .§ § ôo § N V1 o o o oo cO()OOo ,i E co (o r§ (\ E @ (o',t c! æ o o ctr c, E§

cË

dl- Eê ON d-' o\ !4 \o^ ovd o ËcJ §I Eé lür o U)\>ri è3 hoÉË JY o à> E- !v19â .9e -ÉNE+ .ES=El o9, o!Éû o 8d ü.9 -1 00000 ^5 =No@(osl>FF .o ts Nu èô.9 iiH * J.-R. Ioset et al. f Food and Chemical Toxicology 41 (2003) 29-i6 É Eo 8. eaEF ÊtEE "i - É a ha d Eï 19È É IlE :Ii i 9t iit ,à-9' ! dEI

-Ê ,au)

€!; o 6 o o o o

ad 'ô o oË o .É '6 a o t co 6 E Ëg Àt a! o @ .2 É o E' p É o À o E o U o -H Z6 zi a lo l- r) 6o U) Ill» N c, l9 (a d lÈ- 05 lJ) l() cd lo (oN §) È J! ii È t-Ir E € G t6 (E 0, É I IGls o É E o cd t- lÈ F co ,l Ê;§i -§§E .6 '- (Ë q'=§o \t--o cËO 8 PH F.L( tB"ë o(Eo o9l î- t*iào 'ô, -à:(nu (nÉtr utÉ à9 <6ÀÈ' âo

;ï lÈ

ni üe llo * P.05 : M.H. Rafamantanana, E. Rozet, G.E. Raoelison, K. Cheuk, S. Urverg Ratsimamanga, Ph. Hubert, J. Quetin-Leclercq. An improved HPLC-UV method for the simultaneous quantification of triterpenic glycosides and aglycones in leaves of Centella asiatica (L.) Urb (APIACEAE). Journal of Chromatography B 2009, 877, 2396-2402.

Journal of Chromatography B, 877 (2009) 2396–2402

Contents lists available at ScienceDirect

Journal of Chromatography B

journal homepage: www.elsevier.com/locate/chromb

An improved HPLC-UV method for the simultaneous quantification of triterpenic glycosides and aglycones in leaves of Centella asiatica (L.) Urb (APIACEAE)ଝ

M.H. Rafamantanana a,b,∗,E.Rozetc, G.E. Raoelison a, K. Cheuk a, S.U. Ratsimamanga a, Ph. Hubert c, J. Quetin-Leclercq b a IMRA (Institut Malgache de Recherches Appliquées), BP 3833, 101 Antananarivo, Madagascar b Laboratoire de Pharmacognosie, Unité CHAM, LDRI (Louvain Drug Research Institute), UCL (Université catholique de Louvain), Av. E. Mounier 72, 1200 Bruxelles, Belgium c Laboratoire de Chimie Analytique, Département de Pharmacie, CIRM, ULg (Université de Liège), CHU, B36, B-4000 Liège, Belgium article info abstract

Article history: The simultaneous quantification of madecassoside, asiaticoside, madecassic acid and asiatic acid in Received 13 October 2008 Centella asiatica by HPLC-UV is proposed. Asiaticoside was used as reference for the quantification of het- Accepted 16 March 2009 erosides and asiatic acid for aglycones. The evaluation of the extraction efficiency of the four molecules Available online 19 March 2009 led to use Soxhlet extraction for 8 h. The method was validated and was found to be accurate in the concentration range of 1.0–3.0 mg/ml for asiaticoside and 0.5–2.0 mg/ml for asiatic acid with CV <3% for Keywords: all investigated compounds. LOD and LOQ were, respectively, 0.0113 and 1.0 mg/ml for asiaticoside and Centella asiatica 0.0023 and 0.5 mg/ml for asiatic acid. This method was shown to be convenient for routine analysis of Madecassoside Asiaticoside samples of C. asiatica. Madecassic acid © 2009 Elsevier B.V. All rights reserved. Asiatic acid Method validation Quantification

1. Introduction Triterpenoids of C. asiatica (Fig. 1) are components of medici- nal drugs and are much used in cosmetic preparations for skin care Centella asiatica (APIACEAE) is an ethnomedicinal herbaceous [12,13]. Collected during all year long, considerable differences of species, originated from India which grows spontaneously in the triterpenoid contents were observed according to geographic subtropical regions: China, Malaysia, Australia, America, South regions, phenotype and genotype [14,15], so the assessment of a val- Africa and Madagascar. In Madagascar, the plant is largely used by idated analytical method is necessary. This will also help local pop- the local population and is the second medicinal species exported ulation to determine the best cultivating and harvesting conditions. [1]. C. asiatica is claimed to have a number of medicinal properties Most studies led on C. asiatica report the quantification of het- and is used in Ayurvedic medicine for the treatment of leprosy, erosides [16,17], acids [18] or acids and heterosides but with insuffi- skin tuberculosis, wound healing, stomach aches, arthritis, vari- ciently validated methods [14,15,19–21]. Several reported methods cose veins, high blood pressure and as a memory enhancer [2]. were based on HPLC differing only in the mobile phase composition Recently, several studies demonstrated that extracts of the plant or in the detection system. Xingyi et al. [16] proposed a method only possess antioxidant activity [3,4], have antiproliferative effects dedicated for the quantification of madecassoside and asiaticoside in tumor cells [5], improve venous wall alterations in chronic using acetonitrile/water (29/17, v/v) in isocratic mode. Quantifica- venous hypertension and protect the venous endothelium [6]. tion of only madecassoside, asiaticoside and its isomer has been Asiaticoside, one of its active molecules, is reported to cause reported by Zhang et al. [17] using ELSD detector. An addition of ␤- changes in gene expression and to induce type I collagen synthesis cyclodextrin in the mobile phase was used for the quantification of in human fibroblasts [7–9]. Madecassoside was reported to have the sole madecassic acid [18]. The existing methods for the simul- an anti-rheumatoid effect and wound healing properties [10,11]. taneous quantification of madecassoside, asiaticoside, madecassic acid and asiatic acid used either acetonitrile/water [19], acetoni- trile/water with TFA 0.1% [20] or acetonitrile/water each containing ଝ This paper is part of a special issue entitled “Method Validation, Comparison 0.05% of H3PO4 as mobile phase [21], together with detection at and Transfer”, guest edited by Serge Rudaz and Philippe Hubert. 205 nm. Most methods did not give good resolution or were not ∗ Corresponding author at: Laboratoire de Pharmacognosie, Unité CHAM, LDRI suitable for LC–MS and all were insufficiently validated [19–21]. The (Louvain Drug Research Institute), UCL (Université Catholique de Louvain), Av. E. monography of the European Pharmacopeia [22] quantifies total Mounier 72, 1200 Bruxelles, Belgium. E-mail address: [email protected] (M.H. Rafamantanana). triterpenoids (Fig. 2a) but our attempts to reproduce the separa-

1570-0232/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.jchromb.2009.03.018 M.H. Rafamantanana et al. / J. Chromatogr. B 877 (2009) 2396–2402 2397

Table 1 Gradient conditions for HPLC.

Time (Min) Pump A, water (%) Pump B, acetonitrile (%)

08020 15 65 35 30 35 65 35 20 80 40 20 80 45 80 20 55 80 20

Chromatographic separation was performed with a reversed phase RP-18 LiChroCART® column (250 mm × 4 mm I.D.; particle size: 5 ␮m). Mobile phase was a gradient of acetonitrile/water (Table 1), a flow rate of 1 ml/min and detection at 206 nm.

2.3. Standards solutions

Stock solutions of asiaticoside and asiatic acid were prepared in methanol at 5.0 and 2.5 mg/ml, respectively, and stored at 0 ◦C. Dilu- tion was done for each experiment. Three concentrations (m =3)of asiaticoside (0.5, 2.5 and 5.0 mg/ml) and of asiatic acid (0.25, 1.0 and 2.5 mg/ml) were used. Each concentration was analyzed two Fig. 1. Structure of triterpenes of Centella asiatica (Glu: glucose, Rha: rhamnose). times (n =2)for3days(k = 3). The extract solution was diluted with methanol (1:5, v/v) for tion did not allow us to quantify precisely these compounds. Only the preparation of the validation standards and spiked with three the peak corresponding to madecassoside was clearly identified, known concentrations of a stock mixture of asiaticoside and asiatic asiaticoside and the two aglycones eluted during the washing part acid. Each validation standard was analysed three times (n =3)for of the gradient and were not distinguished (Fig. 2b). This was also 3days(k = 3). observed by the European Pharmacopeia expert group in which the monography is under revision. 2.4. Evaluation of the extraction The main objective of this work is thus to improve the existing methods based on HPLC-UV and validate it for the simultaneous The extraction kinetic was established by evaluating the peak quantitation of madecassoside, asiaticoside and their aglycones, area (HPLC analysis) of each compound after 4, 6, 8 and 10 h (n = 3). madecassic and asiatic acids in C. asiatica. The most appropriate Soxhlet extraction time was determined using these data. 2. Experimental 2.5. Validation of the method 2.1. Chemicals and plant material All the reference compounds are commercially available but asi- Asiaticoside (99.2%, HPLC), madecassoside (97.94%, HPLC), asi- aticoside and asiatic acid were selected to achieve the validation of atic acid (99%, HPLC) and madecassic acid (95%, HPLC) were the method and to quantify madecassoside and madecassic acid, purchased from Extrasynthese (Genay, France). Acetonitrile and respectively, because we observed identical response factors for methanol HPLC grade were from Prolabo, VWR (Leuven, Bel- the two osides and the two aglycones in HPLC-UV (Table 2). Con- gium). sequently, choosing only two references reduces the cost of the Fresh leaves of C. asiatica (L.) Urban were collected in Decem- analysis. ber 2007 and January 2008 at the East and High Plateau regions As C. asiatica is a biological matrix, relatively large acceptance of Madagascar. Leaves were separated from stems, dried at 40 ◦C, limits are prescribed [23]. Validation of the method was done for powdered and sifted with a sieve of 355 ␮m meshes [22]. The pow- 3 days by testing the following criteria: response function, linear- dered leaves were stored at ambient temperature in obscurity and ity, trueness, precision (repeatability and intermediate precision), in a dry area. accuracy, limits of detection (LOD) and quantification (LOQ), and One gram dried powdered leaves [22] were extracted by Soxhlet quantification range. for 8 h with 100 ml of methanol. The extracts were evaporated to Statistical analyses of data were done using the e-noval V2.0 dryness under reduced pressure. The dried crude extract was dis- (Arlenda-Liège) software. solved in 10 ml of methanol, filtered through a 0.45 filter (Whatman, New Jersey, USA). 3. Results and discussions

2.2. Apparatus 3.1. Optimization of the extraction

The HPLC Waters 2690 separation module (Waters, Milford, MA, Several extraction modes were proposed such as maceration, USA) used consisted of a pump, an autoinjector, a UV spectrophoto- sonication or Soxhlet. Soxhlet extraction was selected as it is eas- metric detector Kromaton (Angers, France), all controlled by Borwin ier to control. As illustrated in Fig. 3, no significant difference was software (Borwin, Rostock, Germany). observed after 8 and 10 h of extraction. Consequently, an extraction For the determination of mass spectra, a LCQ Advantage Thermo time of 8 h was used for all experiments. This result corroborates Finnigan (Waltham, MA, USA) was used piloted by X-Calibur soft- the European Pharmacopeia findings indicating that the prob- ware. lem observed with the European Pharmacopeia method is due 2398 M.H. Rafamantanana et al. / J. Chromatogr. B 877 (2009) 2396–2402

Fig. 2. (a) Chromatogram of a crude extract of Centella asiatica [22] (1: solvent, 2: madecassoside (TR = 5.8), 3: asiaticoside (TR = 8.1), 4: madecassic acid (TR = 17.6), 5: asiatic acid (TR = 21.7)); (b) chromatogram of a crude extract of Centella asiatica obtained in our lab with European Pharmacopoeia method (1: madecassoside); (c) chromatogram of reference standards with the developed method (1: madecassoside and its isomer (asiaticoside B), 2: asiaticoside, 3: madecassic acid, 4: asiatic acid); (d) chromatogram of a crude extract of Centella asiatica with the developed method. M.H. Rafamantanana et al. / J. Chromatogr. B 877 (2009) 2396–2402 2399

Table 2 Peaks area ratio of the different compounds investigated with asiaticoside.

Compounds

Madecassoside Asiaticoside Madecassic acid Asiatic acid

Reference standards purity 97.94% 99.2% 95% 99% Peak area ratioa (n =2) 1± 0.5b 1.0 ± 0.2b 2.1 ± 1.3b 2.1 ± 0.3b

a Area ratio for each compound was calculated using the response of asiaticoside (major compound) as reference. b RSD (%).

to the chromatographic conditions and not an extraction failure Precision was evaluated in terms of standard deviation (SD, [22]. mg/ml) and relative standard deviation (RSD %) values for repeata- bility and intermediate precision [27,28]. As seen in Table 3, RSD (%) for repeatability and intermediate precision did not 3.2. Method validation exceed 4%. Accuracy allows to evaluate total error, the sum of systematic Selectivity and peak purity were analysed by the comparison and random errors of the tests results [24–28]. For asiaticoside and of retention times and mass spectra with reference compounds. asiatic acid, as illustrated in Fig. 5a and b, their respective accuracy Mass spectra were analysed at three levels (beginning, middle and profiles show that the relative upper and lower 95% ␤-expectation end) of each peak investigated and found to be comparable (Figs. tolerance limits are totally included inside the acceptance limits set 2c–d and 4a–d). Comparison of the chromatogram of an extract at ±20%. The method can thus be considered as accurate between of C. asiatica from European Pharmacopoeia method (Fig. 2a) and the chromatogram obtained with our method (Fig. 2d) shows its good resolution and the interest of this developed method for the quantification of aglycones. We also observed that madecasso- side and its isomer (asiaticoside B also called terminoloside) were slightly separated, but as both are considered to be active and are usually not separated; we added both areas and considered both peaks as one madecassoside peak to allow comparison with other results. Calibration standards of asiaticoside and asiatic acid were prepared without matrix (m =3,n = 2). Different regression models were tested such as: weighted (1/X) quadratic regression, weighted (1/X2) quadratic regression, quadratic regression, weighted (1/X) linear regression, weighted (1/X2) linear regression, linear regres- sion after logarithm transformation, linear regression after square root transformation, weighted (1/X) linear regression, linear regres- sion. Accuracy profiles were plotted to determine the most suitable regression model [24]. Fig. 5a and b shows the accuracy profiles obtained with the quadratic regression as response function for both standards. It was selected as the most adequate one as the 95% ␤-expectation tolerance intervals were totally included inside the ±20% acceptance limits for each concentration level of the val- idation standards for both analytes. Trueness [25,26] is expressed in relative bias (%) at each concen- tration level of the validation standards. Relative bias was less than 3% (Table 3) for asiaticoside and 10% for acid asiatic showing the excellent trueness of the method.

Fig. 4. Positive ion mode mass spectra obtained for peaks of the methanol + extract of Centella asiatica. (a) madecassoside [M + NH4] = 992; (b) asiatico- + + Fig. 3. Madecassoside, asiaticoside, madecassic and asiatic acids responses after side [M + NH4] = 976; (c) madecassic acid [M + NH4] = 522; (d) asiatic acid + different times of Soxhlet extraction of Centella asiatica (n = 3, RSD <1%). [M+NH4] =506. 2400 M.H. Rafamantanana et al. / J. Chromatogr. B 877 (2009) 2396–2402

1.0 and 3.0 mg/ml for asiaticoside and between 0.5 and 2.0 mg/ml imum total error of 20%. As shown in Fig. 5a and b, the LOQ for asiatic acid. are the smallest concentration levels of the validation stan- For asiaticoside and asiatic acid, LOD (the smallest quantity dards, i.e. 1.0 and 0.5 mg/ml for asiaticoside and asiatic acid, of the analyte that can be detectable in the sample, but not respectively. quantifiable) were 0.0113 and 0.0023 mg/ml, respectively. These Uncertainty of the measurement characterises the dispersion results were estimated using the mean intercept of the cali- of the values that could reasonably be attributed to the measurer bration model and the residual variance of the regression. The [29,30]. It was evaluated using expanded uncertainty where true LOQ (the smallest quantity quantifiable in the sample) were value can be observed with a confidence level at 95%. Table 4 determined with the accuracy profiles as they are the smallest shows that relative expanded uncertainties were less than 10% concentration levels where the 95% ␤-expectation tolerance lim- for asiaticoside and asiatic acid which means that the unknown its remain inside the ±20% acceptance limits [24–26]. In other true value is located at a maximum of ±10% around the measured words, they are the smallest concentration levels with a max- result.

Fig. 5. Accuracy profiles of asiaticoside (a) and asiatic acid (b) obtained with quadratic regression. The plain line is the relative bias, dashed lines are the ␤-expectation tolerance limit (ˇ = 95%) and dotted lines represent the acceptance limit (±20%). The dots represent the relative back-calculated concentrations of the validation standards and are plotted according to their targeted concentration. M.H. Rafamantanana et al. / J. Chromatogr. B 877 (2009) 2396–2402 2401

The linearity demonstrated the relationship between intro- duced and calculated concentration [25,26,28] using ␤-expectation tolerance interval approach. The concentrations of the validation standards were back-calculated in order to determine, by concen- tration level, the mean relative bias as well as the upper and lower ␤-expectation tolerance intervals. The acceptance limits were set at ±20%. In order to demonstrate method linearity, a regression line was fitted on the calculated concentrations of the validation stan- dards as a function of the introduced concentrations by applying a linear regression model. The equations obtained for asiaticoside and asiatic acid with their coefficient of determination are pre- sented in Table 3. The slopes values obtained for the two standards were, respec- tively 1.02, 1.023. Fig. 6a and b demonstrate the linearity of the results.

3.3. Application to samples of Centella asiatica

Three samples of C. asiatica were collected in the East and High Plateau regions of Madagascar. Extracts were analyzed and the results obtained by our method and the data described in the literature [14,15] are given in Table 5. Our results are in the same range than those obtained in the previous works. The heterosides are more abundant than aglycones and asiaticoside is often the major compound. Nevertheless, differences occur between samples and those analyzed here seemed less rich than the others, but we have to point out that all these other meth- ods were not validated so the reliability of these results needs caution. This proposed method gives low RSD (%) values and is reliable. These results stress the importance of a good quan- tification method to determine the best culture and harvesting conditions. A validated method for the quantification of asiaticoside, made- cassoside, asiatic and madecassic acids in drug samples of C. asiatica was developed. This method allows the simultaneous quantifica- tion of madecassoside, asiaticoside, madecassic acid and asiatic acid. Most of other validated methods only quantified some of these active molecules. Furthermore, the method described in European Pharmacopeia does not give good signal to noise ratio for the accu- Fig. 6. Linear profiles of asiaticoside (a) and asiatic acid (b). The continuous line rate quantification of aglycones and was not reproducible in our is identity line (y = x), the dotted lines are the upper and lower acceptance lim- laboratory. The method developed and validated was then success- ␤ its in absolute values and the dashed lines are the upper and lower -expectation fully applied to quantify these four compounds in different samples tolerance limits (ˇ = 95%). of C. asiatica collected in Madagascar.

Table 3 Validation results in crude extract of Centella asiatica.

Validation criteria Asiaticoside Asiatic acid

Response Quadratic regression Quadratic regression function Calibration range (3 points) Calibration range (3 points) 0.5–5 mg/ml 0.25–2.5 mg/ml Trueness Concentration (mg/ml) Relative bias (%) Concentration (mg/ml) Relative bias (%) 1 2.6 0.5l 9.7 2 0.7 1 −5.8 3 2.2 2 3 Precision Repeatability (SD mg/ml) Intermediate precision (RSD %) Repeatability (SD mg/ml) Intermediate precision (RSD %) 0.2313 2.9 0.2421 1.5 0.0987 2.5 0.1427 1.8 0.252 1.5 0.1997 3.3 Accuracy ␤-Expectation lower and ␤-Expectation lower and upper upper tolerance limits of the tolerance limits of the relative relative error (%) error (%) −11.5, 16.6 2.5, 16.9 −11.6, 13.0 −14.6, 3.1 −5.0, 9.3 −13.4, 19.3

Linearity Slope 1.02 1.023 Intercept −0.005007 −0.01017 r2 0.998 0.9912 2402 M.H. Rafamantanana et al. / J. Chromatogr. B 877 (2009) 2396–2402

Table 4 Estimates of the measurement uncertainties related to asiaticoside and asiatic acid, at each concentration level investigated in validation using quadratic regression model.

Asiaticoside Concentration Mean introduced Uncertainty of Uncertainty Expanded Uncertainty Relative expanded level (mg/ml) concentration (mg/ml) the bias (mg/ml) (mg/ml) (mg/ml) uncertainty (%)

1.0 1.000 0.01644 0.03294 0.06588 6.6 2.0 2.000 0.02860 0.05723 0.1145 5.7 3.0 3.000 0.02554 0.05146 0.1029 3.4

Asiatic acid Concentration Mean introduced Uncertainty of Uncertainty Expanded uncertainty Relative expanded level (mg/ml) concentration (mg/ml) the bias (mg/ml) (mg/ml) (mg/ml) uncertainty (%)

0.5 0.5000 0.004286 0.008628 0.01726 3.4 1.0 1.000 0.01031 0.02065 0.04131 4.1 2.0 2.000 0.03809 0.07625 0.1525 7.6

Table 5 Comparison of the content of triterpenes in Centella asiatica samples analysed by our method (n = 3) in % of the dry plant with results given in previous works [14,15].

Sample Madecassoside Asiaticoside Madecassic acid Asiatic acid

Sample 1 1.7 ± 0.04 2.0 ± 0.02 0.95 ± 0.039 0.98 ± 0.03 Sample 2 1.27 ± 0.013 1.63 ± 0.04

Mean ± RSD %. CA: Centella asiatica [14].

Acknowledgements [13] D.H. Park, J.S. Lee, G.S. Jung, K.R. Patent No. 2007111785 Rep Kor. Kongkae Taeho Kongbo (2007). [14] D. Randriamampionona, B. Diallo, F. Rakotonirina, C. Rabemanantsoa, K. Cheuk, We wish to thank the “Commission Universitaire pour le A.M. Corbisier, J. Mahillon, S. Ratsimamanga, M. El Jaziri, Fitoterapia 78 (2007) Développement” and the Catholic University of Belgium for finan- 7. cial support. We are grateful to Denis Randriamampionona, Dr [15] T.J. Jacinda, R. Meyer, I.A. Dubery, Plant Cell. Tissue Organ Culture 94 (2008) 1. [16] L. Xingyi, F. Lijuan, Zhongguo Yaoshi (Wuhan, China) 11 (2008) 11. Gabrielle Chataigné, Marie Christine Fayt, Jean Paul Vanhelputte [17] F.L. Zhang, Y.J. Wei, J. Zhu, Z.N. Gong, Biomed. Chromatogr. 22 (2008) 2. and Ramazan Colak for technical helps and assistance. A research [18] P. Jian, K. Guiqing, Y. Chuanxun, Z. Beibei, J. Risheng, Y. Yuan, J. Chromatogr. 25 grant from the Belgium National Fund for Scientific Research (F.R.S.- (2007) 3. FNRS) to E. Rozet is also gratefully acknowledged. [19] B.T. Schaneberg, J.R. Mikell, I.A. Khan, Pharmazie 58 (2003) 6. [20] P.K. Inamdar, R.D. Yeole, A.B. Ghogare, N. de Souza, J. chromatogr. A 742 (1996) 1. References [21] B. Günther, H. Wagner, Phytomedecine 3 (1996) 59. [22] European Pharmacopoeia, 6th ed., 2008 (http://www.edqm.eu). [23] Analytical Methods committee, Analyst 120 (1995) 2303. [1] G. Péchard, M. Antona, S. Aubert, D. Babin, Bois et Forêts des Tropiques 284 [24] E. Rozet, C. Hubert, A. Ceccato, W. Dewé, E. Ziemons, F. Moonen, K. Michail, R. (2005) 2. Wintersteiger, B. Streel, B. Boulanger, Ph. Hubert, J. Chromatogr. A 1158 (2007) [2] P. Boiteau, A.R. Ratsimamanga, Thérapie 1 (1956) 11. 126. [3] G. Jayashree, G.K. Muraleedhara, S. Sudarslal, V.B. Jacob, Fitoterapia 74 (2003) [25] E. Rozet, A. Ceccato, C. Hubert, E. Ziemons, R. Oprean, S. Rudaz, B. Boulanger, 5. Ph. Hubert, J. Chromatogr. A 1158 (2007) 111. [4] A. Gnanapragasam, K. Kuma Ebenezar, V. Sathish, P. Govindaraju, T. Devaki, Life [26] Ph. Hubert, J.J. Nguyen-Huu, B. Boulanger, E. Chapuzet, P. Chiap, N. Cohen, P.A. Sci. 76 (2004) 5. Compagnon, W. Dewe, M. Feinberg, M. Lallier, M. Laurentie, N. Mercier, G.M. [5] M. Yoshida, M. Fuchigami, T. Nagao, H. Okabe, K. Matsunaga, J. Takata, et al., Luzard, C. Nivet, L. valet, J. Pharm. Biomed. Anal. 36 (2004) 579. Biol. Pharm. Bull. 28 (2005) 1. [27] Guidance for industry: Bioanalytical Method Validation, US Department of [6] L. Incadela, M.R. Cesarone, M. Cacchio, M.T. De Sanctis, C. Santavenere, M.G. Health and Human Services, US Food and Drug Administration (FDA), Center D’Auro, M. Bucci, G. Belcaro, Angiology 52 (Suppl. 2) (2001) S9. for Drug Evaluation and Research (CDER), Center for Biologics Evaluation and [7] C.D. Coldren, P. Hashim, J.M. Ali, S.K. Oh, A.J. Sinskey, C. Rha, Planta Med. 69 Research (CBER), Rockville, May 2001. (2003) 8. [28] Ph. Hubert, J.J. Nguyen-Huu, B. Boulanger, E. Chapuzet, P. Chiap, N. Cohen, P.A. [8] L. Lu, K. Ying, S. Wei, Y. Liu, H. Lin, Y. Mao, Br. J. Dermatol. 151 (2004) 3. Compagnon, W. Dewe, M. Feinberg, M. Lallier, M. Laurentie, N. Mercier, G.M. [9] J. Lee, E. Jung, Y. Kim, J. Park, S. Hong, J. Kim, C. Hyun, S.Y. Kim, D. Park, Planta Luzard, L. Valat, E. Rozet, J. Pharm. Biomed. Anal. 45 (2007) 82. Med. 72 (2006) 4. [29] Eurachem/Citac Guide Quantifying the uncertainty in analytical measurement, [10] M. Liu, Y. Dai, X. Yao, Y. Li, Y. Xia, Z. Gong, Int. Immunopharmacol. 8 (2008) 11. 2nd ed., 2000. [11] M. Lui, Y. Dai, Y. Li, F. Huang, Z. Gong, Q. Meng, Planta Med. 74 (2008) 8. [30] CBEA-4/16, EA Guidelines on the Expression of Uncertainty in Quantitative [12] E.K. Park, S.H. Park, K.R. Patent No. 007051377. Rep. Kor. Kongkae Taeho Kongbo Testing, 2004, http://www.european-accreditation.org. (2007). P.06 : MH. Rafamantanana, B. Debrus, EG. Raoelison, E. Rozet, P. Lebrun, S. Ratsimamanga- Urverg, P. Hubert, J. Leclercq-Quetin. Application of design experiments and design space methodology for the HPLC-UV separation optimization of aporphine alkaloids from leaves of Spirospermum penduliflorum Thouars. Journal of Pharmaceuticals and Biomedical Analysis 2012, 62, 23 – 32.

Journal of Pharmaceutical and Biomedical Analysis 62 (2012) 23–32

Contents lists available at SciVerse ScienceDirect

Journal of Pharmaceutical and Biomedical Analysis

jou rnal homepage: www.elsevier.com/locate/jpba

Application of design of experiments and design space methodology for the

HPLC-UV separation optimization of aporphine alkaloids from leaves of

Spirospermum penduliflorum Thouars

a,b,∗ c b c,1 c

Mamy H. Rafamantanana , Benjamin Debrus , Guy E. Raoelison , Eric Rozet , Pierre Lebrun ,

b c a

Suzanne Uverg-Ratsimamanga , Philippe Hubert , Joëlle Quetin-Leclercq

a

Groupe de Recherche en Pharmacognosie, Louvain Drug Research Institute (LDRI), Université catholique de Louvain, Av. Mounier 7230, 1200 Bruxelles, Belgium

b

Institut Malgache de Recherches Appliquées (IMRA), BP 3833, Itaosy, Antananarivo 102, Madagascar

c

Laboratoire de Chimie Analytique, Département de Pharmacie, CIRM, Université de Liège, CHU, Av. de l’hôpital 1, B36, B-4000 Liège, Belgium

a r t i c l e i n f o a b s t r a c t

Article history: Spirospermum penduliflorum Thouars (Menispermaceae) is an endemic species of Madagascar tradition-

Received 14 October 2011

ally used as vasorelaxant. Recently, two aporphine alkaloids known to possess antihypertensive activity

Received in revised form

(dicentrine and neolitsine) were isolated and identified from the leaves of this plant. In the present study,

22 December 2011

a HPLC-UV method allowing the separation of all alkaloids and the quantification of dicentrine in the alka-

Accepted 23 December 2011

loidic extract of leaves was developed using design of experiments and design space methodology. Three

Available online 25 January 2012

common chromatographic parameters (i.e. the mobile phase pH, the initial proportion of methanol and

the gradient slope) were selected to construct a full factorial design of 36 experimental conditions. The

Keywords:

times at the beginning, the apex (i.e. the retention time) and the end of each peak were recorded and

Design of experiments

modelled by multiple linear equations. The corresponding residuals were normally distributed which

Spirospermum penduliflorum

Quantitative determination confirmed that the models can be used for the prediction of the retention times and to optimize the

Validation separation. The optimal separation was predicted at pH 3, with a gradient starting at 32% of methanol

Dicentrine and a gradient slope of 0.42%/min. Good agreement was obtained between predicted and experimental

chromatograms. The method was also validated using total error concept. Using the accuracy profile

approach, validation results gave a LOD and LOQ for dicentrine of 3 ␮g/ml and 10 ␮g/ml, respectively. A

relative standard deviation for intermediate precision lower than 10% was obtained. This method was

found to provide accurate results in the concentration range of 10–75 ␮g/ml of dicentrine and is suitable

for routine analysis.

© 2012 Elsevier B.V. All rights reserved.

1. Introduction plant is rich in alkaloids. The decoction of all parts is tradi-

tionally used as anticholinergic and vasorelaxant [1] and the

Nowadays, the traditional medicine still holds an important

decoction of leaves is also used for the treatment of malaria and

place to cure diseases in developing countries. It is mainly due to

as a chloroquine adjuvant. The dried leaves are also smoked for

the non accessibility of modern medicine by the local population.

pulmonary tuberculosis treatment. The decoction of roots was

The safety, effectiveness and quality of finished herbal medicinal

taken as cholagogue, tonic and for hepatic disorders [2]. Dif-

products depend on the quality of their source materials, which

ferent studies on root extracts of S. penduliflorum allowed the

can include hundreds of natural constituents, and how elements

isolation of some active molecules: columbine (clerodane-type

are handled through production processes.

diterpenoid), palmitine (protoberberine-type quaternary alkaloid)

Spirospermum penduliflorum Thouars (Menispermaceae) was

and limacine (bisbenzylisoquinoline alkaloid) [3]. Recently, Raoeli-

chosen among a list of medicinal plants used by Malagasy pop-

son et al. found a vasorelaxant activity on isolated rat aorta using

ulation because it is an endemic species of Madagascar. This

a methanol/dichloromethane leaves extract (under publication).

Two aporphine alkaloids were then isolated by bioguided frac-

tionation: dicentrine (Fig. 1) and neolitsine both known to possess

∗

Corresponding author at: Groupe de Recherche en Pharmacognosie, Louvain antihypertensive activities. In their experimental model, dicentrine

Drug Research Institute (LDRI), Université catholique de Louvain, Av. Mounier 7230, ± ␮

gave an EC50 value of 0.15 0.04 g/ml on rat aorta relaxation.

1200 Bruxelles, Belgium.

In the previous studies, Teng et al. also demonstrated that

E-mail address: [email protected] (M.H. Rafamantanana).

1 ␣

F.R.S.-FNRS Postdoctoral Researcher (Belgium). dicentrine is a strong vascular -1 adrenoceptor antagonist

0731-7085/$ – see front matter © 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.jpba.2011.12.028

24 M.H. Rafamantanana et al. / Journal of Pharmaceutical and Biomedical Analysis 62 (2012) 23–32

2.2. Plant material

Fresh leaves of S. penduliflorum were collected in December 2009

in the Eastern Region of Madagascar. A voucher specimen (AML

13) was deposited at the herbarium of the “Institut Malgache de

Recherches Appliquées” (IMRA). Leaves were dried at ambient tem-

perature and reduced to powder. In order to have a homogenous

powder, the dried powdered leaves were passed through a sieve of

710 ␮m meshes.

2.3. Apparatus

Chromatographic separations were performed with a reversed-

®

phase C18 LiChroCART column (250 mm × 4 mm i.d.; particle

size: 5 ␮m) (Merck, Darmstadt, Germany), a HPLC Waters 2690

separation module (Waters, Milford, MA, USA) coupled to a UV

spectrophotometric detector Kromaton (Angers, France); all con-

trolled by Borwin software (Borwin, Rostock, Germany). The

◦

column was maintained at 30 C. For the peaks identification, a LCQ

Fig. 1. Chemical structure of dicentrine.

Advantage Thermo Finnigan (Waltham, MA, USA) mass spectrom-

eter with ESI ion source was used, piloted by X-Calibur software.

Method validation experiments were performed using a Merck

Hitachi HPLC system consisting of a pump L6200, an automatic

having antiplatelet [4–6], and antiarrhythmic activities [7]. It con-

injector AS 2000 A, and a UV detector L400; all piloted by the Borwin

firms the traditional use of this plant as an antihypertensive. Lai

software.

et al. [8] and Tsai et al. [9] proposed two HPLC methods for the quan-

tification of dicentrine in rat plasma and urine. The first method was

not validated [8]. The second method [9] was fully validated but the 3. Experimental

separation of some peaks was not complete. Besides, the present

matrix was more complex than plasma and urine. Therefore, a HPLC 3.1. Plant extraction

method coupled with UV and MS detection was developed to sep-

arate and quantify these molecules and to control the quality and One gram of powdered leaves was macerated in a refluxing

◦

effectiveness of this plant. The optimization of chromatographic water bath at 50 C for 1 h with 50 ml of methanol–acetic acid 99%

methods for plant extracts is often intricate and can be time con- (99:1, v/v). The residue was washed with 10 ml of the same sol-

suming. In fact, it is a thorny problem to separate components vent and filtered through a filter paper (No 5, Whatman). These

because of the number of compounds, the similarities between the operations were repeated 4 times in order to have an extraction of

chromatographic behaviours of some of them while some others 4 h.

have widely distinct physico-chemical properties (e.g. polarities, The extracts were combined and evaporated to dryness under

pKa, Log P). Usually, optimization of a chromatographic method is reduced pressure. Dried crude extract was diluted in 50 ml of water

achieved by changing parameters one by one which is commonly acidified with 1% of acetic acid, filtered and washed twice with

called “one factor/variable at a time” methodology. On the contrary, 30 ml of ethoxyethane. Liquid-liquid extraction was performed

design of experiments (DoE) methodology investigates the factors four times with 50 ml of dichloromethane. Organic phases were

effects and the interactions between them by modifying multi- dried over anhydrous sodium acetate and evaporated to dryness.

ple factors at a time. Combined with design space (DS) [10–15], it The dried crude extract was diluted to 40 ml of methanol, filtered

␮

␮

leads to a powerful methodology (DoE-DS) allowing the identifica- through a 0.45 m filter (Whatman). Finally, 20 l of this solution

tion of optimal conditions for the achievement of robust analytical were injected into the HPLC system.

methods. In the present study, DoE-DS methodology was used to

develop a method for the separation of aporphin alkaloids con-

3.2. Experimental design

tained in leaves extract of S. penduliflorum. The method was then

fully validated using the total error concept and the accuracy pro-

Preliminary assays were performed to identify the criti-

file approach for the quantification of dicentrine in these leaves

cal factors (i.e. the factors having the highest effect on the

extracts [16,17].

responses) to be further used during optimization procedure.

These preliminary tests also allowed for the establishment of the

ranges of the factors. Different columns were therefore tested to

®

2. Materials and methods change chromatographic selectivity: a LiChroCART RP-18e col-

®

umn (250 mm × 4 mm i.d.; particle size: 5 ␮m), a LiChroCART

2.1. Chemicals and reagents RP-select B column (250 mm × 4 mm i.d.; particle size: 5 ␮m) and

Dicentrine reference standard was purchased from Sequoia

Table 1

Research Products Ltd. (Pangbourne, Ukraine). The reagents were

Chromatographic parameters studied for the method robust optimization. pci:

purchased as follows: methanol HPLC grade, ethoxyethane from

methanol proportion at the beginning of the gradient.

VWR (Leuven, Belgium), dichloromethane from Sigma Aldrich (St.

Parameter Levels

Louis, MO, USA), ammonia solution (25%) from J.T. Baker (Deven-

ter, Netherlands), acetic acid from Fisher scientific (Erembodegem, pH 2.5 4.5 7

Gradient slope (%/min) 0.4 0.6 1

Belgium), ammonium formate and anhydrous sodium sulphate

pci (%) 20 30 40 50

from Merck (Darmstadt, Germany).

M.H. Rafamantanana et al. / Journal of Pharmaceutical and Biomedical Analysis 62 (2012) 23–32 25

Table 2

Gradient conditions tested for the robust optimization of the HPLC method.

pci (%) Gradient slope (%/min)

0.4 0.6 1

Time (min) MeOH (%) Time (min) MeOH (%) Time (min) MeOH (%)

20 0 20 0 20 0 20

100 60 66.6 60 40 60

110 60 77 60 50 60

111 20 78 20 51 20

30 0 30 0 30 0 30

75 60 50 60 30 60

85 60 60 60 40 60

86 30 61 30 41 30

40 0 40 0 40 0 40

50 60 33.3 60 20 60

60 60 43 60 30 60

61 40 44 40 31 40

50 0 50 0 50 0 50

25 60 16.6 60 10 60

35 60 26 60 20 60

36 50 44 50 21 50

®

−

a Aluspher C8 column (250 mm × 4 mm i.d.; particle size: 3 ␮m) and wr = ktE ktR). These responses were modeled by multiple lin-

were tested. pH and gradient slopes were also varied. ear equations (see Eq. (1)).

The LiChroCART RP-select B column was found to be the most

k 2 2

= ˇ + ˇ · + ˇ · + ˇ · + ˇ · + ˇ ·

adequate for the chromatographic separation offering the best log( tRi) 0 1 pH 2 pH 3 pci 4 pci 5 slope

selectivity among the tested columns. + ˇ 2

6 · slope + ˇ7 · pH · pci + ˇ8 · pH · slope + ˇ9 · pci · slope

The method optimization involved a full factorial design set up

with 3 chromatographic factors: pH and gradient slope both at 3 + ˇ10 · pH · pci · slope + ε (1)

levels and methanol proportion at the beginning of the gradient

(pci) was studied with 4 levels (see Table 1). Table 2 summarizes where log(ktRi) represents the logarithm of the retention times (tR)

the gradient conditions which ensue from the DOE. All the peaks of the ith peak. Logarithms of peaks half-width were modelled by

are eluated at 60% of methanol, then 12 different gradient times the same equations. The ˇ1 to ˇ10 are the parameters of the math-

are defined using pci and gradient slope and use for 3 pH lev- ematic model, pH, pci and slope are the factors of the design of

els. Finally a total of 43 analyses were performed including the 36 experiments and ε is representative of the residual error assumed

conditions defined by the DoE. The conditions at pH 2.5, pci = 30%, normally distributed. Furthermore, a new separation criterion, S,

slope = 1%/min; pH 4.5, pci = 30%, slope = 0.6% and pH 4.5, pci = 50%, was used to ease the predictive error propagation from the mod-

slope = 0.4%/min were carried out in duplicate and 4 independent elled responses to this criterion [10–12,19]. S is defined as the

repetitions were performed at pH 4.5, pci = 40%, slope = 0.6%/min. difference between tB (the beginning) of the second peak and tE (the

These independent repetitions were carried out to estimate the end) of the first peak of the critical pair (i.e. the two most proximate

responses variability over the whole experimental domain: at peaks). The error affecting the modelled responses was estimated

some extreme points (i.e. pH 2.5, pci = 30%, slope = 1%/min and pH and propagated to S using Monte Carlo simulations and allowed to

4.5, pci = 50%, slope = 0.4%/min) and at the two points which are compute the DS. The DS corresponds to a region of the experimental

the closest to the DoE geometrical center (i.e. pH 4.5, pci = 30%, domain where the probability to attain S > 0 (i.e. baseline-resolved

slope = 0.6% and pH 4.5, pci = 40%, slope = 0.6%/min). peaks) is higher than a selected quality level [10–12]. The quality

level and the design space shapes are also representative of the

robustness of a method [20–23]. This is expressed in Eq. (2).

3.3. Design of experiments and design space methodology

DS

= {x0 ∈ |EÂ[P(S > )|Â] ≥ } (2)

DoE provides a structured methodology for the determination

of effects and the interactions between factors affecting a process

where x0 is a point in the experimental domain, .  is the thresh-

and the output response of that process [18]. The most common

old on the separation criterion S and  is the quality level. P and

objective of chromatographic methods development is to deter-

E correspond to the estimators of probability and mathematical

mine the operating conditions which allow obtaining an optimal

expectation, respectively.

separation of all compounds in the shortest analysis time. Fac-

tors which have the highest effects on the response (and therefore

on the separation) must be selected. Usually, the chromatographic 3.4. Chromatographic conditions

resolution (RS) between two adjacent peaks is modelled to assess

separation. Nevertheless, RS cannot be selected as the response The mobile phase consisted of methanol and ammonium for-

of multiple linear models due to discontinuities when selectivity mate buffer (20 mM). Experiments were carried out in gradient

changes and peaks interbred [10,19]. Hence, the beginnings, apex mode (see Table 2 for gradients). The pH of the aqueous solu-

and ends (tB, tR, tE, respectively) of each peak were measured and tion was adjusted to 2.5, 4.5 or 7 with concentrated formic acid

◦

the selected responses were the logarithm of the retention factor or ammonia. The column was thermostated at 30 C, flow rate

(log(ktR) with ktR = (tR − t0)/t0 and t0: the column dead time) and the was 1 ml/min and the detection was performed at a wavelength

− logarithm of the half-widths (log(wl) and log(wr) with wl = ktR ktB of 307 nm.

26 M.H. Rafamantanana et al. / Journal of Pharmaceutical and Biomedical Analysis 62 (2012) 23–32

Fig. 2. Extraction procedure performed to assess extraction efficiency.

Fig. 3. (a) Correlation between predicted and experimental responses of total alkaloids of Spirospermum penduliflorum. (b) Corresponding residual distributions.

M.H. Rafamantanana et al. / Journal of Pharmaceutical and Biomedical Analysis 62 (2012) 23–32 27

Table 3

R2

Adjusted coefficient of determination ( ) for the 3 modeled responses. adj

2 2

Compounds R Compounds R

adj adj

log(ktR) log(wl) log(wr) log(ktR) log(wl) log(wr)

1 0.9995 0.9305 0.8374 8 0.9996 0.9016 0.9816

2 0.9995 0.9278 0.9754 9 0.9996 0.9901 0.9592

3 0.9995 0.8926 0.9763 10 0.9997 0.7972 0.7215

4 0.9996 0.9828 0.6205 11 0.9995 0.7923 0.6341

5 0.9997 0.8521 0.8258 12 0.9995 0.7876 0.8241

6 0.9993 0.8202 0.7595 13 0.9998 0.8588 0.7886

7 0.9995 0.9641 0.9497 Mean 0.9995 0.8844 0.8349

3.5. Determination of extraction efficiency distribution in Fig. 3b highlights the need to perform the logarith-

mic transformation used to predict log(ktR). Indeed, the p-value

One gram of dried and powdered leaves was extracted as computed on the modeled response (log(ktR)) were acceptable

described in Fig. 2. The leaves were first extracted for 1 h. The (>0.05; see Table 4) for 10 of the 13 compounds suggesting that

residue was recuperated and extracted again with fresh solvent the models were suitable for the optimization of the separation.

for 1 h. This was repeated until the exhaustion of leaves in dicen- Then, once the responses are correctly modeled, the error can

trine. Each extract obtained after each extraction was evaporated be propagated from them to the separation criterion S. This step is

to dryness separately and fractionated. The alkaloidic fraction was carried out to estimate the prediction error affecting S and to give

injected separately onto HPLC column. The kinetics of the extrac- confidence in predicted optima. Thus, the predicted criterion was

tion was established by monitoring the peak area of dicentrine. computed over the whole experimental domain using a grid search

approach. Moreover, the probability to reach the criterion thresh-

3.6. Method validation old S > 0 min was also computed from 2000 Monte Carlo simulations

over each point of this grid [10–12]. Fig. 4 displays the probability

Selectivity, response function, trueness, precision (repeatabil- surfaces to at least reach a complete separation for each studied

ity and intermediate precision), accuracy, linearity, range, limits of parameters: P(S > 0 min).

detection (LOD) and quantification (LOQ) were the criteria deter- The optimal separation was predicted at pH 3.0 with a gradi-

mined to validate the method [20,24,25]. The validation of the ent starting at 32% of methanol and whose slope was 0.42%/min.

method was done during three days to evaluate time-dependant In Fig. 4, the DS is hedged in by a red line. The DS corresponds to a

intermediate precision. Data were analyzed with e-noval V3.0 soft- probability of 14% to get a complete separation. This quality level

ware (Arlenda, Liège, Belgium). may seem low, but in the light of the residuals range (between

−

2 and +2 min) all compounds should be separated at the opti-

mal operating conditions. This low quality level and the small DS

3.6.1. Calibration standards

size also suggest that the method will not be very robust. However,

Stock solution of 1 mg/ml of dicentrine was prepared in

as depicted in Fig. 5b, the chromatogram recorded at the above

methanol. This solution was diluted and four concentration lev-

␮ mentioned optimal operating condition allows the complete sep-

els (10, 25, 50 and 75 g/ml) were used to establish the calibration

aration of all the investigated compounds and is in accordance to

curve. Each level of the calibration standards was analyzed in dupli-

cate. the predicted chromatogram (Fig. 5a).

The use of DoE and DS allowed separating of 13 compounds

with UV and MS spectra indicating that they possess an aporphine

3.6.2. Validation standards

skeleton, among them dicentrine, a positional isomer of dicentrine

Validation standards were prepared within matrix. Dried extract

and neolitsine. Other compounds (isocorydine and glaucine) were

of leaves was diluted in 40 ml of methanol. This solution was spiked

identified by the comparison of their mass spectrum with literature

to reach final validation standards at 10, 25, 50, and 75 ␮g/ml of

and with our database [26].

dicentrine. Each level of the validation standards was analyzed

independently in triplicates.

4.2. Extraction efficiency

4. Results and discussion

The kinetic of the extracted amount of dicentrine (Fig. 6)

4.1. Experimental design

showed that total extraction was observed after four 1 h extractions

±

(area = 0). An extraction efficiency of 99.61 0.01% (n = 3) was then

The first step is to verify the validity of the model for each

obtained. For all experiments, four 1 h extractions using each time

response. In total, 13 peaks were identified as aporphine alkaloids

fresh solvent were applied.

in the total alkaloid extract. As depicted in Fig. 3a, the adequacy

between the experimental retention times and the predicted ones

R2 is good. The adjusted coefficients of determination ( ) are sum-

adj Table 4

Shapiro–Wilk normality test p-values for log(k ).

marized in Table 3. Furthermore, the corresponding residuals were tR

normally distributed for most compounds – for each response p-

Compounds p-Value Compounds p-Value

values were >0.05; (p-values for log(ktR) are presented in Table 4)

1 0.243 8 0.524

– and mainly located into the interval [−2,2] minutes (Fig. 3b).

2 0.028 9 0.446

Despite the high number of interactions added in the model 3 0.028 10 0.057

(see Eq. (1)), a curved tendency can be observed on the residu- 4 0.091 11 0.433

5 0.075 12 0.127

als in Fig. 3b. It suggests that some terms of higher orders (e.g.

3 6 0.004 13 0.502

pH ) could have been added in the model to avoid this slight lack

7 0.622

of fit situation. Additionally, the heteroscedasticity of the residuals

28 M.H. Rafamantanana et al. / Journal of Pharmaceutical and Biomedical Analysis 62 (2012) 23–32

Fig. 4. Probability surfaces to reach at least S > 0 min. (a) pH vs gradient slope (%/min), (b) pH vs pci (%) and (c) pci (%) vs gradient slope (%/min). The DS is encircled by the

red lines. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)

Fig. 5. (a) Predicted chromatogram at optimal conditions (see text for details), (b) experimental chromatogram recorded at optimal conditions of total alkaloids extract

of Spirospermum penduliflorum. Peak numbering: 1 = unidentified peak, 2 = isocorydine, 3–6 = unidentified peaks, 7 = glaucine, 8 = dicentrine, 9–11 = unidentified peaks,

12 = neolitsine, 13 = unidentified peak.

4.3. Method validation extracts of S. penduliflorum. The assessed validation criteria were

method selectivity, trueness, precision, linearity, range, LOQ and

The method validation focused on the demonstration of the abil- LOD as well as the accuracy of the generated results. The selection

ity of the optimized method to accurately quantify dicentrine in of the most adequate response function was also performed during

the method validation phase [16,27]. To achieve this, the accuracy

profile methodology was endorsed which allows to decide about

the validity of the method. This accuracy profile is based on the

predictive quality of future results that will be generated by the

method under trial [28,29]. In this framework two values have to

be defined a priori. The first one is the acceptance limits. They rep-

resent the maximum relative error that is acceptable for the future

application of the method being validated. These acceptance limits

were set at ±30% as commonly done for plant matrices [25,30,31].

The second one is the guarantee or minimum probability to have

future results included within these limits. It was set at 90%.

4.3.1. Selectivity

The selectivity of the method was insured by the comparison of

the retention time and mass spectra with the reference standard

of dicentrine. Mass spectra at the beginning, the middle and the

end of the chromatographic peak of dicentrine were recorded and

Fig. 6. Dicentrine peak area responses with increasing times of extraction obtained

found to be comparable (Fig. 7).

with the same powdered leaves of Spirospermum penduliflorum (n = 3).

M.H. Rafamantanana et al. / Journal of Pharmaceutical and Biomedical Analysis 62 (2012) 23–32 29

Fig. 7. Positive ion mode mass spectra of standard dicentrine (a), and in total alkaloid extract (b) of Spirospermum penduliflorum [M+H = 340].

30 M.H. Rafamantanana et al. / Journal of Pharmaceutical and Biomedical Analysis 62 (2012) 23–32

Table 5

Validation results of the HPLC-UV method for the determination of dicentrine in leaves extract of Spirospermum penduliflorum (3 series, 3 repetitions).

Validation criteria Dicentrine

2

Response function Weighted (1/X ) linear regression

Calibration range (4 levels)

10–75 ␮g/ml

Trueness Concentration (␮g/ml) Relative bias (%)

10 −9.4

25 −5.6

50 −2.8

75 −1.4

␮

Precision Concentration ( g/ml) Repeatability (RSD%) Intermediate precision (RSD%)

10 9.1 9.6

25 5.6 5.9

50 2.8 7.9

75 1.7 2.8

Accuracy Concentration (␮g/ml) 90% ˇ-expectation lower and upper

tolerance limits of the relative error (%)

10 −28.6, 9.9

25 −17.5, 6.3

50 −26.5, 20.9

75 −8.3, 5.6

Linearity

Slope 0.9995

Intercept −1.176 r2 0.991

2

4.3.2. Response function weighted 1/X quadratic regression and weighted 1/X quadratic

Calibration standards without matrix were prepared at four regression.

2

concentration levels of dicentrine. Analyses were performed in According to accuracy index [17], the weighted (1/X ) linear

duplicate. Validation standards were prepared independently with regression model was the most adequate calibration model. Indeed,

the matrix and also contained known concentrations of dicen- the 90% ˇ-expectation tolerance intervals are included inside the

±

trine at the same four concentration levels. They were analyzed in 30% acceptance limits within the range of 10–75 ␮g/ml of dicen-

triplicate. These validation standard concentration levels were con- trine [16,27,29]. These tolerance intervals are intervals one can

sidered as conventionally true values. The concentration for both claim to include each future result with 90% probability, thus giving

␮

calibration and validation standards ranged from 10 to 75 g/ml. a high level of guarantee about the suitability of the method for its

Different regression models were fitted on the calibration stan- intended purpose [27,29].

dards such as: simple linear regression, weighted (1/X) linear

2

regression, weighted (1/X ) linear regression, linear regression

4.3.3. Trueness

through 0 fitted with the highest calibration level only, linear

The systematic error between the accepted true values of the

regression after logarithmic transformation, quadratic regression,

validation standards and the mean experimental one represents

2

Fig. 8. Accuracy profile of dicentrine obtained with a weighted (1/X ) linear regression model as calibration curve. The plain line is the relative bias (%), the dashed lines are

the ˇ-expectation tolerance limits (ˇ = 90%) and the dotted lines represent the acceptance limits (±30%).

M.H. Rafamantanana et al. / Journal of Pharmaceutical and Biomedical Analysis 62 (2012) 23–32 31

± ␮

Fig. 9. Linearity profile of dicentrine. The continuous line is the identity line y = x, the dotted lines are the 30% acceptance limits expressed in concentration units ( g/ml

of dicentrine) and the dashed lines are the upper and lower ˇ-expectation tolerance limits (ˇ = 90%) also expressed in ␮g/ml of dicentrine.

the trueness of the method [16,29]. It can be expressed in relative limits [16,20,29]. Consequently, the range over which the method

− ␮

bias which did not exceed 10% (see Table 5) showing the adequate is valid extends from 10 to 75 g/ml of dicentrine as shown in Fig. 8.

trueness of the method.

4.3.7. Linearity

4.3.4. Precision The linearity of an analytical method is the ability within a

Precision represents the variation of the results of dicentrine definite range to obtain results directly proportional to the concen-

in intraday (or repeatability) and interday conditions (intermedi- tration of the analyte in the sample [16,20,29]. Good relationship

ate precision) [20,24,29]. Precision was assessed during three days was shown between introduced and back calculated concentration.

(p = 3) with three independent repetitions (n = 3) at each of the four A regression line fitted on the back-calculated results obtained with

concentration levels of the validation standards, leading to 9 results the selected calibration curve and the concentration levels of the

per level of dicentrine and a total amount of 36 results. It was validation standards led to the equation:

evaluated in terms of relative standard deviation (RSD%) values.

Y

= − . + . X.

These values did not exceed 10% for repeatability as well as for 1 176 0 9995

intermediate precision.

The linearity of the results generated by the method is demon-

ˇ

strated in Fig. 9 showing that the 90% -expectation tolerance limits

4.3.5. Accuracy

expressed in concentration units are totally included between the

Accuracy refers to the closeness of agreement between the test ±

30% acceptance limits.

result and the accepted reference value, namely the convention-

ally true value [20,24,29]. Accuracy takes into account the sum of

4.4. Sample analyses

systematic and random errors giving the total error [29]. The accep-

tance limits have been set at ±30% according to the complexity of

Two samples of leaves of S. penduliflorum collected in the same

the matrix (plant extract) [25,30,31].

region but in different period were analyzed 3 times. Sample 1 was

Fig. 8 shows that the 90% ˇ-expectation upper and lower tol-

collected in March 2007 while the second one in December 2009.

erance limits are totally included inside the acceptance limits for

The amounts of dicentrine obtained using the newly developed and

the whole concentration range of 10–75 ␮g/ml of dicentrine when

2 validated method were respectively 1.42 ± 0.06% (m/m) for S1 and

a weighted (1/X ) linear regression model is used as calibration

1.015 ± 0.007% for S2.

curve. This demonstrates the accuracy of the results obtained by the

developed method [16,27,28]. This dicentrine concentration inter-

val is very small because of the closeness of the retention times of 5. Conclusion

dicentrine and of its isomer. A high concentration level of dicen-

trine may cause a coelution with its isomer, impairing its accurate A novel analytical method for the quantification of dicentrine

quantification. in leaves of S. penduliflorum was developed using DoE and DS

methodology. This methodology allowed the separation of dicen-

4.3.6. LOD, LOQ and range trine and its positional isomer without using specific column or

Limits of detection (LOD) and quantification (LOQ) were mobile phase. 11 other compounds were also separated and some

␮ ␮

3.0 g/ml and 10.0 g/ml respectively. The LOD was estimated of them were identified by MS detection, showing the advantage

using the mean intercept of the calibration model and the resid- of hyphenated techniques for peak identification and to confirm

ual variance of the regression. For the LOQ it was obtained from the the selectivity of the analytical method. The encountered opti-

accuracy profile (Fig. 8) as the smallest concentration where the 90% mal chromatographic method was not very robust but peaks

ˇ ±

-expectation tolerance limits remain inside the 30% acceptance were sufficiently separated. This method was then successfully

32 M.H. Rafamantanana et al. / Journal of Pharmaceutical and Biomedical Analysis 62 (2012) 23–32

validated using the accuracy profile methodology within the range [12] B. Debrus, P. Lebrun, A. Ceccato, G. Caliaro, B. Govaerts, B.A. Olsen, E. Rozet,

B. Boulanger, P. Hubert, A new statistical method for the automated detection

of 10–75 ␮g/ml of dicentrine.

of peaks in UV-DAD chromatograms of a sample mixture, Talanta 79 (2009)

This analytical method is suitable for routine analyses and was

77–85, http://hdl.handle.net/2268/13236.

used for the quantification of dicentrine, the major vasorelaxing [13] B. Debrus, P. Lebrun, E. Rozet, I. Nistor, A. Ceccato, G. Caliaro, R. Oprean, B.

Boulanger, P. Hubert, Nouvelle méthodologie pour le développement automa-

compound, in two plant samples, giving results of 1.42 and 1.015%.

tisé de méthodes analytiques en chromatographie liquide pour l’analyse

These preliminary results show the variability of the active princi-

de mélanges de composés inconnus, Spectra Analyse 268 (2009) 28–33,

ple content in plants according to their harvesting season. It also http://hdl.handle.net/2268/26406.

[14] I. Molnár, H.-J. Rieger, K.E. Monks, Aspects of the Design Space in high pressure

shows the necessity of a quantification of dicentrine to standard-

liquid chromatography method development, J. Chromatogr. A 1217 (2010)

ize the activity of the plant extract, and then of the development of

3193–3200.

such a validated method. Moreover, the developed method allowed [15] J.J. Peterson, A posterior predictive approach to multiple response surface opti-

the separation of several alkaloids and molecules whose structures mization, J. Qual. Technol. 36 (2004) 139.

[16] P. Hubert, J.-J. Nguyen-Huu, B. Boulanger, E. Chapuzet, N. Cohen, P.-A. Com-

were analyzed by MS. Nevertheless, MS did not allow to determine

pagnon, W. Dewé, M. Feinberg, M. Laurentie, N. Mercier, G. Muzard, L. Valat,

the stereochemistry and to distinguish between isomers, but the

E. Rozet, Harmonization of strategies for the validation of quantitative analyt-

analytical method can also be the basis of preparative purifications ical procedures A SFSTP proposal. Part IV, examples of application, J. Pharm.

Biomed. Anal. 48 (2008) 760–771, http://hdl.handle.net/2268/6171.

of crude extract of S. penduliflorum to provide enough pure com-

[17] E. Rozet, V. Wascotte, N. Lecouturier, V. Préat, W. Dewé, B. Boulanger, P.

pounds for NMR analysis. The determination of the structures and

Hubert, Improvement of the decision efficiency of the accuracy profile by

stereochemistries of these compounds could be finally envisaged. means of a desirability function for analytical methods validation: applica-

tion to a diacetyl-monoxime colorimetric assay used for the determination of

urea in transdermal iontophoretic extracts, Anal. Chim. Acta 591 (2007) 239,

Acknowledgments http://hdl.handle.net/2268/6022.

[18] International Conference on Harmonization (ICH) of Guidance for Industry,

Pharmaceutical Development, Topic Q8 (R2), 2009.

We are thankful to the «Coopération Universitaire pour le

[19] W. Dewé, R.D. Marini, P. Chiap, P. Hubert, Guidance for robustness: ruggedness

Développement» and «l’Université Catholique de Louvain» for the

tests in method validation J. Crommen, B. Boulanger, Chemom. Intell. Lab. Syst.

financial supports. We acknowledge Dr Gabrielle Chataigné, Marie 74 (2004) 263–268, http://hdl.handle.net/2268/6036.

[20] International Conference on Harmonization (ICH) of Technical Requirements

Christine Fayt, Jean Paul Vanhelpute, Ramazan Colak, Aubin Ran-

for registration of Pharmaceuticals for Human Use, Topic Q2 (R1): Validation

drianirina for technician assistance and Benja Rakotonirina for

of Analytical Procedures: Text and Methodology, Geneva, 2005.

botanical identification. [21] Y. Vander Heyden, A. Nijhuis, J. Smeyers-Verbeke, B.G.M. Vandeginste, D.L. Mas-

sart, Guidance for robustness: ruggedness tests in method validation, J. Pharm.

The authors would like to thank the Walloon Region of Belgium

Biomed. Anal. 24 (2001) 723–753.

for the PPP convention funds No 917007 and Arlenda SA for partial

[22] B. Dejaegher, Y. Vander Heyden, Ruggedness and robustness testing, J. Chro-

funding of Optimal-DS project. A research grant from the Belgium matogr. A 1158 (2007) 138–157.

National Fund for Scientific Research (F.R.S.-FNRS) to E. Rozet is also [23] R.D. Marini, E. Rozet, Y. Vander Heyden, E. Ziemons, B. Boulanger, A. Bouklouze,

A.-C. Servais, M. Fillet, J. Crommen, P. Hubert, Robustness testing of a chiral

gratefully acknowledged.

NACE method for R-timolol determination in S-timolol maleate and uncertainty

assessment from quantitative data, J. Pharm. Biomed. Anal. 44 (2007) 640–651,

References http://hdl.handle.net/2268/21170.

[24] Guidance for industry: Bioanalytical Method Validation, US Department of

Health and Human Services, US Food and Drug Administration (FDA), Center

[1] A. Zafera Rabesa, Pharmacopée de L’Alaotra, Antananarivo, 1986.

for Drug Evaluation and Research (CDER), Center for Biologics Evaluation and

[2] G.H. Schmelzer, A. Gurib-Fakim Fondation, Ressources végétales de l’Afrique

Research (CBER), Rockville, May 2001.

tropicale 11(1) Plantes médicinales 1, Pays-Bas, 2008.

[25] C.T. Viswanathan, S. Bansal, B. Booth, A.J. DeStefano, M.J. Rose, J. Sailstad,

[3] P. Rasoanaivo, S.U. Ratsimamanga, A. Rakoto-Ratsimamanga, Isoquinoline alka-

V.P. Shah, J.P. Skelly, P.G. Swann, R. Weiner, Workshop/conference report—

loid constituents of Spirospermum penduliflorum and Strychnopsis thouarsii

quantitative bioanalytical methods validation and implementation: best prac-

(Menispermaceae), Biochem. Syst. Ecol. 23 (1995) 679–680.

tices for chromatographic and ligand binding assays, AAPS J. 9 (2007) E30–E42.

[4] C.-M. Teng, S.-M. Yu, F.-N. Ko, C.C. Chen, Y.-L. Huang, T.-F. Huang, Dicentrine,

[26] C. Stevigny, J.-L. Habib Jiwan, R. Rozenberg, E. de Hoffmann, J.Q. Leclercq, Key

␣

a natural vascular 1-adrenoreceptor antagonist isolated from Lindera mega-

fragmentation patterns of aporphine alkaloids by electrospray ionization with

phylla, Br. J. Pharmacol. 104 (1991) 651–656.

multistage mass spectrometry, Rapid Commun. Mass Spectrom. 18 (2004)

[5] S.-M. Yu, S.-Y. Hsu, F.-N. Ko, C.-C. Chen, Y.-L. Huang, T.-F. Huang, C.-M. Teng,

␣ 523–528.

Haemodynamic effects of dicentrine, a novel 1-adrenoceptor antagonist:

[27] A. Bouabidi, E. Rozet, M. Fillet, E. Ziemons, E. Chapuzet, B. Mertens, R. Klinken-

comparison with prazosin in spontaneously hypertensive and normotensive

berg, A. Ceccato, M. Talbi, B. Streel, A. Bouklouze, B. Boulanger, P. Hubert,

Wistar-Kyoto rats, Br. J. Pharmacol. 106 (1992) 797–801.

Critical analysis of several analytical method validation strategies in the frame-

[6] M.R. Mustafa, F.I. Achike, Dicentrine is preferentially antagonistic to rat aortic

work of the fit for purpose concept, J. Chromatogr. A 1217 (2010) 3180–3192,

than splenic alpha 1-adrenoceptor stimulation, Acta Pharmacol. Sin. 21 (2000) http://hdl.handle.net/2268/29471.

1165–1168.

[28] B. Boulanger, E. Rozet, F. Moonen, S. Rudaz, P. Hubert, A risk-based anal-

[7] M.-L. Young, M.-J. Su, M.-H. Wu, C.-C. Chen, The electrophysiological effects

ysis of the AAPS conference report on quantitative bioanalytical methods

of dicentrine on the conduction system of rabbit heart, Br. J. Pharmacol. 113

validation and implementation, J. Chromatogr. B 877 (2009) 2235–2243,

(1994) 69–76.

http://hdl.handle.net/2268/18777.

[8] Y.-C. Lai, T.-F. Kuo, C.-K. Chen and S.-S. Lee, Dicentrine metabolites char-

[29] E. Rozet, A. Ceccato, C. Hubert, E. Ziemons, R. Oprean, S. Rudaz, B.

acterized in miniature pig urine by HPLC-SPE-NMR, http://gra103.aca.ntu.

Boulanger, P. Hubert, Analysis of recent pharmaceutical regulatory docu-

edu.tw/gdoc/97/D95423002a.pdf.

ments on analytical method validation, J. Chromatogr. A 1158 (2007) 111–125,

[9] T.-H. Tsai, T.-R. Tsai, C.-J. Chou, C.-F. Chen, Determination of dicentrine in rat

http://hdl.handle.net/2268/3646.

plasma by high performance liquid chromatography and its application to

[30] M.H. Rafamantanana, E. Rozet, G.E. Raoelison, K. Cheuk, S.U. Ratsimamanga,

pharmacokinetics, J. Chromatogr. B 681 (1996) 277–281.

P. Hubert, J. Quetin-Leclercq, An improved HPLC-UV method for the simul-

[10] P. Lebrun, B. Govaerts, B. Debrus, A. Ceccato, G. Caliaro, P. Hubert, B. Boulanger,

taneous quantification of triterpenic glycosides and aglycones in leaves of

Development of a new predictive modelling technique to find with confidence

Centella asiatica (L.) Urb (APIACEAE), J. Chromatogr. B 877 (2009) 2396–2402,

equivalence zone and design space of chromatographic analytical methods,

http://hdl.handle.net/2268/18776.

Chemom. Intell. Lab. Syst. 91 (2008) 4–16, http://hdl.handle.net/2268/1640.

[31] D.S.S. Kpoviessi, F. Gbaguidi, J. Gbenou, G. Accrombessi, M. Moudachirou, E.

[11] B. Debrus, P. Lebrun, A. Ceccato, G. Caliaro, E. Rozet, I. Nistor, R. Oprean, F.J.

Rozet, P. Hubert, J. Quetin-Leclercq, Validation of a method for the determi-

Rupérez, C. Barbas, B. Boulanger, P. Hubert, Application of new methodologies

nation of sterols and triterpènes in the aerial part of Justicia anselliana (Nees)

based on design of experiments, independent component analysis and design

T. Anders by capillary gas chromatography, J. Pharm. Biomed. Anal. 48 (2008)

space for robust optimization in liquid chromatography, Anal. Chim. Acta 691

1127–1135, http://hdl.handle.net/2268/18778.

(2011) 33–42, http://hdl.handle.net/2268/88084. P.07 : S. Afoulous, H. Ferhout, EG. Raoelison, A. Valentin, B. Moukarzel, F. Couderc, J. Bouajila. Chemical composition and anticancer, antiinflammatory, antioxidant and antimalarial activities of leaves essential oil of Cedrelopsis grevei. Food and Chemical Toxicology 2013, 56, 352- 362.

This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier’s archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/authorsrights Author's personal copy

Food and Chemical Toxicology 56 (2013) 352–362

Contents lists available at SciVerse ScienceDirect

Food and Chemical Toxicology

journal homepage: www.elsevier.com/locate/foodchemtox

Chemical composition and anticancer, antiinflammatory, antioxidant and antimalarial activities of leaves essential oil of Cedrelopsis grevei

Samia Afoulous a, Hicham Ferhout b, Emmanuel Guy Raoelison c, Alexis Valentin d, Béatrice Moukarzel d, ⇑ François Couderc a, Jalloul Bouajila a, a Université de Toulouse, Laboratoire des Interactions Moléculaires et Réactivité Chimique et Photochimique, UMR CNRS 5623, Université Paul-Sabatier, 118 route de Narbonne, F-31062 Toulouse, France b Nat’Ex Biotech. Bat 7, 55 avenue Louis Breguet, 31400 Toulouse, France c Laboratoire de Phytochimie et Standardisation, IMRA, BP 3833, Antananarivo 101, Madagascar d Université de Toulouse, UMR 152 IRD-UPS, Pharma-DEV, Université Paul Sabatier Toulouse 3, Faculté de Pharmacie, 35 chemin des maraîchers, 31062 Toulouse Cedex 9, France article info abstract

Article history: The essential oil from Cedrelopsis grevei leaves, an aromatic and medicinal plant from Madagascar, is Received 19 December 2012 widely used in folk medicine. Essential oil was characterized by GC–MS and quantified by GC–FID. Accepted 7 February 2013 Sixty-four components were identified. The major constituents were: (E)-b-farnesene (27.61%), Available online 28 February 2013 d-cadinene (14.48%), a-copaene (7.65%) and b-elemene (6.96%). The essential oil contained a complex mixture consisting mainly sesquiterpene hydrocarbons (83.42%) and generally sesquiterpenes Keywords: (98.91%). The essential oil was tested cytotoxic (on human breast cancer cells MCF-7), antimalarial Cedrelopsis grevei (Plasmodium falciparum), antiinflammatory and antioxidant (ABTS and DPPH assays) activities. C. grevei Leaves essential oil essential oil was active against MCF-7 cell lines (IC = 21.5 mg/L), against P. falciparum, (IC = 17.5 - Chemical composition 50 50 Biological activities mg/L) and antiinflammatory (IC50 = 21.33 mg/L). The essential oil exhibited poor antioxidant activity Correlations against DPPH (IC50 > 1000 mg/L) and ABTS (IC50 = 110 mg/L) assays. A bibliographical review was carried out of all essential oils identified and tested with respect to antiplasmodial, anticancer and antiinflamma- tory activities. The aim was to establish correlations between the identified compounds and their biological activities (antiplasmodial, anticancer and antiinflammatory). According to the obtained corre- lations, 1,4-cadinadiene (R2 = 0.61) presented a higher relationship with antimalarial activity. However, only (Z)-b-farnesene (R2 = 0.73) showed a significant correlation for anticancer activity. Ó 2013 Elsevier Ltd. All rights reserved.

1. Introduction Cedrelopsis was used in traditional medicine to treat malaria, fe- ver and fatigue (Mulholland et al., 1999). C. grevei trunk bark has Cedrelopsis grevei is endemic species to Madagascar. This tree of wide-ranging empirical uses such as to relieve muscle fatigue the vernacular, named ‘katafa’ or ‘katrafay’ (katra-bitter, fay-juice), and reduce capillary fragility, as an ingredient of a cough syrup may reach 15 m in height, it has branches in gray bark and fragrant against persistent catarrh and also as febrifuge and antihyperten- (Smell of cedar) (Cabanis et al., 1969; Courchet, 1906). Flowering is sive beverages (Um et al., 2003). The bark essential oil is also used from September to December (after the first rains) and followed by to cure rheumatism and muscular pains and is known to exert fruiting from October to December (Rasoanaivo and De La Gorce, antifungal and antibiotic activities. All these popular uses may be 1998). The panniculus is the highly branched inflorescence bearing explained by the presence of biologically active volatile constitu- a numerous small yellow flowers polygamous giving each a brown ents. The widespread use of C. grevei in traditional medicine stim- capsule containing five valves fruit dry. This species is mainly pres- ulated us to explore its potential biological activity. ent in dense forests or dry bush (South and West Island), at 900 m To the best of our knowledge, no previous study of the antican- above sea level (Rasoanaivo and De La Gorce, 1998), on different cer, antimalarial, antiinflammatory and antioxidant activities of soil types: silica, limestone, sand, etc. The wood is rot resistant in- the essential oil of C. grevei have been reported. We report here sect and it is used in the production of royal tombs (Sakalava). In the chemical composition of the leaves essential oil of C. grevei addition, Cedrelopsis produces hardwood and sought for the and its anticancer, antimalarial antiinflammatory and antioxidant construction of houses and traditional cabinetry. activities. Moreover, we reviewed bibliographical of all essential oils having an activity against P. falciparum, MCF-7 cell line and ⇑ Corresponding author. Tel.: +33 562256825; fax: +33 562256826. 5-Lipoxygenase in order to identify, by correlation, the main active E-mail address: [email protected] (J. Bouajila). compounds.

0278-6915/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.fct.2013.02.008 Author's personal copy

S. Afoulous et al. / Food and Chemical Toxicology 56 (2013) 352–362 353

2. Experimental part 2.4. Antioxidant activity

2.1. Extraction of the essential oil 2.4.1. Free radical scavenging activity: DPPH test Antioxidant activity was studied using 1,1-diphenyl-2-picrylhydrazyl free rad- The leaves of C. grevei were collected in Antananarivo, Madagascar (June 2008). ical (DPPH) as described by Blois (1958) with some modifications. 1.5 mL of various The leaves are dried outdoors in the shade. Steam-distillation was used to extract dilutions of essential oil (EO) was mixed with 1.5 mL of a 0.2 mmol/L methanolic the essential oil according to the protocol of the European Pharmacopeia (European DPPH solution. After an incubation period of 30 min at 25 °C, the absorbances at

Pharmacopoeia, 1983). The essential oil was dried by anhydrous sodium sulfate, fil- 520 nm (the wavelength of maximum absorbance of DPPH) were recorded as A(sam- tered and stored in sealed vials at 4 °C, prior to analyses. ple), using a Helios spectrophotometer (Unicam, Cambridge, UK). A blank experi- ment was also carried out applying the same procedure to a solution without the

test material and the absorbance was recorded as A(blank). The free radical-scaveng- 2.2. Chemicals used ing activity of each solution was then calculated as percent inhibition according to the following equation: All chemicals used were of analytical reagent grade. All reagents were pur- chased from Sigma–Aldrich–Fluka (Saint-Quentin France). % inhibition ¼ 100ðAðblankÞ AðsampleÞÞ=AðblankÞ

Antioxidant activity of the essential oil was expressed as IC50, defined as the 2.3. Gas chromatography and gas chromatography–mass spectrometry concentration of the test material required to cause a 50% decrease in initial DPPH concentration. Ascorbic acid was used as a standard. All measurements were per- Quantitative and qualitative analysis of the essential oil was carried out by gas formed in triplicate. chromatography–flame ionization detection (GC–FID) and gas chromatography– mass spectrometry (GC–MS), respectively. Gas chromatography analyses were car- 2.4.2. ABTS radical-scavenging assay ried out on a Varian Star 3400 Cx chromatograph (Les Ulis, France) fitted with a fused The radical scavenging capacity of the samples for the ABTS (2,20-azinobis-3- silica capillary DB-5MS column (5% phenylmethylpolysyloxane, 30 m 0.25 mm, ethylbenzothiazoline-6-sulphonate) radical cation was determined as described film thickness 0.25 lm). Chromatographic conditions were 60–260 °C temperature by Re et al. (1999) with some modifications. ABTS radical cation was generated rise with a gradient of 5 °C/min and 15 min isotherm at 260 °C. A second gradient by mixing a 7 mmol/L of ABTS at pH 7.4 (5 mmol/L NaH PO , 5 mmol/L Na HPO was applied to 340 °Cat40°C/min. Total analysis time was 57 min. For analysis pur- 2 4 2 4 and 154 mmol/L NaCl) with 2.5 mmol/L potassium persulfate (final concentration) poses, the essential oil was dissolved in petroleum ether. One microliter of sample followed by storage in the dark at room temperature for 16 h before use. The mix- was injected in the split mode ratio of 1:10. Helium (purity 99.999%) was used as car- ture was diluted with water to give an absorbance of 0.70 ± 0.02 units at 734 nm rier gas at 1 mL/min. The injector was operated at 200 °C. The GC–MS system (Varian using spectrophotometer (Helios). For sample, solutions of the essential oil in meth- Saturn 2000 ion trap GC/MS with CP-3800 GC) was used with the same chromato- anol (100 lL) were allowed to react with fresh ABTS solution (900 lL), and then the graphic conditions for GC–FID. Mass spectrometer was adjusted for an emission cur- absorbance was measured 6 min after initial mixing. Ascorbic acid was used as a rent of 10 lA and electron multiplier voltage between 1400 and 1500 V. Trap standard and the capacity of free radical scavenging was expressed by IC (mg/ temperature was 250 °C and that of the transfer line was 270 °C. Mass scanning 50 L). The capacity of free radical scavenging (IC ) was determined using the same was from 40 to 650 amu. 50 previously used equation for the DPPH method. All measurements were performed Compounds were identified by (i) comparison of their retention indices (RI) rel- in triplicate. ative to C5-C24 n-alkanes obtained on a nonpolar DB-5MS column, with those pro- All data of antioxidant activity were expressed as means ± standard deviations vided in the literature and (ii) by comparison of their mass spectra with those (SD) of triplicate measurements. The confidence limits were set at p < 0.05. SD did recorded in NIST 08 (National Institute of Standards and Technology), or reported not exceed 5% for the majority of the values obtained. in published articles or by co-injection of available reference compounds. The sam- ples were analyzed in duplicate. The percentage composition of the essential oil was computed by the normali- 2.5. Antiplasmodial activity zation method from the GC peak areas, assuming identical mass response factor for all compounds. Results were calculated as mean values of two injections from The chloroquine-resistant FcB1-Columbia strain of P. falciparum (IC50 for chloro- essential oil, without using correction factors. All determinations were performed quine: 186 nM) was cultured continuously according to Trager and Jensen (1976) in triplicate and averaged. with modifications (Desoubzdanne et al., 2008). The IC50 values for chloroquine

Fig. 1. Chromatograms of leaves essential oil of Cedrelopsis grevei (1: a-pinene; 2: b-pinene; 3: o-cymene; 4: m-cymene; 5: sylvestrene; 6: (E)-pinocarveol; 7: isopinocampheol; 8: p-cymen-8-ol; 9: a-terpineol; 10: myrtenal; 11: d-elemene; 12: a-cubebene; 13: a-longipinene; 14: isoledene; 15: a-copaene; 16: b-bourbonene; 17: b- elemene; 18: isoitalicene; 19: cyperene; 20: 9,10-dehydro-isolongifolene; 21: a-gurjunene; 22: a-cedrene; 23: b-caryophyllene; 24: b-gurjunene; 25: (Z)-b-farnesene; 26: (Z)-b-farnesene; 27: neoclovene; 28: (E)-b-farnesene; 29: a-amorphene; 30: c-muurolene; 31: a-curcumene; 32: ar-curcumene; 33: b-selinene; 34: b-guaiene; 35: viridiflorene; 36: a-muurolene; 37: a-bulnesene; 38: b-dihydroagarofuran; 39: c-cadinene; 40: d-cadinene; 41: 1,4-cadinadiene; 42: 4,5,9,10-dihydro-isolongifolene; 43: a- calacorene; 44: elemol; 45: spathulenol; 46: lauric acid; 47: germacrene D-4-ol; 48: viridiflorol; 49: epicedrol; 50: (E)-isolongifolanone; 51: 2-p-tolyl-6-methyl-5-hepten-2- ol; 52: hinesol; 53: daucol; 54: b-eudesmol; 55: a-bisabobloxide B; 56: bulnesol; 57: valeranone; 58: 5-allyl-4,7-dimethoxy-1,3-benzodioxole; 59: a-bisabolol; 60: c- dodecalactone; 61: juniper camphor; 62: (Z)-b-santalol; 63: alloevodionol; 64: verticiol). Author's personal copy

354 S. Afoulous et al. / Food and Chemical Toxicology 56 (2013) 352–362 were checked every 2 months, and we observed no significant variations. The par- Table 1 asites were maintained in vitro in human red blood cells (O±; EFS; Toulouse, France), Chemical composition of leaves essential oil of Cedrelopsis grevei. diluted to 4% hematocrit in RPMI 1640 medium (Lonza; Emerainville, France) sup- No. RI Compounds % Area plemented with 25 mM Hepes and 30 M NaHCO3 and complemented with 7% hu- + man AB serum (EFS). 1 927 a-Pinene 0.14 Parasites cultures were synchronized by combination of magnetic enrichment 2 971 b-Pinene 0.20 followed by D-sorbitol lysis (5% of D-sorbitol in sterile water) as described by Lamb- 3 1015 o-Cymene 0.13 ros and Vanderberg (1979). The antimalarial activity of essential oil was evaluated 4 1018 m-Cymene 0.18 by a radioactive micromethod described elsewhere (Benoit-Vical et al., 1999). Tests 5 1022 Sylvestrene 0.06 were performed in triplicate in 96-well culture plates (TPP) with cultures mostly at 6 1137 E-pinocarveol 0.11 ring stages (synchronization interval, 16 h) at 0.5–1% parasitemia (hematocrit, 7 1175 Isopinocampheol 0.03 1.5%). Parasite culture was incubated with each sample for 48 h. Parasite growth 8 1178 p-Cymen-8-ol 0.03 3 was estimated by [ H]-hypoxanthine (Perkin–Elmer; Courtaboeuf, France) incorpo- 9 1187 a-Terpineol 0.08 ration, which was added to the plates 24 h before freezing. After 48 h incubation, 10 1196 Myrtenal 0.16 plates were frozen-defrosted and each well was harvested on a glass fiber filter. 11 1334 d-Elemene 0.96 3 Incorporated ( H)-hypoxanthine was then determined with a b-counter (1450- 12 1346 a-Cubebene 0.72 Microbeta Trilux; Wallac-Perkin Elmer). The control parasite cultures, free from 13 1351 a-Longipinene 0.09 any sample, was referred to 100% growth. IC50 were determined graphically in con- 14 1365 Isoledene 1.01 centration versus percent inhibition curves. Chloroquine diphosphate was used as 15 1375 a-Copaene 7.67 positive control. 16 1382 b-Bourbonene 0.17 The antimalarial activity of essential oil was expressed by IC50, representing the 17 1390 b-Elemene 6.98 concentration of drug that induced a 50% parasitaemia decrease compared to the 18 1396 Isoitalicene 0.15 positive control culture referred to as 100% parasitaemia (Muñoz et al., 1999). 19 1400 Cyperene 1.27 According to the literature concerning plant antiplasmodial activities a sample is 20 1404 9,10-Dehydro-isolongifolenea 0.21 very active if IC50 < 5 mg/L, active if IC50 between 5 and 50 mg/L, weakly active if 21 1409 a-Gurjunene 0.36 IC50 between 50 and 100 mg/L and inactive if IC50 > 100 mg/L (Ouattara et al., 2006). 22 1413 a-Cedrene 0.19 23 1419 b-Caryophyllene 0.59 2.6. Cytotoxicity evaluation 24 1428 b-Gurjunene 0.34 25 1432 (E)-a-bergamotene 0.15 Cytotoxicity of essential oil was estimated on human breast cancer cells (MCF- 26 1443 (Z)-b-farnesene 2.31 7). The cells were cultured in the same conditions as those used for P. falciparum, 27 1453 Neoclovene 3.83 except for the 10% human serum, which was replaced by 10% foetal calf serum (Lon- 28 1469 (E)-b-farnesene 27.67 za). For the determination of essential oil activity, cells were distributed in 96-well 29 1471 a-Amorphene 0.56 plates at 3 104 cells/ well in 100 lL, and then 100 lL of culture medium contain- 30 1475 c-Muurolene 1.61 ing sample at various concentrations were added. Cell growth was estimated by 31 1479 a-Curcumene 1.69 (3H)-hypoxanthine incorporation after 48 h incubation exactly as for the P. falcipa- 32 1481 ar-Curcumene 0.10 rum assay. The (3H)-hypoxanthine incorporation in the presence of essential oil was 33 1484 b-selinene 0.77 compared with that of control cultures without sample (positive control being 34 1487 b-guaiene 0.17 doxorubicin) (Cachet et al., 2009). 35 1494 Viridiflorene 2.80 36 1498 a-Muurolene 2.58 37 1504 a-Bulnesene 0.62 2.7. Antiinflammatory activity 38 1510 b-Dihydroagarofurana 0.27 39 1513 c-Cadinene 0.87 5-Lipoxygenase is known to catalyze the oxidation of unsaturated fatty acids 40 1523 d-Cadinene 14.52 containing 1,4-pentadiene structures. In the body, arachidionic acid (biological sub- 41 1532 1,4-Cadinadiene 0.31 strate) is oxidized to hydroperoxyeicosatetraenoic acid (HPETE’s) by the 5-Lipoxy- 42 1539 4,5,9,10-Dihydro-isolongifolene 0.53 genase. Linoleic acid (substrate) is oxidized in vitro to a conjugate diene by 5- 43 1544 a-Calacorene 0.99 Lipoxygenase, the activity is evaluated by the spectrophotometric measurement 44 1549 Elemol 0.79 of the conjugated diene at 234 nm. 20 lL of various concentrations (0.23–3.16 g/ 45 1556 Spathulenola 0.25 L) of essential oil was mixed individually with sodium phosphate buffer (pH 7.4) 46 1565 Lauric acid 0.18 containing 5-LOX (500 U) and 60 lL of linoleic acid (3.5 mM), yielding a final vol- 47 1571 Germacrene D-4-ol 0.57 ume of 1 mL. However, the blank does not contain the substrate, but will be added 48 1588 Viridiflorol 1.23 30 lL of buffer solution. The mixture was incubated at 25 °C for 10 min, and the 49 1610 Epicedrol 2.50 absorbance was determined at 234 nm. The percentage of enzyme activity was 50 1616 E-isolongifolanone 0.44 plotted against concentration of the essential oil. The IC value is the concentration 50 51 1627 2-p-Tolyl-6-methyl-5-hepten-2-ola 0.25 of essential oil that caused 50% enzyme inhibition (Bylac and Racine, 2003). 52 1636 Hinesol 1.26 53 1639 Daucol 0.33 2.8. Statistical analysis 54 1651 b-Eudesmol 0.27 55 1654 a-Cadinol 1.62 All data were expressed as mean ± standard deviation of triplicate measure- 56 1664 Bulnesol 0.14 ments. The confidence limits were set at p < 0.05. Standard deviations (SD) did 57 1668 Valeranone 0.18 not exceed 5% for the majority of the values obtained. 58 1677 5-Allyl-4,7-dimethoxy-1,3-benzodioxolea 0.15 59 1683 a-Bisabolol 0.05 60 1688 c-Dodecalactone 2.05 3. Results and discussion 61 1693 Juniper camphor 0.09 62 1709 (Z)-b-santalol 2.14 3.1. Chemical composition of the essential oil 63 1874 Alloevodionola 1.32 64 1886 Verticola 0.03 The essential oil yield of C. grevei obtained from hydrodistillation Total 100 of leaves was 0.73%. Gauvin et al. (2004) have quantified the yield of Monoterpenes hydrocarbons 0.7 leaves essential oils, which were between 0.1% and 0.3%. The yield Monoterpenes oxygenated 0.41 Sesquiterpenes hydrocarbons 82.8 of our work was far superior to that of Gauvin et al. (2004). Sesquiterpenes oxygenated 13.19 Sixty-seven components of leaves essential oil were identified Others 2.90 (Fig. 1). This essential oil contained a complex mixture consisting a Tentatively identified supported by good match of MS. mainly sesquiterpene hydrocarbons (82.80%) and oxygenated ses- quiterpenes (13.19%) (Table 1). We had a little portion of oxygen- ated monoterpenes (0.41%) and monoterpene hydrocarbons (E)-b-farnesene (27.67%), d-cadinene (14.5%), a-copaene (7.67%) (0.70%). The major essential oil constituents characterized were: and b-elemene (6.98%). Author's personal copy

S. Afoulous et al. / Food and Chemical Toxicology 56 (2013) 352–362 355

One study in the literature has cited the chemical composition It is supposed that this difference of chemical composition be- of essential oil leaves of C. grevei. Gauvin et al. (2004) have shown tween our study and Gauvin et al. (2004) work quoted with the that the chemical composition of leaves essential oil of C. grevei top is due to the several factors like the process extraction to was dominated by (E)-b-farnesene (35.6%), b-pinene (12.8%), (Z)- use, the genetic factors and environmental (climatic, seasonal, geo- sesquisabinene-hydrate (9.8%) and ar-curcumene (8.6%). Their graphical, and geological). essential oil contained monoterpene (hydrocarbons (16.5%) and oxygenated (0.5%)), who were very poor in our work. On the other 3.2. Antioxidant activity hand, we found that our oil contained more d-cadinene (14.52%), a- copaene (7.67%), b-elemene (6.98%), (Z)-b-farnesene (2.31%) com- Antioxidant activity of essential oil of C. grevei leaves has been pared to Gauvin et al. (2004) respectively 0.4%, trace, 1.1% and determined by DPPH and ABTS assays. To our knowledge there is 1%. Also, b-pinene (0.2%), b-caryophyllene (0.59%), and ar-curcum- no published literature discussing the antioxidant activity of C. gre- ene (0.1%) were obtained in lowest concentration compared to vei essential oil and no other Cedrelopsis’ species. All results were study of Gauvin et al. (2004): 12.8%, 3.5% and 0.1% respectively. presented in Table 2. The low reactivity by DPPH assay (IC50 > 1000 - We identified new compounds not reported in the previous mg/L) was explained by the high concentration of sesquiterpene study Gauvin et al. (2004). The total content of these new com- hydrocarbons in essential oil (83.43%). The ABTS assay showed a pounds (Fig. 2) varied between 1.32% and 3.83% like neoclovene moderate antioxidant activity of essential oil (IC50 = 110 ± 3.6 mg/ (3.83%), viridiflorene (2.80%), a-muurolene (2.58%), epicedrol L). Although, the DPPH test is widely applied in the literature, it is (2.50%), (Z)-b-santalol (2.14%), c-dodecalactone (2.05%) and allo- based on the use of a very crowded radical. However, the use of a evodionol (1.32%). low crowded radical such as ABTS can be more suitable. In addition, We compared the composition of the essential oil of our leaves the DPPH test is performed in methanol while the ABTS test is car- of C. grevei with those of bark from the same species. Cavalli et al. ried out mainly in water, conditions closer to the physiological con- (2003) have obtained (E)-b-farnesene (9.3%), a-copaene (7.7%), a- ditions. The differences between the two results can be explained by selinene (5.8%), b-selinene (4.5%) and d-cadinene (4.9%) as major the mechanism of the involved reaction. The ABTS radical reactions compounds, the contents completely different from those of our involve electron transfer and take place at a much faster rate com- essential oil, respectively, 27.67%, 7.67%, 0%, 0.77% and 14.58%. pared to DPPH radicals (Ennajar et al., 2009) whose degree of discol- The bark contained 114 compounds identified but our essential oration is attributed to hydrogen donating ability of tested oil consists of 64 compounds. Another study was conducted by compounds. Rakotobe et al. (2008), dealing the essential oils of the bark of C. grevei from several parts of Madagascar. The compounds have var- 3.3. Cytotoxic activity ied with sampling sites but the major compounds (>20%) were a- pinene, ishwarane, (c,d)-cadinene, a-copaborneol and eudesmol, The essential oil showed a pronounced effect against the cell knowing that the total percentage of essential oil compounds iden- line MCF-7 with the IC50 values of 21.5 ± 2 mg/L (Table 2). A cyto- tified are between 23.9% and 71.2%. This composition was mark- toxic activity of C. grevei essential oil was showed for the first time. edly different from our essential oil from the leaves. To our knowledge, the abundant compounds ((E)-b-farnesene, d-

OH

OH

neoclovene (3.83%) (Z )- -santalol (2.14%) epicedrol (2.50%)

O

O O O OH -dodecalactone (2.05%) O -muurolene (2.58%) alloevodinol (1.32%)

Fig. 2. Structures of new abundant compounds identified in leaves essential oil of Cedrelopsis grevei.

Table 2

Antioxidant, anticancer, antimalarial and antiinflammatory activities (IC50 (mg/L)) of leaves essential oil of Cedrelopsis grevei.

Sample Antioxidant activity (DPPH assay) Antioxidant activity (ABTS assay) Anticancer activity Antimalarial activity Antiinflammatory activity Essential oil >1000 110.0 ± 3.6 21.5 ± 2.0 17.5 ± 1.0 21.3 ± 0.5 Control 3.75 ± 0.01a 1.84 ± 0.03a 0.22 ± 0.04b 0.14 ± 0.09c 1.23 ± 0.14d

a Ascorbic acid. b Doxorubicin. c Chloroquine. d Nordihydroguaiaretic acid (NDGA). Author's personal copy 356

Table 3 Cytotoxic activity (IC50 (mg/L)) and chemical composition of essential oils. Our work, I: Cedrelopsis grevei. El Babili et al. (2011), II: Origanum campactum. Imelouane et al. (2010), III: Lavandula dentata (flower); IV: Lavandula dentata (leaf). Li et al. (2009),V:Schefflera heptaphylla. Al-Kalaldeh et al. (2010), VI: Laurus nobilis; VII: Origanum syriacum; VIII: Origanum vulgare; IX: Salvia triloba. Bendaoud et al. (2010),X:Schinus molle; XI: Schinus terebinthifolius. Sibanda et al. (2004), XII: Heteropyxis dehniae. El Hadri et al. (2010), XIII: Salvia of ficinalis. Hussain et al. (2010a), XIV: Rosmarinus officinalis. Liu et al. (2009), XV: Melaleuca alternifolia. Monajemi et al. (2005), XVI: Citrus sinensis; XVII: Citrus medica; XVIII: Citrus limon. Haber et al. (2008), XIX: Talouma gloriensis. Hussain et al. (2010b), XX: Mentha arvensis (Summer); XXI: Mentha arvensis (Winter); XXII: Mentha piperita (Summer); XXIII: Mentha piperita (Winter); XXIV: Mentha longifolia (Summer); XXV: Mentha longifolia (Winter); XXVI: Mentha spicata (Summer); XXVII: Mentha spicata (Winter). Chabir et al. (2011), XXVIII: Melaleuca armillaris. Loizzo et al. (2007), XXIX: Satureia thymbra; XXX: Sideritis perfoliata; XXXI: Laurus nobilis (leaves); XXXII: Laurus nobilis (fruits); XXXIII: Pistacia palestina; XXXIV: Salvia officinalis. Sharopov and Setzer (2012), XXXV: Artemsia scoparia. Sonboli et al. (2010), XXXVI: Dracocephalum surmandinum. Cole et al. (2007), XXXVII: pure compounds).

Essential oil I II III IV V VI VII VIII IX X XI XII XIII XIV XV XVI XVII XVIII XIX Cytotoxic activity 21.5 >100 98.5 ± 1.02 101 ± 4.4 7.3 101.7 ± 7.9 130 ± 52.2 30.1 ± 1.14 174.3 ± 73.04 54 ± 10 47 ± 9 35.95 ± 8.11 a 554.4 ± 1.5 190.1 ± 6.0 310 a 0.5 1 10 14.1 Compounds a-Pinene 0.14 0.62 8.38 7.78 5.82 0.38 0.64 3.35 4.34 6.49 3.18 12.3 2.06 0.2 1.5 0,2 1 b-Pinene 0.20 30,06 27.8 22.24 4.55 0.6 1.3 8.98 4.96 3.09 2.57 0.2 0.91 16.3 3.7 E-pinocarveol 0.11 8.59 14.77 p-Cymen-8-ol 0.03 0.14 0.21 0.11 0.2 a-Terpineol 0.08 0,5 0.62 8.38 5.6 3.6 0.27 2.3 11.3 Myrtenal 0.16 6.81 8.18 0.1 352–362 (2013) 56 Toxicology Chemical and Food / al. et Afoulous S. a-Cubebene 0.72 0.25 0.39 0.21 a-Copaene 7.67 0.02 0.11 0.19 0.6 0.1 b-Bourbonene 0.17 0.06 b-Elemene 6.98 0.09 0.28 a-Gurjunene 0.36 0.02 0.08 a-Cedrene 0.19 0.04 0.29 b-Caryophyllene 0.59 0.78 5.61 0.37 2.5 1.79 2.23 0.04 0.29 1.6 1.68 1.12 7.56 0.9 b-Gurjunene 0.34 0.25 2.21 (E)-a-bergamotene 0.15 0.38 (Z)-b-farnesene 2.31 1.3 (E)-b-farnesene 27.67 0.27 2.37 0.49 0.2 1.13 3.4 c-Muurolene 1.61 0.02 0.05 0.07 0.9 0.2 b-Selinene 0.77 0.5 0.59 1.1 0.3 a-Muurolene 2.58 0.02 0.1 0.7 1.7 c-Cadinene 0.87 0.07 0.07 18.04 1.1 1.6 d-Cadinene 14.52 0.1 0.27 0.69 2.8 3.3 1,4-Cadinadiene 0.31 0.8 a-Calacorene 0.99 0.01 Elemol 0.79 0.03 0.11 Spathulenol a 0.25 0.03 0.76 0.29 0.2 Viridiflorol 1.23 0.6 0.79 Epicedrol 2.50 0.42 0.33 0.82 3.1 0.46 0.2 b-Eudesmol 0.27 0.68 a-Cadinol 1.62 0.01 0.61 0.64 Bulnesol 0.14 0.03 0.64

XX XXI XXII XXIII XXIV XXV XXVI XXVII XXVIII XXIX XXX XXXI XXXII XXXIII XXXIV XXXV XXXVI XXXVII 55.3 ± 1.9 59.7 ± 2.2 75.2 ± 2.9 80.8 ± 3.2 45.2 ± 2.0 50.6 ± 2.0 80.0 ± 2.4 80.6 ± 2.0 12 ± 1 >400 >400 >400 >400 >400 >400 1000 14 20.6 19.7 69.6 a-Pinene 3.53 1.31 1.95 10.15 8.66 5.72 3.67 6.81 4.72 5.4 2.4 100 b-Pinene 5.7 4.3 2.01 2.42 0.33 2.9 8.9 3.46 2.14 6.48 3.01 21.3 0.1 E-pinocarveol 0.08 p-Cymen-8-ol 0.2 0.21 a-Terpineol 0.71 0.67 6.13 0.19 0.07 1.32 0.49 0.16 1.53 0.27 2.42 0.4 2.43 3.18 0.2 0.2 Myrtenal 0.11 0.1 a-Cubebene a-Copaene 0.41 0.22 0.81 0.41 1.67 0.12 0.17 0.2 0.3 100 b-Bourbonene 0.23 0.17 1.82 1.12 0.61 0.31 0.93 0.56 0.24 0.92 b-Elemene 0.84 0.99 0.21 0,1 1 0.1 a-Gurjunene 0.34 0.24 0.51 Author's personal copy

S. Afoulous et al. / Food and Chemical Toxicology 56 (2013) 352–362 357

c- adinene and a-copaene) of this oil are not assessed against cancer. Li et al. (2010) showed that b-elemene has an anticancer activity

against the cell line MCF-7 (IC50 = 93.0 ± 7.8 mg/L). This compound was identified to 6.96% in our essential oil. In our results, b-elem-

ene has an IC50 higher than that of the essential oil. So, b-elemene is not the responsible for anticancer activity in the essential oil. We have the same trends for these three compounds a-pinene, b- pinene and a-bisabolol as b-elemene, which have respectively

IC50 = 172.2, 164.1 and 41.8 mg/L against the cell line MCF-7 (Van 0.5 1 10 14.1 Zyl et al., 2006). a The anticancer activity of the essential oil might be due to the synergic effects of all the terpenes in the oil, or perhaps there are some other active compounds responsible for the anticancer activ- ity of the essential oil, which deserves attention in future study.

3.4. Antimalarial activity

The in vitro inhibitory effect of the C. grevei essential oil against the chloroquine-resistant P. falciparum (FCR-3) was showed in Ta- 554.4 ± 1.5 190.1 ± 6.0 310 0.13 0.1 ble 2. The antimalarial activity (IC50 values) of the essential oil was a 17.5 ± 1 mg/L. Since the value of IC50 was found between 5 and 50 mg/L, we can be considered that C. grevei essential oil has a good activity against P. falciparum (Ouattara et al., 2006). This high value

of IC50 compared to that of chloroquine (IC50 = 0.1 ± 0.09 mg/L) can be explained by the low concentration of the active compound(s) since the essential oil is a multi-components mixture. To our 0.3 knowledge and according to the literature, none of majority com- pounds in the essential oil is known for antimalarial activity. Van Zyl et al. (2006) showed that a-pinene has an antimalarial activity 0.78 1.17 0.11 0.37 (IC50 = 1.20 ± 0.2 mg/L). This compound was identified to 0.14% in our essential oil. In our results, IC50 of the antimalarial activity for the essential oil was higher than those for b-pinene

(IC50 = 294.7 mg/L) and a-bisabolol (IC50 = 307.3 mg/L), reported 0.12 0.11 0.09 0.02 by Van Zyl et al. (2006).

3.5. Antiinflammatory activity

1.35 0.67 0.1 The essential oil displayed in vitro 5-Lipoxygenase inhibitory

activity with an IC50 value of 21.33 ± 0.5 mg/L (Table 2). An essen- tial oil with a 10 6 IC50 6 30 mg/L is defined as a good 5-Lipoxyge- nase inhibitor (Bylac and Racine, 2003). Therefore, we can conclude that the essential oil of C. grevei is a good 5-Lipoxygenase inhibitor. It has been reported that terpenes such as a-pinene and sesqui- terpenes such as b-caryophyllene and a-bisabolol exhibited activ- ity in the in vitro 5-Lipoxygenase assay (Bylac and Racine, 2003),

but no IC50 reported for these three compounds. The presence of these components may contribute to the activity of this essential oil as observed in the 5-Lipoxygenase assay.

3.6. Correlation

To study the role of the various compounds of an essential oil for the biological activities obtained, we performed an assessment 0,47 0.78 0.09 0.46 0.37 0.22 0.61 0.88 0.31 2 ) a

1,4-cadinadiene (Z)- -farnesene continued calculated from percentage. ( -bergamotene -farnesene -farnesene a b b 50 )- )- )- -Muurolene-Calacorene 0.73-Cadinol 0.77 0.08 0.16 -Gurjunene -Eudesmol 0.32 0.22 0.13 0.17 -Selinene -Muurolene-Cadinene 0.95 0.99 0.57IC 0.94 0.29 0.59 0.16 0.02 0.21 0.75 0.12 0.36 -Cadinene 1.21 1.02 0.36 3.11 0.57 0.14 1.51 0.1 E Z E Fig. 3. Compounds with good correlations for anticancer ((Z)-b-farnesene) and b ( Epicedrol b Viridiflorol Elemol Spathulenol Bulnesol Essential oilCytotoxic activity 21.5 >100 I 98.5 ± 1.02 II 101 ± 4.4c 7.3 IIIa 101.7 ± 7.9c 130 ± 52.2 30.1 ± 1.14 IVa 174.3 ± 73.04 54 ± 10 47 ± V 9 35.95 ± 8.11 VIa VII VIII IX X XI XII XIII XIV XV XVI XVII XVIII XIX 1,4-Cadinadiene b d ( ( a antimalarial (1,4-cadinadiene) activities. Table 3 Author's personal copy

Table 4 358 Antimalarial activity (IC50 (mg/L)) and chemical composition of essential oils. Our work, I: Cedrelopsis grevei. El Babili et al. (2011), II: Origanum camactum. Boyom et al. (2003), III: Xylopia phloiodora; IV: Pachypodanthium confine;V: Antidesma laciniatum; VI: Xylopia aethiopica; VII: Hexalobus crispiflorus. Ortet et al. (2010), VIII: Artemisia gorgonum. Tabanca et al. (2005), IX: Arnica longifolia;X:Aster hesparius; XI: Chrysothamnus nauseosus. Kamatou et al., 2005, XII: Salvia stenophylla; XIII: Salvia runcinata; XIV: Salvia repens. Lopes et al. (1999), XV: Virola surinamensis. Kamatou et al. (2007), XVI: Salvia albicaulis; XVII: Salvia dolomitica. Liu et al. (2009), XVIII: Melaleuca armillaris. Lanfranchi et al. (2010), XIX: Daucus crinitus Desf. Van Zyl et al. (2006), XX: pure compounds (a-pinene, b-pinene and a-bisabolol).

Essential oil I II III IV V VI VII VIII IX X XI XII XIII XIV XV XVI XVII XVIII XIX XX Antimalarial activity 21.5 34 17.9 16.6 29.4 17.8 2 5.2 ± 0.7 No activity No activity No activity 4.38 ± 1.07 1.23 ± 0.31 1.68 ± 0.26 9 a 6.4 ± 2.0 4.8 ± 0.7 27.0 ± 2.0 26.7 1.2 ± 0.20 294.77 307.30 Compounds a-Pinene 0.14 0.62 0.58 4.05 1.6 1.1 2.7 1.8 6.6 49.7 1.9 1.95 9.9 95.48 b-Pinene 0.20 0.68 10.07 0.12 0.7 19.8 0.7 0.8 3 1.5 0.33 0.2 96.32 E-pinocarveol 0.11 0.4 0.3 0.5 Isopinocampheol 0.03 1.23 0.31 5.42 p-Cymen-8-ol 0.03 0.14 1.58 0.1 0.2 a-Terpineol 0.08 4.99 0.2 0.2 0.3 0.1 1 0.2 2.7 6.2 0.16 Myrtenal 0.16 0.28 2.85 0.1 0.1 0.2 a-Cubebene 0.72 0.52 1.04 0.36 0.1 0.1 a-Copaene 7.67 0.02 0.53 7.06 2.2 4.07 13.27 0.2 0.1 4.6 0.2 b-Bourbonene 0.17 0.5 0.2 0.4 352–362 (2013) 56 Toxicology Chemical and Food / al. et Afoulous S. b-Elemene 6.98 0.58 0.73 1.34 1.92 0.3 Isoitalicene 0.15 0.1 Cyperene 1.27 0.34 15.54 3.95 11.53 a-Gurjunene 0.36 0.02 1.86 0.64 0.2 b-Caryophyllene 0.59 0.78 0.28 5.2 1.67 1.33 0.5 0.5 0.1 0.1 7.3 11.4 1.6 2.1 0.3 0.05 5.4 (E)-a-bergamotene 0.15 0.46 0.18 0.1 (E)-b-farnesene 27.67 0.2 0.2 0.2 3.4 c-Muurolene 1.61 0.02 0.35 0.7 2.64 1.93 0.2 0.1 0.3 0.02 0.6 ar-Curcumene 0.10 0.7 0.6 1.1 b-Selinene 0.77 0.28 0.3 0.12 Viridiflorene 2.80 0.3 3.5 1.4 a-Muurolene 2.58 0.02 1.5 1.84 1.29 0.2 0.5 0.5 0.1 c-Cadinene 0.87 0.07 11.27 3.51 0.3 2.53 0.1 0.2 0.2 0.12 d-Cadinene 14.52 0.1 0.4 0.8 0.5 0.3 2.2 0.36 1,4-Cadinadiene 0.31 0.67 0.1 a-Calacorene 0.99 0.01 0.89 0.84 7.82 0.1 0.1 0.2 Elemol 0.79 2.04 1.24 1.09 1.3 0.09 Spathulenol a 0.25 0.03 1.02 2.16 1.4 6.33 1.97 3.3 2 0.2 0.3 1.0 2.0 1.5 0.22 Viridiflorol 1.23 0.3 24.5 Epicedrol 2.50 5.07 7.24 8.5 1.99 2.54 0.3 4 3.9 0.2 5.6 b-Eudesmol 0.27 1.15 1.08 0.1 0.2 2.5 7.7 a-Cadinol 1.62 0.01 0.5 1.16 3 1.41 0.1 0.3 0.8 0.02 a-Bisabolol 0.05 1.1 8.2 0.8 90.32

a IC50 calculated from percentage. Author's personal copy

Table 5 Antiinflammatory activity (IC50 (mg/L)) and chemical composition of essential oils. Our work, I: Cedrelopsis grevei. Lourens et al. (2004), II: Helichrysum dasyanthum; III: Helichrysum excisum; IV: Helichrysum felinum;V:Helichrysum petiolare. Benites et al., 2009, VI: Lantana Camara. Ashour et al. (2009), VII: Bupleurum marginatum. Albano et al. (2012), VIII: Dittrichia viscosa; IX: Foeniculum vulgare;X:Origanum vulgare; XI: Salvia officinalis; XII: Thymbra capitata; XIII: Thymus camphoratus; XIV: Thymus carnosus; XV: Thymus mastichina. Mulyaningsih et al. (2010), XVI: Kadsura longipedunculata. Dongmo et al. (2010), XVII: Canarium schweinfurthii (Lolodorf); XVIII: Aucoumea klaineana (Lolodorf); XIX: Canarium schweinfurthii (Mbouda). Hamdan et al. (2010), XX: Citrus jambhiri; XXI: Citrus pyriformis. Nyiligira et al. (2004), XXII: Vitex pooara; XXIII: Vitex rehmannii; XXIV: Vitex obovata ssp. Abovata; XXV: Vitex obovata ssp. Wilmsii; XXVI: Vitex zeyheri.

Essential oil I II III IV V VI VII VIII IX X XI XII XIII Antiinflammatory activity 21.33 ± 0.5 31.24 ± 1.31 27.62 ± 0.43 22.89 ± 7.59 25.03 ± 0.57 81.5 63.64 291.2 ± 2 2. 67.7 ± 2.3 264.2 ± 20.7 827.9 ± 60.6 93.3 ± 1 0.5 334.3 ± 43.6 Compounds a-Pinene 0.14 16.6 2.8 6.8 0.1 0.2 25.8 1.1 8.4 1.6 11.9 b-Pinene 0.20 6.2 1.1 0.4 0.3 6.8 3.2 0.1 0.9 (E)-pinocarveol 0.11 0.17 0.8 p-Cymen-8-ol 0.03 0.1 0.2 0.3 a-Terpineol 0.08 0.5 0.8 5.1 1.08 0.4 1.3 1.5 0.1 0.5 a-Cubebene 0.72 1.96 0.1 a-Copaene 7.67 0.1 2.5 4 1.3 0.52 0.7 0.1 b-Elemene 6.98 0.2 0.34 a-Gurjunene 0.36 0.1 1 0.1 .Aolu ta./Fo n hmclTxclg 6(03 352–362 (2013) 56 Toxicology Chemical and Food / al. et Afoulous S. b-Caryophyllene 0.59 13.3 5.7 27.6 14 0.1 5.53 1.7 0.9 1.1 (E)-a-bergamotene 0.15 0.1 0.9 0.1 (E)-b-farnesene 27.67 0.1 0.44 c-Muurolene 1.61 1.2 b-Selinene 0.77 Viridiflorene 2.80 0.1 a-Muurolene 2.58 2.2 0.18 c-Cadinene 0.87 0.1 2 1.69 3.7 0.7 d-Cadinene 14.52 5.7 0.2 0.1 1,4-Cadinadiene 0.31 0.1 a-Calacorene 0.99 1.33 0.4 Elemol 0.79 Spathulenol a 0.25 0.6 0.2 4.4 Viridiflorol 1.23 18.2 0.7 0.9 0.78 0.5 a-Cadinol 1.62 a-Bisabolol 0.05 0.1 XIV XV XVI XVII XVIII XIX XX XXI XXII XXIII XXIV XXV XXVI 544.3 ± 64.5 1084.5 ± 146.1 38.58 ± 3.8 62.6 ± 4.2 No activity No activity 40 ± 1.36 38 ± 0.82 25 40.5 42 48 64 a-Pinene 4.9 7 1.55 1.7 29.3 2.6 0.01 0.26 1.69 0.02 2.22 0.02 2.53 b-Pinene 2.8 5.3 0.81 0.4 0.8 1.2 0.44 0.92 0.16 1.57 3.01 0.23 (E)-pinocarveol 0.2 0.1 p-Cymen-8-ol a-Terpineol 0.8 3.4 0.21 34.4 3.1 18 0.15 0.51 3.43 a-Cubebene 1.2 0.06 a-Copaene 2.83 7.39 11.79 0.3 b-Elemene 1.5 0.28 a-Gurjunene 1.16 b-Caryophyllene 0.4 0.5 0.19 3.82 3.64 (E)-a-bergamotene 0.32 0.28 (E)-b-farnesene 0.01 c-Muurolene 1.41 0.02 0.42 3.19 4.9 5.08 b-Selinene 1.74 0.04 0.88 14.41 1.47 Viridiflorene 0.48 a-Muurolene 2.55 1.36 1.69 c-Cadinene 0.03 1.58 1.71 2.28 d-Cadinene 0.67 0.43 0.25 1,4-Cadinadiene 0.94 a-Calacorene 1.37 0.11 1.07 Elemol 1.1 0.5 359 (continued on next page) Author's personal copy

360 S. Afoulous et al. / Food and Chemical Toxicology 56 (2013) 352–362

of the literature of all essential oils from various plants tested for biological activities (anticancer, antimalarial and antiinflamma- tory). Based on these results, we established correlations be- tween biological activities and concentrations of compounds. 334.3 ± 43.6 The aim of the study was to find correlation(s) between com- pound(s) quantity and biological activity.

3.6.1. Correlation for cytotoxic activity Data in Table 3 showed the cytotoxic activity against the MCF-

7 cells (IC50 mg/L) of all essential oils in literature and their com- pounds. Essential oils were extracted from Cedrelopsis grevei, Origanum campactum, Lavandula dentata (flower and leaf), Sche- fflera heptaphylla, Laurus nobilis, Origanum syriacum, Origanum vulgare, Salvia triloba, Schinus molle, Schinus terebinthifolius, Het-

2.85 3.27 3.57 eropyxis dehniae, Salvia officinalis, Rosmarinus officinalis, Melaleuca alternifolia, Citrus sinensis, Citrus medica, Citrus limon, Talouma gloriensis, Mentha arvensis (Summer and Winter), Mentha piperita (Summer and Winter), Mentha longifolia (Summer and Winter),

0.34 1.58 0.38 1.52 Mentha spicata (Summer and Winter), Melaleuca armillaris, Satur- eia thymbra, Sideritis perfoliata, Laurus nobilis (leaves and fruits), Pistacia palestina, Salvia officinalis, Artemsia scoparia, Dracoceph- alum surmandinum and pure compounds (a-pinene, a-copaene and b-caryophyllene). We established correlations between compound contents and anticancer activity against the MCF-7 cells. (Z)-b-farnesene showed good correlation, R2 = 0.73. To our knowledge, no litera- ture cited this compound ((Z)-b-farnesene) for any anticancer activity (Fig. 3).

3.6.2. Correlations for antimalarial activity Data in Table 4 summarize chemical composition and the

antimalarial activity (IC50 (mg/L)) of all essential oils cited in lit- erature. These essential oils have been obtained from Cedrelopsis grevei, Origanum camactum, Xylopia phloiodora, Pachypodanthium confine, Antidesma laciniatum, Xylopia aethiopica, Hexalobus crisp- iflorus, Artemisia gorgonum, Arnica longifolia, Aster hesparius, Chrysothamnus nauseosus, Salvia stenophylla, Salvia runcinata, Sal- via repens, Virola surinamensis, Salvia albicaulis, Salvia dolomitica, Melaleuca armillaris, Daucus crinitus and pure compounds (a- pinene, b-pinene and a-bisabolol). We established correlations between compounds contents

and IC50 of antimalarial activity. 1,4-cadinadiene showed good correlation (R2 = 0.61). To our knowledge, no literature cited this compound (1,4-cadinadiene) for any antimalarial activity.

0.1 3.07 3.6.3. Correlations for antiinflammatory activity

Data in Table 5 showed the antiinflammatory activity (IC50 (mg/L)) of all essential oils in literature and their chemical compo- sition. Essential oils were extracted from Cedrelopsis grevei, Heli- chrysum dasyanthum, Helichrysum excisum, Helichrysum felinum, Helichrysum petiolare, Lantana Camara, Bupleurum marginatum, Dittrichia viscose, Foeniculum vulgare, Origanum vulgare, Salvia offi- cinalis, Thymbra capitata, Thymus camphorates, Thymus carnosus, Thymus mastichina, Kadsura longipedunculata, Canarium schwein- furthii (Lolodorf and Mbouda), Aucoumea klaineana (Lolodorf), Cit- ) rus jambhiri, Citrus pyriformis, Vitex pooara, Vitex rehmannii, Vitex a obovata ssp. Abovata, Vitex obovata ssp. Wilmsii and Vitex zeyheri. We established correlations between compounds contents continued calculated from percentage. (

50 and IC50 of antiinflammatory activity. This study showed that -Cadinol -Bisabolol 0.3 IC Viridiflorol 0.2 0.75 Spathulenol Essential oilAntiinflammatory activity 21.33 ± 0.5 31.24 ±a 1.31a I 27.62 ± 0.43 22.89 ± 7.59 25.03 ± 0.57 81.5 II 63.64 III 291.2 ± 2 2. 67.7 ± 2.3 264.2 ± 20.7 IV 827.9 ± 60.6 93.3 ± 1 0.5 Vthere VI is no correlation VII VIII between IX constituents X and the XI IC ofthe XII XIII

a 50

Table 5 antiinflammatory activity. Author's personal copy

S. Afoulous et al. / Food and Chemical Toxicology 56 (2013) 352–362 361

4. Conclusion of Plasmodium falciparum enzyme targets from a New Caledonian deep water sponge. J. Nat. Prod. 71, 1189–1192. Dongmo, P.M.J., Tchoumbougnang, F., Ndongson, B., Agwanande, W., Sandjon, B., In conclusion, we have identified all volatile constituents of C. Zollo, P.H.A., Menut, C., 2010. Chemical characterization, antiradical, grevei leaves essential oil and evaluated the antioxidant, antican- antioxidant and antiinflammatory potential of the essential oils of Canarium cer, antimalarial and antiinflammatory activities. Our results schweinfurthii and Aucoumea klaineana (Burseraceae) growing in Cameroon. Agric. Biol. J. North Am. 1, 606–611. clearly show that this essential oil is active against cells line tested El Babili, F., Bouajila, J., Souchard, J.P., Bertrand, C., Bellvert, F., Fouraste, I., Moulis, C., (tumor MCF-7), parasite (FCR-3), and inhibitory 5-Lipoxugenase. Valentin, A., 2011. Oregano: chemical analysis and evaluation of its This anticancer activity against MCF-7 could be explained, in part, antimalarial, antioxidant, and cytotoxic activities. J. Food Sci. 76, 512–518. El Hadri, A., Gómez del Río, M.A., Sanz, J., González Coloma, A., Idaomar, M., Ozonas, by the presence of b-elemene. However other compounds may B.R., González, J.B., Sánchez Reus, M.I., 2010. Cytotoxic activity of a-humulene have anticancer and antimalarial activity like the majority com- and trans-caryophyllene from Salvia officinalis in animal and human tumor cells. pounds which are never evaluated. Based on established correla- An. Real Acad. Farmacia. 76, 343–356. Ennajar, M., Bouajila, J., Lebrihi, A., Mathieu, F., Abderraba, M., Raies, A., Romdhane, tions for antimalarial and anticancer activity, compounds such as M., 2009. Chemical composition, antimicrobial and antioxidant activities of 1,4-cadinadiene and (Z)-b-farnesene, could be the best candidates essential oils and various extracts of Juniperus phoenicea L. (cupressacees). J. for further analysis. Food Sci. 74 (7), M364–371. European Pharmacopoeia, 1983. Maissoneuve SA, Sainte Ruffine. The results presented here can be considered as the first infor- Gauvin, A., Ravaomanarivo, H., Smadja, J., 2004. Comparative analysis by gas mation on the anticancer, antimalarial and antiinflammatory prop- chromatography–mass spectrometry of the essential oils from bark and leaves erties of C. grevei. In a future study, we will identify and investigate of Cedrelopsis grevei Baill, an aromatic and medicinal plant from Madagascar. J. Chromatogr. A 1029, 279–282. the components responsible for the anticancer, antimalarial and Haber, W.A., Agius, B.R., Stokes, S.L., Setzer, W.N., 2008. Bioactivity and chemical antiinflammatory activities of C. grevei. Consequently, those results composition of the leaf essential oil of Talauma gloriensis Pittier (Magnoliaceae) can be considered as a preliminary study in order to show the from Monteverde, Costa Rica Rec. Nat. Prod. 2, 1–5. importance and the originality of this endemic species. Hamdan, D., El-Readi, M.Z., Nibret, E., Sporer, F., Farrag, N., El-Shazyl, A., Wink, M., 2010. Chemical composition of the essential oils of two Citrus species and their biological activities. Pharmazie 65, 141–147. Hussain, A.I., Anwar, F., Chatha, S.A.S., Jabbar, A., Mahboob, S., Nigam, P.S., 2010a. Conflict of Interest Rosmarinus officinalis essential oil: antiproliferative, antioxidant and antibacterial activities. Braz. J. Microbiol. 41, 1070–1078. The authors declare that there are no conflicts of interest. Hussain, A.I., Anwar, F., Nigam, P.S., Ashraf, M., Gilani, A.H., 2010b. Seasonal variation in content, chemical composition and antimicrobial and cytotoxic activities of essential oils from four Mentha species. J. Sci. Food Agric. 90, 36– References 1827. Imelouane, B., Tahr, I.M., Ankit, M., Khdid, K., Amhamdi, H., Dubois, J., Elbachiri, A., 2010. The essential oil of eastern moroccan Rosmarinus officinalis: chemical Albano, S.M., Lima, A.S., Miguel, M.G., Pedro, L.G., Barroso, J.G., Figueiredo, A.C., composition, in-vitro antimicrobial and antioxidant activities. Rev. Microbiol. 2012. Antioxidant, anti-5-lipoxygenase and antiacetylcholinesterase activities Indig. Sant. Environ. 4, 120–141. of essential oils and decoction waters of some aromatic plants. Rec. Nat. Prod. 6, Kamatou, G.P.P., Viljoen, A.M., Figueiredo, A.C., Tilney, P.M., Van Zyl, R.L., Barroso, 35–48. J.G., Pedro, L.G., Van Vuuren, S.F., 2007. Trichomes, essential oil composition and Al-Kalaldeh, J.Z., Abu-Dahab, R., Afifi, F.U., 2010. Volatile oil composition and biological activities of Salvia albicaulis Benth and S. dolomitica Codd, two species antiproliferative activity of Laurus nobilis, Origanum syriacum, Origanum vulgare, from the Cape region of South Africa. S. Afr. J. Bot. 73, 102–108. and Salvia triloba against human breast adenocarcinoma cells. Nutr. Res. 30, Kamatou, G.P.P., Viljoen, A.M., Gono-Bwalya, A.B., Van Zyl, R.L., Van Vuuren, S.F., 271–278. Lourens, A.C.U., Baser, K.H.C., Demirci, B., Lindsey, K.L., Van Staden, J., Ashour, M.L., El-Readi, M., Youns, M., Mulyaningsih, S., Sporer, F., Efferth, T., Wink, Steenkamp, P., 2005. The in vitro pharmacological activities and a chemical M., 2009. Chemical composition and biological activity of the essential oil investigation of three South African Salvia species. J. Ethnopharmacol. 102, 382– obtained from Bupleurum marginatum (Apiaceae). J. Pharm. Pharmacol. 61, 390. 1079–1087. Lambros, C., Vanderberg, J.P., 1979. Synchronization of Plasmodium falciparum Bendaoud, H., Romdhane, M., Souchard, J.P., Cazaux, S., Bouajila, J., 2010. Chemical erythrocytic stages in culture. J. Parasitol. 65, 418–420. composition and anticancer and antioxidant activities of Schinus molle L. and Lanfranchi, D.A., Laouer, H., El Kolli, M., Prado, S., Maulay-Bailly, C., Baldovini, N., Schinus terebinthifolius Raddi Berries Essential Oils. J. Food Sci. 75, 466–472. 2010. Bioactive phenylpropanoids from Daucus crinitus Desf. from Algeria. J. Benoit-Vical, F., Valentin, A., Mallié, M., Bastide, J.M., Bessière, J.M., 1999. In vitro Agric. Food Chem. 58, 2174–2179. antimalarial activity and cytotoxicity of Cochlospermum tinctorium and C. Li, Q.Q., Wang, G., Huang, F., Banda, M., Reed, E., 2010. Antineoplastic effect of beta- planchonii leaf extracts and essential oils. Planta Med. 65, 378–381. elemene on prostate cancer cells and other types of solid tumour cells. J. Pharm. Benites, J., Moiteiro, C., Miguel, G., Rojo, L., Lopez, J., Venância, F., Ramalho, L., Feio, Pharmacol. 62, 1018–1027. S., Dandlen, S., Casanova, H., Torres, I., 2009. Composition and biological activity Li, Y.L., Yeung, C.M., Chiu, L.C.M., Cen, Y.Z., Ooi, V.E.C., 2009. Chemical composition of the essential oil of peruvian Lantana camara. J. Chil. Chem. Soc. 54, 379–384. and antiproliferative activity of essential oil from the leaves of a medicinal herb, Blois, M.S., 1958. Antioxidant determination by use of free radical stable. Nature Schefflera heptaphylla. Phytother. Res. 23, 140–142. 181, 1199–1200. Liu, X., Zu, Y., Fu, Y., Yao, L., Gu, C., Wang, W., Efferth, T., 2009. Antimicrobial activity Boyom, F.F., Ngouana, V., Amvam Zollo, P.H., Menut, C., Bessiere, J.M., Gut, J., and cytotoxicity towards cancer cells of Melaleuca alternifolia (tea tree) oil. Eur. Rosenthal, P.J., 2003. Composition and anti-plasmodial activities of essential oils Food Res. Technol. 229, 247–253. from some Cameroonian medicinal plants. Phytochemistry 64, 1269–1275. Loizzo, M.R., Tundis, R., Menichini, F., Saab, A.M., Statti, G.A., Menichini, F., 2007. Bylac, S., Racine, P., 2003. Inhibition of 5-lipoxygenase by essential oils and other Cytotoxic activity of essential oils from labiatae and lauraceae families against natural fragrant extracts. Int. J. Aromather. 13, 138–142. in vitro human tumor models. Anticancer Res. 27, 3293. Cabanis, Y., Chabouis, L., Chabouis, F., 1969. Végétaux et groupements végétaux de Lopes, N.P., Kato, M.J., Andrade, E.H.de A., Maiaa, J.G.S., Yoshida, M., Planchart, A.R., Madagascar et des Mascareignes. BDPA 70, Tananarive. Katzin, A.M., 1999. Antimalarial use of volatile oil from leaves of Virola Cachet, N., Hoakwie, F., Bertani, S., Bourdy, G., Deharo, E., Stien, D., Houel, E., surinamensis (Rol.) Warb. by Waiaˇpi Amazon Indians. J. Ethnopharmacol. 67, Gornitzka, H., Fillaux, J., Chevalley, S., Valentin, A., Jullian, V., 2009. Antimalarial 313–319. activity of simalikalactone E, a new quassinoid from Quassia amara L. Lourens, A.C.U., Reddy, D., Basßer, K.H., Viljoen, A.M., Van Vuuren, S.F., 2004. In vitro (Simaroubaceae). Antimicrob. Agents Chemother. 53, 4393–4398. biological activity and essential oil composition of four indigenous South Cavalli, J.F., Tomi, F., Bernardini, A.F., Casanova, J., 2003. Composition and chemical African Helichrysum species. J. Ethnopharmacol. 95, 253–258. variability of the bark oil of Cedrelopsis grevei H. Baillon from Madagascar. Flav. Monajemi, R., Oryan, S., Haeri-Roohani, A., Ghannadi, A., Jafarian, A., 2005. Cytotoxic Fragr. J. 18 (6), 532–538. effects of essential oils of some Iranian Citrus peels. Iran. J. Pharm. Res. 3, 183– Chabir, N., Romdhane, M., Valentin, A., Moukarzel, B., Ben Marzoug, H.N., Ben 187. Brahim, N., Mars, M., Bouajila, J., 2011. Chemical study and antimalarial, Mulholland, D.A., Kotsos, M., Mahomed, H.A., Randrianarivelojosia, M., 1999. The antioxidant, and anticancer activities of Melaleuca armillaris (Sol Ex Gateau) Sm chemistry of the Ptaeroxylaceae. Nig. J. Nat. Prod. Med. 3, 13–16. essential oil. J. Med. Food 14, 1383–1388. Mulyaningsih, S., Youns, M., El-Readi, M.Z., Ashour, M.L., Nibret, E., Sporer, F., Cole, R.A., Bansal, A., Moriarity, D.M., Haber, W.A., Setzer, W.N., 2007. Chemical Herrmann, F., Reichling, J., Wink, M., 2010. Biological activity of the essential oil composition and cytotoxic activity of the leaf essential oil of Eugenia of Kadsura longipedunculata (Schisandraceae) and its major components. J. zuchowskiae from Monteverde, Cota Rica. J. Nat. Med. 61, 414–417. Pharm. Pharmacol. 62, 1037–1044. Courchet, L., 1906. Recherches morphologiques et anatomiques sur le Katafa ou Muñoz, V., Sauvain, M., Mollinedo, P., Callapa, J., Rojas, I., Gimenez, A., Valentin, A., Katrafay de Madagascar (Cedrelopsis grevei H. Baillon). Ann. Inst. Colon. Mallié, M., 1999. Antimalarial activity and cytotoxicity of () roemrefidine Marseille 4, 29–118. isolated from the stem bark of Sparattanthelium amazonum. Planta Med. 65, Desoubzdanne, D., Marcourt, L., Raux, R., Chevalley, S., Dorin, D., Doerig, C., Valentin, 448–449. A., Ausseil, F., Debitus, C., 2008. Alisiaquinones and alisiaquinol, dual inhibitors Author's personal copy

362 S. Afoulous et al. / Food and Chemical Toxicology 56 (2013) 352–362

Nyiligira, E., Viljoen, A.M., Ozek, T., Baser, K.H.G., Vuuren, V.S., 2004. Essential oil Sharopov, S.F., Setzer, W.N., 2012. The essential oil of Salvia sclarea L. from composition and in vitro antimicrobial and antiinflammatory activity of South Tajikistan. Nat. Prod. Commun. 6, 75–79. African Vitex species. S. Afr. J. Bot. 70, 611–617. Sibanda, S., Chigwada, G., Poole, M., Gwebu, E.T., Noletto, J.A., Schmidt, J.M., Rea, A.I., Ortet, R., Thomas, O.P., Regalado, E.L., Pino, J.A., Filippi, J.J., Fernandez, M.D., 2010. Setzer, W.N., 2004. Composition and bioactivity of the leaf essential oil of Composition and biological properties of the volatile oil of Artemisia gorgonum Heteropyxis dehniae from Zimbabwe. J. Ethnopharmacol. 92, 107–111. Webb. Chem. Biodivers. 7, 1325–1333. Sonboli, A., Esmaeili, M.A., Gholipour, A., Kanani, M.R., 2010. Composition, Ouattara, Y., Sanon, S., Traoré, Y., Mahiou, V., Azas, N., Sawadogo, L., 2006. cytotoxicity and antioxidant activity of the essential oil of Dracocephalum Antimalarial activity of Swartzia madagascarensis desv. (leguminosae), surmandinum from Iran. Nat. Prod. Commun. 5, 341–344. Combretum glutinosum Guill & Perr. (combretaceae) and Tinospora bakis miers Tabanca, N., Demirci, B., Crockett, S.L., Baser, K.H.C., Wedge, D.E., 2005. Chemical (menispermaceae) Burkina fasso medicinal plants. Afr. J. Trad. Compl. Alter. composition and antifungal activity of Arnica longifolia, Aster hesperius, and Med. 3, 75–81. Chrysothamnus nauseosus essential oils. J. Agric. Food Chem. 55, 8430–8435. Rakotobe, M., Menut, C., Andrianoelisoa, H.S., Rahajanirina, V., Collas de Trager, W., Jensen, J.B., 1976. Human malarial parasites in continuous culture. Chatelperron, P., Roger, E., Danthu, P., 2008. The bark essential oil Science 193, 673–675. composition and chemotaxonomical appraisal of Cedrelopsis grevei H. Baillon Um, B.H., Lobstein, A., Weniger, B., Spiegel, C., Yice, F., Rakotoarison, O., from Madagascar. Nat. Prod. Commun. 3 (7), 1145–1150. Andriantsitohaina, R., Anton, R., 2003. New coumarins from Cedrelopsis grevei. Rasoanaivo, P., De La Gorce, P., 1998. Essential oils of economic value in Fitoterapia 74, 638–642. Madagascar: present state of knowledge. HerbalGram 43, 31–59. Van Zyl, R.L., Seatlholo, S.T., Van Vuuren, S.F., Viljoen, A.M., 2006. The biological Re, R., Pellegrini, N., Proteggente, A., Pannala, A., Yang, M., Rice-Evans, C., 1999. activity of 20 nature identical essential oil constituents. J. Essent. Oil Res. 18, Antioxidant activity applying an improved ABTS radical cation decolorization 129–133. assay. Free Rad. Biol. Med. 26, 1231–1237. P.08 : AT H. Mossa, TM. Heikal, M. Belaiba, EG. Raoelison, H. Ferhout, J. Bouajila. Antioxidant activity and hepatoprotective potential of Cedrelopsis grevei on cypermethrin oxidative stress in liver damage in male mice. BMC Alternative and Complementary Medicine 2015, 15, 251- 260.

Mossa et al. BMC Complementary and Alternative Medicine (2015) 15:251 DOI 10.1186/s12906-015-0740-2

RESEARCH ARTICLE Open Access Antioxidant activity and hepatoprotective potential of Cedrelopsis grevei on cypermethrin induced oxidative stress and liver damage in male mice Abdel-Tawab H. Mossa1*, Tarek M. Heikal1, Meriam Belaiba2, Emmanuel Guy Raoelison3, Hicham Ferhout4 and Jalloul Bouajila2*

Abstract Background: The liver is the most sensitive and main target organ of pesticide toxicity and damage, they play an essential role in metabolism and detoxification of pesticides. Due to these functions, hepatotoxicity continues to be among the main threats to public health, and they remain problems throughout the world. Therefore, the present study was designed to evaluate the antioxidant and hepatoprotective effects of Cedrelopsis grevei leaves against cypermethrin (Cyp) induced oxidative stress and liver damage in male mice. Methods: The extracts were subjected to different analyses (phenolics, tannin, flavonoids, antioxidant activity and reducing power assays). For hepatoprotective evaluation, male mice were daily exposed to Cyp and/or C. grevei by gavages for 28 days. Hepatoprotective effects were demonstrated by significant alterations in serum liver dysfunction biomarker enzymes, liver lipid peroxidation and antioxidant enzymes.

Results: The antioxidant activity of C. grevei methanolic extract was the highest with an IC50 <225μg/ml by DPPH assay. The high dose of methanolic extract (300 mg/kg. b.wt.) was effective to attenuate the perturbations in the tested enzymes. Histopathological examination in the liver tissue of those mice, demonstrated that a co-administration of methanolic extract (150 & 300 mg/kg/day) showed marked improvement in its histological structure in comparison to Cyp-treated group alone and represented by nil to moderate degree in inflammatory cells. Conclusions: In view of the data of the present study, it can deduce that cypermethrin caused oxidative damage and liver dysfunction in male mice. C. grevei extract has protective effects on cypermethrin-induced lipid peroxidation, oxidative stress and liver damage. Results indicated that administration of C. grevei is useful, easy, and economical to protect humans against pesticide toxicity. The results presented here can be considered as the first information on the hepatoprotective and antioxidant properties of C. grevei extracts. In a future study, we will identify and investigatethecomponentsresponsibleforthehepatoprotectiveandantioxidantactivitiesofC. grevei. Keywords: Cedrelopsis grevei, Antioxidant, Hepatoprotective, Cypermethrin, Oxidative stress, Liver damage, Mice

*Correspondence:[email protected]; [email protected] 1Environmental Toxicology Research Unit (ETRU), Pesticide Chemistry Department, National Research Centre (NRC), 33 El Bohouth Street (former El Tahrir St.), P.O. 12622, Dokki, Giza, Egypt 2Faculté de pharmacie de Toulouse, Laboratoire des Interactions Moléculaires et Réactivité Chimique et Photochimique UMR CNRS 5623, Université de Toulouse, Université Paul-Sabatier, 118 route de Narbonne, F-31062 Toulouse, France Full list of author information is available at the end of the article

© 2015 Mossa et al. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http:// creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Mossa et al. BMC Complementary and Alternative Medicine (2015) 15:251 Page 2 of 10

Background phenolic compounds of C. grevei are playing an essential Pesticides have been applied in agriculture and household role in neutralizing free radical, quenching singlet and to protect plants, animals and human from insects and triplet oxygen, decomposing peroxides, stabilizing lipid vector diseases. The negligent and random uses of pesti- peroxidation and protecting the cells against oxidative cides can cause environmental damage, food, water con- damage [18]. tamination, and health problems (e.g. cancer, nerve disease, Currently, some synthetic antioxidant use to prevent birth defects). Cypermethrin (Cyp), a class II pyrethroid free radical damage can induce side effects [19]. So, the pesticide, first synthesized in 1974, widely used to control dietary intake of natural products is considered very many pest species in agriculture, animal breeding and the important for preventing a wide variety of diseases, in- household [1]. It has been reported that Cyp residues were cluding allergies, cardiovascular disease, certain forms of found in the air, on walls and furniture after three months cancer, hepatic diseases, and inflammation, which involve of household treatments [2]. Cyp was accumulated in adi- free radical–mediated damage in pathologically generating pose tissue, brain and liver of rats [3, 4] and has hepato- processes [20]. Therefore, that is an essential research toxic potential in rodents [2, 5]. It crosses the blood–brain about suitable herbal drugs, that could replace the chem- barrier and induces neurotoxicity and motor deficits [6]. ical ones [21]. However, the widespread use of C. grevei in However, due to the low toxicity of pyrethroids, persistence traditional medicine stimulated us to explore its potential of these insecticides in mammalian tissues may be danger- biological activity. To the best of our knowledge, no ous [3]. previous study of the antioxidant and hepatoprotective In fact, one possible mechanism of pesticide-induced activities of C. grevei leaves extract have been reported. toxicity is the production of reactive oxygen species (ROS) Therefore, the current study was designed to evaluate in the cell. ROS can alter oxidant/prooxidants statues and the antioxidant activity and hepatoprotective effect of C. antioxidant defense system by increasing lipid peroxidation grevei leaves against Cyp induced oxidative stress, lipid (LPO) and depleting the antioxidants in cell (enzymatic peroxidation and liver damage in male mice. and non-enzymatic) which leading to a condition of oxida- tive stress [7]. It has been reported that ROS were involved Materials and methods in the toxicity of organophosphate insecticides (OPIs) such Chemicals and reagents as chlorpyrifos [7] and pyrethroid insecticides such as The assay kits used for biochemical measurements of prallethrin [8, 9] and a positive correlation with the liver aspartate aminotransferases (AST; EC 2.6.1.1.), alanine damage has been reported. ROS, especially superoxide aminotransferases (ALT; EC 2.6.1.2), alkaline phosphat- anion and hydrogen peroxide, are important signaling ase (ALP; EC 3.1.3.1), lactate dehydrogenase (LDH; EC molecules in developing and proliferating cells, but also 1.1.1.27), catalase (CAT; EC 1.11.1.6), superoxide dismut- in the induction of programmed cell death [10]. ROS ase (SOD; EC 1.15.1.1), glutathione-s-transferase (GST; EC are transient species due to its high chemical reactivity 2.5.1.18), glutathione reduced (GSH), malondialdehyde that leads to the LPO and a massive protein oxidation (MDA) and total protein were purchased from Biodiag- and degradation [11, 12]. The author reported that ROS nostic Company, Dokki, Giza, Egypt. Cypermethrin (95.1 %) cause DNA damage and strand breaks as a result of modi- was obtained from Jiangsu Yangnong Chemical Co., fying purines and pyrimidines bases by superoxide anion Ltd, China. All other chemicals used were of analytical •- radical (O2 ), hydrogen peroxide (H2O2), and hydroxyl reagent grade. All reagents were purchased from Sigma– • radical (HO ). Aldrich–Fluka (Saint-Quentin France). Cedrelopsis grevei is an endemic species in Madagascar. It was used in traditional medicine to treat malaria, fever In vitro studies and fatigue [13]. Our previous study [14] showed that Plant material and extraction sixty-four components were identified in C. grevei essen- The leaves of C. grevei was collected from Toliara, tial oil by GC-MS. The major constituents were: (E)-β- Madagascar and identified by Benja RAKOTONIRINA farnesene (27.61 %), δ-cadinene (14.48 %), α-copaene (Institut malgache de recherches appliquées (IMRA), (7.65 %) and β -elemene (6.96 %). It exhibited antioxi- Madagascar). A voucher specimen (No. TLR 07) was dant activities and concentration-dependent inhibitory deposited at the Herbarium of IMRA, Madagascar. The • effects on DPPH and ABTS. Previous investigations of leaves of C. grevei collected in Antananarivo, Madagascar the trunk [15] and stem [16] bark of C. grevei showed were ground to fine powder and successively extracted that five coumarins i.e. norbraylin, methyl-O-cedrelopsin, with solvents of increasing polarity (cyclohexane, dichlo- cedrecoumarin A, scoparone and braylin were isolated. romethane, ethyl acetate, ethanol and finally methanol). Atmaca et al. [17] showed that hepatoprotective effect of Thus, 200 g of leaves powder were placed in steeping with coumarins against oxidative stress and liver damage in- cyclohexane (2 L) for 4 h under frequent agitation at duced by carbon tetrachloride in male rats. In addition, ambient temperature and pressure. The mixture was Mossa et al. BMC Complementary and Alternative Medicine (2015) 15:251 Page 3 of 10

then filtrated using Wattman Paper (GF/A, 110 mm). Antioxidant activity ● The solvent was evaporated by rotary evaporation under The free radical scavenging activities by DPPH and • vacuum at 30 °C. Then, the same powder extracted with ABTS + assays were evaluated according to the method cyclohexane was extracted again with the next solvent, di- cited by Afoulous et al. [14]. Total reducing capacity of chloromethane, under the same conditions. The same C. grevei extracts was determined according to the method protocol was applied for ethyl acetate, ethanol and metha- of Oyaizu [23]. One mL of extract at different concentra- nol. Extracts were subsequently stored in sealed and tions (250–1000 μg/ml) were mixed with 2.5 ml phosphate amber vials at 4 °C for the following analysis. buffer (0.2 M, pH 6.6) and 2.5 mL potassium ferricyanide [K3 Fe (CN)6] (1 %). The mixture was incubated at 50 °C for 20 min, and then a portion (2.5 ml) of TCA (10 %) was Photochemical study Determination of phenolic contents added to the mixture, which was centrifuged for 10 min at 1000 x g. The upper layer of solution (2.5 ml) was mixed The phenolic content of each extract was determined by with distilled water (2.5 ml) and 0.5 mL FeCl (0.1 %). Folin-Ciocalteu method [22]. Briefly, a mixture of diluted 3 Then the absorbance was measured at 700 nm. Ascorbic solution (20 μL) of each extract and 100 μL of Folin- acid (0.5-10 μg/ml) was used as the reference compound. Ciocalteu reagent (0.2 N) were prepared. The mixture was incubated for 5 min at room temperature and then 80 μL of sodium carbonate solution (75 g/L in water) In vivo studies was added. The absorbance was read at 765 nm, after Animals and treatments 1 h against water blank. A standard calibration curve was Healthy adult male mice weighing 22.5 ± 1.0 g were ob- plotted using gallic acid (0–300 mg/L). Results were tained from the Animal Breeding House of the National expressed as g of gallic acid equivalents (GAE)/ Kg of dry Research Centre (NRC), Dokki, Giza, Egypt. Animals mass. were housed in clean plastic cages in the laboratory animal room (23 ± 2 °C) on the standard pellet diet Determination of flavonoids contents and tap water ad-libitum, a minimum relative humidity of 40 % and a 12 h dark/light cycle. Mice were allowed to ac- The flavonoids were determined by Dowd method [22]. climate to laboratory conditions for 7 days prior to dosing. A mixture of diluted solution (100 μL) of each extract The experimental work on mice was performed with the and 100 μL of aluminum trichloride (AlCl ) in methanol 3 approval of the Animal Care & Experimental Committee, (2 %) were prepared. The absorbance of the mixture was National Research Centre, Giza, Egypt, and international measured at 415 nm, after 15 min. At the same time, a guidelines for the care and use of laboratory animals. blank sample formed by methanol (100 μL) and extract Animals were randomly divided into six groups (six (100 μL) without AlCl . The results were expressed as g 3 mice each), control, Cyp, extract (150 and 300 mg/kg. of quercetin equivalents (QE)/Kg of dry mass from cali- b.wt), Cyp plus extract (150 or 300 mg/kg. b.wt) groups. bration curve. Cyp and C. grevei extracts were administered by gavages in a fixed volume of 0.1 ml/mice daily for 28 days. Cyp Determination of anthocyanin content was dissolved in corn oil and given via oral route at dose The absorbance of the extract was measured at 510 and 13.8 mg/kg b.wt 1/10 LD50 [24]. C. grevei extracts were 700 nm in buffers at pH 1.0 (hydrochloric acid–potas- dissolved in water and given via oral route at dose 150 sium chloride, 0.2 M) and 4.5 (acetic acid–sodium acetate, and 300 mg/kg b.wt. Dosages were freshly prepared, ad- 1 M) [22] using 96-well plates. 20 μl extract solution was justed weekly for body weight changes, while a control mixed with 180 μl of buffers and the wavelength reading group was received corn oil. was performed after 15 min of incubation. Anthocyanin contents were calculated using a molar extinction coeffi- Sample collections cient (ε) of 29600 (cyanidin-3-glucoside) and absorbance At the end of the experiment, blood samples were drawn of A =[(A – A ) – (A – A ) ]. 510 700 pH1.0 510 700 pH4.5 from the retro-orbital venous plexus of the animals in glass tubes. Within 20 min of blood collection, the sera Determination of tannin content were drawn after centrifugation at 3500 rpm for 10 min Proanthocyanidins reactive to vanillin were analyzed by at 4°C. The sera were kept in a deep freezer (−20 °C) the vanillin method [22]. One milliliter of extract solution for biochemical analysis. Portions of liver from all animals was placed in a test tube together with 2 mL of vanillin in each group were homogenized in 50 mM Tris–HCl (1 % in 7 M H2SO4) in an ice bath and then incubated at buffer (pH 7.4) containing 1.15 % potassium chloride. The 25 °C. After 15 min, the absorbance of the solution was homogenates were centrifuged at 10,000 g for 15 min read at 500 nm. at 4 °C. The collected supernatants were used for the Mossa et al. BMC Complementary and Alternative Medicine (2015) 15:251 Page 4 of 10

Table 1 Solvents extract and yield of Cedrelpsis grevei Solvents Cyclohexane Dichloromethane Acetate ethyl Ethanol Methanol N13 221 Yield (g/100 g) 2.1 4.6 0.44 2.2 0.8 N, number of extractions; C. grevei leaves weights 200 g dry weight; solvent = 2 liter; extraction after 4h of steeping estimation of the activities of CAT, SOD, GST enzymes test for comparison between different treatment groups. and the contents of GSH and LPO. Statistical significance was set at P ≤ 0.05.

Body weights and relative liver weights Results During the experimental period, body weight changes of The highest yield of C. grevei was obtained with the di- male mice were recorded weekly. At the end of treatments, chloromethane extract (4.6 %) of the plant material. the mice were sacrificed by cervical dislocation. Liver of While, the low yield was observed with the ethyl acetate mice was quickly removed, cleaned, weighed and used for extract (0.44 %). This variation in the yields of different biochemical and histological studies. Then, relative weight extracts can be explained by the difference of polarity for of liver was calculated. compounds present in the plant (Table 1). Methanolic and ethanolic extracts of C. grevei were found to have high total phenolic content, 41.67 ± 0.19 and 39.12 ± 0.44 g Biochemical measurements GAE/Kg ES, respectively (Table 2). So, in the present All biochemical measurements (AST, ALT, ALP, LDH, study, the polar extracts containing the highest amount SOD, CAT, GST, MDA, and GSH) were determined using of phenolic (EtOH and MeOH) were selected for the fur- commercial kits in accordance with manufacturers’ in- ther analysis. The ethanol extract is the richest by tannins structions using a spectrophotometer (Shimadzu UV–VIS with 22.14 ± 1.03 g EC/kg ES, followed by methanol ex- Recording 2401, PC, Japan). tract with 12.69 ± 0.63 g EC/Kg ES. The ethanolic and methanol extracts have obtained intermediate values for Histological study anthocyanins (respectively 1.86 ± 0.13 and 1.35 ± 0.08 g Small pieces of liver samples from each group were dis- C3GE/Kg ES) (Table 2), while flavonoids not detected. sected and fixed in 10 % neutral formalin, dehydrated in Figure 1 illustrates a decrease in the concentration of ascending grades of alcohol and embedded in paraffin DPPH radical due to the scavenging ability of C. grevei μ wax. Paraffin sections (5 m thick) were stained for routine extracts. On DPPH radical, EtOH and MeOH had scav- histological study using haematoxylin and eosin (H&E). enging effects with increasing concentration in the range Two slides were prepared for each mice; each slid content of 95–400 μg/mL, and the scavenging effect of EtOH two sections. Ten field areas for each section were selected was lower than MeOH. The IC50 value of ascorbic acid and examined for histopathological changes (x160) under was 4.7 μg/mL. MeOH extract (400 μg/mL) exhibited Michael light microscope. According to [25], the liver the highest inhibition of about 69.45 %, but this is lower − fields were scored as follows: normal appearance ( ), min- in EtOH extract whose percentage of inhibition is 61.27 %. imal cellular disruption in less than 1 % of field area (+), ABTS assay is shown in Fig. 2. C. grevei extracts (100 μg/ mild cellular disruption of 1-30 % of field area (++), mod- mL) exhibited the highest inhibition of 76.81 % and erate cellular disruption of 31-60 % of field area (+++), se- 78.35 % of MeOH and EtOH extracts. As illustrated in vere cell disruption of 61-90 % of field area (++++) and Fig. 3, Fe3+ was transformed to Fe2+ inthepresenceof very severe cellular disruption of 91-100 % of field area MeOH and EtOH extracts of C. grevei to measure the (++++). reductive capability. C. grevei extracts had significant in- hibition of reducing power with increasing concentration Statistical analysis in the range of 0.10-0.125 mg/mL. The results were expressed as means ± S.D. All data were In the present study, the body weight loss was mark- done with the Statistical Package for Social Sciences (SPSS edly observed of mice treated with Cyp compared to the 17.0 for windows). The results were analyzed using one control group (Fig. 4A). The difference between the two way analysis of variance (ANOVA) followed by Duncan’s groups was statistically significant (30.38 g vs. 32.82 g,

Table 2 Chemical composition of C. grevei leaves extracts Extract Phenolics (g GAE/Kg ES) Flavonoids (g QE/Kg ES) Tannins (g CE/Kg ES) Anthocyanins (g C3GE/Kg ES) Ethanol 39.12±0.44b nd 22.14±1.03a 1.86±0.13b Methanol 41.67±0.19a nd 12.69±0.63b 1.35±0.08d Nd: not detected Mossa et al. BMC Complementary and Alternative Medicine (2015) 15:251 Page 5 of 10

Fig. 1 DPPH radical scavenging activity (%) of MeOH and EthOH extracts of C. grevei. Bars represent the mean ± SD

P ≤ 0.05). Co- administration of C. grevei (MeOH extract) resulted in a partial recovery in the above-mentioned pa- at 300 mg/kg b.wt. to mice of Cyp group restored body rameters (AST, ALT, ALP and LDH) in a dose dependent weight to normal range (33.81 g vs. 32.82 g). As shown in manner. The high dose of C. grevei extract (300 mg/kg) Fig. 4B, significant decrease in relative liver weight was ob- was more effective than low dose (150 mg/kg) to attenuate served after treatment of mice with Cyp compared to con- the perturbations in the tested enzymes (Table 3). trol group (5.44 % g vs. 4.88 % g). Co-administration of C. In the liver homogenates of Cyp-treated mice, the ac- grevei at 150 & 300 mg/kg b.wt. with Cyp modulated sig- tivities of SOD, CAT, GST and the level of GSH de- nificantly relative live weight to the normal control value creased significantly (P ≤ 0.05) by 20.5 %, 38.5 %, 18.6 % (4.98 %, 4.91 % vs. 4.88 %). and 30.1 %, respectively, while the level of MDA increased The oral administration of the two tested doses of C. significantly (P ≤ 0.05) by 51.0 % when compared to the grevei extract (150 & 300 mg/kg/day) to normal mice pro- corresponding controls (Table 4). In this study, MDA level duced no changes in all serum biochemical parameters of increased in mice exposed to Cyp. Co-administration of liver compared to the normal control mice (Table 3). In C. grevei extract to Cyp-treated mice resulted in a partial contrast, male mice exposed to Cyp (13.80 mg/kg/day) in- recovery in the above-mentioned parameters in a dose duced a severe hepatic damage in serum enzyme activities dependent manner. The high dose of C.greveiextract of ALT, AST and ALP levels comparable to control (300 mg/kg) was more effective than low dose (150 mg/kg) (Table 3). Also in Cyp-treated mice, serum LDH activity to attenuate the perturbations in the tested enzymes was increased by 37.53 %, while it was decreased by (Table 4). −27.84 % in the liver homogenate. However, co- The representative picutures of histopathological exam- administration of C. grevei extract to Cyp-treated mice ination of the liver tissue are shown in Fig. 5 (A-D) and

Fig. 2 ABTS•+ Scavenging activity (%) of MeOH and EthOH extracts of C. grevei. Bars represent the mean ± SD Mossa et al. BMC Complementary and Alternative Medicine (2015) 15:251 Page 6 of 10

Fig. 3 Reducing power (absorbance at 700 nm) of MeOH and EthOH extracts of C. grevei. Values are mean ± SD

Fig. 4 Body (a) and relative liver (b) weights of rats exposed to cypermethrin (Cyp) and the protective effect of C. grevei (MeOH extract). Bars represent the mean of 6 mice ± SD; values are not sharing superscripts letters (a, b, c, d, e) differ significantly at P ≤ 0.05. Relative liver weight (%) = (liver weight/body weight) X 100. Con: control; Cyp: cypermethrin; Ext 150: 150 mg/kg b.wt of C. grevei extract; Ext 300: 300 mg/kg b.wt of C. grevei extract (MeOH extract) Mossa et al. BMC Complementary and Alternative Medicine (2015) 15:251 Page 7 of 10

Table 3 Effect of C. grevei MeOH extract on SOD, CAT, GST, LPO and GSH in liver of mice exposed to cypermethrin Treatments SOD(U/g tissue)) CAT(U/g tissue) GST(U/g tissue) LPO(nmol/g tissue) GSH(mmol/g tissue) Control 664.20±8.58c 298.68±4.47c 2018.51±63.92c 295.42±8.12a 155.11±3.02c Ext 150 668.74±17.49c 292.59±8.34c 2093.45±31.66c 298.03±9.45a 157.18±2.85c Ext 300 676.82±20.47c 297.95±5.17c 2157.52±32.78c 292.38±7.25a 159.99±3.07c Cyp 527.93±19.46a 183.67±6.38a 1642.73±52.93a 446.19±13.78c 108.48±3.93a Cyp+ Ext 150 609.69±17.84b 272.15±6.78b 1710.54±53.95ab 409.15±20.25c 120.03±6.18a Cyp+ Ext 300 632.40±22.02bc 287.87±9.92bc 1808.70±52.66b 369.50±14.58b 135.57±4.39b Each value is a mean of 6 animals ± SD.; a, b, c, dvalues are not shared superscripts letters (a, b, c, d) differ significantly at P≤ 0.05; Con: control; Cyp: cypermethrin; Ext 150 & Ext 300; 150 and 300 mg/ kg b.wt of C. grevei extract. SOD: superoxide dismutase; CAT: catalase; GST: glutathione-S-transferase; LPO: lipid peroxidation; GSH: glutathione reduced the semi quantitative histological scoring of liver damage study of Paris and Debray [26], which showed that the is presented in Table 5. Liver sections from the control leaves of C. grevei contain 2 % flavonoids as glycosides of group mice showed normal hepatic cytoarchitecture. They flavonols. The ethanol extract is the richest by tannins, formed of hepatocytes radiating from central vein to the followed by methanol extract. In addition, the ethanolic periphery of the lobules (Fig. 5A). Liver lobules of Cyp- and methanol extracts have obtained intermediate values treated mice showed severe degeneration in hepatocytes, for anthocyanins [26]. necroses, dilation of portal vein with inflammatory cell Free radicals are continuously produced in body and infiltration in the portal area (Fig. 5B). Liver sections of cause the oxidation of biomolecules (e.g., protein, amino mice treated with Cyp + Ext 150 (Fig. 5C) showed mild acids, lipid and DNA) which leads to cell injury and death. to moderate dilatation of central vein and ballooning Our results revealed that EtOH and MeOH extract of C. degeneration in surrounding hepatocytes. Whereas, in grevei had DPPH scavenging activity in a concentration cypermethrin + Ext 300 treated mice (Fig. 5C) showed depending manner. C. grevei extracts (100 μg/mL) exhib- only moderate dilatation of portal vein. ited the highest inhibition of ABTS. In reducing power assay, Fe3+ was transformed to Fe2+ in the presence of Discussion MeOH and EtOH extracts of C. grevei to measure the re- The natural antioxidants present in many plants reduce ductive capability. C. grevei extracts had significant inhib- oxidative damage and help prevent mutagenesis, carcino- ition of reducing power with increasing concentration in genesis and aging due to their radical scavenging activities the range of 0.10-0.125 mg/mL. The result obtained by [8, 20]. Therefore, phenolic compounds of plants are very this method confirms those of tests DPPH and ABTS. important because their hydroxyl groups confer anti- In toxicological studies, changes in the body weight and oxidant activity. Results of the present study showed relative organs weights are important criteria for evaluation that phenolics content of methanol and ethanol extract of organ toxicity and were used as a valuable index of of C. grevei were similar quantities. However, no work in insecticide-related organ damage [8]. Liver is the first the literature cited the quantification of phenolic com- organ to face any foreign molecule that is carried through pound of that family. We have not detected flavonoids. portal circulation and it is subjected to most damage. It These results are contradictory to those presented by the has been previously reported that during liver damage

Table 4 Effect of C. grevei MeOH extract on the activity of ALT, AST, ALP and LDH in serum and LDH in liver homogenate of mice exposed to cypermethrin Treatments ALT AST ALP LDH (U/L) (U/L) (U/L) Serum(U/L) Liver(U/g tissue) Control 44.36±0.92a 53.33±0.32a 89.88±1.77a 176.83±6.96a 158.30±11.83b Ext 150 45.23±0.92a 52.82±0.24a 90.97±1.29a 172.87±8.38a 160.55±10.32b Ext 300 43.84±1.02a 53.45±0.57a 88.99±1.80a 178.32±10.05a 161.90±8.98b Cyp 60.62±1.43c 69.39±2.33c 131.45±3.32d 243.21±8.23c 114.23±4.86a Cyp+ Ext 150 55.88±1.26b 63.88±1.32b 113.84±3.05c 222.61±8.95bc 115.04±4.91a Cyp+ Ext 300 53.16±1.18b 60.70±2.23b 103.10±2.25b 203.32±7.88b 133.12±2.92a Each value is a mean of 6 animals ± S.E.; a, b, c, dvalues are not shared superscripts letters (a, b, c, d) differ significantly at P≤ 0.05; Con: control; Cyp: cypermethrin; Ext 150 & Ext 300; 150 and 300 mg/ kg b.wt of C. grevei MeOH extract. ALT: alanine aminotransferases; AST: aspartate aminotransferases; ALP: alkaline phosphatase; LDH: lactate dehydrogenase Mossa et al. BMC Complementary and Alternative Medicine (2015) 15:251 Page 8 of 10

the liver damage [27–29]. It may be attributed to the ef- fect of insecticides on gastrointestinal tract resulting in decreased appetite and absorption of nutrients from gut [30]. Other investigations have reported the reduction in body weight and change in relative organs weights in Cyp-treated rats [31, 32] and in rabbits [33] Cyp treatment induced a severe hepatic damage in serum enzyme activities of ALT, AST, ALP and LDH of male mice. The increase in serum AST, ALT and ALP en- zymes may be due to liver dysfunction and disturbance in the biosynthesis of these enzymes with alteration in the permeability of the liver membrane takes place [34]. The increase in serum LDH activity may be due to the hepato- cellular necrosis leading to leakage of the enzyme into the blood stream [35]. However, co-administration of C. grevei extract to Cyp-treated mice resulted in a partial recovery in the above-mentioned parameters (AST, ALT, ALP and LDH). The high dose of C. grevei extract (300 mg/kg) was more effective than low dose (150 mg/kg) to attenuate the perturbations in the tested enzymes. SOD and CAT are known to play an important role in scavenging ROS. SOD catalyzes the destruction of the Fig. 5 Paraffin sections of liver stained by haematoxylin and eosin (H&E) superoxide radicals to H2O2, while CAT reduces the for histopathological changes. Control group and extract groups [A] H O into water and oxygen to prevent oxidative stress showing the normal histological structure of the central vein (CV) and 2 2 surrounding hepatocytes (H) (x40). Cypermethrin group showing [B] and in maintaining cell homeostasis. In addition, GST play degeneration in hepatocytes and dilation of portal vein with essential role in the detoxification process. In the liver inflammatory cells infiltration (m) in the portal area (x80). Cypermethrin homogenates of Cyp-treated mice, the activities of SOD, plus extract at 150 mg/kg. b.wt. [C] showing dilatation of central vein CAT, GST and the level of GSH decreased significantly, (CV) ballooning degeneration (h) in surrounding hepatocytes (x80). while the level of LPO increased significantly. The change Cypermethrin plus extract at 300 mg/kg. b.wt. [D] showing dilatation of portal vein (x80) in SOD and CAT might be in response to increased oxida- tive stress. Considering that GST are detoxifying enzymes that catalyze the conjugation of a variety of electrophilic there was an observed decrease in antioxidant defenses substrates to the thiol group of GSH, producing less toxic in the liver [26]. In the present study, the body weights forms [36], the significant decrease of GST activity may in- of mice treated with Cyp was markedly less while rela- dicate insufficient detoxification of Cyp in intoxicated tive liver weight was increased compared to the control mice. group. Administration of C. grevei extract at 300 mg/kg Malondialdehyde (MDA) is a major oxidation product b.wt. to Cyp-intoxicated mice restored body weight to of peroxidized polyunsaturated fatty acids and increased normal range and modulated significantly relative live MDA content is an important indicator of lipid peroxida- weight to the normal control value. The decrease in tion (LPO) [37]. In the present study, MDA level increased body weight of Cyp treated mice was considered the re- in mice exposed to Cyp. However, when a condition of sult of direct toxicity and/or indirect toxicity related to oxidative stress strongly establishes, the defense capacities

Table 5 The severity of the reaction in liver tissue of different groups according to the histopathological alterations. Histopathological alterations Treatment Con Ext 150 Ext 300 Cyp Cyp+ Ext 150 Cyp+ Ext 300 Inflammatory cell infiltration in portal area - - - +++ - - Inflammatory cell infiltration in hepatic parenchyma - - - +++ - - Degeneration in hepatocytes - - - ++++ ++ - Congestion in portal vein - - - ++++ - ++ Congestion in central vein - - - +++ + - Con: control; Cyp: cypermethrin; Ext 150 & Ext 300; 150 and 300 mg/ kg b.wt of C. grevei MeOH extract. ++++ Very sever; +++ Sever; ++ moderate; + mild; - nil Mossa et al. BMC Complementary and Alternative Medicine (2015) 15:251 Page 9 of 10

against ROS becomes insufficient, in turn ROS also affect in male mice. C. grevei extract has protective effects on the antioxidant defense mechanisms, reduces the intracel- Cyp-induced lipid peroxidation, oxidative stress and liver lular concentration of GSH, alters the activity of antioxi- damage. C. grevei was found to be an effective antioxi- dant enzymes e.g., SOD & CAT and increase MDA. These dant in many in vitro assays such as DPPH, ABTS and indirectly suggest an increased production of oxygen free reducing power. These results indicated that administra- radicals in mice. Highly reactive oxygen metabolites, espe- tion of C. grevei is useful, easy, and economical to protect cially hydroxyl radicals, act on unsaturated fatty acids humans against pesticides toxicity. The results presented of phospholipid components of membranes to produce here can be considered as the first information on the malondialdehyde, an LPO product [38]. However, co- hepatoprotective and antioxidant properties of C. grevei administration of C. grevei extract to Cyp-treated mice extracts. resulted in a partial recovery in the above-mentioned parameters in a dose dependent manner. Conflict of interests The authors declare that there is no conflict of interests regarding the The histopathological changes in liver showed that ex- publication of this paper. posure of mice to Cyp resulted in degenerative changes in the liver, including degeneration, necrosis, and inflam- Authors’ contributions matory cells in the portal area. The biomarkers of liver AHM and TMH carried out the some in vitro study (reducing power, DPPH and ABTS) and in vivo study on mice (biochemical and histopathological dysfunction corroborated with the histopathological le- studies). AHM and JB drafted the manuscript. MB, EGR, HF and JB carried out sions observed in the current study. These observations the plant extracts and phytochemicals determinations. AHM and JB carried indicated marked changes in the overall histoarchitec- out the final approval of the version to be published. All author’s read and ture of liver in response to Cyp, which could be due to approves the final manuscript. its toxic effects primarily by the generation of reactive Acknowledgments oxygen species causing damage to the various membrane The authors thank the Academy of Scientific Research & Technology (ASRT) components of the cell. Co-administration of C. grevei and National Research Centre, Egypt and IMRCP UMR 5623, Faculty of extract showed marked improvement in its histological Pharmacy, University Paul Sabatier (France), for supporting this study within the IMHOTEP program (Project no. 109/EGY/FR 8–02). In addition, this work structure in comparison to Cyp-treated group alone and was supported by the France Government through a fellowship granted by represented by nil to moderate degree in inflammatory French Embassy in Egypt (Institut Francais d’Egypte). The authors are grateful cell infiltration in portal area, degeneration in hepatocytes to Professor Dr. Adel Mohamed Bakeer Kholoussy, Professor of Pathology, Faculty of Veterinary Medicine, Cairo University, for reading the histopathological and congestion in the portal vein and central vein. sections. The present study has demonstrated that C. grevei exert have a hepatoprotective effect against Cyp-induced Author details 1Environmental Toxicology Research Unit (ETRU), Pesticide Chemistry oxidative damage and hepatotoxicity in mice. Normalized Department, National Research Centre (NRC), 33 El Bohouth Street (former El antioxidant enzymes and reduction in MDA level are Tahrir St.), P.O. 12622, Dokki, Giza, Egypt. 2Faculté de pharmacie de Toulouse, likely to be the major mechanisms by which C. grevei Laboratoire des Interactions Moléculaires et Réactivité Chimique et Photochimique UMR CNRS 5623, Université de Toulouse, Université prevented development of the liver damage. Supporting Paul-Sabatier, 118 route de Narbonne, F-31062 Toulouse, France. 3Laboratoire this hypothesis, we observed significant increases in de Phytochimie et Standardisation, IMRA, BP, 3833 Antananarivo 101, 4 ’ SOD, CAT, and GST enzyme activity and decreases in Antananarivo, Madagascar. Nat Ex Biotech. Bat 7, 55 avenue Louis Breguet, 31400 Toulouse, France. the levels of MDA in liver tissue concomitant with in- significant changes in liver serum dysfunction biomarkers Received: 25 February 2015 Accepted: 23 June 2015 (AST, ALT, ALP and LDH) of Cyp-treated mice by the administration of C grevei especially at high dose (300 mg/ References kg/day). This might be due to the antioxidant activity and 1. Elliot M, Janes NF. Synthetic pyrethroids a new class of insecticides. Chem hydroxyl radical scavenging effect of C. grevei,which Soc Rev. 1978;7:473–05. sported by DPPH, ABTS scavenging activity and redu- 2. Cox C. Insecticide factsheet, Cypermethrin. Journal of Pesticide Reform. 1996;16:15–20. cing power measurements. However, the antioxidant ac- 3. Crawford MJ, Croucher A, Huston DH. Metabolism of cis and trans-cypermethrin tivity of an antioxidant compound have been attributed in rats. Balance and tissue retention study. Journal of Agriculture and Food to various mechanisms, among which are prevention of Chemistry. 1981;29:130–5. 4. Marei AEM, Ruzo LO, Casida JE. Analysis and persistance of permethrin, chain initiation, binding of transition metal ion catalysts, cypermethrin, deltamethrin and fenvalerate in the fat and brain of treated decomposition of peroxides, prevention of continued rats. Journal of Agriculture and Food Chemistry. 1982;30:558–62. hydrogen abstraction, reductive capacity and radical scav- 5. Desi I, Dobronyi I, Varga L. Immuno-, neuro-, and general toxicologic animal studies on a synthetic pyrethroid: cypermethrin. Ecotoxicol Environ Saf. enging [38]. 1986;12:220–32. 6. Baselt R. Disposition of toxic drugs and chemicals in Man, 8th edition. Conclusion Foster City, CA: Biomedical Publications; 2008. p. 389–91. 7. Mansour SA, Mossa AH. Lipid peroxidation and oxidative stress in rat erythrocytes In view of the data of the present study, it can deduce induced by chlorpyrifos and the protective effect of zinc. Pestic Biochem Physiol. that Cyp caused oxidative damage and liver dysfunction 2009;93:34–9. Mossa et al. BMC Complementary and Alternative Medicine (2015) 15:251 Page 10 of 10

8. MossaAT,RefaieAA,RamadanA,BouajilaJ. Amelioration of prallethrin-induced 32. Sangha GK, Kaur K, Khera KS, Singh B. Toxicological effects of cypermethrin oxidative stress and hepatotoxicity in rat by the administration of Origanum on female albino rats. Toxicol Int. 2011;18:5–8. majorana essential oil. Biomed Research International. 2013;2013:1–11. 33. Lakkawar AW, Chattopadhyay SK, Somuanshi R. Experimental cypermethrin 9. Mossa AH, Swelam ES, Mohafrash SMM. Sub-chronic exposure to fipronil toxicity in rabbits- a clinical and patho-anatomical study. Folia Veterinaria. induced oxidative stress, biochemical and histotopathological changes in 2004;48:3–8. the liver and kidney of male albino rats. Toxicology Reports. 2015; http:// 34. De Bruin A. Biochemical toxicology of environmental agents. Amsterdam, dx.doi.org/10.1016/j.toxrep.2015.02.009 North Holland: Elsevier/North Holland Biochemical Press; 1979. p. 457–784. 10. Gutteridge JM, Halliwell B. Antioxidants: molecules, medicines, and myths. 35. Wang X, Zhai W. Cellular and biochemical in bronchoalveolar lavage fluids Biochem Biophys Res Commun. 2010;393:561–4. of rats exposed to fenvalerate. Zhongguo Yaolixue Yu Dulixue Zoghi. 11. Mimić-Oka J, Simić T, Djukanović L, Reljić Z, Davicević Z. Alteration in 1988;2:271–6. plasma antioxidant capacity in various degrees of chronic renal failure. Clin 36. Hayes D, Pulford D. The glutathione S-transferase supergene family: regulation Nephrol. 1999;51:233–41. of GST and the contribution of the isoenzymes to cancer chemoprotection 12. Nice D. Antioxidant based nutraceuticals. In: Yalpani M, editor. New and drug resistance. Crit Rev Biochem Mol Biol. 1995;30:445–600. technologies for healthy foods and nutraceuticals. Shrewsbury: Science 37. Kelly SA, Harvilla KM, Brady TC, Abrano KH, Leveir ED. Oxidative stress in Publishers; 1997. p. 105–23. toxicology: established mammalian and emerging piscine model systems. 13. Mulholland DA, Kotsos M, Mahomed HA, Randrianarivelojosia M. The Environmental Health Perspective. 1998;106:375–84. chemistry of the Ptaeroxylaceae. Nigerian Journal of Natural Products and 38. Mossa AH, Nawwar GA. Free radical scavenging and antiacetylcholinesterase Medicine. 1999;3:13–6. activities of Origanum majorana L. essential oil. Human & Experimental 14. Afoulous S, Ferhout H, Raoelison GE, Valentin A, Moukarzel B, Couderc F, Toxicology. 2011;30:1501–13. et al. Chemical composition and anticancer, antiinflammatory, antioxidant and antimalarial activities of leaves essential oil of Cedrelopsis grevei. Food Chem Toxicol. 2013;56:352–62. 15. Rakotoarison O, Rabenau I, Lobstein A, Um BH, Schott C, Anton R, et al. Vasorelaxing properties and Bio-guided fractionation of Cedrelopsis grevei. Planta Med. 2003;69(2):179–81. 16. Mulholland DA, Kotsos M, Mahomed HA, Koorbanally NA, Randrianarivelojosia M, van Ufford LQ, et al. Coumarins from Cedrelopsis grevei (Ptaeroxylaceae). Phytochemistry. 2002;61(8):919–22. 17. Atmaca M, Bilgin HM, Obay BD, Diken H, Kelle M, Kale E. The hepatoprotective effect of coumarin and coumarin derivates on carbon tetrachloride-induced hepaticinjurybyantioxidativeactivitiesinrats.JPhysiolBiochem. 2011;67(4):569–76. 18. Mossa AH, Heikal TM, Mohafrash SMM, Refaie AA. Antioxidant potential and hepatoprotective activity of Origanum majorana leaves extract against oxidative damage and hepatotoxicity induced by pirimiphos-methyl in male mice. J Appl Sci. 2015;15(1):69–79. 19. Conwell DG, Jones KH, Jiang Z, Lantry LE, Keely PSW, Kohar I, et al. Cytotoxicity of tocopherols and their quinines in drug-sensitive and multidrug-resistant leukemia cells. Lipids. 1998;33:295–301. 20. Cook NC, Samman S. Flavonoids chemistry, metabolism, cardioprotective effects, and dietary sources. J Nutr Biochem. 1996;7:66–76. 21. Bruck R, Hershkoviz R, Lider O, Aeed H, Zaidel L, Matas Z, et al. Inhibition of experimentally-induced liver cirrhosis in rats by a non peptidic mimetic of the extracellular matrix associated Arg-Gly-Asp epitope. J Hepatol. 1996;24:731–8. 22. El Kar C, Ferchichi A, Attia F, Bouajila J. Pomegranate (Punica granatum) juices: chemical composition, micronutrient cations, and antioxidant capacity. J Food Sci. 2011;76:C795–800. 23. Oyaizu M. Studies on product of browning reaction prepared from glucose amine. Jpn J Nutr. 1986;44:07–315. 24. Tomlin CDS. The e-Pesticide Manual, 13thed, CDROM version 3.1. Farnham, UK: The British Crop Protection Councill; 2004. 25. Michael JD. The toxicologist’s pocket handbook. 2nd ed. New York: Informa Healthcare USA, Inc; 2008. p. 44. 26. Paris RR, Debray M. Sur les polyphénols (acides-phénols, Flavonoïdes) des feuilles de deux Méliacées malgaches : Cedrelopsis grevei Baillon et Neobeguea Leroy. Plantes Méd Phyto. 1972;6:311–9. 27. Seven A, Güzel S, Seymen O, Civelek S, Bolayirh M, Uncu M, et al. Effects of vitamin E supplementation on oxidative stress in streptozotocin induced Submit your next manuscript to BioMed Central diabetic rats: investigation of liver and plasma. Yonsei Med J. 2004;45:703–10. and take full advantage of: 28. Abbassy MA, Mossa AH. Haemato-biochemical effects of formulated and technical cypermethrin and deltamethrin insecticides in male rats. J Pharmacol • Convenient online submission Toxicol. 2012;7:312–21. 29. Sankar P, Telang AG, Manimaran A. Protective effect of curcumin on • Thorough peer review cypermethrin-induced oxidative stress in Wistar rats. Experimental Toxicology • No space constraints or color figure charges Pathology. 2012;64:487–93. 30. Venkateshwarlu P, Sharma BJR, Kalakumar B, Reddy KS, Ravikumar P. • Immediate publication on acceptance Comparative evaluation of toxicity of carbaryl, cypermethrin and malathion • Inclusion in PubMed, CAS, Scopus and Google Scholar – of testis in mice. Indian Journal of Toxicology. 1997;4:33 7. • Research which is freely available for redistribution 31. Hussain S, Khan MZ, Khan A, Javed I, Asi MR. Toxico-pathological effects in rats induced by concurrent exposure to aflatoxin and cypermethrin. Toxicon. 2009;53:33–41. Submit your manuscript at www.biomedcentral.com/submit P.09 : S. Afoulous, H. Ferhout, EG. Raoelison, A. Valentin, B. Moukarzel, F. Couderc and J. Bouajila. Helichrysum gymnocephalum essential oil : chemical composition and cytotoxic, antimalarial and antioxidant activities, attribution of the activity origin by correlations. Molecules 2011, 16 (10), 8273- 8291.

Molecules 2011, 16, 8273-8291; doi:10.3390/molecules16108273

OPEN ACCESS molecules ISSN 1420-3049 www.mdpi.com/journal/molecules Article Helichrysum gymnocephalum Essential Oil: Chemical Composition and Cytotoxic, Antimalarial and Antioxidant Activities, Attribution of the Activity Origin by Correlations

Samia Afoulous 1, Hicham Ferhout 2, Emmanuel Guy Raoelison 3, Alexis Valentin 4, Béatrice Moukarzel 4, François Couderc 1 and Jalloul Bouajila 1,*

1 Laboratoire des Interactions Moléculaires et Réactivité Chimique et Photochimique UMR CNRS 5623, University of Toulouse, University of Paul-Sabatier, 118 route de Narbonne, F-31062 Toulouse, France 2 Nat’Ex Biotech. Bat 7, 55 avenue Louis Breguet, 31400 Toulouse, France 3 Laboratoire de Phytochimie et Standardisation, IMRA, BP: 3833 Antananarivo 101, Madagascar 4 Faculté of Pharmacy, University of Toulouse, UMR 152 IRD-UPS Pharma-DEV, University of Paul Sabatier Toulouse 3, 35 chemin des maraîchers, 31062 Toulouse Cedex 9, France

* Author to whom correspondence should be addressed; E-Mail: [email protected]; Tel.: +33-561558333; Fax: +33-561558155.

Received: 19 September 2011; in revised form: 23 September 2011 / Accepted: 23 September 2011 / Published: 29 September 2011

Abstract: Helichrysum gymnocephalum essential oil (EO) was prepared by hydrodistillation of its leaves and characterized by GC-MS and quantified by GC-FID. Twenty three compounds were identified. 1,8-Cineole (47.4%), bicyclosesquiphellandrene (5.6%), γ-curcumene (5.6%), α-amorphene (5.1%) and bicyclogermacrene (5%) were the main components. Our results confirmed the important chemical variability of H. gymnocephalum. The essential oil was tested in vitro for cytotoxic (on human breast cancer cells MCF-7), antimalarial (Plasmodium falciparum: FcB1-Columbia strain, chloroquine-resistant) and antioxidant (ABTS and DPPH assays) activities. H. gymnocephalum EO was found to be

active against MCF-7 cells, with an IC50 of 16 ± 2 mg/L. The essential oil was active

against P. falciparum (IC50 = 25 ± 1 mg/L). However, the essential oil exhibited a poor

antioxidant activity in the DPPH (IC50 value > 1,000 mg/L) and ABTS (IC50 value = 1,487.67 ± 47.70 mg/L) assays. We have reviewed the existing results on the anticancer activity of essential oils on MCF-7 cell line and on their antiplasmodial activity against the P. falciparum. The aim was to establish correlations between the identified compounds and Molecules 2011, 16 8274

their biological activities (antiplasmodial and anticancer). β-Selinene (R² = 0.76), α-terpinolene (R² = 0.88) and aromadendrene (R² = 0.90) presented a higher relationship with the anti-cancer activity. However, only calamenene (R² = 0.70) showed a significant correlation for the antiplasmodial activity.

Keywords: Helichrysum gymnocephalum; GC-MS; antimalarial activity; cytotoxic activity; antioxidant activity

1. Introduction

The flora of Madagascar is particularly rich, with a specific diversity and more importantly, it contains endemic species. In Madagascar, more than 400 species described in other parts of the World, especially in Southern Africa, Western Europe and Australia, and 115 species, all endemic of the large Island, are currently known [1]. The gender Helichrysum (from the Greek "helios" sun and "chrysos" gold) which belongs to Asteraceae family is one of the endemic species. Helichrysum gymnocephalum (Immortal with naked head) is a shrub from 1 to 4 meters in height whose ultimate branches are covered by a white and very fine tomentum. The leaves are lengthily attenuated towards the base in short petioles, covered on the two faces by a tomentum and having three quite visible principal veins in lower part. Flowering lasts from February till May, seeds are also papillate [2]. Some medicinal properties of Helichrysum species have been reported, and various studies have demonstrated the antimicrobial and antiseptic properties of many species from the genus Helichrysum [3-5]. H. gymnocephalum has been traditionally used therapeutically as a tea or syrup prepared from the leaves to treat gingivitis or buccal ulcers [6]. It is also used as ointment for rheumatism or gout by mixing crushed leaves with bacon. The leaves and flowers have properties as diuretics, stimulants and would relieve neuralgia [7]. Analgesic, aphrodisiac, antiseptic, antiscorbutic, deodorant, tonic and anthelmintic applications have been reported for H. gymnocephalum [8,9]. H. gymnocephalum is also used in the manufacture of toothpaste for the local market. The leaves are also used to embalm dead bodies. The essential oil (EO), a complex mixture of compounds, is considered among the most important antimicrobial agents [3] present in these plants, and may also have antioxidant and anti-inflammatory activities. These last years, an overflow of information has appeared on the role of the oxidative stress in the emergence of a certain number of diseases, such as cancers, cardiovascular diseases and degenerative diseases related to ageing, in parallel the possible therapeutic role of antioxidants in these diseases was emphasized [10]. More than a third of the World’s population (about two billion people) lives in malaria-endemic areas. The majority of deaths caused by falciparum malaria occur in sub-Saharan Africa, primarily among children younger than 5 years and pregnant women living in remote rural areas with limited access to health services [11]. Nearly all malaria deaths and a large proportion of morbidity are caused by Plasmodium falciparum. In the last decades, resistance to several antimalarial drugs became widely disseminated, while the cost of effective treatments was prohibitive for the majority of the populations Molecules 2011, 16 8275 in the endemic regions. Thus, rapid development of resistance by P. falciparum to the conventional drugs necessitates search for new antimalarial drugs [11]. Breast cancer is second only to lung cancer as the most common cancer in women. Roughly 180,000 women are diagnosed with this disease each year, of which 44,000 or almost 20% will die [12]. With increased awareness and increased use of routine mammograms, more women are diagnosed in the earlier stages of this disease, at which time a cure may be possible. The disease is more common in women after the age of 40. It is also more frequent in women of a higher social-economic class [12]. Cancer diseases are characterized by an uncontrolled proliferation of cells. They constitute the second cause of mortality behind cardiovascular diseases in developed countries and the third after infectious and cardiovascular diseases in developing countries [13]. The use of plant extracts and derived products in the treatment of cancer is of exceptional value in the control of malignancies, due to the fact that most of the anticancer drugs severely affect the normal cells [13,14]. The aim of the present study was the evaluation of H. gymnocephalum leaves essential oil for its possible antioxidant and biological activities. In this study, we: (i) examined the chemical components of the essential oil by GC-MS and GC-FID; (ii) evaluated their cytotoxic, antimalarial and antioxidant activities and (iii) reviewed researches for essential oils having an activity against P. falciparum and/or on MCF-7 cell line in order to identify, by correlation, the main active compounds.

2. Results and discussion

2.1. Chemical Composition

The essential oil yield of H. Gymnocephalum obtained from hydrodistillation of leaves was 0.40%. Two previous works have quantified the yield of this essential oil. They studied the shoot part of the plant mainly bark and leaves; extraction yields for essential oil were 0.5% and 0.41% according to Mollenbeck et al. [26] and Cavalli et al. [27], respectively. In the present work, only the leaves were used and a similar essential oil yield was obtained. Twenty three components have been identified in H. gymnocephalum essential oil by GC-MS (Figure 1 and Table 1). Sesquiterpene hydrocarbons and oxygenated monoterpenes were the major groups of compounds (32.4% and 54.0%, respectively), while oxygenated sesquiterpenes were not present. The major monoterpenes found were 1,8-cineole (47.4%), p-cymene (4.3%), (E)-β-ocimene (2.4%), 2,3-dihydro-1,8-cineole (2.1%) and α-terpinolene (1,3%). Among the sesquiterpene hydrocarbons bicyclosesquiphellandrene (5.6%), γ-curcumene (5.6%), α-amorphene (5.1%) and bicyclogermacrene (5%) were present in appreciable quantities. On the other hand, this essential oil contained a phenolic compound (2,3-di-tert-butylphenol) at 0.5%. The chemical composition of H. gymnocephalum essential oil has been reported elsewhere. Mollenbeck et al. [26] have shown that 1,8-cineole (66.7%) was the major compound of the essential oil, a level which was higher than in our work (47.7%). Also, in our work, the major compounds were β-caryophyllene (3.3%), γ-terpinene (3.1%), β-pinene (2.7%), α-pinene (2.5%) and ρ-cymene (2.3%). Molecules 2011, 16 8276

Figure 1. Chromatograms GC-MS of H. gymnocephalum essential oil (1: α-thujene; 2: sabinene; 3: β-pinene; 4: 2,3-dihydro-1,8-cineole; 5: α-terpinene; 6: p-cymene; 7: limonene; 8: 1,8-cineole; 9: (E)-β-ocimene; 10: α-terpinolene; 11: α-phellandrene; 12: terpinen-4-ol; 13: α-terpineol; 14: α-copaene; 15: aromadendrene; 16: bicyclo- sesquiphellandrene; 17: γ-curcumene; 18: β-selinene; 19: bicyclogermacrene; 20: α-amorphene; 21: 2,3-di-tert-butylphenol; 22: calamenene; 23: δ-cadinene).

8

15-21 23 7 12 14 22 6 13 9 11 4 5 10 1 23

Table 1. Chemical composition of H. gymnocephalum essential oil. N RI Compounds (%) 1 928 α-Thujene 1.0 2 967 Sabinene 0.3 3 971 β-Pinene 1.1 4 984 2,3-Dihydro-1,8-cineole 2.1 5 1010 α-Terpinene 1.3 6 1018 p-Cymene 4.3 7 1022 Limonene 0.5 8 1026 1,8-Cineole 47.4 9 1052 (E)-β-Ocimene 2.4 10 1084 α-Terpinolene 1.3 11 1164 α-Phellandrene 0.2 12 1175 Terpinen-4-ol 2.7 13 1187 α-Terpineol 1.8 14 1373 α-Copaene 0.4 15 1438 Aromadendrene 2.0 Molecules 2011, 16 8277

Table 1. Cont. N RI Compounds (%) 16 1470 Bicyclosesquiphellandrene 5.6 17 1473 γ-Curcumene 5.6 18 1485 β-Selinene 3.3 19 1494 Bicyclogermacrene 5.0 20 1497 α-Amorphene 5.1 21 1502 2,3-di-tert-butylphenol 0.5 22 1512 Calamenene 1.8 23 1521 δ-Cadinene 3.6 Identified components 99.3 Monoterpene hydrocarbons 8.1 Monoterpenes oxygenated 54.0 Sesquiterpenes hydrocarbons 32.4 Others 4.8

Cavalli et al. [27] have shown that the chemical composition of the essential oil of H. gymnocephalum was dominated by 1,8-cineole (59.7%), p-cymene (6.3%), α-pinene (4.9%), β-pinene (3.3%) and terpinen-4-ol (2.6%). These published compositions of H. gymnocephalum essential oil indicated the presence of oxygenated sesquiterpenes, which were not present in our work. On the other hand, we found more α-thujene (1%) and δ-cadinene (3.6%) compared to Cavalli et al. [27] 0.5% and 0.5%, respectively. For terpinen-4-ol, we obtained 2.6% which was similar to the result of Cavalli et al. [27] (2.7%) but higher than the result of Mollenbeck et al. [26] (1.3%). An intermediate concentration (4.3%) of p-cymene was obtained compared to Cavalli et al. [27] and Mollenbeck et al. [26], with 6.3% and 2.3%, respectively. On the other hand, β-pinene (1.1%) was obtained in the lowest concentration compared to these two studies of Cavalli et al. [27] and Mollenbeck et al. [26] (3% and 2.7%, respectively). We identified some new compounds which were not reported in these two previous studies. The total yield of these new compounds varied between 2.1 and 5.6% (Figure 2). The presence of 2,3-dihydro-1,8-cineole (2.1%), (E)-β-ocimene (2.4%), β-selinene (3.3%), bicyclogermacrene (5.0%), α-amorphene (5.1%), bicyclosesquiphellandrene (5.6%) and γ-curcumene (5.6%) was evidenced.

Figure 2. Structures of new abundant compounds identified in H. gymnocephalum essential oil compared to other studies of same essential oil.

O

(E)-β-ocimene

2,3-dihydro-1,8-cineole Molecules 2011, 16 8278

Figure 2. Cont.

β-selinene bicyclogermacrene

α-amorphene bicyclosesquiphellandrene

γ-curcumene

As shown in Table 1, the differences between our work and previously published results concerned extraction yield, the number of identified compounds and their respective contents. In our investigation, H. gymnocephalum plants were harvested in July 2008, and we treated only leaves for the essential oil extraction. Indeed, H. gymnocephalum was harvested in March 1997 by the team of Mollenbeck et al. [26] and between November–December 1994 by Cavalli et al. [27].

2.2. Antioxidant Activity

Data presented here is the first bibliographical report on the antioxidant activity of H. gymnocephalum. Furthermore, our results (Table 2) demonstrated that the essential oil of + H. gymnocephalum has poor antioxidant activity against DPPH (IC50 value > 1,000 mg/L) and ABTS

(IC50 = 1,487.67 ± 47.70 mg/L). These results may be attributed to a low antioxidant activity (in these two tests) of the compounds identified in the essential oil of H. gymnocephalum. Four other Helichrysum species (H. dasyanthum, H. excisum, H. felinum and H. petiolare) were also reported to exhibit also a low antioxidant activity against test DPPH (IC50 > 100 mg/L) [28].

Table 2. Anticancer, antimalarial and antioxidant activities (IC50 (mg/L)) of H. gymnocephalum essential oil. Anticancer Antiplasmodial Antioxidant activity Antioxidant activity Sample activity activity (DPPH assay) (ABTS assay) Essential oil 16 ± 2 25 ± 1 >10000 1487.67±47.70 Control 0.218 a ± 0.04 0.10 b ± 0.09 3.75 c ± 0.01 1.84 c ± 0.03 a Doxorubicin; b Chloroquine; c Ascorbic acid. Molecules 2011, 16 8279

2.3. Cytotoxic Activity

In recent years, considerable attention has been focused to identifying naturally occurring substances able to inhibit, delay or reverse the process of multistage carcinogenesis. Plant essential oils are believed to reduce the risk of cancer when used in prevention [29]. In this work, the cytotoxic activity of H. gymnocephalum leaf essential oil against human breast cancer cells MCF-7 (IC50 = 16 ± 2 mg/L) was reported for the first time.

Data in Table 3 shows the anti-cancer activity against MCF-7 cells (IC50 mg/L) of several essential oils reported in the literature and their main components. Essential oils were extracted from H. gymnocephalum, Schefflera heptaphylla [30], Laurus nobilis, Origanum syriacum, Origanum vulgare, Salvia triloba [31], Heteropyxis dehniae [32], Schinus molle, Schinus terebenthifolius [33], Salvia officinalis [34], Melaleuca alternifolia [35], Citrus limon, Citrus medica, Citrus sinensis [36] and Talauma gloriensis [37].

Based on the bibliographical review, we can consider that the IC50 of our essential oil (16 mg/L ± 2) is very well positioned among these studied oils in the Table 3. Among the studied essential oils only five showed higher anti-cancer activities than H. gymnocephalum’s essential oil. Indeed, nine essential oils had lower activities for the same test (Table 3).

2.4. Antimalarial Activity

Many studies on the antiplasmodial activity of crude essential oils have been reported [38,39]. The in vitro antiplasmodial activity of H. gymnocephalum essential oil was determined against the FcB1 chloroquine-resistant strain of P. falciparum (Table 2). The antimalarial activity of the essential oil

(IC50 values) was 25 ± 1 mg/L. Since the value of IC50 was found between 5 and 50 mg/L, it can be considered that H. gymnocephalum essential oil has a good activity against P. falciparum [24]. This rather high value of IC50 compared to that of chloroquine (IC50 = 0.1 ± 0.09 mg/L) can be explained by the low concentration of the active compound(s) since the essential oil is a multi-components mixture.

Data in Table 4 summarize the antimalarial activity [IC50 (mg/L)] of all essential oils cited in the literature and their components. These EO have been obtained from H. gymnocephalum, Xylopia phloiodora, Pachypodanthium confine, Antidesma laciniatum, Xylopia aethiopica, Hexalobus crispiflorus[40], Salvia stenophylla, Salvia runcinata, Salvia repens [41], Salvia albicaulis, Salvia dolomitica [42], Lippia multiflora [43], Helichrysum cymosum [44], Artemisia gorgonum Webb [45], Arnica longifolia, Aster hesperius and Chrysothamnus nauseosus [46]. The results based on this bibliographical review (Table 4) showed that the IC50 of our essential oil (25 mg/L ± 1) was at an interesting level, quite well positioned among these studied oils. We found seven essential oils which have higher antimalarial activity than that of H. gymnocephalum essential oil. In addition, 11 essential oils had lower activities in similar tests (Table 4). Molecules 2011, 16 8280

Table 3. Anticancer activity and chemical composition of essential oils (I: Helichrysum gymnocephalum; II: Schefflera heptaphylla [30]; III: Laurus nobilis; IV: Origanum syriacum; V: Origanum vulgare; VI: Salvia triloba [31]; VII: Heteropyxis dehniae [32]; VIII: Schinus molle; IX: Schinus terebenthifolius [33]; X: Salvia officinalis [34]; XI: Melaleuca alternifolia [35]; XII: Citrus limon; XIII: Citrus medica; XIV: Citrus sinensis [36]; XV: Talauma gloriensis [37]). Essential oil I II III IV V VI VII VIII IX

Anticancer activity (IC50 mg/L) 16 ± 2 7.3 101.7 ± 7.9 130 ± 52.2 30.1 ± 1.14 174.3 ± 73.04 150 * 54 ± 10 47 ± 9 Component α-Thujene 1.0 0.62 0.47 0.54 0.38 Sabinene 0.3 6.92 0.35 5.56 0.24 0.02 0.02 β-Pinene 1.0 22.24 4.55 0.6 1.3 8.89 4.96 3.09 2,3-Dihydro-1,8-cineole 2.0 α-Terpinene 1.2 0.7 0.37 4.46 0.44 p-Cymene 4.2 0.74 30.22 3.83 0.34 1.5 2.49 7.34 Limonene 0.5 3.61 2.1 1.68 0.9 1,8-Cineole 47.4 40.91 0.27 45.16 0.4 (E)-β-Ocimene 2.4 α-Terpinolene 1.3 0.35 0.43 α-Phellandrene 0.2 46.52 34.38 Terpinen-4-ol 2.7 1.55 1.09 0.07 0.03 α-Terpineol 1.8 0.62 3.6 8.38 5.6 α-Copaene 0.4 0.6 0.11 0.19 Aromadendrene 2.0 0.49 0.19 Bicyclosesquiphellandrene 5.6 γ-Curcumene 5.6 β-Selinene 3.3 0.3 1.1 Bicyclogermacrene 5.0 1.05 α-Amorphene 5.1 2,3-Di-tert-butylphenol 0.5 Calamenene 1.8 δ-Cadinene 3.6 2.8 0.27 0.69 Molecules 2011, 16 8281

Table 3. Cont. Essential oil X XI XII XIII XIV XV [37]

Anticancer activity (IC50 mg/L) 554.4 ± 1.5 310 * 10 1 0.5 14.1 Component α-Thujene 0.31 Sabinene 0.41 0.1 β-Pinene 2.57 0.91 16.3 3.7 2,3-Dihydro-1,8-cineole α-Terpinene 0.2 5.76 0.1 p-Cymene Limonene 1.7 98.4 56.6 98.4 0.8 1,8-Cineole 17.52 19.29 (E)-β-Ocimene 0.1 α-Terpinolene α-Phellandrene Terpinen-4-ol 1.01 42.62 α-Terpineol 0.27 11.3 α-Copaene 0.1 Aromadendrene Bicyclosesquiphellandrene γ-Curcumene β-Selinene Bicyclogermacrene 2.1 α-Amorphene 2,3-Di-tert-butylphenol Calamenene δ-Cadinene 3.3 Molecules 2011, 16 8282

Table 4. Antipaludic activity and chemical composition of essential oils (I: Helichrysum gymnocephalum; II: Xylopia phloiodora; III: Pachypodanthium confine; IV: Antidesma laciniatum; V: Xylopia aethiopica; VI: Hexalobus crispiflorus [40]; VII: Salvia stenophylla; VIII: Salvia runcinata; IX: Salvia repens [41]; X: Salvia albicaulis; XI: Salvia dolomitica[42]; XII: Lippia multiflora [43]; XIII: Helichrysum cymosum [44]; XIV: Artemisia gorgonum Webb [45]; XV: Arnica longifolia; XVI: Aster hesperius; XVII: Chrysothamnus nauseosus [46];

XVIII: Essential oil [39]). * IC50 was calculated from percentage; na: no activity. Essential oil I II III IV V VI VII VIII IX X XI

Antipalaudic activity (IC50 mg/L) 25 ± 1 17.9 16.6 29.4 17.8 2.0 4.38 ± 1.07 1.23 ± 0.31 1.68 ± 0.26 6.4 ± 2.0 4.8 ± 0.7 Component α-Thujene 1.0 0.59 0.61 Sabinene 0.3 0.59 0.46 0.1 0.2 0.1 β-Pinene 1.0 0.68 10.07 0.12 0.7 0.8 3.0 2,3-Dihydro-1,8-cineole 2.0 α-Terpinene 1.2 0.48 0.43 0.3 0.2 p-Cymene 4.2 0.35 0.32 1.72 0.2 0.7 2.5 0.2 Limonene 0.5 5.3 0.6 9.8 9.4 0.6 1,8-Cineole 47.4 2.0 9.4 (E)-β-Ocimene 2.4 1.93 1.13 0.8 1.5 α-Terpinolene 1.3 α-Phellandrene 0.2 Terpinen-4-ol 2.7 0.16 0.49 0.16 0.8 α-Terpineol 1.8 4.99 2.7 6.2 α-Copaene 0.4 0.53 7.06 2.2 4.07 13.27 0.1 Aromadendrene 2.0 1.08 2 Bicyclosesquiphellandrene 5.6 γ-Curcumene 5.6 β-Selinene 3.3 0.28 2.2 Bicyclogermacrene 5.0 α-Amorphene 5.1 2,3-Di-tert-butylphenol 0.5 Calamenene 1.8 2.32 0.93 1.09 δ-Cadinene 3.6 15.11 8.06 1.3 4.3 10.07 0.5 0.3 Molecules 2011, 16 8283

Table 4. Cont. Essential oil XII XIII XIV XV XVI XVII XVIII

Antipalaudic activity (IC50 mg/L) 2–4 1.25 ± 0.77 5.2 ± 0.77 na na na 40.15 * 205.19 * 10.82 * 72.68 * Component α-Thujene 0.8 0.8 Sabinene 0.3 0.2 β-Pinene 3.7 96.32 2,3-Dihydro-1,8-cineole α-Terpinene 1.2 1.0 p-Cymene 0.2 2.4 0.1 0.1 0.6 93.19 Limonene 0.2 7.2 4.7 92.98 1,8-Cineole 3.1 20.4 0.1 0.1 93.13 (E)-β-Ocimene 3.6 0.3 α-Terpinolene α-Phellandrene 1.8 Terpinen-4-ol 0.2 α-Terpineol 0.3 2.6 0.2 0.2 0.1 0.3 α-Copaene 0.3 1.2 0.2 Aromadendrene 1.5 Bicyclosesquiphellandrene γ-Curcumene 0.5 β-Selinene 0.3 Bicyclogermacrene α-Amorphene 2,3-Di-tert-butylphenol Calamenene 0.1 0.2 δ-Cadinene 0.2 0.4 0.8 Molecules 2011, 16 8284

To study the role of the various components of an essential oil in the biological activities obtained, we performed a complete survey of the activities and the percentage of compounds present in order to assign the origin of the activity. The aim of the study was to find correlations between each component present in our literature search of all essential oils of various plants tested for biological activities (anticancer and antimalarial). Correlations between our essential oil, other oils and their components are listed in Tables 3 and 4, with the IC50 on the tested biological target (MCF-7 cell line or P. falciparum).

2.5. Cytotoxic Activity Correlations

We established correlations between compound contents and anti-cancer activity against the MCF-7 cells. Aromadendrene, α-terpinolene and β-selinene showed good correlations, respectively R² = 0.90, 0.88 and 0.76 (chemical structures are presented in Figure 3).

Figure 3. Structures of compounds which have good correlation for anticancer (a–c) and antimalarial (d) activities.

α-terpinolene (a) aromadendrene (b)

β-selinene (c) calamenene (d)

To our knowledge, no literature has cited any anticancer activity of these three compounds (β-selinene, aromadendrene and α-terpinolene). We will carry out a deeper study to focus on their specific biological activity against MCF-7 cell line. Based on this correlation study, we noticed that six components were present only in our essential oil, namely bicyclosesquiphellandrene (5.6%), α-amorphene (5.1%), 2,3-dihydro-1,8-cineole (2%), limonene (0.5%), and 2,3-di-tert-butylphenol (0.5%). In addition, for bicyclogermacrene (5.0%) and (E)-β-ocimene (2.4%), the correlation has not been established because we did not have enough data (only two essential oils). A deep study of the anticancer activity of aromadendrene, α-terpinolene, β-selinene, bicyclo- sesquiphellandrene, α-amorphene, 2,3-dihydro-1,8-cineole, calamenene, 2,3-di-tert-butylphenol, bicyclogermacrene and (E)-β-ocimene will have to be carried out to determine the origin of the anticancer activity of H. gymnocephalum. On the same target, we should evaluate the synergy between these compounds and their own contribution to the final anticancer activity. Molecules 2011, 16 8285

2.6. Correlations for Antimalarial Activity

We established correlations between both major and minor components and IC50 (antimalarial activity). Calamenene showed the highest correlation R2 = 0.7 (Figure 3). Our results are in accordance with those of Van Zyl et al. [39]; the former showed that p-cymene, limonene and β-pinene had low antiplasmodial activity (IC50 = 205.20, 72 and 40.16 mg/L, respectively). Indeed, they also showed that there are significant synergies (the sum of the fractional inhibitory concentrations (∑FIC) < 0.5) between p-cymene and other compounds such as E- and Z-(±)-nerolidol, carvacrol and γ-terpinene (0.09, 0.02 and 0.37, respectively). We noticed that bicyclosesquiphellandrene (5.6%), α-amorphene (5.1%), bicyclogermacrene (5.0%), 2,3-dihydro-1,8-cineole (2.0%), and 2,3-di-tert-butyl-phenol (0.5%) were present only in our essential oil. It is possible that these compounds may be involved in the activity of our essential oil.

3. Experimental

3.1. Extraction of the Essential Oil

The leaves of H. gymnocephalum were collected in Antananarivo, Madagascar (July 2008). Steam- distillation was used to extract the essential oil according to the European Pharmacopoeia protocol [15]. The essential oil was dried with anhydrous sodium sulphate, filtered and stored in sealed vials at 4 °C prior to analyses.

3.2. Chemicals

All chemicals used were of analytical reagent grade. All reagents were purchased from Sigma- Aldrich-Fluka (Saint-Quentin, France).

3.3. Gas Chromatography and Gas Chromatography-Mass Spectrometry

Quantitative and qualitative analysis of the essential oil was carried out by gas chromatography- flame ionization detection (GC-FID) and gas chromatography-mass spectrometry (GC-MS). Gas chromatography analyses were carried out on a Varian Star 3400 Cx chromatograph (Les Ulis, France) fitted with a fused silica capillary DB-5MS column (5% phenylmethylpolysyloxane, 30 m × 0.25 mm, film thickness 0.25 µm). Chromatographic conditions were 60 °C to 260 °C temperature rise with a gradient of 5 °C/min and 15 min isotherm at 260 °C. A second gradient was applied to 340 °C at 40 °C/min. Total analysis time was 57 min. For analysis purposes, the essential oil was dissolved in petroleum ether. One microliter of sample was injected in the split mode ratio of 1:10. Helium (purity 99.999%) was used as carrier gas at 1 mL/min. The injector was operated at 200 °C. The mass spectrometer (Varian Saturn GC/MS/MS 4D) was adjusted for an emission current of 10 µA and electron multiplier voltage between 1,400 and 1,500 V. Trap temperature was 150 °C and that of the transfer line was 170 °C. Mass scanning was from 40 to 650 amu. Compounds were identified by comparison of their Kovats indices (KI) obtained on a nonpolar

DB-5MS column relative to C5-C24 n-alkanes, with those provided in the literature, by comparison of their mass spectra with those recorded in NIST 08 (National Institute of Standards and Technology) Molecules 2011, 16 8286 and reported in published articles and by co-injection of available reference compounds. The samples were analyzed in duplicate. The percentage composition of the essential oil was computed by the normalization method from the GC peak areas, assuming identical mass response factors for all compounds. Results were calculated as mean values of two injections from essential oil, without using correction factors. All determinations were performed in triplicate and averaged.

3.4. Antioxidant Activity

Free Radical Scavenging Activity: DPPH Test

Antioxidant scavenging activity was studied using the 1,1-diphenyl-2-picrylhydrazyl free radical (DPPH) assay as described by Blois [16] with some modifications. Various dilutions of EO (1.5 mL) were mixed with a 0.2 mmol/L methanolic DPPH solution (1.5 mL). After an incubation period of 30 min at 25 °C, the absorbance at 520 nm (the wavelength of maximum absorbance of DPPH) were recorded as A(sample), using a Helios spectrophotometer (Unicam, Cambridge, UK). A blank experiment was also carried out applying the same procedure to a solution without the test material and the absorbance was recorded as A(blank). The free radical-scavenging activity of each solution was then calculated as percent inhibition according to the following equation:

% inhibition = 100 (A(blank) − A(sample)) / A(blank)

Antioxidant activities of test compounds or the essential oil were expressed as IC50 values, defined as the concentration of the test material required to cause a 50% decrease in initial DPPH concentration. Ascorbic acid was used as a standard. All measurements were performed in triplicate.

3.5. ABTS Radical-Scavenging Assay

The radical scavenging capacity of the samples for the ABTS (2,2′-azinobis-3-ethylbenzothiazoline- 6-sulphonate) radical cation was determined as described by Re et al. [17] with some modifications.

ABTS was generated by mixing a 7 mmol/L solution of ABTS at pH 7.4 (5 mmol/L NaH2PO4,

5 mmol/L Na2HPO4 and 154 mmol/L NaCl) with 2.5 mmol/L potassium persulfate (final concentration) followed by storage in the dark at room temperature for 16 h before use. The mixture was diluted with ethanol to give an absorbance of 0.70 ± 0.02 units at 734 nm using a spectrophotometer (Helios). For samples, solutions of the essential oil in methanol (100 μL) were allowed to react with fresh ABTS solution (900 μL), and then the absorbance was measured 6 min after initial mixing. Ascorbic acid was used as a standard and the capacity of free radical scavenging was expressed by IC50 (mg/L). IC50 values were calculated as the concentration required for scavenging 50% of ABTS radicals. The capacity of free radical scavenging (IC50) was determined using the same previously used equation for the DPPH method. All measurements were performed in triplicate. All data of antioxidant activity were expressed as means ± standard deviations (SD) of the triplicate measurements. The confidence limits were set at P < 0.05. SD did not exceed 5% for the majority of the values obtained. Molecules 2011, 16 8287

3.6. Antiplasmodial Activity

The chloroquine-resistant FcB1-Columbia strain of Plasmodium falciparum (IC50 for chloroquine: 186 nM) was cultured continuously according to Trager and Jensen [18], with modifications [19]. The

IC50 values for chloroquine were checked every 2 months, and we observed no significant variations. The parasites were maintained in vitro in human red blood cells (O±; EFS; Toulouse, France), diluted to 4% hematocrit in RPMI 1640 medium (Lonza, Emerainville, France) supplemented with 25 mM + Hepes and 30 M NaHCO3 and complemented with 7% human AB serum (EFS). Parasite cultures were synchronized by combination of magnetic enrichment [20] followed by D-sorbitol lysis (5% of D-sorbitol in sterile water) as described by Lambros and Vanderberg [21]. The antimalarial activity of essential oil was evaluated by a radioactive micromethod described elsewhere [22]. Tests were performed in triplicate in 96-well culture plates (TPP) with cultures mostly at ring stages (synchronisation interval, 16 h) at 0.5–1% parasitemia (hematocrit, 1.5%). Parasite culture was incubated with each sample for 48 h. Parasite growth was estimated by [3H]-hypoxanthine (Perkin- Elmer, Courtaboeuf, France) incorporation, which was added to the plates 24 h before freezing. After 48 h incubation, plates were frozen-defrosted and each well was harvested on a glass fiber filter. Incorporated (3H)-hypoxanthine was then determined with a β-counter (1450-Microbeta Trilux, Wallac-Perkin Elmer). The control parasite cultures, free from any sample, was referred to 100% growth.

IC50 were determined graphically in concentration versus percent inhibition curves. Chloroquine diphosphate was used as positive control. The antimalarial activity of sample was expressed by IC50, representing the concentration of drug that induced a 50% parasitaemia decrease compared to the positive control culture referred to as 100% parasitaemia [23]. According to the literature concerning plant antiplasmodial activities a sample is very active if IC50 < 5 mg/L, active if IC50 between 5 and

50 mg/L, weakly active if IC50 between 50 and 100 mg/L and inactive if IC50 > 100 mg/L [24].

3.7. Cytotoxicity Evaluation

Cytotoxicity of sample was estimated on human breast cancer cells (MCF-7). The cells were cultured in the same conditions as those used for P. falciparum, except for the 10% human serum, which was replaced by 10% foetal calf serum (Lonza). For the determination of pure compound activity, cells were distributed in 96-well plates at 3 × 104 cells/well in 100 µL, and then 100 µL of culture medium containing sample at various concentrations were added. Cell growth was estimated by (3H)-hypoxanthine incorporation after 48h incubation exactly as for the P. falciparum assay. The (3H)-hypoxanthine incorporation in the presence of sample was compared with that of control cultures without sample (positive control being doxorubicin) [25].

3.8. Statistical Analysis

All data were expressed as mean ± standard deviation of triplicate measurements. The confidence limits were set at P < 0.05. Standard deviations (SD) did not exceed 5% for the majority of the values obtained. Molecules 2011, 16 8288

4. Conclusions

In conclusion, we have identified all volatile constituents of H. gymnocephalum leaf essential oil and evaluated its anticancer, antimalarial, and antioxidant activities. Our results clearly showed that this essential oil was active against the tumor cell lines MCF-7 and the FcB1 strain of P. falciparum. Based on established correlations, compounds such as α-terpinolene, aromadendrene and β-selinene against MCF-7, could be the best candidates for further analysis. Purification of these compounds is under development to test them separately. In addition, an in depth study could determine the mechanisms by which these compounds exert their biological activities. The results presented here can be considered as the first information on the anticancer and antimalarial properties of H. gymnocephalum. This may also contribute to our knowledge of the genus Helichrysum. These in vitro results provide some scientific validation for the widespread use of plants from this genus in traditional medicine.

References

1. Humbert, H. Flore de Madagascar et des Comores. 189e Famille, Composées (Compositae); Firmin-Didot et Cie: Paris, France, 1962; pp. 339-622. 2. Zechmeister, L.; Herz, W.; Grisebach, H.; Kirby, G.W.; Chang, C.W.J.; Flament, I. The Chemistry of Organic Natural Products; Springer-Verlag: New York, NY, USA, 1979; p. 302. 3. Ramanoelina, A.R.P.; Terrom, G.P.; Bianchini, J.P.; Coulanges, P. Contribution à l’étude de l’action antibactérienne de quelques huiles essentielles de plantes malgaches. Arch. Inst. Past. Madag. 1987, 53, 217-226. 4. De Medici, D.; Pieretti, S.; Salvatore, G.; Nicoletti, M.; Rasoanaivo, P. Chemical analysis of essential oils of malagasy medicinal plants by gas chromatography and NMR spectroscopy. Flav. Fragr. J. 1992, 7, 275-281. 5. Afolayan, A.J.; Meyer, J.J.M. The antimicrobial activity of 3,5,7-trihydroxyflavone isolated from the shoots of Helichrysum aureonitens. J. Ethnopharmacol. 1997, 57, 177-181. 6. Boiteau, P.; Allorge-Boiteau, L. Plantes médicinales de Madagascar; ACCT: Paris, France, 1993; pp. 135-136. 7. Debray, M.; Jacquemin, H.; Razafindrambao, R. Contribution à l’inventaire des plantes médicinales de Madagascar; ORSTOM: Paris, France, 1971; p. 150. 8. Cabanis, Y.; Chabouis, L.; Chabouis, F. Végétaux et groupements végétaux de Madagascar et des Mascareignes; BDPA: Tananarive, Madagascar, 1969; Tome I–IV, p. 1342. 9. Boiteau, P. Médecine traditionnelle et pharmacopée, précis de matière médicale malgache; ACCT: Paris, France, 1986; p. 141. 10. Cavalli, J.F. Caractérisation par CPG/IK, CPG/SM et RMN du Carbone-13 d’huiles essentielles de Madagascar. Ph.D. Thesis. Faculty of Science and Technology, Pascal Paoli University of Corsica: Corse, France, 2002. 11. Guerin, J.P.; Olliaro, P.; Nosten, F.; Druilhe, P.; Laxminarayan, R.; Binka, F.; Kilama, W.L.; Ford, N.; White, N.J. Malaria, current status of control, diagnosis, treatment, and a proposed agenda for research and development. Lancet Infect. Dis. 2002, 2, 564-573. Molecules 2011, 16 8289

12. Greenlee, R.T.; Murray, T.; Bolden, S.; Wingo, P.A. Cancer statistics. Cancer J. Clin. 2000, 50, 7-33. 13. Cordell, G.A.; Beecher, C.W.; Pezzut, J.M. Can ethnopharmacology contribute to development of new anti-cancer? J. Ethnopharmacol. 1991, 32, 117-133. 14. Popoca, J.; Aguilar, A.; Alonso, D.; Villarreal, M.L. Cytotoxic activity of selected plant used as antitumorals in Mexican traditional medicine. J. Ethnopharmacol. 1998, 59, 173-177. 15. European Pharmacopoeia; Maissoneuve SA: Sainte Ruffine, France, 1983. 16. Blois, M.S. Antioxidant determination by use of free radical stable. Nature 1958, 181, 1199-1200. 17. Re, R.; Pellegrini, N.; Proteggente, A.; Pannala, A.; Yang, M.; Rice-Evans, C. Antioxidant activity applying an improved ABTS radical cation decolorization assay. Free Radic. Biol. Med. 1999, 26, 1231-1237. 18. Trager, W.; Jensen, J.B. Human malarial parasites in continuous culture. Science 1976, 193, 673-675. 19. Desoubzdanne, D.; Marcourt, L.; Raux, R.; Chevalley, S.; Dorin, D.; Doerig, C.; Valentin, A.; Ausseil, F.; Debitus, C. Alisiaquinones and alisiaquinol, dual inhibitors of Plasmodium falciparum enzyme targets from a New Caledonian deep water sponge. J. Nat. Prod. 2008, 71, 1189-1192. 20. Ribaut, C.; Berry, A.; Chevalley, S.; Reybier, K.; Morlais, I.; Parzy, D.; Nepveu, F.; Benoit-Vical, F.; Valentin, A. Concentration and purification by magnetic separation of the erythrocytic stages of all human Plasmodium species. Malar. J. 2008, 7, 45. 21. Lambros, C.; Vanderberg, J.P. Synchronization of Plasmodium falciparum erythrocytic stages in culture. J. Parasitol. 1979, 65, 418-420. 22. Benoit-Vical, F.; Valentin, A.; Mallié, M.; Bastide, J.M.; Bessière, J.M. In vitro antimalarial activity and cytotoxicity of Cochlospermum tinctorium and C. planchonii leaf extracts and essential oils. Planta Med. 1999, 65, 378-381. 23. Muñoz, V.; Sauvain, M.; Mollinedo, P.; Callapa, J.; Rojas, I.; Gimenez, A.; Valentin, A.; Mallié, M. Antimalarial activity and cytotoxicity of (−) roemrefidine isolated from the stem bark of Sparattanthelium amazonum. Planta Med. 1999, 65, 448-449. 24. Ouattara, Y.; Sanon, S.; Traoré, Y.; Mahiou, V.; Azas, N.; Sawadogo, L. Antimalarial activity of swartzia madagascarensis desv. (leguminosae), combretum glutinosum Guill. & Perr. (combretaceae) and tinospora bakis miers. (menispermaceae) Burkina Fasso medicinal plants. Afr. J. Trad. Cam. 2006, 3, 75-81. 25. Cachet, N.; Hoakwie, F.; Bertani, S.; Bourdy, G.; Deharo, E.; Stien, D.; Houel, E.; Gornitzka, H.; Fillaux, J.; Chevalley, S.; Valentin, A.; Jullian, V. Antimalarial activity of simalikalactone E, a new quassinoid from Quassia amara L. (Simaroubaceae). Antimicrob. Agents Chemother. 2009, 53, 4393-4398. 26. Mollenbeck, S.; Konig, T.; Schreier, P.; Schwab, W.; Rajaonarivony, J.; Ranarivelo, L. Chemical composition and analyses of enantiomers of essential oils from Madagascar. Flav. Frag. J. 1997, 12, 63-69. 27. Cavalli, J.F.; Ranarivelo, L.; Ratsimbason, M.; Bernardini, A.F.; Casanova, J. Constituents of the essential oil of six Helichrysum species from Madagascar. Flav. Fragr. J. 2001, 116, 253-256. Molecules 2011, 16 8290

28. Lourens, A.C.U.; Reddya, D.; Baser, K.H.C.; Viljoena, A.M.; Van Vuuren, S.F. In vitro biological activity and essential oil composition of four indigenous South African Helichrysum species. J. Ethnopharmacol. 2004, 95, 253-258. 29. Sharma, P.R.; Dilip, M.; Mondhe, D.M.; Muthiah, S.; Pal, H.C.; Ashok, K.; Shahi, A.K.; Ajit, K.; Saxena, A.K.; Qazi, G.N. Anticancer activity of an essential oil from Cymbopogon flexuosus. Chem.-Biol. Inter. 2009, 179, 160-168. 30. Li, Y.L.; Yeung, C.M.; Chiu, L.C.M.; Cen, Y.Z.; Ooi, V.E.C. Chemical composition and antiproliferative activity of essential oil from the leaves of a medicinal herb, Schefflera heptaphylla. Phytother. Res. 2009, 23, 140-142. 31. Al-Kalaldeh, J.Z.; Abu-Dahab, R.; Afifi, F.U. Volatile oil composition and antiproliferative activity of Laurus nobilis, Origanum syriacum, Origanum vulgare, and Salvia triloba against human breast adenocarcinoma cells. Nutr. Res. 2010, 30, 271-278. 32. Sibanda, S.; Chigwada, G.; Poole, M.; Gwebu, E.T.; Noletto, J.A.; Schmidt, J.M.; Rea, A.I; Setzer, W.N. Composition and bioactivity of the leaf essential oil of Heteropyxis dehniae from Zimbabwe. J. Ethnopharmacol. 2004, 92, 107-111. 33. Bendaoud, H.; Romdhane, M.; Souchard, J.P.; Cazaux, S.; Bouajila, J. Chemical composition and anticancer and antioxidant activities of Schinus Molle L. and Schinus Terebinthifolius Raddi Berries essential oils. J. Food Sci. 2010, 75, 466-472. 34. El Hadri, A.; Gómez del Río, M.A.; Sanz, J.; González Coloma, A.; Idaomar, M.; Ozonas, B.R.; González, J.B.; Sánchez Reus, M.I. Cytotoxic activity of α-humulene and trans-caryophyllene from Salvia officinalis in animal and human tumor cells. An. R. Acad. Nac. Farm. 2010, 76, 343-356. 35. Liu, X.; Zu, Y.; Fu, Y.; Yao, L.; Gu, C.; Wang, W.; Efferth, T. Antimicrobial activity and cytotoxicity towards cancer cells of Melaleuca alternifolia (tea tree) oil. Eur. Food Res. Technol. 2009, 229, 247-253. 36. Monajemi, R.; Oryan, S.; Haeri-Roohani, A.; Ghannadi, A.; Jafarian, A. Cytotoxic effects of essential oils of some Iranian citrus peels. Iranian J. Pharm. Res. 2005, 3, 183-187. 37. Haber, W.A.; Agius, B.R.; Stokes, S.L.; Setzer, W.N. Bioactivity and chemical composition of the leaf essential oil of Talauma gloriensis Pittier (Magnoliaceae) from Monteverde, Costa Rica. Rec. Nat. Prod. 2008, 2, 1-5. 38. Lopes, N.P.; Kato, M.J.; De A Andrade, E.H.; Maiaa, J.G.S.; Yoshida, M.; Planchart, A.R.; Katzin, A.M. Antimalarial use of volatile oil from leaves of Virola surinamensis (Rol.) Warb. by Waiǎpi Amazon Indians. J. Ethnopharmacol. 1999, 67, 313-319. 39. Van Zyl, R.L.; Seatlholo, S.T.; Van Vuuren, S.F.; Viljoen, A.M. The biological activity of 20 nature identical essential oil constituents. J. Essential Oil Res. 2006, 18, 129-133. 40. Boyom, F.F.; Ngouana, V.; Amvam Zollo, P.H.; Menut, C.; Bessiere, J.M.; Gut, J.; Rosenthal, P.J. Composition and anti-plasmodial activities of essential oils from some Cameroonian medicinal plants. Photochemistry 2003, 64, 1269-1275. 41. Kamatou, G.P.P.; Viljoen, A.M.; Gono-Bwalya, A.B.; Van Zyl, R.L.; Van Vuuren, S.F.; Lourens, A.C.U.; Baser, K.H.C.; Demirci, B.; Lindsey, K.L.; Van Staden, J.; Steenkamp, P. The in vitro pharmacological activities and a chemical investigation of three South African Salvia species. J. Ethnopharmacol. 2005, 102, 382-390. Molecules 2011, 16 8291

42. Kamatou, G.P.P.; Viljoen, A.M.; Figueiredo, A.C.; Tilney, P.M.; Van Zyl, R.L.; Barroso, J.G.; Pedro, L.G.; Van Vuuren, S.F. Trichomes, essential oil composition and biological activities of Salvia albicaulis Benth. and S. dolomitica Codd, two species from the Cape region of South Africa. S. Afr. J. Bot. 2007, 73, 102-108. 43. Valentin, A.; Pélissier, Y.; Benoit, F.; Marion, C.; Kone, D.; Mallie, M.; Bastide, J.M.; Bessiere, J.M. Composition and antimalarial activity in vitro of volatile components of lippia multiflora. Photochemistry 1995, 40, 1439-1442. 44. Van Vuuren, S.F.; Viljoen, A.M.; Van Zyl, R.L.; Van Heerden, F.R.; Baser, K.H.C. The antimicrobial, antimalarial and toxicity profiles of helihumulone, leaf essential oil and extracts of Helichrysum cymosum (L.) D. Don subsp. Cymosum. S. Afr. J. Bot. 2006, 72, 287-290. 45. Ortet, R.; Thomas, O.P.; Regalado, E.L.; Pino, J.A.; Filippi, J.J.; Fernandez, M.D. Composition and biological properties of the volatile oil of Artemisia gorgonum Webb. Chem. Biodivers. 2010, 7, 1325-1333. 46. Tabanca, N.; Demirci, B.; Crockett, S.L.; Baser, K.H.C.; Wedge, D.E. Chemical composition and antifungal activity of Arnica longifolia, Aster hesperius, and Chrysothamnus nauseosus essential oils. J. Agric. Food Chem. 2007, 55, 8430-8435.

Sample Availability: Samples of the compounds are available from the authors.

© 2011 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/3.0/). P.10 : D. Ben Hassine, D. Khlifi, H. Ferhout, G. Raoelison, J. Bouajila. Chapter 44 : Curry Plant (Helichrysum sp.) Oils dans Essential Oils in Food Preservation, Flavor and Safety. Academic Press, First Edition Elsevier, 2016, 395-403, ISBN : 9780124166417.

Author's personal copy

Provided for non-commercial research and educational use only. Not for reproduction, distribution or commercial use.

This chapter was originally published in the book Essential Oils in Food Preservation, Flavor and Safety. The copy attached is provided by Elsevier for the author's benefit and for the benefit of the author's institution, for non-commercial research, and educational use. This includes without limitation use in instruction at your institution, distribution to specific colleagues, and providing a copy to your institution's administrator.

All other uses, reproduction and distribution, including without limitation commercial reprints, selling or licensing copies or access, or posting on open internet sites, your personal or institution’s website or repository, are prohibited. For exceptions, permission may be sought for such use through Elsevier’s permissions site at: http://www.elsevier.com/locate/permissionusematerial

From Ben Hassine, D., Khlifi, D., Ferhout, H., Raoelison, E., Bouajila, J., 2016. Curry Plant (Helichrysum sp.) Oils. In: Preedy, V.R. (Ed.), Essential Oils in Food Preservation, Flavor and Safety. Academic Press, 395–403. ISBN: 9780124166417 Copyright © 2016 Elsevier Inc. All rights reserved. Academic Press Author's personal copy

Chapter 44

Curry Plant (Helichrysum sp.) Oils

D. Ben Hassine1,2, D. Khlifi3, H. Ferhout4, E.G. Raoelison5, J. Bouajila1 1UMR CNRS 5623, Université Paul-Sabatier, Laboratoire des Interactions Moléculaires et Réactivité Chimique et Photochimique, Faculté de Pharmacie de Toulouse, Toulouse, France; 2IPEST, Laboratoire Matériaux, Molécules et Applications, La Marsa, Tunisia; 3Université de Carthage, Laboratoire d’Ecologie et de Technologie Microbienne, Institut National des Sciences Appliquées et de la Technologie (INSAT), Tunis, Tunisia; 4Nat’Ex Biotech, Toulouse, France; 5IMRA, Laboratoire de Phytochimie et Standardisation, Antananarivo, Madagascar

INTRODUCTION Madagascar’s flora is one of the richest in the world and has many medicinal and aromatic plants. The essential oils that are extracted are products with high added value that could contribute to economic island development. Among these aromatic and medicinal plants, we will focus on the genus Helichrysum. Varietal differences occur worldwide, with more than 600 species (Bigovic et al., 2010). The genus Helichrysum (from the Greek helios, meaning sun, and chrysos, meaning gold) belongs to the Asteraceae family which incudes many species rich in essential oils and aromatic compounds. The Asteraceae family, also known as the Com- positae family, includes 1600–1700 genera and 24,000–30,000 species (Funk et al., 2005). All species are well distinguished by the arrangement of the florets and the fruit. Plants belonging to this family may appear on different forms, as it includes herbs, succulents, lianas, epiphytes, trees, or shrubs. They are located everywhere, except Antarctica (Funk et al., 2005). The uses of several members of the Asteraceae family are in medicine, as ornaments, and for trade. Among species we can cite those of commercial impact as the food crops Lactuca sativa (lettuce), Cichorium intybus (chicory), Cynarascoly- mus (globe artichoke), Smallanthus sonchifolius (yacon), and Helianthus tuberosus (Jerusalem artichoke). Seeds of some species are used for the production of cooking oil such as the seeds of Helianthus annuus (sunflower) and Carthamus tinctorius (safflower) (Gao et al., 2010). Figure 1 illustrates some flowering plants which belong to the Asteraceae family. Several researchers have focused on the chemical composition of the essential oil of some species belonging to this family. These volatile compounds may be applied in edible, medicinal, and herbal plants which make them safe when used in food products. Essential oils and their constituents have been largely employed as flavoring agents in foods and much of them exhibited a large antimicrobial activity (Alzoreky and Nakahara, 2002). Raala et al. (2011) studied the content and composition of the essential oils of five Asteraceae species from Estonia, namely Chamomilla recutita, Chamomilla suaveolens, Matricaria perforate, Anthemistinctoria, and Leucanthemum vul- gare. The oil yields ranged from trace amounts up to 0.2%. The chemical composition was established using gas chro- matography (GC)–flame ionization detector and GC–mass spectrometry (MS) methods. The identified compounds varied significantly from one species to another. The main constituents of the essential oils of the studied Asteraceae species were as follows: C. recutita: bisabolol oxide A (39.4%), bisabolone oxide A (13.9%), (Z)-En-yne-dicycloether (11.5%), bisabo- lol oxide B (9.9%), α-bisabolol (5.6%), and chamazulene (4.7%); C. suaveolens: (Z)-En-yne-dicycloether (37.2%), gera- nylisovalerate (22.9%), (E)-β-farnesene (15.6%); Anthemistinctoria: α-eudesmol (10.2%), γ-cadinol (8.7%), γ-cadinene (4.0%); M. perforata: (Z,Z)-matricaria ester (77.9%), (E)-β-farnesene (3.5%), matricaria ester isomer (3.5%), and mat- ricaria lactone (3.0%); and L. vulgare: (E)-β-farnesene (7.3%), hexadecahydrocyclobuta[1,2:3,4]dicyclooctene (5.3%), decanoic acid (4.9%), and γ-eudesmol (4.5%).

BOTANICAL ASPECTS The genus Helichrysum (Asteraceae family) is abundantly prevalent throughout the world (Europe, Africa, Australia, North America) and is represented by more than 600 species. One hundred and fifteen of them are located in Madagascar; most are endemic, and some are used in folk medicine. Essential oils and extracts are obtained from the whole plant or from several parts of the plant.

Essential Oils in Food Preservation, Flavor and Safety. http://dx.doi.org/10.1016/B978-0-12-416641-7.00044-4 Copyright © 2016 Elsevier Inc. All rights reserved. 395

Essential Oils in Food Preservation, Flavor and Safety, First Edition, 2016, 395-403 Author's personal copy

396 PART | II Named Essential Oils

FIGURE 1 Flowering plants (Asteraceae family) (Funk et al., 2005).

This chapter will study of essential oils of seven Malagasy species, namely Helichrysum bracteiferum (DC) H. Hum- bert, Helichrysum gymnocephalum (DC), H. Humbert, Helichrysum selaginifolium Vig. Humboldt, Helichrysum hypnoides (DC) Vig. Humboldt, Helichrysum cordifolium (DC), and Helichrysum faradifani Sc Ell. We will start with a botanical description of each one, then focus on the chemical composition of their essential oils. The botanical description and embranchment is summarized in Table 1. It is important to note that both species H. bracteiferum and H. gymnocephalum are often confused because of their great morphological similarity. They both share the same common name Rambia- zina, but they can be differentiated according to the size of their leaves. The first, with small leaves is classified as the male and called lahy, while the second, with broader leaves, is classified as the female and called vavy (Rasoanaivo and De La Gorce, 1998).

Essential Oils in Food Preservation, Flavor and Safety, First Edition, 2016, 395-403 Author's personal copy

Curry Plant (Helichrysum sp.) Oils Chapter | 44 397

TABLE 1 Botanical Description of Helichrysum gymnocephalum

The Reign Plant Class Dicotyledons Sub-class Asteridae Order Asterales Family Asteraceae Genus Helichrysum

Species gymnocephalum

Helichrysum bracteiferum (DC) H. Humbert, vernacular name is Rambiazina, is a shrub of 1–3 m high, with ultimate branches covered with dense brown tawny tomentum formed with small fine hairs. Leaves are sessile, attenuate at the base, and small (20–30 × 4–5 mm). Flowers are among two or three, all contained in flower heads, and are all homogamous. The fruits, called Achenes, are papillose (Humbert, 1962). This species is found in several places of Madagascar, particularly the siliceous rock in the middle of the ericoid vegetation peaks in the massif of Tsaratanana (North), the massif of Andran- govalo (East Lake Alaotra), or the massif of Ankaratra (south of Antananarivo). Helichrysum selaginifolium Vig. and Humboldt is a branching shrub ericoid,1–1.5 m in height, or it is a suffrutescent plant of 10–60 cm in places subject to periodic fires. It has very slender twigs (ultimate branches of 1–1.5 mm in diameter), prepared, and tomentose. The leaves are narrowly deltoid-acuminate (long May to September, mm wide and 0.5 mm at the base). Heterogamous flower heads are very small and grouped in globular clusters of 10–20, many of which are female. Flowering lasts from August to October. The achenes are generally papillose. Helichrysum hypnoides (DC) R. Vig. and H. Humboldt is an extremely branching ericoid shrub, 10–80 cm tall, with thin but highly lignified stems (brittle) and covered with a fine tomentum greyish white cobweb blanc. The leaves are ses- sile, lanceolates, and very small (3–5 mm long and 0.5–1.5 mm wide at the base). Heterogamous flower heads are all or most solitary, or two or three on top of twigs or partially close in small terminal clusters of three to 15 and are arranged in heads (several are female). The petals are red or brownish in color, especially at the top. Flowering lasts from March to October. The achenes are papillose. Helichrysum faradifani Sc. Ell., vernacular name ahibalala (grasshoppers’s grass), is a shrub measuring up to 60 cm in height and is abundant on the whole island of Madagascar. It is particularly present in rocky or sandy places. This species is used in traditional medicine. Flowering lasts all year. Helichrysum cordifolium DC, common name Fotsiavadika Tsimanandra, is a very rower suffrutescent plant (0.5–1.2 m high and sometimes up to 2–3 m, supported on the bushes). It has juvenile branches covered with loose tomentum. There are about 15 flowers, one to five of which are female. Flowering lasts all year following localities. It is found in forest edges, in ravines, etc. It is located in the center of the island and there are also a few specimens in the Comoros (probably introduced as a medicinal plant). We will next focus on the botanical description of four species growing in Greece, namely Helichrysum orientale, Helichrysum heldreichii, Helichrysum italicum subsp. Microphyllum, and Helichrysum doeryeri.

l Helichrysum orientale (L.) is a non aromatic species with flat leaves; 20–60 mm at base and 7–10 mm in diameter. l Helichrysum heldreichii Boiss is rare and endemic to West Crete. This species grows on vertical cliffs. The leaves are linear and long (more than 30 mm). l Helichrysum italicum (Roth) G. Don fil is a Mediterranean species. It is a shrub with yellow flowers, with a height of 50–70 cm. This species grows on dry cliffs and sandy soil. H. italicum consists of two subspecies: subsp. microphyllum (Willd.) Nyman (Sardinia, south of Corsica), and subsp. Italicum (Tuscany, Corsica). Some phenotypic differences occur between the subspecies, for example, the height, ramification type, and size of the leaves (Bianchini et al., 2003). l Helichrysum italicum (Roth) G. Don in Loudon ssp. microphyllum (Willd.) Nyman is an aromatic shrub. The leaves are very small (10 mm in length) with revolute margins. This species grows on cliffs and in rocky places of the islands in the Mediterranean region (Roussis et al., 2000). This species is also located in Tuscan Archipelago Islands, in Italy. l Helichrysum doeryeri Rech. Fil is rare and endemic to east Crete. This species grows on mountain cliffs. It is a perennial plant characterized by very short stems and lower leaves. The capitulars are usually large with white outer bracts and form small bundles. Roussis et al. (2000) published the first analysis of essential oils in the Greek Helichrysum species which is described in the section related to the chemical composition of Helichrysum essential oils (see Usage and Applications).

Essential Oils in Food Preservation, Flavor and Safety, First Edition, 2016, 395-403 Author's personal copy

398 PART | II Named Essential Oils

FIGURE 2 Helichrysum italicum essential oil in a clear glass vial.

In addition to Madagascar, Greece, and Italy, other Helichrysum species have been found, such as Helichrysum cymosum (L.), an aromatic perennial herb with yellow flowers and characteristic odors, widely distributed in southern tropical Africa. This species is used in traditional medicine to treat respiratory ailments and wound infections (Van Vurrenet al., 2006). Libya is rich in Helichrysum stoechas species, but there is not sufficient botanical information concerning this species. In Iran, 19 species of the genus Helichrysum were found, eight of them were endemic Helichrysum oligocephalum DC, spread over the entire country, and were most common in the west of Iran. As far as we know, there is only one report describing the phytochemical composition of the essential oil of H. oligocephalum (Ebrahim Sajjadi et al., 2009).

Usage and Applications Several studies have reported that there is great phenotypic variability in different species of the genus Helichrysum. There are also the problems of intra- and interspecific variability, all leading to significant variation in chemical composition. Some studies have analyzed the chemical composition of various Helichrysum species located in several places in the world (Madagascar; Turkey; Italy, Libya, Greece, Iran). The obtained essential oils varied significantly. This variability may be attributed to the climatic conditions, harvested time, location, and geographic conditions. Afoulous et al. (2011) used gymnocephalum leaves harvested in July 2008; however, this plant was harvested in March 1997 by Möllenbeck et al. (1997) and during November–December 1994 by Cavalli et al. (2001). Therefore, it is important to note that despite the significant quantitative changes in H. gymnocephalum essential oil composition, 1,8-cineole remains the major component of Helichrysum essential oil. Roussis et al. (2000) showed that variability in the chemical constituents of the essential oil of Helichrysum species was detected according to the period of anthesis (before and after anthesis). Figure 2 shows a picture of a vial of H. italicum essential oil. 1,8-cineol has known antimicrobial properties (Pattnaik et al., 1997) and may possibly contribute to the antimicrobial effect of the H. cymosum oil. Table 2 shows the major components found in the different tested essential oils. As we mentioned above, essential oils obtained from aerial parts of various Helichrysum species present a mixture of volatile compounds. Their chemical profiles vary with seasons, geographical conditions, plant parts used, the harvesting time, and even method of isolation. These secondary metabolites are known by their biological activities. In fact, many studies focused on the ability of essential oils to inhibit the development of food spoilage organisms, especially food-borne

Essential Oils in Food Preservation, Flavor and Safety, First Edition, 2016, 395-403 Author's personal copy

Curry Plant (Helichrysum sp.) Oils Chapter | 44 399

TABLE 2 Major Compounds of Helichrysum Essential Oil

Helichrysum Species and Localization Major Compounds of Helichrysum Essential Oil References Helichrysum gymnocephalum 1,8-cineole (17.2–14.6%), borneol (16.2–10.4%), (E)-caryophyllene De Medici et al. (1992) (Madagascar) (12.7–9.9%), β-pinene (8.2–6.4%), and eugenol (2.2–13.5%) 1, 8-cineole (20.4–14.1%), 3-ethyl-2,5-dimethylhexan-1,3-diene Theron et al. (1994) (17.2–2, 8%), α-farnesene (12.7–1.8%), β-pinene (3.1–10, 6%), (E)-caryophyllene (1.2–9.5%) and α-humulene (0.6–13.2%) 1, 8-cineole (66.7%) Möllenbeck et al. (1997) 1, 8-cineole (59.7%) Cavalli et al. (2001)

1, 8-cineole (47.4%), bicyclosesquiphellandrenea (5.6%), γ-curcumenea Afoulous et al. (2011) (5.6%), α-amorphenea (5.1%), and bicyclogermacrenea (5%) Helichrysum cymosum (L.) Trans-caryophyllene (27.02%), caryophyllene oxide (7.65%), p-cymene Bougatsos et al. (2004) Less. (7.55%), Δ-3-carene (6.84%), and α-fenchene (6.25%) (Tanzania)

Helichrysum fulgidum (L.) Caryophyllene oxide (12.45%), β-pinene (8.72%), spathulenol (7.88%), Bougatsos et al. (2004) (Tanzania) t-muurolol (7.31%), β-bourbonene (7.11%) and camphor (5.35%)

Helichrysum selaginifolium β-pinene (38.2%), α-pinene (16.3%), caryophyllene (7.5%), 1,8-cineole Cavalli et al. (2001) (Madagascar) (7.1%)

Helichrysum bracteiferum 1,8-Cineole (27.3%), β-pinene (11.9%), α-pinene (5.9%), α-humulene Cavalli et al. (2001) (Madagascar) (10.1%), and (E)-caryophyllene (7.1%) Helichrysum cordifolium (E)-caryophyllene (55.6%) Cavalli et al. (2001) (Madagascar) Helichrysum faradifani (E)-Caryophyllene (34%) and linalool (16.1%) Cavalli et al. (2001) (Madagascar) Helichrysum hypnoides (E)-Caryophyllene (35%) and 1,8-cineole (13.4%) Cavalli et al. (2001) (Madagascar) Helichrysum orientale Nonacosane (11.1%) Roussis et al. (2000) (Greece) Helichrysum heldreichii (E)-caryophyllene (38.5%) and caryophyllene derivatives (a total of Roussis et al. (2000) (Greece) 21.5%)

Helichrysum italicum subsp. β-selinene (17.2%) and γ-curcumene (13.7%) Roussis et al. (2000) microphyllum (Greece) Helichrysum doeryeri Four eudesmol isomers (31.4%) Roussis et al. (2000) (Greece)

Helichrysum cymosum α-pinene (12.4%) and 1,8-cineole (20.4%) Van Vurren et al. (2006) (southern tropical Africa)

Helichrysum cymosum Trans-caryophyllene (27.02%), caryophyllene oxide (7.65%), p-cymene Bougatsos et al. (2004) (Tanzania) (7.55%), Δ-3-carene (6.84%), and α-fenchene (6.25%).

Helichrysum stoechas α-pinene (59%), limonene (16,7%), and α-bisabolol (9.6%) Sobhy and El feky (Libiya) (2007) Helichrysum oligocephalum Thymol (14.4%) Ebrahim Sajjadi et al. (Iran) (2009)

Helichrysum graveolens α-cubebene (10.5%), β-caryophyllene (9.4%), caryophyllene oxide Bagci et al. (2013) (Bieb.) (8.2%) and azulene-octahydro (7.5%) (Turkey) aNewly identified compounds.

Essential Oils in Food Preservation, Flavor and Safety, First Edition, 2016, 395-403 Author's personal copy

400 PART | II Named Essential Oils

TABLE 3 The Application Fields of the Major Components of Helichrysum Essential Oils

Major Components Application References 1,8-cineole Antimicrobial properties against Staphylococcus aureus, Bacillus Ayala-zavala et al. (2009) cereus, Micrococcus luteus, Enterococcus faecalis and aroma characteristics, such as fresh, herb, spice Antibacterial properties in vitro or in food models Burt (2004) Antioxidant activities Wang et al. (2008)

α-pinene Antibacterial properties in vitro or in food models Burt (2004) Antimicrobial agent against Sarcina spp. Holley and Patel (2005) Antioxidant activity Wang et al. (2008) and Dai et al. (2013) Antibacterial, antifungal, antiinflammatory, insecticidal, and anti- Tantaoui-Elaraki et al. (1993) oxidant properties, also traditionally used as a flavoring agent and antimicrobial material in food

β-pinene Antibacterial properties in vitro or in food models Burt (2004) Antioxidant activities Wang et al. (2008) (E)-Caryophyllene Antimicrobial and antioxidant activities Mimika-Dukic et al. (2004)

Trans-caryophyllene Antioxidant activity Legault and Pichette (2007) Antibacterial activity Rahman et al. (2008) thymol Antibacterial properties in vitro or in food models with an impor- Burt (2004) tant minimal inhibitory concentration against pathogens such as Escherichia coli, Salmonella typhimurium, S. aureus, Listeria monocytogenes, and B. cereus

Antimicrobial properties against B. cereus, Clostridium botuli- Ayala-zavala et al. (2009) num, E. faecalis, E. coli, S. aureus, L. monocytogenes, Aspergillus flavus, Aspergillus niger, Penicillium corylophilum, K. pneumoniae, Pseudomonas aeruginosa, S. aureus, Salmonella sp., and aroma characteristics such as spice, citrus, and wood

pathogens (Burt, 2004). Nowadays, there is a great interest in natural products to provide solutions for food protection against microorganisms, agriculture products, preservation of raw materials, and crop products against any chemicals sub- stances that are toxic and damage health (Ayala-Zavala et al., 2009). To have great idea about some biological activities, we developed Table 3 which shows an overview of the fields of application of the major compounds found in Helichrysum essential oils.

Usage and Applications in Food Science Species belonging to the genus Helichrysum have various preservative capacities, mainly antimicrobial and antiseptic effects on food products. Particularly, leaves of H. gymnocephalum have been used as tea (Boiteau and Allorge-Boiteau, 1993). Plants that belong to the genus Helichrysum are often used for food, and the leaves are cooked and eaten in Africa (Mathekga et al., 2000). Figure 3 shows a photo of H. gymnocephalum. The species H. italicum is known as “immortal of Italy”. This plant is very aromatic, and is used in food because of its flavour, close to curry. However, this plant is not used in curry, which is an Indian blend of spices, but its scent is very similar. Its leaves pleasantly perfumed for all kinds of food preparation. In fact, its aroma of curry marries well with south Asian culinary habitudes, and Middle Eastern grilled meats, fresh cheeses, salads. Helichrysum essential oil does not tolerate long cooking. When used with potatoes, it gives a taste of the Maquis to the preparation. The essential oil aroma of several Helichrysum species goes well with garlic dressing. Also, fresh or dried

Essential Oils in Food Preservation, Flavor and Safety, First Edition, 2016, 395-403 Author's personal copy

Curry Plant (Helichrysum sp.) Oils Chapter | 44 401

FIGURE 3 Photo of Helichrysum gymnocephalum. leaves, as well as essential oil, can flavor many savory dishes such as rice, vegetables, poultry, meat, and fish and it goes well with marinades for the barbecue. Mancini et al. (2011) used GC and GC–MS to study the chemical composition of the essential oil of H. italicum (Roth) Don subsp. italicum collected in the National Park of Cilento and Diano Valley, Southern Italy. Forty four com- pounds of 45 constituents were identified in the oil, and the oxygenated sesquiterpenes were preponderant. The essential oil was evaluated for its possible in vitro phytotoxic activity against germination and early radical elongation of radish and garden cress. The essential oil contributed to the significant inhibition of the radical elongation of radish when applied at the highest doses. This example elucidated the use of H. italicum (Roth) Don subsp. Italicum essential oil in agriculture. Bougatos et al. (2004) investigated the chemical composition of two essential oils obtained from the aerial parts of H. cymosum and Helichrysum fulgidum from Tanzania using GC and GC–MS. Sixty-five compounds, representing 92.4% and 88.2%, respectively, of the two oils, were identified. Trans-caryophyllene, caryophyllene oxide, β-pinene, p-cymene, spathulenol, and β-bourbonene were found to be the main components. Moreover, the oils were tested against two Gram-positive bacteria, Staphylococcus aureus (ATCC 25923) and Staphylococcus epidermidis (ATCC 12228), four Gram-negative bacteria Escherichia coli (ATCC 25922), Enterobacter cloacae (ATCC 13047), Klebsiella pneu- moniae (ATCC13883), and Pseudomonas aeruginosa (ATCC 227,853). Essential oils were also tested against the patho- genic fungi Candida albicans (ATCC 10231), Candida tropicalis (ATCC 13801), and Candida glabrata (ATCC 28838). Results showed that the oil of H. fulgidum exhibited significant antimicrobial activity, while the oil of H. cymosum did not exhibit any antimicrobial activity. Afoulous et al. (2011) showed that the essential oil of H. gymnocephalum had poor antioxidant activity against two tests: 2,2-diphenyl-1-picrylhydrazyl (IC50 value >1000 mg/L) and 2,2’-azino-bis(3-ethylbenzothiazoline-6-sulphonic acid) (IC50 = 1487.67 ± 47.70 mg/L). The antimicrobial activity of the essential oils of four Helichrysum species growing in Greece (H. orientale, H. held- reichii, H. italicum subsp. Microphyllum, and H. doeryeri) were investigated. H. italicum subsp. microphyllum exhibited the strongest antibacterial activity against S. aureus, S. epidermidis, P. aeruginosa, Es.coli, and En.cloacae. While only the oil of H. oriental significantly inhibited Klebsiella pneumonia (Roussis et al., 2000). Van Vurren et al. (2006) studied the antimicrobial and toxicity activities of the essential oil of H.cymosum. The anti- microbial activity of the essential oil was determined using the p-iodonitrotetrazolium chloride (INT) microplate method, where the essential oil was tested against 10 pathogens. The result showed that the minimum inhibitory concentration ranged from 1 to 8 mg/mL. However, Bougatsos et al. (2004) reported that essential oil of H. cymosum was inactive against all pathogens tested (six bacteria and three pathogenic fungi).

Essential Oils in Food Preservation, Flavor and Safety, First Edition, 2016, 395-403 Author's personal copy

402 PART | II Named Essential Oils

Sobhy and EL Feky (2007) reported the antimicrobial activity of essential oil of H. cymosum against bacteria and fungi such as S. aureus, S. epidermis, Es. coli, and C. albicans.

SUMMARY POINTS

l Helichrysum essential oil is obtained from the different species of the genus Helichrysum which belongs to the Asteraceae family. l The principal fields of use of Helichrysum essential oil are pharmacology, medicine, and the food industry. l The chemical composition of Helichrysum essential oil varies according its geographical provenance, the period of essen- tial oil isolation, and the parts of the plant used, harvested time, as well as the method of extraction. l The main components of Helichrysum essential oil are 1,8-cineole, borneol, (E)-caryophyllene, β-pinene, eugenol, 3-ethyl- 2,5-diméthylhexan-1,3-diene, α-farnesene, α-humulene, bicyclosesquiphellandrene, γ-curcumene, bicyclogermacrene, α-amorphene, trans-caryophyllene, caryophyllene oxide, p-cymene, spathulenol, β-bourbonene, camphor, α-pinene, lin- alool, and thymol. l Biological activities such as antioxidant, antifungal, and antibacterial activity contribute to the application of Helichrysum essential oil in the food industry and agriculture. l The antimicrobial activity of Helichrysum essential oil has been attributed to 1,8-cineol for both fungi and bacteria. l Helichrysum has many food uses, such as in poultry and salads etc.

REFERENCES Afoulous, S., Ferhout, H., Raoelison, E.G., Valentin, A., Moukarzel, B., Couderc, F., Bouajila, J., 2011. Helichrysum gymnocephalum essential oil: chemi- cal composition and cytotoxic, antimalarial and antioxidant activities, attribution of the activity origin by correlations. Molecules 16 (10), 8273–8291. Alzoreky, N.S., Nakahara, K., 2002. Antimicrobial activity of extracts from some edible plants commonly consumed in Asia. Int. J. Food Microbiol. 80, 223–230. Ayala-Zavala, J.F., González-Aguilar, G.A., Del-Toro-Sánchez, L., 2009. Enhancing safety and aroma appealing of fresh-cut fruits and vegetables using the antimicrobial and aromatic power of essential oils. J. Food Sci. 74 (7), 84–91. Bagci, E., Elkiran, O., Evren, H., 2013. Constituents of the essential oils of Helichrysum graveolens (Bieb.) sweet from Turkey. Asian J. Chem. 25, 7254–7256. Bianchini, A., Tomi, P., Bernardini, A.F., Morelli, L., Flamini, G., Cioni, P.L., 2003. Comparative study of volatile constituents of two Helichrysum itali- cum (Roth) Guss. Don Fil subspecies growing in Corsica (France), Tuscany and Sardinia (Italy). Flavour Fragrance J. 18, 487–491. Bigovic, D., Brankovic, S., Kitic, D., Radenkovic, M., Jankovic, T., Savikin, K., Slavoljub, Z., 2010. Relaxant effect of the ethanol extract of Helichrysum plicatum (Asteraceae) on isolated rat ileum contractions. Molecules 15, 3391–3401. Boiteau, P., Allorge-Boiteau, L., 1993. Plantes médicinales de Madagascar. Accticsn-Karthala, Paris. Bougatsos, C., Ngassapa, O., Runyoro, D.K.B., Chinoua, I.B., 2004. Chemical composition and in vitro antimicrobial activity of the essential oils of two Helichrysum species from Tanzania. Z. Naturforsch. C. 59, 368–372. Burt, S., 2004. Review essential oils: their antibacterial properties and potential applications in foods. Int. J. Food Microbiol. 94, 223–253. Cavalli, J.F., Ranarivelo, L., Ratsimbason, M., Bernardini, A.F., Casanova, J., 2001. Constituents of the essential oil of six Helichrysum species from Madagascar. Flavour Fragrance J. 116, 253–256. Dai, J., Zhu, L., Yang, L., Qiu, J., 2013. Chemical composition, antioxidant and antimicrobial activities of essential oil from Wedelia prostrate. EXCLI J. 12, 479–490. De Medici, D., Pieretti, S., Salvatore, G., Nicoletti, M., Rasoanaivo, P., 1992. Chemical analysis of essential oils of Malagasy medicinal plants by gas chromatography and NMR spectroscopy. Flav. Fragr. J. 7, 275–276. Dukic, N.M., Bozin, B., Sokovic, M., Simin, N., 2004. Antimicrobial and antioxidant activities of Melissa officinalis L. (Lamiaceae) essential oil. J. Agric. Food Chem. 52, 2485–2489. Ebrahim Sajjadi, S., Jafari, A., Naderian, M., 2009. Chemical composition of the essential oil of Helichrysum oligocephalum. Chem. Nat. Comp. 45 (2), 269–271. Funk, V.A., Bayar, R.J., Keeley, S., Chan, R., Watson, L., Gemeinholzer, B., Schilling, E., Panero, J.L., Baldwin, B.G., Garcia-Jacas, N., Susanna, A., Jansen, R.K., 2005. Everywhere but Antarctica: using a supertree to understand the diversity and distribution of the compositae. Biol. Skr. 55, 343–374. Gao, T., Yao, H., Song, J., Zhu, Y., Liu, C., Chen, S., 2010. Evaluating the feasibility of using candidate DNA barcodes in discriminating species of the large Asteraceae family. BMC Evol. Biol. 10 (324), 1–7. Holley, R.A., Patel, D., 2005. Improvement in shelf-life and safety of perishable foods by plant essential oils and smoke antimicrobials. Food Microbiol. 22, 273–292. Humbert, H., 1962. Flore de Madagascar et des Comores. 189e Famille: Composées -Tome II. Museum National d’Histoire Naturelle, Paris.

Essential Oils in Food Preservation, Flavor and Safety, First Edition, 2016, 395-403 Author's personal copy

Curry Plant (Helichrysum sp.) Oils Chapter | 44 403

Legault, J., Pichette, A., 2007. Potentiating effect of β–caryophyllene on anticancer activity of α humulene, isocaryophyllene and paclitaxel. J. Pharm. Pharmacol. 59 (12), 1643–1647. Mancini, E., De Martino, L., Marandino, A., Scognamiglio, M.R., De Feo, V., 2011. Chemical composition and possible in vitro phytotoxic activity of Helichrsyum italicum(Roth) Don ssp. Ital. Mol. 16, 7725–7735. Mathekga, A., Meyer, J.J.M., Horn, M.M., Drewes, S.E., 2000. An acylated phloroglucinol with antimicrobial properties from Helichrysum caespititium. Phytochemistry 53, 93–96. Möllenbeck, S., König, T., Schreier, P., Schwab, W., Rajaonarivony, J., Ranarivelo, L., 1997. Chemical composition and analyses of enantiomers of essen- tial oils from Madagascar. Flavour Fragrance J. 12, 63–69. Pattnaik, S., Subramanyam, V.R., Bapaji, M., Kole, C.R., 1997. Antibacterial and antifungal activity of aromatic constituents of essential oils. Microbios 89, 39–46. Raala, A., Kaura, H., Oravb, A., Araka, E., Kailasb, T., Tallinn, M.M., 2011. Content and composition of essential oils in some Asteraceae species. Proc. Est. Acad. Sci. 60 (1), 55–63. Rahman, M.M., Garvey, M.I., Piddock, L.J.V., Gibbon, S., 2008. Antibacterial terpenes from the oleo-resin of Commiphora molmol (Engl.). Phytother. Res. 22 (10), 1356–1360. Rasoanaivo, P., De La Gorce, P., 1998. Essential oils of economic value in Madagascar: present state of knowledge. Herb. Gram. 43, 31–59. Roussis, V., Tsoukatou, M., Petrakis, P., Chinou, I., Skoula, M., Harborne, J.B., 2000. Volatile constituents of four Helichrysum species grown in Greece. Biochem. Syst. Ecol. 28, 163–175. Sobhy, E.A., El-Feky, S.S., 2007. Chemical constituents and antimicrobial activity of Helichrysum stoechas. Asian J. Plant Sci. 6, 692–695. Tantaoui-Elaraki, A., Lattaoui, N., Errifi, A., Benjilali, B., 1993. Composition and antimicrobial activity of the essential oils ofThymus broussonettii, T. zygis and T. satureioides. J. Essent. Oil Res. 5, 45–53. Theron, E., Holeman, M., Potin-Guatier, M., Pinel, R., 1994. Etudes du vieillissement d’huiles essentielles malgaches riches en 1,8-cineole. Partie I: Helichrysum gymnocephalum- Partie II: Ravensara aromatica. Rivista Ital. EPPOS 76, 33–38. Van Vuuren, S.F., Viljoen, A.M., Van Zyl, R.L., Van Heerden, F.R., Husnu, K., Baser, C., 2006. The antimicrobial, antimalarial and toxicity profiles of helihumulone, leaf essential oil and extracts of Helichrysum cymosum (L.) D. Don subsp. cymosum. S. Afr. J. Bot. 72, 287–290. Wang, W., Wu, N., Zu, Y.G., Fu, Y.J., 2008. Antioxidative activity of Rosmarinus officinalis L. essential oil compared to its main components. Food Chem. 108, 1019–1022.

Essential Oils in Food Preservation, Flavor and Safety, First Edition, 2016, 395-403 P.11 : DJR. Rabehaja, G. Raoelison, H. Ihandriharison, PAR. Ramanoelina, J. Casanova, F. Tomi. Volatile Components from Cymbopogon giganteus (Hochst) Chiov var. madagascariensis (A. Camus). Journal of Essential Oil Bearing Plants 2010, 13 (5), 522-527.

Jeobp 13 (5) 2010 pp 522 - 527 522

ISSN 0972-060X

Volatile Components from Cymbopogon giganteus (Hochst) Chiov var. madagascariensis (A. Camus)

Delphin J.R. Rabehaja 1,3, Guy Raoelison 1, Harilala Ihandriharison 1, Panja A.R. Ramanoelina 2, Joseph Casanova 3 and Félix Tomi 3*

1 Laboratoire d’Analyses d’Huiles Essentielles, Institut Malgache de Recherches Appliquées (IMRA), BP 3833 Antananarivo, 101 Antananarivo, Madagascar 2 Laboratoire des Industries Agricoles et Alimentaires, Ecole Supérieure des Sciences Agronomiques, Université d’Antananarivo, BP 175, 101 Antananarivo, Madagascar 3 Université de Corse-CNRS, UMR 6134 SPE, Equipe Chimie et Biomasse, Route des Sanguinaires, 20000 Ajaccio, France Received 01 February 2010; accepted in revised form 26 June 2010

Abstract: The composition of essential oil isolated by hydrodistillation (Clevenger type apparatus) and by vapour distillation (industrial-scale equipment) from the aerial parts of Cymbopogon giganteus (Hochst) Chiov. var. madagascariensis (A. Camus) growing wild in South-Western Madagascar was investigated by GC (RI), GC-MS and 13C NMR. The composition of the Malgasy oil, dominated by p-menthadienol isomers and limonene, is similar to that of African C. giganteus oils.

Key words: Cymbopogon giganteus var. madagascariensis, Poaceae, Essential oil, 13C NMR, GC-MS, Madagascar.

Introduction: The Cymbopogon genus (Poaceae family) comprises 56 species 1 among which four species, Cymbopogon citratus, C. flexuosus, C. martini and C. giganteus grow wild in Madagascar. More precisely, the Malagasy plant growing in arid south-western Madagascar is Cymbopogon giganteus (Hochst) Chiov var. madagascariensis (A. Camus). This variety differs from the African C. giganteus by its lower height (0.5 - 1.5 m instead of 2 - 3 m). Leaves are glaucous green, linear and flexible, with little rounded or no rounded at the base. Conversely, in African species, leaves are flat and their edges are rough to the touch. Finally, the inflorescences are smaller in diameter (10 - 30 cm instead of 30 - 40 cm) 2,3. Aerial parts of the African C. giganteus produce an essential oil whose composition

*Corresponding author (Felix Tomi) E- mail: < [email protected] > Félix Tomi et al. / Jeobp 13 (5) 2010 pp 522 - 527 523 has been extensively investigated. The major componenents of C. giganteus oils from Benin 4, Burkina Faso 5, Mali 6, Côte d’Ivoire 7 and Cameroon 8 are p-menthane derivatives such as cis- and trans-p-mentha-2,8-dien-1-ol, cis- and trans-p-mentha-1(7),8-dien-2-ol, cis- and trans-isopiperitenol, limonene and carvone. The aim of the present work was to characterise the essential oil of C. giganteus var. madagascariensis from Madagascar and to compare its composition with that of C. giganteus oils from continental Africa.

Experimental Plant material: Aerial parts (leaves, stems) of Cymbopogon giganteus var. madagascariensis were collected in Befoly, Toliara, in the South-West region of Madagascar, in March - July 2009. A voucher specimen was deposited at the herbarium of the Department of Botany, Institut Malgache de Recherches Appliquées (IMRA).

Distillation of volatile components: Aerial parts of C. giganteus var. madaga- scariensis were subjected to hydrodistillation in our laboratories for 3 h using a Clevenger- type apparatus (Samples L1-L4) or by steam distillation for 3h in industrial equipment (Inox Alembic of 2000 L, samples I1-I4). The essential oils were dried over anhydrous sodium sulphate and kept at 4°C prior analysis.

Analytical GC: The GC analysis was carried out with a Clarus 500 Perkin- Elmer Autosystem apparatus equipped with two flame ionisation detectors (FID) and fused capillary columns (50 m x 0.22 mm i.d., film thickness 0.25 μm), BP-1 (polymethylsiloxane) and BP-20 (polyethylene glycol). Carrier gas: helium; linear velocity: 0.8 mL.min-1. The oven temperature was programmed from 60°C to 220°C at 2°C.min-1 and then held iso- thermal (20 min). Injector temperature was 250°C (injection mode: split 1/60). Detector temperature: 250°C. The relative proportions of the essential oil constituents were expressed as percentages obtained by peak area normalization, all relative response factors being taken as one.

GC-MS analysis: The essential oil (sample I1) was analysed with a Perkin-Elmer TurboMass detector (quadrupole), directly coupled to a Perkin-Elmer Autosystem XL, equipped with a fused-silica capillary column (60 m x 0.22 i.d., film thickness 0.25 μm), Rtx- 1 (polydimethylsiloxane). Carrier gas: helium at 1 mL/min; split, 1/80: injection volume, 0.2 μL; injection temperature, 250°C; oven temperature programmed from 60°C to 230°C at 2°C/min and then held isothermal (45 min); Ion source temperature, 150°C; ionisation energy, 70 eV; electron ionisation mass spectra were acquired over the mass range 35-350 Da.

Carbon-13 NMR analysis: NMR spectra were recorded on a Bruker AVANCE 400 Fourier Transform spectrometer operating at 100.13 MHz for 13C, equipped with a 5

mm probe, in deuterated chloroform (CDCl3), with all shifts referred to internal tetramethylsilane (TMS). 13C NMR spectra were recorded with the following parameters: Félix Tomi et al. / Jeobp 13 (5) 2010 pp 522 - 527 524 pulse width (PW), 4 μs (flip angle 45°); acquisition time, 2.7 s for 128 K data table with a spectral width (SW) of 24 000 Hz (240 ppm); digital resolution 0.183 Hz/pt. The number of accumulated scans was 3000 for each sample (50-60 mg of oil in 0.5 mL of CDCl3).

Identification of components: Identification of the individual components was based: (i) on comparison of their GC retention indices (RI) for both apolar and polar columns- calculated by linear interpolation to n-alkanes retention times-, with those of standard compounds; (ii) on computer matching with a laboratory-made (Laboratory “Chimie des Produits Naturels) and commercial mass spectral libraries 9,11 and comparison with spectra of reference compounds or literature data12-14; (iii) by 13C NMR spectroscopy, following a computerized method developed in our laboratories, using a home-made software, by comparison of the chemical shift values of the signal in the oil spectrum with those of reference compounds compiled in a laboratory-built library 15-17. This technique allows the identification of individual components until a content about 0.5 % without any previous purification.

Results and discussion: In order to characterise the essential oil of Cymbopogon giganteus variety madagascariensis from Madagascar, eight oil samples were isolated from plants collected in Befoly (Toliara-Madagascar) during March - July 2009 period. The first four oil samples (L1-L4) were obtained at the laboratory (hydrodistillation in a Clevenger type apparatus), the last four samples (I1-I4) were prepared using an industrial scale equipment (vapour distillation). The yield calculated (w/w on the fresh weight basis) varied substantially from sample to sample with respect to the mode of isolation and the time of harvest: 0.10 to 1.15 %, laboratory scale distillation and 0.05- 0.59 %, industrial scale disti- llation. Integrated analysis of C. giganteus var. madagascariensis was carried out using various techniques recently reviewed 18. All the samples were analyzed using by GC in combination with retention indices on two columns of different polarity and by 13C NMR following a method developed in our laboratories 9,10. One oil sample (I1) was also analysed by GC-MS. The composition of all oil samples was largely dominated by p-menthadienols: trans- p-mentha-1(7),8-dien-2-ol (14.7 - 22.4 %), cis-p-mentha-1(7),8-dien-2-ol (12.3 - 19.0 %), trans-p-mentha-2,8-dien-1-ol (11.2 - 19.0 %) and cis-p-mentha-2,8-dien-1-ol (8.3 - 9.8 %) (Figure 1). Other components were present at appreciable contents: carvone (3.1 - 4.1 %), trans-isopiperitenol (2.3 - 8.9 %), cis-isopiperitenol (2.1 - 3.8 %). The content of trans- carveol (2.5 - 3.2 %) was always higher than that of its cis isomer (0.2 - 0.8 %). Limonene was also among the major components during the period March - June (11.6 - 24.0 %). Its content dropped drastically early July (6.3 % lab samples and 5.1 % industrial samples). The contents of the major components varied slightly for the lab-made samples in comparison with industrial samples. Ascaridole was identified by NMR in various samples. In conclusion, the composition of the essential oil from aerial parts of Cymbopogon giganteus (Hochst) Chiov var. madagascariensis (A. Camus) largely dominated by p- menthadienol isomers is quite similar to that reported for African C. giganteus. This is appreciable information for local producers and purchasers. Félix Tomi et al. / Jeobp 13 (5) 2010 pp 522 - 527 525

trans-p-mentha-1(7),8-dien-2-ol cis-p-mentha-1(7),8-dien-2-ol

trans-p-mentha-2,8-dien-1-ol cis-p-mentha-2,8-dien-1-ol

Figure 1. Structure of p-menthadienols

Acknowledgements: R.D. expresses his gratitude to the French Ministère des Affaires Etrangères for financial support (Project MADES) and the Society SOIMANGA (Rakotomanga Pascal) for her collaboration.

References 1. Mabberley, D.J. (1999). The Plant Book, Cambridge University Press. Cambridge. UK. 2. Bosser, J. (1969). Graminées des pâturages et des cultures à Madagascar. Orstom Ed. Paris. 3. Letouzey, R. (1972). Manuel de Botanique Forestière Afrique Tropicale, Centre Technique Forestier Tropical, Tome 2B, 123-435. 4. Ayedoun, M.A., Moudachirou, M. and Lamaty, G. (1997). Composition chimique des huiles essentielles de deux espèces de Cymbopogon du Bénin exploitables Félix Tomi et al. / Jeobp 13 (5) 2010 pp 522 - 527 526

industriellement. Bioressources, Energies, Développement, Environnement. 8: 4-6. 5. Menut, C., Bessière, J.M., Samate, D., Djibo, A.K., Buchbauer, G. and Schopper, B. (2000). Aromatic Plants of Tropical West Africa. XI Chemical Composition, Antioxidant and Antiradical Properties of the Essential Oils of Three Cymbopogon Species from Burkina Faso. J. Essent. Oil Res, 12: 207-212. 6. Sidibé, L., Chalchat, J.C., Garry, R.P. and Lacombe L. (2001). Aromatic Plants of Mali (IV): Chemical Composition of Essential Oils of Cymbopogon citratus (DC) Stapf and Cymbopogon giganteus (Hochst) Chiov. J. Essent. Oil Res. 13: 110-112. 7. Boti, J.B., Muselli, A., Tomi, F., Koukoua, G., N’Guessan, Y.T., Costa, J. and Casanova, J. (2006). Combined Analysis of Cymbopogon giganteus Chiov. Leaf Oil from Ivory Coast by GC(RI), GC-MS and 13C-NMR. C.R. Chimie, 9: 164-168. 8. Jirovetz, L., Buchbauer, G., Eller, G., Ngassoum, M.B., and Maponmetsem, P.M. (2007). Composition and Antimicrobial Activity of Cymbopogon giganteus (Hochst) Chiov. Essential Flower, Leaf and Stem Oils from Cameroon.. J. Essent. Oil Res. 19: 485-489. 9. McLafferty, F.W. and Stauffer, D.B. (1994). Wiley Registry of Mass Spectral Data. 6th ed. Mass Spectrometry Library Search System Benchtop/PBM, version 3.10d. Palisade CO: Newfield. 10. König, A., Hochmuth, D.H. and Joulain, D. (2001). Terpenoids and Related Constituents of Essential Oils, Library of MassFinder 2.1. Institute of Organic Chemistry, Hamburg. 11. NIST/EPA/NIH, National Institute of Standards and Technology, (1999). (PC Version 17 of the Mass Spectral Library), Perkin-Elmer Corporation, Saint Quentin. 12. Adams, R.P. (2001). Identification of Essential Oil Components by Gas Chromato- graphy/Quadrupole Mass Spectroscopy. Allured Publishing Co, Carol Stream, USA. 13. Joulain, D. and König, W.A. (1998). The Atlas of Spectral Data of Sesquiterpene Hydrocarbons, E.B.-Verlag, Hamburg. 14. McLafferty, W. and Stauffer, D.B. (1988). The Wiley/NBS Registry of Mass Spectral Data (4th edn), Wiley-Interscience, New York. 15. Tomi, F., Bradesi, A., Bighelli, A. and Casanova, J. (1995). Computer-aided identification of indivudial components of essential oils using Carbon-13 NMR spectroscopy. J. Magn. Reson. Anal. 1: 25-34. 16. Tomi, F. and Casanova, J. (2006). 13C-NMR as a tool for identification of individual components of essential oils from Labiatae. A review. Acta Horticulturae. 723: 185- 192. 17. Blanc, M.C., Bradesi, P., Gonçalves, M.J., Salgueiro, L. and Casanova, J. (2006). Essential Oil of Dittrichia viscosa ssp. viscosa: Analysis by 13C-NMR and Antimicrobial Activity. Flavour Fragr. J. 21: 324-332. 18 Bighelli, A. and Casanova, J. (2009). Analytical tools for analysing Cymbopogon oils. Chapter 8 in “Essential Oil Bearing Grasses - The genus Cymbopogon”, Ed, Akhila A, CRC Press, Taylor and Francis, London, 195-221. Table 1. Chemical composition of Cymbopogon giganteus var. madagascariensis essential oils

Samples Compounds RI a RI p L1 L2 L3 L4 I1 I2 I3 I4 Identification p-Cymene 1010 1274 0.3 0.2 0.2 0.2 0.3 0.4 0.4 0.5 RI MS 13C NMR 13 Limonene 1022 1205 24.0 15.3 20.6 6.3 14.5 17.3 11.6 5.1 RI MS C NMR Félix Tomi p-Cymenene 1072 1440 0.3 0.3 0.2 0.3 0.4 0.6 0.6 0.9 RI MS 13C NMR trans-p-Mentha-2,8-dien-1-ol 1103 1632 16.3 15.7 18.3 11.2 16.6 17.9 18.1 19.0 RI MS 13C NMR cis-p-Mentha-2,8-dien-1-ol 1115 1673 8.3 8.4 8.7 8.6 8.6 9.0 9.3 9.8 RI MS 13C NMR 13 trans-Limonene-1,2-oxide 1119 1447 1.1 0.8 0.8 0.5 0.4* 0.4* 0.2* 0.1* RI MS C NMR et al. cis-Limonene-1,2-oxide 1123 1459 0.4 0.3 0.2 0.4 0.4 0.4 0.5 0.6 RI MS 13C NMR trans-p-Mentha-1(7),8-dien-2-ol 1169 1794 14.9* 14.7* 14.9* 22.1* 16.6* 17.3* 19.5* 22.4* RI MS 13 527 C NMR /Jeobp13(5)2010pp522-527 3,9-Oxy-mentha-1,8(10)-diene 1169 1561 1.5* 1.8* 1.9* 1.9* 2.1* 2.1* 2.4* 2.7* RI 13C NMR cis-Dihydroperillaldehyde 1172 1604 0.6 0.8 0.3 0.5 0.3 0.4 0.5 0.4 RI MS 13C NMR trans-Dihydroperillaldehyde* 1175 1608 0.7 1.1 0.9 0.8 0.3 0.2 0.7 0.7 RI 13C NMR trans-Isopiperitenol 1179 1751 2.3 3.8 2.7 8.9 3.2 3.2 3.8 3.9 RI MS 13C NMR cis-Isopiperitenol* 1196 1745 2.2* 2.6* 2.1* 3.8* 2.3* 2.6* 2.7* 2.7* RI 13C NMR trans-Carveol* 1196 1828 2.5* 2.6* 2.7* 3.2* 2.7* 2.8* 3.0* 3.0* RI 13C NMR cis-p-Mentha-1(7),8-dien-2-ol 1203 1886 12.3 13.7 14.9 19.0 15.2 13.7 15.4 16.2 RI MS 13C NMR cis-Carveol 1208 1884 0.2 0.6 0.6 0.8 0.7 0.8 0.6 0.5 RI MS 13C NMR Carvone 1214 1728 3.9 3.9 3.1 4.1 3.7 3.4 3.5 3.6 RI MS 13C NMR Perillaldehyde 1244 1783 0.4 0.4 0.3 0.4 0.3 0.3 0.4 0.4 RI MS 13C NMR Ascaridole 1273 1886 0.1 0.1 0.1 0.2 0.2 0.1 0.1 0.3 RI # 13C NMR Phenylethyl hexanoate 1610 2168 0.3 0.3 0.3 1.0 - 0.3 0.4 0.5 RI MS 13C NMR

Order of elution and percentage are given on apolar column (BP-1) except those with asterisk (polar column) RIa and RIp: retention indices measured on apolar (BP-1) and polar (BP-20) columns L1-L4: oil samples obtained in the laboratory using a Clevenger-type apparatus I1-I4: oil samples isolated using an industrial scale apparatus # MS do not allows to differentiate ascaridole and isoascaridole. P.12 : DJR. Rabehaja, EG. Raoelison, H. Ihandriharison, PA. Ramanoelina, A. Bighelli, J. Cassanova, F. Tomi. Combined analysis by chromatographic and spectroscopic techniques : Composition of the essential oil of ‘Andriambolamena’, a wild aromatic plant from Madagascar. TACL Analytical Chemistry Letters 2014, 4(1), 57-64.

This article was downloaded by: [41.74.26.218] On: 13 June 2014, At: 00:52 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Analytical Chemistry Letters Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/tacl20 Combined Analysis by Chromatographic and Spectroscopic Techniques: Composition of the Essential Oil from “Andriambolamena”, A Wild Aromatic Plant from Madagascar Delphin J.R. Rabehajaab, Guy Raoelisonb, Harilala Ihandriharisonb, Panja A.R. Ramanoelinac, Ange Bighellia, Joseph Casanovaa & Félix Tomia a Université de Corse-CNRS, Equipe Chimie et Biomasse, UMR CNRS 6134, Route des Sanguinaires, 20000 Ajaccio, France b Laboratoire de Contrôle Qualité et Standardisation des Phytomédicaments, Institut Malgache de Recherches Appliquées, B.P. 3833, 101 Antananarivo, Madagascar c Laboratoire des Industries Agricoles et Alimentaires, Ecole Supérieure des Sciences Agronomiques, Université d'Antananarivo 101, B.P. 175, 101 Madagascar Published online: 01 May 2014.

To cite this article: Delphin J.R. Rabehaja, Guy Raoelison, Harilala Ihandriharison, Panja A.R. Ramanoelina, Ange Bighelli, Joseph Casanova & Félix Tomi (2014) Combined Analysis by Chromatographic and Spectroscopic Techniques: Composition of the Essential Oil from “Andriambolamena”, A Wild Aromatic Plant from Madagascar, Analytical Chemistry Letters, 4:1, 57-64, DOI: 10.1080/22297928.2014.905751 To link to this article: http://dx.doi.org/10.1080/22297928.2014.905751

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions TACL 4 (1) 2014 pp 57 - 64 57 ISSN Print: 2229-7928 ISSN Online: 2230-7532

Combined Analysis by Chromatographic and Spectroscopic Techniques: Composition of the Essential Oil from “Andriambolamena”, A Wild Aromatic Plant from Madagascar

Delphin J.R. Rabehaja 1,2, Guy Raoelison 2, Harilala Ihandriharison 2, Panja A.R. Ramanoelina 3, Ange Bighelli 1, Joseph Casanova 1 and Félix Tomi 1*

1 Université de Corse-CNRS, Equipe Chimie et Biomasse, UMR CNRS 6134, Route des Sanguinaires, 20000 Ajaccio, France 2 Laboratoire de Contrôle Qualité et Standardisation des Phytomédicaments, Institut Malgache de Recherches Appliquées, B.P. 3833, 101 Antananarivo, Madagascar 3 Laboratoire des Industries Agricoles et Alimentaires, Ecole Supérieure des Sciences Agronomiques, Université d’Antananarivo 101, B.P. 175, 101 Madagascar Received 17 October 2013; accepted in revised form 12 December 2013

Abstract: “Andriambolamena” is an aromatic plant of the Croton genus, growing wild in Madagascar. The chemical composition of the essential oil isolated from aerial parts is reported for the first time. Analysis has been carried out by combination of chromatographic (CC, GC) and spectroscopic (MS, 13C NMR) techniques. In total, 63 compounds that accounted for 96.6 % of the whole composition have been identified. Essential oil composition was dominated by monoterpenes, limonene (21.6 %) and β-pinene (11.0 %) being the major components, beside 1,8-cineole (8.7 %) and linalool (5.9 %). (E)-β-Caryophyllene (8.1 %) was the main sesquiterpene hydrocarbon while oxygenated sesquiterpenes were mainly represented by epi-α-bisabolol (2.6 %). The combination of chromatographic [CC, GC(RI)] and spectroscopic techniques (GC-MS and 13C NMR) appeared really useful for the identification of minor oxygenated sesquiterpenes and particularly compounds that co-eluted on the chromatographic column and diastereo isomers that exhibited insufficiently differentiated mass spectra and close retention indices. Downloaded by [41.74.26.218] at 00:52 13 June 2014 Key words: Croton sp, chemical composition, 13C NMR, “Andriambolamena”, Madagascar.

Introduction families have been isolated and their structure The genus Croton L. (Euphorbiaceae familly) elucidated: diterpenes, triterpenes, steroids, constitutes one of the largest genera with more alkaloids, proanthocyanidines, flavonoids and than one thousand species. It is characterized by other phenolic substances, coumarins, the production of a wide range of secondary cembranolides, etc. The composition of essential metabolites, particularly alkaloids and terpenoids. oils isolated from Croton species of different Saps and essential oils are widely used in the folk origins (Central Africa, Central and South medicine, in Amazonia 1,2 , in Venezuela and America) has been widely investigated and it Columbia 3 , in China 4 and Viet-Nam 5, in West varied drastically from species to species. and Central Africa 6-8. Individual components of The genus Croton is well represented in solvent extracts belonging to various chemical Madagascar with about two hundred species

*Corresponding author (Felix Tomi) E-mail: < [email protected] > © 2014, Har Krishan Bhalla & Sons Delphin J.R. Rabehaja et al., / TACL 4 (1) 2014 57 - 64 58 growing wild, numerous species being endemic scales. In the folk medicine, this plant is used as to the island. Some Croton species are used in anticonvulsant and it reduces the temperature the folk medicine. For instance, the leaves of C. caused by fever. kimosorum Leandri are used against cough and The aim of the present work was to isolate the plant is considered as an anti-spasmodic 9. essential from aerial parts of Andriambolamena Only two papers concerned the composition of and to give some insight on its composition, in essential oil isolated from Croton species growing order to compare it with other Madagascan wild in Madagascar 10,11: essential oils of Croton species and eventually to i) Leaf and stem oil from C. antanosiensis con- help with the plant identification. The tained mainly monoterpenes, α-pinene (32.8 %), identification by conventional techniques of and β-pinene (16.4 %) being the major compo- various minor components remained uncertain nents after conventional analysis and needed ii) Mono and sesquiterpenes were the major confirmation by NMR. Therefore, the analysis components of leaf and stem oil from C. geayi: of the essential oil was conducted by combination 1,8-cineole (15.7 %), (E)-β-caryophyllene (14.8 of chromatographic and spectroscopic techniques %), α-terpineol (14.1 %) and τ-muurolol (6.6 %); and it will be reported in detail. iii) The chemical composition of C. decaryi leaf oil was dominated by (E)-β-caryophyllene (26.7 Experimental %), α-pinene (21.2 %) and α-humulene (19.0 %) Plant material while stem oil contained mostly α-pinene (26.1 Aerial parts of Andriambolamena were collec- %), borneol (13.3 %), camphene (11.5 %) and ted in Andatabo-Toliara, South-West Madagascar, (E)-β-caryophyllene (8.6 %); in January 2011. The plant has been authenticated iv) Leaves of C. sakamaliensis produced a by Benja Rakotonirina, botanist at the Herbarium sesquiterpene-rich essential oil dominated by (E)- of Institut Malgache de Recherches Appliquées β-caryophyllene (28.3 %) accompanied by its (IMRA), Antananarivo, Madagascar. Voucher oxide (12.5 %) while the major components of specimen was deposited at IMRA. stem oil were monoterpenes, 1,8-cineole (37.9 %) and β-phellandrene (14.7%). Isolation of essential oils v) The essential oil isolated from aerial parts Aerial parts (leaves and stems, 475 g) of of Croton kimosorum Leandri, an endemic Andriambolamena were subjected to hydro- species to Madagascar, contained mainly linalool distillation for 3 h using a Clevenger-type (21.6 %), sabinene (10.4 %), 1,8-cineole (6.3 %), apparatus. They yielded 2.56 g of EO. Downloaded by [41.74.26.218] at 00:52 13 June 2014 β-pinene (6.2 %), (E)-β-caryophyllene (5.9 %), 11 terpinen-4-ol (4.8 %), and geraniol (4.5 %) . Fractionation of the essential oil Although extensive work has been done in the An aliquot (1.00 g) of oil obtained from aerial botanical field (morphology) on the characteri- parts was fractionated by column chromatography zation of croton species and sub-species growing (silica gel, 63-200 μm, 35g) and four fractions wild in Madagascar, the classification of a few were eluted with a gradient of solvents pentane/ plants of Croton family remains uncertain, for diethyl ether from 100/0 to 0/100. instance a species whose vernacular name is, “Andriambolamena”, frequently encountered in GC (FID) and GC-MS analysis association with C. kimosorum. It is a shrub of GC(FID) analyses were performed on a Perkin- 100-150 cm in height; growing on poor soil in Elmer Clarus 500 (FID) equipped with two fused sub-arid zone. Branches have circular section silica capillary columns (50 m x 0.22 mm x 0.25 with grey-light color. Internodes have 2.5 to 3.0 μm film thickness), BP-1 (polydimethyl siloxane) cm length. Branchlets have a principal axe rapidly and BP-20 (polyethylene glycol). The oven defoliated. On the young twigs, petioles (3-6 mm temperature program was: 60-220°C at a rate of length) and leaves are covered with circular 2°C/min and then held isothermal at 220°C for Delphin J.R. Rabehaja et al., / TACL 4 (1) 2014 57 - 64 59 20 min. Helium was the carrier gas at a 1.0 mL. Results and discussion min-1 flow rate. The injector and detector In continuation of our work on the characteri- temperature was 250°C. Samples were injected zation of aromatic and medicinal plants from with a 1:60 split ratio. The relative amounts of Madagascar 11,14, the aim of the present work was individual components were expressed as to investigate the chemical composition of the percentages, obtained by peak area normalization, essential oil isolated from Andriambolamena, a without response factor correction. wild plant of the Croton species, collected in GC-MS was performed on a Perkin-Elmer South-West Madagascar. An oil sample has been TurboMass detector, directly coupled to a Perkin- isolated from aerial parts, using a Clevenger-type Elmer Autosystem XL, equipped with a fused- apparatus, with a yield of 0.54 %, calculated on silica capillary column (60 m x 0.22 mm x 0.25 fresh mass basis (w/w). The essential oil was μm film thickness) of Rtx-1 (polydimethyl- analyzed by combination of chromatographic and siloxane). Helium was the carrier gas at a flow spectroscopic techniques. Indeed, direct analysis of 1.0 mL.min-1; 1:80 split ratio; 0.2 μL injection of the bulk sample by GC(RI), GC-MS and by volume. The oven temperature was programmed 13C NMR, following a computerized method rising from 60° to 230° at 2°C/ min and then held developed in our laboratory 13 allowed the identi- isothermal for 45 min. MS were taken at 70 eV fication of the major components by the three and fragments from 35-350 Da. techniques. The essential oil isolated from aerial parts of Andriambolamena contained mostly 13C NMR analysis monoterpenes (66.2 %) and among these, the Analysis was performed on a Bruker AVANCE monoterpene hydrocarbons predominated (47.9 400 Fourier Transform spectrometer operating at %), limonene (21.6 %) and β-pinene (11.0 %) 100.623 MHz for 13C, equipped with a 5 mm being the major components, beside 1,8-cineole

probe, in CDCl3, with all shifts referred to internal (8.7 %) and linalool (5.9 %), the main oxygenated TMS. 13C NMR spectra were recorded with the monoterpenes (Table 1). Other monoterpenes following parameters: pulse width (PW), 4 μs present at appreciable contents were α-pinene (flip angle 45°); acquisition time, 2.73 s for 128 (2.6 %), sabinene (2.4 %), α-terpineol (2.3 %), K data table with a spectral width (SW) of 22000 myrcene (2.2 %), α-phellandrene (2.1 %), p- Hz (220 ppm). The number of accumulated scans cymene (1.9 %), γ-terpinene (1.4 %), terpinen-4- was 3000 (50 mg of oil or fraction of CC in 0.5 ol (1.2 %) and α-thujene (0.7 %). (E)-β-Caryo-

mL of CDCl3). phyllene (8.1 %) was the main sesquiterpene hydrocarbon beside germacrene-D (2.2 %), Downloaded by [41.74.26.218] at 00:52 13 June 2014 Identification of components bicyclogermacrene (1.8 %), δ-cadinene (1.8 %), Identification of the components was based: (i) α-copaene (1.6 %), α-humulene (1.3 %), allo- on comparison of their GC retention indices (RI) aromadendrene (0.8 %), and (E)-β-farnesene (0.7 on polar and apolar columns, determined relative %). Oxygenated sesquiterpenes (8.8 %) were α to the retention times of C7-C27 n-alkane series mainly represented by epi- -bisabolol (2.6 %) with linear interpolation, with those of authentic beside caryophyllene oxide (0.8 %), τ-cadinol compounds; (ii) on computer search using digital (0.8 %), β-elemol (0.7 %) and spathulenol (0.6 libraries of mass spectral data 11 and comparison %). with published data 12 and (iii) by comparison of Identification of other minor components was the signals in the 13C NMR spectra of essential suggested by GC(RI) and GC-MS. However, oils and all the fractions of chromatography with unambiguous identification required confir- those of reference spectra compiled in the labo- mation by NMR. Moreover, for several peaks no ratory spectral library, with the help of a labora- proposal was made by computer matching against tory-made software 13. Individual components commercial mass spectra libraries. Therefore, the were identified by 13C NMR at contents as low oil sample was fractionated on silica gel column as 0.4 %. chromatography (CC) and the four fractions were Delphin J.R. Rabehaja et al., / TACL 4 (1) 2014 57 - 64 60 Table 1. Volatile components of the essential oil isolated from aerial parts of Andriambolamena (Croton sp.), a wild aromatic plant from Madagascar

No. Components RIa RIp % Identification

1 α-thujene 922 1028 0.7 RI, MS, 13C NMR 2 α-pinene 930 1026 2.6 RI, MS, 13C NMR 3 camphene 943 1066 0.4 RI, MS, 13C NMR 4 1-octen-3-ol 959 1144 0.1 RI, MS, 13C NMR 5 sabinene 965 1124 2.4 RI, MS, 13C NMR 6 β-pinene 970 1114 11.0 RI, MS, 13C NMR 7 myrcene 980 1162 2.2 RI, MS, 13C NMR 8 α-phellandrene 997 1167 2.1 RI, MS, 13C NMR 9 α-terpinene 1009 1182 0.3 RI, MS, 13C NMR 10 p-cymene 1011 1274 1.9 RI, MS, 13C NMR 11 1,8-cineole# 1020 1211 8.7 RI, MS, 13C NMR 12 limonene# 1021 1203 21.6 RI, MS, 13C NMR 13 β-phellandrene# 1021 1211 0.6 RI, MS, 13C NMR 14 (Z)-β-ocimene 1024 1233 0.2 RI, MS 15 (E)-β-ocimene 1036 1250 0.2 RI, MS 16 γ-terpinene 1048 1246 1.4 RI, MS, 13C NMR 17 trans-sabinene hydrate 1052 1459 0.1 RI, MS, 13C NMR 18 cis-linalool oxyde THF 1056 1438 tr RI, MS, 13C NMR 19 trans-linalool oxyde THF 1071 1467 tr RI, MS, 13C NMR 20 terpinolene 1078 1284 0.3 RI, MS, 13C NMR 21 linalool 1082 1542 5.9 RI, MS, 13C NMR 22 borneol 1148 1697 0.1 RI, MS, 13C NMR 23 terpinen-4-ol 1161 1597 1.2 RI, MS, 13C NMR 24 α-terpineol 1171 1690 2.3 RI, MS, 13C NMR 25 α-copaene 1375 1491 1.6 RI, MS, 13C NMR 26 β-elemene 1387 1588 0.5 RI, MS, 13C NMR 27 α-gurjunene 1409 1529 0.3 RI, MS, 13C NMR Downloaded by [41.74.26.218] at 00:52 13 June 2014 28 (E)-β-caryophyllene 1417 1595 8.1 RI, MS, 13C NMR 29 aromadendrene 1440 1606 0.1 RI, MS 30 (E)-β-farnesene 1445 1663 0.7 RI, MS, 13C NMR 31 α-humulene 1450 1666 1.3 RI, MS, 13C NMR 32 allo-aromadendrene 1457 1636 0.8 RI, MS, 13C NMR 33 γ-muurolene 1470 1685 0.3 RI, MS, 13C NMR 34 germacrene-D 1475 1705 2.2 RI, MS, 13C NMR 35 β-selinene 1481 1710 0.1 RI, MS 36 4-epi-cubebol 1486 1880 0.2 RI, MS, 13C NMR 37 bicyclogermacrene 1490 1729 1.8 RI, MS, 13C NMR 38 α-muurolene 1492 1720 0.4 RI, MS, 13C NMR 39 β-bisabolene 1499 1723 0.2 RI, MS, 13C NMR 40 cubebol 1505 1931 0.9 RI, MS, 13C NMR 41 γ-cadinene 1508 1755 0.1 RI, MS, 13C NMR 42 δ-cadinene 1513 1753 1.8 RI, MS, 13C NMR 43 β-sesquiphellandrene 1519 1766 0.5 RI, MS, 13C NMR Delphin J.R. Rabehaja et al., / TACL 4 (1) 2014 57 - 64 61 table 1. (continued).

No. Components RIa RIp % Identification

44 α-calacorene 1524 1912 0.3 RI, MS, 13C NMR 45 (E)-α-bisabolene 1531 1771 0.4 RI, MS, 13C NMR 46 β-elemol 1533 2071 0.7 RI, MS, 13C NMR 47 spathulenol* 1563 2113 0.6 RI, MS, 13C NMR 48 palustrol* 1563 1920 tr RI, 13C NMR 49 caryophyllene oxyde * 1568 1975 0.8 RI, MS, 13C NMR 50 germacra-1(10),5-diene-4β-ol* 1568 2020 0.1 RI, 13C NMR 51 epi-cubenol 1583 2059 tr RI, MS, 13C NMR 52 humulene oxyde 1592 2030 0.2 RI, MS, 13C NMR 53 1,10-di-epi-cubenol 1601 2050 0.2 RI, MS, 13C NMR 54 γ-eudesmol* 1620 2159 0.1 RI, MS, 13C NMR 55 τ-cadinol* 1624 2161 0.8 RI, 13C NMR 56 τ-muurolol* 1624 2176 0.3 RI, 13C NMR 57 caryophylla-4(14),8(15)dien-5α-ol 1628 2286 tr RI, 13C NMR 58 β-eudesmol 1634 2218 0.4 RI, MS, 13C NMR 59 α-cadinol 1637 2221 0.5 RI, MS, 13C NMR 60 α-eudesmol 1639 2233 0.3 RI, MS, 13C NMR 61 β-bisabolol 1648 2159 0.1 RI, MS, 13C NMR 62 α-bisabolol 1665 2208 tr RI, MS, 13C NMR 63 epi-α-bisabolol 1667 2211 2.6 RI, MS, 13C NMR Total 96.6 Monoterpene hydrocarbons 47.9 Oxygenated monoterpenes 18.3 Sesquiterpene hydrocarbons 21.5 Oxygenated sesquiterpenes 8.8 Other 0.1

Order of elution and percentages are given on polar column (BP-1) otherwise indicated by an asterisk*,

Downloaded by [41.74.26.218] at 00:52 13 June 2014 percentage on polar column (BP-20); # Content calculated by combination of GC and 13C NMR. RIa and RIp: retention indices measured on apolar (BP-1) and polar (BP-20) column, respectively. Trace tr <0.05%. 13C NMR (italic): compounds identified on the spectra of the fractions of chromatography. 13C NMR : compounds also identified on the spectrum of the essential oil. analyzed by GC(RI), GC-MS and 13C NMR cubenol, γ-eudesmol, β-eudesmol, α-cadinol, α- (Table 2). Various minor components were eudesmol, β-bisabolol. 13C NMR data of all these identified in the fractions of CC by the three compounds were compiled in our laboratory- techniques: camphene, 1-octen-3-ol, α-terpinene, made library. A few components were identified trans-sabinene hydrate, cis-linalool oxyde THF by GC(RI) and GC-MS: (Z)-β-ocimene, (E)-β- trans-linalool oxyde THF terpinolene, borneol, ocimene, aromadendrene and β-selinene. β-elemene, α-gurjunene, γ-muurolene, α- Identification of α-bisabolol 62 and epi-α- muurolene, 4-epi-cubebol, β-bisabolene, γ- bisabolol 63 is hazardous by GC(RI) and GC- cadinene, β-sesquiphellandrene, α-calacorene, MS. Indeed, both epimers possess almost super- (E)-α-bisabolene, humulene oxyde, 1,10-di-epi- imposable mass spectra and very close retention Delphin J.R. Rabehaja et al., / TACL 4 (1) 2014 57 - 64 62 Table 2. Fractionation of Andriambolamena essential oil. Compounds identified by NMR in every fraction F1-F4 of CC

Fractions F1 F2 F3 F4 Mass mg 580 mg 96 mg 159 mg 140 mg P/E 100/0 95/5 90/10 100/0 %a %a %a %a

1 (0.8%) 11 (69.0%) 4 (1.0%) 17 (1.5%) 2 (2.7%) 48 (1.8%) 21 (41.2%) 18 (0.6%) 3 (0.4%) 49 (8.1%) 23 (8.7%) 19 (0.5%) 5 (2.8%) 50 (1.0%) 36 (0.6%) 22 (2.9%) 6 (12.9%) 52 (1.2%) 40 (1.1%) 24(31.5%) 7 (2.8%) 53 (1.6%) 47 (3.3%) 36 (0.5%) 8 (2.8%) 49 (1.2%) 40 (1.2%) 9 (0.4%) 50 (0.8%) 46 (7.3%) 10 (2.5%) 51 (0.6%) 47 (2.7%) 12 (30.5%) 52 (1.0%) 54 (1.4%) 13 (0.6%) 53 (0.6%) 57 (1.3%) 16 (1.9%) 55 (6.5%) 58 (9.3%) 20 (0.4%) 56 (1.9%) 59 (7.4%) 25 (2.3%) 62 (0.8%) 60 (4.8%) 26 (0.6%) 63 (19.8%) 61 (1.8%) 27 (0.4%) 28 (12.1%) 30 (1.2%) 31 (2.0%) 32 (1.2%) 33 (1.8%) 34 (3.1%) 37 (2.6%) 39 (0.5%)

Downloaded by [41.74.26.218] at 00:52 13 June 2014 41 (1.2%) 42 (2.9%) 43 (0.6%) 44 (0.6%) 45 (0.6%)

Numbering of components as in Table 1 P = pentane E = Diethyl ether indices on both capillary columns ((RI apol/pol time in an oil sample of Croton genus 11. = 1665/2208 and 1667/2211, respectively). They Caryophylla-4(14),8(15)dien-5α-ol 57 has been were unambiguously differentiated by 13C NMR. identified by comparison of the chemical shifts This example perfectly illustrated the usefulness in the 13C NMR spectrum of fraction F4 of CC of 13C NMR for the differentiation of diastereo- with those of the reference compound compiled isomers. It could be mentioned that epi-α- in our NMR data library. This compound has been bisabolol as been recently identified for the first first isolated and identified as a component of Delphin J.R. Rabehaja et al., / TACL 4 (1) 2014 57 - 64 63 the essential oil of Achillea eriphora 15. (1.0 %) and its content in the EO has been A special comment concerns the identification evaluated on the chromatogram of the polar of compounds that were co-eluted on the apolar column (RI apol/pol = 1568/2020). column used for GC(RI) and GC-MS analyses: τ-Cadinol 55 (RI apol/pol = 1624/2161) and t- For instance: muurolol 56 (RI apol/pol = 1624/2176) accounted β-phellandrene 13 was overlapped with for 6.5 % and 1.9 %, respectively in fraction F3 limonene 12 on (apolar and polar columns) and of CC. Their identification by MS in the EO and with 1,8-cineole (polar column). Characteristic in fraction F3 was uncertain. In contrast, both signals of the three compounds were observed in compounds were identified by 13C NMR, the first the 13C NMR spectra of the EO and in the frac- one in the EO and in fraction F3 of CC, the second tions of CC. Quantification of both compounds one in the fraction F3. was achieved by combination of GC(FID) and In conclusion, 63 compounds have been identi- 13C NMR, taking into account the relative mean fied in total in the essential oil isolated from aerial intensity of their respective signals. parts of Andriambolamena, a wild aromatic plant Spathulenol 47 (RI apol/pol = 1563/2113, 3.6 from Madagascar. They accounted for 96.6 % of % in fraction F3) was identified by MS and 13C the whole composition. The combination of NMR. Palustrol 48 (RI apol/pol = 1563/1920, chromatographic (CC, GC(RI) and spectroscopic 0.7%% in fraction F2) has been identified only techniques (GC-MS and 13C NMR) appeared by 13C NMR. really useful for the identification of minor Similarly, caryophyllene oxide 49 (RI apol/pol oxygenated sesquiterpenes. Particularly, 13C = 1568/1975) has been identified by MS and 13C NMR associated with CC is a powerful tool for NMR in the EO (0.8 %) and in fraction F2 of identification of diastereoisomers as well as CC(8.1 %). Germacra-1(10),5-diene-4β-ol has compounds that co-elute on GC. been identified by 13C NMR in the same fraction

References 1. Simionatto, S., Bonani, V.F.L., Morel, A.F., Poppi, N.R., Júnior, J.L.R., Stuker, C.Z., Peruzzo, G.M., Peres, M.T.L. and Hess, S.C. (2007). Chemical Composition and Evaluation of Anti- microbial and Antioxidant Activities of the Essential oil of Croton urucurana Baillon (Euphorbi- aceae) Stem bark. J. Braz. Chem. Soc., 18(5): 879-885. 2. Campos, A.R., Albuquerque, F.A.A., Rao, V.S.N., Maciel, M.A.M. and Pinto, A.C. (2002). Investigation on the antinociceptive activity of crude extracts from Croton cajura leaves in Downloaded by [41.74.26.218] at 00:52 13 June 2014 mice. Fitoterapia, 73: 117-120. 3. Suárez, A.I., Blanco, Z., Delle Monache, F., Compagnone, R.S. and Arvelo, F. (2004). Three new glutarimide alkaloids from Croton cuneatus. Nat. Prod. Res. 18(5): 421-426. 4. Tsai, J.C., Tsai, S.L. and Chang, W.C. (2004). Effect of ethanol extracts of tree Chinese medi- cinal plants with laxative properties on ion transport of the rat intestinal epithelia. Biol. Pharm. Bull., 27(2): 162-165. 5. Giang, P.M., Jin, H.Z., Son, P.T., Hong, J.H. and Lee, J.J. (2003). ent-Kaurane Diterpenoids from Croton tonkinensis Inhibit LPS-Induced NF-kB Activation and NO Production. J. Nat. Prod., 66: 1217-1220. 6. Ngadjui, B.T., Abegaz, B.M., Keumedjio G.N. and Kapche, G.W. (2002). Diterpenoids from the stem bark of Croton zambesicus. Phytochem., 60: 345-349. 7. Block, S., Stévigny, C., De Pauw-Gillet, M.C., De Hoffmann, E. and Llabrés, G., Adjakidjé, V., Quetin-Leclercq, J. (2002). ent Trachyloban-3β-ol, a New Cytotoxic Diterpene from Croton zambesicus. Planta medica, 68: 647-649. 8. Block, S., Bacelli, C., Tinant, B., Meervelt, L.V., Rozenberg, R., Jiwan, J.L.H., Llabrès, G., De Pauw-Gillet, M.C. and Quetin-Leclercq, J. (2004). Diterpens from the leaves of Croton Delphin J.R. Rabehaja et al., / TACL 4 (1) 2014 57 - 64 64 zambesicus. Phytochem., 65: 1165-1171. 9. Schmelzer, G.H., Gurib-Fakim, A. (2008). Ressources végétales de l’Afrique tropicale 11(1); Plantes médicinales 1. PROTA : 220. 10. Radulovic, N., Mananjarasoa, E., Harinantenaina, L. and Yoshinori, A. (2006). Essential oil composition of four Croton species from Madagascar and their chemotaxonomy. Biochem. System. and Ecol., 34: 648-653. 11. Rabehaja, D.J.R., Ihandriharison, H., Ramanoelina, P.A.R., Benja, R., Ratsimamanga- Urverg, S., Bighelli, A., Casanova, J. and Tomi, F. (2013) Chemical composition of the essential oil from Croton kimosorum Leandri, an endemic species to Madagascar. Nat. Prod. Commun., in press. 12. NIST/EPA/NIH. (1999). National Institute of Standards and Technology. PC Version 17 of the Mass Spectral Library. Perkin-Elmer Corporation: Saint Quentin, France 13. König, W.A., Hochmuth, D.H. and Joulain, D. (2001). Terpenoids and Related Constituents of EssentialOils. Library of MassFinder 2.1. Institute of Organic Chemistry, Hamburg. 14. McLafferty, F. W. and Stauffer, D.B. (1994). Mass Spectrometry Library Search System Bench- Top/PBM, version 3.10d. Wiley Registry of Mass Spectral Data, 6th edn. Palisade: Newfield. 15. Joulain, D. and König, W. A. (1998). The Atlas of Spectral Data of Sesquiterpene Hydrocarbons. E. B.-Verlag: Hamburg. 16. Adams, R.P. (2007) Identification of Essential Oils Components by Gas Chromatography/Mass Spectroscopy. 4th edition. Allured, Carol Stream, IL. 17. Tomi, F., Bradesi, A., Bighelli, A. and Casanova, J. (1995). Computer-aided identification of indivudial components of essential oils using Carbon-13 NMR spectroscopy. Journal of Magnetic Resonance Analysis, 1: 25-34. 18. Tomi, F. and Casanova, J. (2006). 13C-NMR as a tool for identification of individual components of essential oils from Labiatae. A review. In “The Labiatae: Advances in Production, Bio- 19. technology and Utilisation” Eds: Cervelli, C., Ruffoni, B., Dalla Guda, C. Acta Horticulturae, 723: 185-192. 20. Duquesnoy, E., Castola, V. and Casanova, J. (2007). Identification and quantitative deter- mination of triterpenes in the hexane extract of Olea europaea L. leaves using 13C NMR spectro- scopy. Phytochemical Analysis, 18: 347-353. 21. Rabehaja, D.J.R., Ihandriharison, H., Randrianirina, A.Y., Ramanoelina, P.A.R., Bighelli, A., Casanova, J. and Tomi, F. (2011). Combined analysis by GC(RI), GC-MS and 13C NMR of Downloaded by [41.74.26.218] at 00:52 13 June 2014 the essential oil from Tana bojeriana (Apiaceae), an endemic species of Madagascar. Analytical Chemistry Letters, 1: 115-122. 22. Rabehaja, D.J.R., Raoelison, G., Ihandriharison, H., Ramanoelina, P.A.R., Casanova, J. and, Tomi, F. (2010). Volatile components from Cymbopogon giganteus (Hochst) Chiov var. madagascariensis (A. Camus). J. Essent. Oil Bearing Plants, 13: 522-527; 23. Weyersthahl, P., Marschall, H., Seelmann, I. and Rustaiyan, A. (1997). Constituents of the essential oil of Achillea eriphora Dc. Flavour Fragr. J., 12: 71-78. P.13 : EF. Rakotonirina, G. Chataigné, EG. Raoelison, C. Rabemanantsoa, F. Munhaut, EJ Mondher, S. Urverg-Ratsimamanga, J. Marchand-Brynaert, AM. Corbisier, S. Deklerck, J. Quetin- Leclercq. Characterization of an endophytic whorlforming Streptomyces from Catharanthus roseus stems producing polyene macrolide antibiotic. Canadian Journal of Microbiology 2012, 58(5), 617-627.

P.14: EF. Rakotoniriana, JF. Rajaonarison, GE. Raoelison, JP. Rajaonarivelo, N. Manga, M. Solofoniaina, B. Rakotonirina, D. Randriamampionona, C. Rabemanantsoa, K. Cheuk, S. Urverg Ratsimamanga, J. Quetin Leclercq. Antimicrobial activity of 23 endemic plants in Madagascar. Tropical Journal of Pharmaceutical Research 2010, 9 (2), 165-171.

Rakotoniriana et al

Tropical Journal of Pharmaceutical Research April 2010; 9 (2): 165-171 © Pharmacotherapy Group, Faculty of Pharmacy, University of Benin, Benin City, 300001 Nigeria. All rights reserved .

Available online at http://www.tjpr.org Research Article

Antimicrobial Activity of 23 Endemic Plants in Madagascar

Erick Francisco Rakotoniriana 1,2* , Jean François Rajaonarison 1, Emmanuel Guy Raoelison 1, Jacob Philémon Rajaonarivelo 1, Nia Manga 2, Marcellin Solofoniaina 1, Benja Rakotonirina 1, Denis Randriamampionona 1, Christian Rabemanantsoa 1, Kiban Cheuk 1, Suzanne Urveg- Ratsimamanga 1 and Joëlle Quetin Leclercq 2 1Laboratoire de microbiologie et de standardisation des médicaments, Institut Malgache de Recherches Appliquées, BP3833, Avarabohitra, Antananarivo, Madagascar, 2Laboratoire de Pharmacognosie, Unité CHAM, Louvain Drug Research Institute, Université catholique de Louvain, 72, UCL7230, Av. E. Mounier- 1200 Bruxelles, Belgium

Abstract

Purpose : To screen the crude methanol extracts obtained from 23 endemic plants in Madagascar for antimicrobial activity. Method s: In order to assess the antimicrobial properties of the extracts, their minimum inhibitory concentrations (MICs) were obtained using the broth microdilution method. The six test pathogenic species used were Bacillus subtilis, Staphylococcus aureus, Escherichia coli, Salmonella typhi, Pseudomonas aeruginosa and Candida albicans. Bioautography agar overlay test and phytochemical screening were also performed on the most active extracts. Results : From the 23 plants tested, 16 of which are used in traditional medicine, Poivrea phaneropetala (Combretaceae), Koehneria madagascariensis (Lythraceae) and Rhopalopilia perrieri (Opiliaceae) exhibited the broad spectrum of activity, being active against all the test organisms, while Monoporus clusiifolius (Myrsinaceae) showed the strongest antifungal activity against Candida albicans with a minimal inhibitory concentration of 0.250 mg/ml. Bioautography and phytochemical analysis of the five active extracts against bacterial strains and of one active extract against C. albicans indicate that the active compounds responsible for antimicrobial activity may be mainly flavonoids and/or terpenes. Conclusion : These preliminary results are the first antimicrobial studies on these plants and lend support for the use of some of them in traditional medicine.

Keywords: Antimicrobial properties, Traditional medicine, Microdilution assay, Bioautography, Madagascar.

Received: 20 September 2009 Revised accepted: 21 February 2010

*Corresponding author: E-mail: [email protected]; Tel: +32 4 96 431 884

Trop J Pharm Res, April 2010; 9 (2): 165 Rakotoniriana et al

INTRODUCTION Plant names and their folkoric use are given in Table 1. Dried plant materials were ground Madagascar is host to approximately 12,000 into fine powders and preserved at the IMRA vegetable species, more than 80 % of which herbarium. are endemic to the island [1]. Owing to environmental degradation, increasing Table 1: List of 23 Madagascan endemic plants deforestation, slash and burn agriculture in used in the study and their folkoric use primary forest, this unique patrimony is threatened by extinction. Only 9 % of the Family Species name Folkloric use original areas are currently available and the Annonaceae Xylopia buxifolia Baill. Tonic, country is among 25 most critical regions for jaundice plant life protection in the world [2]. Apocynaceae Mascarenhasia n.i lisianthiflora A. DC. Asteraceae Pluchea grevei Humbert Headache, In Madagascar, herbal medicines are often tonic used as the first line of treatment of various Asteropeiaceae Asteropeia densiflora n.i diseases. The practice of traditional medicine Baker Celastraceae Evonymopsis longipes Headache is well-embedded in the lifestyle of the H. Perrier eighteen indigenous tribes of Madagascar. Clusiaceae Symphonia clusioides Hair ointment Previous investigations on Madagascan flora Baker mainly dealt with ethnobotanical practices of Combretaceae Poivrea phaneropetala Vermifuge (Baker) H. Perrier plants in folk medicine [3-5]. Screening Combretaceae Poivrea grandidieri Vermifuge, surveys have already shown some biological (Baill.) H. Perrier icterus activities such as antiplasmodial [6], antiviral Combretaceae Poivrea obscura (Tul.) Vermifuge, [7], and cytotoxic activities [8]. However, very H. Perrier diuretic Dilleniaceae Hibbertia coriacea Baill. Vaginitis, few scientific studies have been carried out urethritis on the putative antimicrobial properties of Elaeocarpaceae Elaeocarpus sericeus n.i endemic plants in the island although many Baker of them have been claimed by local Fabaceae Piptadenia pervillei Antimalarial Vatke traditional healers to be effective in the Lythraceae Koehneria n.i treatment of infectious diseases. madagascariensis (Baker) The aim of this study was to investigate the S.A.Graham, H.Tobe & P.Baas antimicrobial properties of 23 endemic plants Melastomaceae Dichaetanthera Diarrhea, obtained from different parts of the country oblongifolia Baker dysentery and to identify the phytochemical class of Moraceae Pachytrophe dimepate Jaundice active components. Bureau Myrsinaceae Monoporus clusiifolius n.i H. Perrier EXPERIMENTAL Opiliaceae Rhopalopilia perrieri Antiseptic, Cavaco & Keraudren wound healing Plant materials Rhamnaceae Bathiorhamnus louvelii Purgative (H. Perrier) Capuron The plants were collected from various Sapindaceae Conchopetalum n.i locations in Madagascar and were madagascariense Radlk Sarcolaenaceae Leptolaena pauciflora Venereal authenticated by Dr Armand Rakotozafy, the Baker disease curator of the Department of Botany at the Sarcolaenaceae Leptolaena diospyroidea Impotence to Institut Malgache de Recherches Appliquées (Baill.) Cavaco man (IMRA), Antananarivo, Madagascar. Voucher Sterculiaceae Rulingia n.i madagascariensis Baker specimens were deposited at the herbarium Thymelaeaceae Peddiea involucrata Antimalarial of the Parc Botanique et Zoologique de Baker Tsimbazaza, Antananarivo, Madagascar. n.i = no previous ethnomedical indication reported

Trop J Pharm Res, April 2010; 9 (2): 166 Rakotoniriana et al

Preparation of plant extracts membrane filters. A known volume (100 µl) of each solution was deposited in the wells. This Dried and powdered plants (10 g) were was followed by the addition of 100 µl of the macerated with agitation in 50 ml methanol inoculum (approximately 10 6 CFU/ml for (MeOH) overnight at room temperature. After bacteria and 10 5 CFU/ml for C. albicans ) was filtration, the solvent was eliminated by added to each well. The microplates were evaporation at reduced pressure and crude incubated overnight at 37 °C for bacteria and extracts were dried at 45 °C using a speed- 30 °C for 48 h for C. albicans . After vac concentrator (Savant), and stored at the incubation, 40 µl of 0.2 mg/ml aqueous IMRA bank at 4 °C. solution of methylthiazoyltetrazolium chloride (MTT) was added to each well and further Microorganisms incubated for 30 min at room temperature. MIC was defined as the lowest concentration The bacterial strains used for the in which no transformation of MTT was investigation were Bacillus subtilis , observed. Streptomycin sulfate and nystatin Staphylococcus aureus, Escherichia coli , were used as positive controls and their MICs Salmonella typhi and Pseudomonas were determined using the same process. All aeruginosa , and were all obtained as clinical samples were tested in triplicate and the tests isolates from patients at the Microbiology were repeted twice. Department, IMRA. The yeast strain, Candida albicans (MUCL 31360), was obtained from Bioautography agar-overlay assay with B. the Mycothèque de l’Université Catholique de subtilis and C. albicans Louvain (MUCL), Belgium. Stock cultures were maintained at 4 °C on slopes of nutrient Plant extracts showing significant agar (Difco) for bacteria and on Sabouraud antimicrobial activity with MICs values close dextrose agar (SDA, Difco) for the yeast, to 1 mg/mL against B. subtilis or C. albicans prior to their use. were investigated by thin layer chroma- tography (TLC) bioautographic agar-overlay Determination of minimum inhibitory according to the method of Rahalison et al concentration (MIC) [10] with minor modifications. Twenty microlitres of different solutions of the The broth microdilution method was used to methanol plant extract (400 µg) were applied assess the MIC of the different plant extracts to precoated Silica gel GF254 plates (Merck in a 96-well microplate using a modified KGaA, Darmstadt, Germany). TLC plates method [9]. Microbial suspensions were first were developed with ethyl acetate/methanol prepared from an overnight culture of 1/1 (v/v) for B. subtilis and ethyl bacterial and fungal cells grown in flasks, acetate/methanol/water 10/10/3 (v/v/v) for C. each containing 10 ml of Mueller-Hinton albicans and dried thoroughly overnight to Broth (Oxoid) for bacteria and Sabouraud achieve complete removal of the solvents. dextrose broth (Difco) for yeast at 37 °C and The developed TLC plates were thinly 30 °C, respectively. The turbidity of the overlaid with molten malt extract agar and microbial suspensions was adjusted to 0.5 with SDA inoculated with an overnight culture McFarland using Densicheck (BioMerieux). of B. subtilis and C. albicans , respectively. Stock solutions of the different extracts were The plates were incubated in a dark and prepared by re-suspending the crude humid chamber at 25 °C for 24 h for B. methanol extracts in 10 % dimethyl sulfoxide subtilis and 48 h for C. albicans . After (DMSO) to produce concentrations in the incubation, cultures were sprayed with MTT range of 1.25 – 160 mg/ml. These solutions and further incubated for 30 min at room were further diluted 10-fold with water and temperature. Microbial growth inhibition then sterilized by filtration through 0.22 µm appeared as clear zones around active

Trop J Pharm Res, April 2010; 9 (2): 167 Rakotoniriana et al compounds against a purple background. MIC lower or equal to 2 mg/ml, Monoporus The plates were in duplicate. One set was clusiifolius H. Perrier (Myrsinaceae) showed used for bioautography experiment and the the most interesting inhibitory activity against other was intended for the reference C. albicans with a MIC of 0.25 mg/ml chromatogram. The experiments were repeated twice. DISCUSSION

Phytochemical screening This study showed that all but two plant extracts possessed some degree of activity Plant extracts previously selected for TLC against at least one tested microorganism, at bioautographic assays were subjected to the highest tested dose of 8 mg/ml. Overall, preliminary phytochemical screening to Gram positive bacteria ( B.subtilis and S. determine the major chemical groups aureus) were more sensitive to plant extracts corresponding to active compounds than Gram negative bacteria ( E. coli, S. typhi responsible for the observed inhibition zones. and P.aeruginosa ). Among them, P. TLC was first developed with the same phaneropetala, K. madagascariensis and R. solvent systems as previously mentioned perrieri demonstrated a broad spectrum of above. For the detection of alkaloid, activity against all the tested organisms while coumarins, flavonoids and terpenes, M. clusiifolius was the most active against C. phytochemical screening was performed albicans . R. perrieri showed the most using standard procedures [11]. Observations promising antibacterial activity, confirming its were made from two independent use in traditional medicine as antiseptic and experiments. wound healing. The present investigation also confirmed the anti-infective property of RESULTS H. coriacea in the treatment of vaginitis and urethritis and that of D. oblongifolia which is A total of 23 methanol plant extracts of used to treat diarrhoea and dysentery in folk Madagascan endemic species belonging to medicine. In contrast, the use of Leptolaena 20 different botanical families, among which pauciflora Baker (Sarcolaenaceae) as anti- 16 are used in traditional medicine (Table 1), infective agent for the treatment of venereal were tested for their antimicrobial activities diseases was not supported by our against 5 bacterial species and one yeast. investigation. Furthermore, the oral use of The results of these tests are summarized in this plant containing several phenolic Table 2. Eight extracts inhibited the growth of compounds could be toxic [12]. B. subtilis at MIC values lower or equal to 1 mg/ml. Among them, the extract of Based on this investigation, we are unable to Rhopalopilia perrieri Cavaco & Keraudren say categorically that extracts having the (Opiliaceae) exhibited the lowest MIC value same Rf values contained the same bioactive (0.13 mg/ml). Four of them were also found compounds. It is possible that the observed to be effective against S. aureus at 1 mg/ml. inhibition was likely due to one or more Five extracts were moderately active against compounds sharing the same Rf in the E. coli and S. typhi with MIC values ranging solvent system used, particularly for low Rf from 2 to 4 mg/ml. None of the tested plant spots. Preliminary phytochemical screening extracts displayed activity against P. revealed the presence of flavonoids, alkaloids aeruginosa (MIC > 8 mg/ml) except Poivrea and terpenes as the probable active phaneropetala (Baker) H. Perrier compounds present in the crude extracts. It is (Combretaceae) , Koehneria madagasca- well known that numerous members of these riensis (Baker) S.A.Graham, H.Tobe and phytochemical groups have already P.Baas (Lythraceae) and R. perrieri . Of the 6 demonstrated antimicrobial activity [13]. plant extracts that inhibited C. albicans at

Trop J Pharm Res, April 2010; 9 (2): 168 Rakotoniriana et al

Table 2: Antimicrobial activity of 23 Madagascan endemic plants

Plant Voucher no. Microorganisms a/ MIC b (mg/ml)

B.s S.a E.c S.t P.a C.a Xylopia buxifolia AML6 ------Mascarenhasia lisianthiflora MOR23 2 - - - - - Pluchea grevei TUL16 4 - - - - 4 Asteropeia densiflora TUL8 0.5 2 - - - 2 Evonymopsis longipes MOR33 2 8 - - - 8 Symphonia clusioides AMB34 ------Poivrea phaneropetala MOR12 0.5 1 4 4 8 2 Poivrea grandidieri TUL10 4 8 8 8 - 8 Poivrea obscura MOR26 1 1 4 4 - - Hibbertia coriacea MKR44 1 2 4 2 - 4 Elaeocarpus sericeus M447 1 4 - - - 4 Piptadenia pervillei AA17 2 8 8 8 - - Koehneria madagascariensis TUL5 0.5 1 4 4 8 2 Dichaetanthera oblongifolia BP31 1 2 - - - 2 Pachytrophe dimepate VAM282 2 - - - - - Monoporus clusiifolius AMB43 2 8 - - - 0.25 Rhopalopilia perrieri TUL35 0.13 1 4 4 8 2 Bathiorhamnus louvelii AA68 2 - - - - - Conchopetalum madagascariense VAT652 2 - - - - - Leptolaena pauciflora MKR692 4 8 8 8 - 8 Leptolaena diospyroidea MOR56 4 - - - - - Rulingia madagascariensis BP14 4 8 - - - 8 Peddiea involucrata AA48 4 8 - - - - References (µg/ml) Streptomycine 3.9 3.9 31.3 31.3 - nd Nystatine nd nd nd nd nd 3.13 a MIC: minimal inhibitory concentration (mg/ml) b Microorganisms: B.s = Bacillus subtilis; S.a = Staphylococcus aureus; E.c = Escherichia coli; S.t = Salmonella typhi; P.s = Pseudomonas aeruginosa; C.a = Candida albicans (-): MIC > 8 mg/ml for plant extracts and MIC > 64 µg/ml for streptomycin; nd = not determined Overall, five plant extracts with MICs lower than 1 mg/ml against either B. subtilis or C. albicans were subjected to TLC bioautographic agar overlay test. Rf value of the inhibition zones and the probable chemical group of compound responsible for the inhibition are listed in Table 3. The tests based on colour development suggested that the active compounds belong mainly to flavonoid or terpene groups. Furthermore, a positive test with Dragendorf reagent suggests that Asteropeia densiflora Baker (Asteropeiaceae) and R. perrieri may contain alkaloids at Rf of 0.02 (at base line).

Table 3: Phytochemical screening of active compounds responsible for microbial inhibition based on bioautography assays

Plant Rf Phytocompound a value A C F Te Monoporus clusiifolius b 0.68 - - + + Poivrea phaneropetala c 0.1 - - + + Koehneria madagascariensis c 0.1 - - + + Asteropeia densiflora c 0.02 + - + + 0.75 - - + + Rhopalopilia perrieri c 0.02 + - + + a A= Alkaloids; C= Coumarins; F= Flavonoids; Te= Terpenes b Test performed on TLC bioautographic agar-overlay with solvent system EtOAc/MeOH/H 2O (10/10/3) using C. albicans as organism test c Test performed on TLC biautographic agar-overlay with solvent system EtOAc/MeOH (1/1) using B. subtilis as organism test

Trop J Pharm Res, April 2010; 9 (2): 169 Rakotoniriana et al

In many cases, these substances serve as serve as a basis for further pharmaceutical plant defense mechanisms against investigation. The findings also lend support aggression by microorganisms, insects, and for the use of some of these plants in herbivores either synthesized during the plant traditional medicine in Madagascar. normal development (constitutive resistance factors) or induced only after contact with the ACKNOWLEDGMENT pathogen (induced resistance factor) [14]. The authors are grateful to the Coopération To the best of our knowledge, no previous Universitaire Belge pour le Développement antimicrobial survey has been reported on (CUD) for financial support which enabled these plants. However, some of them were this survey to be carried and a grant to N. previously reported for other biological Manga. We wish to thank Prof. Philippe activities. Literature searches indicate that Rasoanaivo for helpful comments on the Evonymopsis longipes (H. Perrier) H. Perrier manuscript. (Celastraceae) was particularly potent and could completely inactivate the Herpes Simplex Virus at a concentration lower than REFERENCES

25 µg/ml [7]. Peddiea involucrata Baker 1. Robinson JG. An island of evolutionary (Thymelaeaceae) and Piptadenia pervillei exuberance. Science, 2004; 304 (5667): Vatke (Fabaceae) were said to possess 53. antimalarial activity [15-16]. From this latter 2. Myers N, Mittermeier RA, Mittermeier CG, Da Fonseca GAB, Kent J. Biodiversity was isolated (+)-catechin 5-gallate and (+)- hotspots for conservation priorities. Nature, catechin 3-gallate which demonstrated, in 2000; 403: 853-858. vitro , high activity against the chloroquine- 3. Beaujard P. Plants and traditional medicine in resistant strain FcB1 of Plasmodium Southeast Madagascar. J Ethnophar- macol, 1988; 23: 165-265. falciparum as well as (+)-catechinant 4. Novy JW. Medicinal plants of the eastern region of ethylgallate which were less effective [16] . Madagascar. J Ethnopharmacol, 1997; 55: With regard to antibacterial activity, howerver, 119-226. these plants were less effective, being active 5. Norscia I, Borgognini-Tarli SM. Ethnobotanical reputation of plant species from two forests only against B. subtilis with MIC ranging from of Madagascar: a preliminary investigation. 2 to 4 mg/ml. S Afr J Bot, 2006; 72 (4): 656-660. 6. Rasoanaivo P, Ramanitrahasimbola D, Rafatro H, While no ethnomedicinal practices were Rakotondramanana D, Robijaona B, Rakotozafy A, Ratsimamanga-Urveg S, reported to date for 7 of the 23 plants, the Labaied M, Grellier P, Allorge L et al. study revealed that A. densiflora and K. Screening extracts of Madagascan plants madagascariensis were found to possess in search of antiplasmodial compounds. effective antibacterial activity; in addition, the Phytother Res, 2004; 18(9): 742-747. 7. Hudson JB, Lee MK, Rasoanaivo P. Antiviral extract from M. clusiifolius may be a good activities in plants endemic to Madagascar. candidate in the search for potential Pharmaceut Biol, 2000; 38 (1): 36-39. antifungal compounds. Further stuydies on 8. Williams RB, Norris A, Miller JS, Razafitsalama LJ, these plants are being undertaken in our Andriantsiferana R, Rasamison VE, Kingston DG. Two new cytotoxic laboratory to isolate and elucidate the naphthoquinones from Mendoncia cowanii compounds responsible for the observed from the rainforest of Madagascar. Planta anti-candidiasis and antibacterial activities. Med, 2006; 72: 564-566. 9. Eloff JN. A sensitive and quick microplate method to determine the minimal inhibitory CONCLUSION concentration of plant extracts for bacteria. Planta Med, 1998; 64: 711-713. 10. Rahalison L, Hamburger M, Hostettmann K, This survey contributes new data to the Monod M, Frenk E. A bioautographic agar existing knowledge of antimicrobial activity of overlay method for the detection of the endemic flora of Madagascar and may

Trop J Pharm Res, April 2010; 9 (2): 170 Rakotoniriana et al

antifungal compounds from higher plants. 14. Castro MS, Fontes W. Plant defense and Phytochem Anal, 1991; 2:199-203. antimicrobial peptides. Protein Pept Lett, 11. Wagner H, Bladt S. Plant drug analysis. A thin 2005; 12: 11-16. layer chromatography atlas. Springer- 15. Randrianarivelojosia M, Rasidimanana VT, Verlag, Berlin Heidelberg; 1996.384p. Rabarison H, Cheplogoi PK, Ratsimbason 12. Paris R, Jacquemin H, Linard A. Madagascar M, Mulholland DA, Mauclere P.. Plants plants. XV. Chlaenaceae from Malagasy. traditionally prescribed to treat tazo Leptolaena pauciflora, L. diospyroidea (malaria) in the eastern region of cavaco var tampoketsensis, and Madagascar. Malar J, 2003; 2: 25. Sarcolaena multiflora. Presence of 16. Ramanandraibe V, Grellier P, Martin MT, Deville A, myricetol heterosides. Plantes Méd et Joyeau R, Ramanitrahasimbola D, Mouray Phyto, 1975; 9: 230-237. E, Rasoanaivo P, Mambu L. 13. Cowan MM. Plant products as antimicrobial Antiplasmodial phenolic compounds from agents. Clin Microbiol Rev, 1999; 12 (4): Piptadenia pervillei. Planta Medica, 2008; 564-582. 74: 417-421.

Trop J Pharm Res, April 2010; 9 (2): 171 P.15: O. Rangasamy, G. Raoelison, E.F. Rakotoniriana, K. Cheuk, S. Urverg-Ratsimamanga, J. Quetin-Leclercq, A. Gurib-Fakim and A.H. Subratty. Screening for anti-infective properties of several plants of the Mauritian flora. Journal of Ethnopharmacology 2007, 109, 331-337.

{Jirr:,r*i::-r:* .'.:'':il '.', l'i.:;'rti}!i-i:*xjlrT#j'{:is; r$$*itrc}: "*r:* *,:fe,:{::i," ,:i., } l";i..,i §,;:r {",rr§il,ii j-: ,' tt{jii r}f ,j :,',l}utti;i: *, i:.i-}tilt"§:{*i.{.1;i"'Tr iti<{i.r

ÿdre §m &if t*.:wïiâ*St r*r4T

ffi§ {}- (),&#tY

(Ïftitàr*i J*r*rlwl ol il* " lnÈ+:nrxlitnxË XüNià"l1' [sr Ë*hurrp&tnrrrtrr:Ërrg3

À§l lirtèr$;${tri}iuxr+r1 ^}àdril;it i}*vw{"n§ 1.,} ie*i**§}+il$ *, îs}*

i :i;;,' :.i;, ..liS -"ri#irinl;,;rtr"i:l;.i1p1-, il: i1 i :ir,tir, r.' ';.;.;,*.'4lJ âl';t;i,,,iür. ü:,:ri X*r-t "*q't#*h")# â".1{}*}ur ttii çn:'*tt{ti:'i g-q; [{'69'71,111- T{,i !i?* .t:iir:r : Ii l.: ;.:r-jiàl lirl{"i,{i:}t ('iti.i;i i.:r§;iÈ"ri;!§'ji frt à; :: .,, ;.., g i:l{":,; :,ir:i",. lili ;',:'i':-i:iiirr:i:lii..'iliiiii t;"y;.rr;;} ti:.t;\.-ri\'#.tt'4tê.*::1l.it; ifri ir., ': 1;;1'.1:: ' 'iii'}i.:t:'l: 1i:t.tiiri.i;;r, rlti[ ,i] :l"}$ir;.j{)li,::r\ iii }d{j;Ji'il":+}:.ïr:tir:rr :rl;l-:.{iiir{-i,i ltr l.}1}+;.,":i,".. - ;,1 .'. :.1, ,. . fii.", ;-ril frI,.,-.:. .jîj.J l, ,,i: T it ff .. ,n : ' :5i ,:lTii;;i. .,:. .i _t . ? \! ".r,-'qi i:.i

;'ri. li ';..i]r{i\t:1§r}#" i"{r,I,:fr*r{);,;:lLit}i? rcttiJ li}:ili'tir,j;i :' ,, .'-J ;;rr/ .;;È}:.l,ttil lfr: .lilir. i,iT.:1',:i{r,i i;iiriirlb- *f .frl:.3 t.j jt.iÿr". i... 1.;t 1,.+,... ij;.., ,,

, : ..:'1.-.., * :r. *ü#:i trit*,riil:. :'lÉ*,; ui'r{". }i{};1.x.;i;;. ; t i;.::i'iiJ,i..;'i r:,

ti].;t',i ), :r{âilltli}* i{"rr *r:i-:11 i.li:t{i:,}t*Ljqlh i::ilir,., : ; ..i's'r;.i1çlq; -; i.lii;r;)ï

i':1lf:: .,r '.r!,rriÈti\5Éï$l*{.i-r*ll,rl*i":.i:1ilt:,4i,,iïpi.91,1;r:iti..itir.tiiii,liï11;.i#i

f Available online at www.sciencedirect.com *; "S.i"n."Direct

Journal of Ethnopharmacology 109 (2007) 331-j37 www.clscvicr.com/locatc/j cthpharm

Screening for anti-infective properties of several medicinal plants of the Mauritians flora

Oumadevi Rangasamyu, Guy Raoelisonb, Francisco E. Rakotoniriânab, b, b, t, Kiban Cheuk Suzanne Urverg-Ratsimamanga Joelle Quetin-Leclercq Ameenatr Gurib-Fakimd, Anwar Hussein Subratty a'*

a Departrnent ofHealth Sciences, Factlty of Science, Uniÿersiry of Mauitius, Reduit, Mauritius b Department olPh)tochemistü andMicrobiotost, ,;;ii;#fi#des Recherches Apptiquees (IMM),

c La.boratoire de Pharmacognosie, (lnité CHAM, Université Catholique de ltwain, Avenue E. Mourier 72, '" d Departutent orchemistry, ,":,':;ffi:,tr::'i:;::;lî otruou,i,iu,, Reduit, Mauitiu., Received 6 Februarv 2*,; "ïil,*î'"ï,'#rïî,::.î#à:*6; accepted I August 2ü)6

Abstract

Several plants ofthe Mauriüan flora alleged to possess anti-infectife properties were studied against different strains ofpathogenic bacteria and fungi. The grounded dried plant materials were extracæd with differerl and screened for anti-microbial activity using the disk diffusion "*x6çtants and the micro-dilution techniques. Preliminary screening rëvealed that the methanol extracts weÉ most active . Salmonella enteritidis, Enrerobacter cloacae and Bacillus subtilis werethe three test organisms, which were found to be susceptible to all the crude methanolic extracts of the different plants investigated (lN% susceptibility), followed by Escherichia coli (57.17o) and Pseudomonns aeruginosa (57.lVo), and Staphylococcus aureus (28.6Vo\ The lowest minimum inhibitory concentration recmded for the different crude methanol extracts agailrtst Staphylococcus aureus, Escherichia coli, Salmonelln enteritidis, Enterobacter cloacac, Bacillus subtilis and the mould fingæ Candidn albicans were 500, l0(n, 125, 250, l00O and 125 pg/ml, respectively. Bioautography wing Cladosporium cucumzinurn rcvealeÀ that dichloromethane (DCM) extracts had the highest activity againstthephytopathogenic fungus.It was also notedthatthe DCM extracts ofMichelia champacaandAntidesmamadagascariense yielded the maximum number of growth inhibiting compounds against Cladasporium cucumerinum. Actlity of the different crude extracts was also investigated against several phytopathogenic filamentous fungi, Colletotichwn glocospotoides, Rhizoctonia solani, Sclerotinia sclerotium, Guignardia sp. arrd Fusaium orysporum.It was found that crude hexane extracts as well as crude DCM extracts exhibited marked activity against several strains of fungi, especially Colletotrichum glocosporoides, Sclerotinia sclerutiuru and Guignardia sp. O 2006 Elsevier Ireland Ltd. All rights reserved.

Keywords: Medicinal plants; Mauritius; Infections; Bioautography; Anti-microbial activity; Micro-dilution assay

1. Intruduction to which resistance had not been observed (Eloff, 2000). This is indeed quite alarming when considering that in 1990, out of the Anti-microbial agents are undeniably one of the most impor- 39.5 million of death in the developing world, 9.2 million were tant therapeutic discoveries of the 20th century. However, with estimated to have been caused by infectious and parasitic dis- the 'antibiotic era' barely five decades old, mankind is now eases, and that 987o of deühin children in developing countries faced with the global problem of emerging resistance in vir- resulted mostly from infectious diseases (Murray and Lopez, tually all pathogens (Pêterson and Dalhoff, 2004). Surveys have 1997). Bacterial resistance is beyond doubt the consequence of revealed that alrnost no group of antibiotics has been introduced years of widespread indiscriminate use, incessant misuse and abuse of antibiotics (Peterson and Dalhoff, 2004)l In human medicine alone, the US Centre for Disease Control and Preven- * 150 million Corresponding author. Tel.: +23O 454lû4l; Tax: +23O 4656928. tion estimates that approximately one-third of the E- mai I oilre s s : subratty @ uom.ac.mu (A.H. Subratty). prescriptions for antibiotics written each year were not needed.

0378-E741/$ - see front maner @ 2006 Elsevier Ireland Ltd. All rights reserved. doi: l0.l 0l 6/jjep.2006.08.002 F O. Rangasamy et al. / Journal of Ethrwphnrmacology 109 (2007) 331-i37

Because of the limited life span of antibiotics, it is of utmost from trees grown in the local Botanical Garden while leaves importance to find appropriate solutions to impede, or perhaps of Moringa oleifera, Mangiftra indica and Melia azedarach even reduce, the development ofdrug resistance associated with were collected from trees growing in the hot northwestem part many microbial species (Martini and Eloff, 1998). of the island, mainly, the capital, Pon Louis. The Curator of Since time immemorial, medicinal plants had been a depend- the Botanical Garden identif,ed the different. plants and voucher able source of therapeutics for the treatment of various ailments specimen were collected for all the plants, i. e., Antidesma mada- (Hoareau and Da Silvet 1999) but since the advent of the use gascariense (23,376), Aphloia theifurmis (24,L21), Erythrox- of fermentation-based antibiotics work on anti-microbial agents ylum hurtfolium (24,045), Mangifera ind.ica (19,368), Melia from plants sources has been greatly overshadowed (Mitscher azedarach (15,549), Michelia champaca (11,664) and Moringa et al., 1987). The rapid propagation in antibiotic resistance and oleiftra (10,122) and deposited at the Department of Chem- the increasing interest in natural products, however, have placed istry then transferred to join the vast collection of the National medicinal plants back in the front lights as a reliable source for Herbarium of the Mauritius Sugar Industry Research Institute, the discovery of active anti-microbial agents and possibly even MSIRI, after confirmation of their identities from comparison novel classes ofantibiotics (Shultes, 1992). with botanical descripüons and the collaboration ofthe botanist Plants are complex chemical storehouses of undiscovered in charge of the herbarium. biodynamic compounds with unrealized potential for use in modem medicine (Plotkin, 1988). It has long been established 2.2. Preparation of plant maturtd and extraction that naturally occurring substances in plants have anti-bacterial and anti-fungal activities. In Mauritius, medicinal plants, for Atl the plant materials used were in the form of f,nely centuries, have been used for the treatment of a wide range of grounded dried powder. The different plants collected were ailments, many of which are still in use today and hold favored processed similarly. After their authentication, the plants were positions among local tradi-practitioners. Situated between the collected in large quantity, thoroughly washed with water and southern latitude of 19o50' and2032t and longitude 57'18/ and dried in a dryirrg cabinet at about 40 "C for several days till com- 57"46t, Mauritius is a tropical island, which havemeiged some plete removal of water then processed to a fine powder using a 8 million years ago from the Indian Ocean. Certain conditions Jankel and Kiinkel Model A10 mill. The dried powdered plant such as the topography of the land and the rain distribution have , materials were then extracted via maceration in a serial man- ensured the island a diverse microclimatic regime, which has had ner using hexane, dichloromethane (DCM) and methanol (10:l a direct consequence on both the endemic and exotic ve§etation. solvent to dry weight ratio) for two successive 24-h periods. Its old age and geographic isolation has provided the Mauritian The extracts were f,ltered, combined and dried under reduced flora a high degree of endemism. The island possesses seven pressure. phanerogams all of which are endemic (Gurib-Fakim,2002| For that reason, Mauritius is a rich source of natural products of 2.3. TLC analysis great therapeutic value that wait to be uncovered. lVhich is why emphasis in this work was laid on several plants of the Mauritian Thin layer chromatography (5pl of a l00mgextrac'lJ flora. ml solution) was on Siüca Gel 60 coated on glass plates Aiming for new compounds responsible for anti-infective (Merck TLC F254) with hexane/ethyl acetate 1/1 (v/v) and properties, a thorough literature search was undertaken through DCNUmethanoUwater 6513510.5 (v/v/v) as eluants. The sepa- ethnobotanical published data looking for plants used in rated componènts were visualised under visible and ultraviolet Mauritius to combat fever and diseases caused by bacteria light (254 and 360nm, Camag Universal IfV lamp TL-600) or and fungi. The plants selected were as follows: Antidesma using spray reagents such as 57o anisaldehyde h a 5Vo sulphuric madagascariense (Lam.) (Euphorbiaceae), Aphlnia theiformis acid in ethanol solution, vanillin and Dragendorff (Martini and (Vahl.) (Aphloiaceae), Erythrorylum laurifolium (Lam.) (Ery- Elotr, 1998). throxylaceae), Man gift ra indica (L.) (Anacardiace ae), M elia azedarach (Meliaceae), Michelia champaca (L.) (Magnoli- 2.4. Microorganisms aceae) and Moringa oleifera (Lan) (Moringaceae). The use of medicinal plants towards certain types of illnesses The test organisms used were Bacillas subtilis, Enterobacter has roots in the Mauritian traditional pharmacopoeia. This cloacae, Escherichia coli, Salmonclla enteitidis, Staphylococ- study was undertaken to determine possible inhibitory effects of cus aureus, the yeast mould Candidn albicans and filamentous some plants that are used against common infectious diseases phytopathogenic fungi; Colletotrichum glocosporoides, Cla- in Mauritius. dosporium cucumertnum, Fusarium orysporum, Guignnrdia sp., Rhipctonia solani and Sclerotinia sclerotium. Thedifferent 2. Material and methods bacteria were obtained as clinical isolates from patients of the Department of Microbiology, Institut Malgache des Recherches 2.1. Plant mateial Appliquées (MRA), Antananarivo, Madagascar, as well as the yeast mould Candidn albicans while the filamentous fungi Fresh leaves of all the plants except leaves of Moringa and Enterabacter cloacae were isolated from different plant oleifera, Mangifura indica allrd Melia azedarach were collected specres. F O. Rangasarny et al. / Ioumal of Ethnopharmacology 109 (2N7) 331-337 333

2.5. Bioassay ton agar, colonies (5-7) were then transferred aseptically from the second transfer plate into individual tubes containing sterile 2.5.1. Preliminary screening using the disk diffusion nutrient broth (10 ml). The tubes were incubated for a period of technique 8-12 h at 37 "C to ensure that the bacteria were in the log phase. 2.5.1.1. Procedure usedfor the different bacterta and Candida Subsequently, the bacterial suspensions were visually adjusted albicans. The disk diffrrsion method was used as a prelimi- to 0.5 McFailand and üen further diluted 1 : 100 with fresh sterile nary test to find out 'if the plant extracts were active (Matsen, broth to yield starting inoculums of approximately 106 CFU/ml. 1979). Stock solutions ofthe different extracts at a known con- Stock solutions of the different extracts at a concentration of centration (8 mÿml) were prepared in the solvents used for 16 mÿml were prepared in the solvent used for extraction. A exffaction and suitably stored. Loops full ofthe different bacte- known volume (100 p,l) of each solution was placed in the flrst ria and Candida albicans were transferred aseptically into test well of a 96-well microplate and two-fold serially diluted wiü tubes containing peptone water (10 ml) and incubated at 37 "C sterile distilled water (Klepser et al., 1996). A known volume of for 24h. After the incubation period, the turbidity of the solu- the inoculum (100 p"l) was then added to each well. The plates tion was adjusted to 0.5 McFarland. Hundred microlitres of this were then incubated at 37 "C for 24h. After incubation, 40 pl inoculum (5 x 105 CFUimD was then transferred onto the sur- of MTT (0.2 mÿml) was added to each well and incubated for face of solidif,ed Muller Hinton agar and spread evenly across a further 10-15 m. Bacterial growth is denoted by a blue col- the whole surface of the agar in the petri dish (85 mm). In the oration of the wells. The well of lowest concentration in which case of Candida albicans sabouraud dextrose agar media was no blue coloration is observed is taken as the MIC. Streptomycin used §CCLS , 1992). Sterile paper disks (6 mm diameter, pre- sulphate and Gentamicin sulphate were used to cbmpare the sus- pared from Whatman numbçr 1 filter paper) were then dipped ceptibiliÿ of the different microorganisms. The procedure used into the stock solution of the extract (concentration 8 mÿml) and was as described above, only instead of doing serial dilution of transferred ascetically onto the surface of the agar bearing the the extracts; serial dilution of the standard antibiotics was done bacteria. The tests tvere run in duplicate. Similarly, paper disks (Bonjar, 2004). containing standard concentration (8 mÿmD of Ambicillin were used as positive control. The petri dishes were then incubated 2. 5. 3. Quantitative evaluation of anti-mic robial activ ity at 37 "C for 24.h. The results recorded were the average of the There exist different ways of expressing the biological duplicated test. activity of plant extracts based on the technique used. The agar I diffusion method'led to results being given in terms of width 2,5.1.2. Procedure usedfor the filamentous fungi. The differ- of the inhibition zone (mm or cm) while the micro-dilution ent fungi were sub-cultured on potato dextrose agar (PDA) and method yield MIC values, the minimum concentration at which incubated in a humid atmosphere at 26"C for 48h or until inhibition is observed (mÿml). In this work new ways of the petri dishes were completely invaded by the fungi. Cubes expressing anü-bacterial efficiency as comparative numerical (0.5 cm x 0.5 cm) were cut aseptically from themotherpetri dish values are used. Beside results being recorded in terms of MIC and depositedin the centre offreshpetri dishes containing sterile (mg/ml), total activity values as described by Eloff (2000) was PDA and these were then incubated in a humid atmosphere at a employed, as well as percent activity values which demon- temperature of 26"C until a growth diameter of about 2 cm was strates the total anti-microbial potency of particular extracts observed. The sterile disks impregnated with the plant extracts and bacterial susceptibility index (BSI) as described by Bonjar were then deposited onto the PDA in circle about 10mm from (2004), which is used to compare the relative susceptibility the growing fungi. The Petri dished were incubated for 48 h and among the bacterial strains: zone of inhibition around the disks if any were recorded. The tests were run in duplicate and the results recorded were the o Total activity average of the duplicated test. Anti-microbial acrivities of the plant extracts were expressed in terms of: + (positive with cor- quantity of material extracted from rected value for diameter of inhibition with respect to blank less 1 g of plant material than 4mm); ++ (positive total activity : with corrected value for diameter of MIC inhibition with respect to blank more than 4 mm); (negative, - These values would indicate the largest volume to which when no distinct zone of inhibition is observed). Blanks were biologically active compounds in I g of plant material can be prepared by dipping the filærpaper disk in the different solvenrs growth (Elotr, 200a). used for extraction. diluted and still inhibits the of bacteria o Percent activity: 2.5.2. Determination of minimum inhibitory concentration actlity (Vo) (MIC) using the micro-d,ilution technique on the different 100 x no. ofsusceptible stains to a specific extract strains of bacteria and Candida albicans _ total no. oftested bacterial strains MIC analyses were conducted via broth micro-dilution tech- niques according to the National Committee for Clinical Labora- The percent activity demonstrates the total anti-microbial tory Standards procedures for aerobic testing (NCCLS, 1990), potensy of particular extracts. It shows number of bacteria Each of the bacteria were sub-cultured twice on Muller Hin- found susceptible to one particular extract. F 334 O. Rangasarny et al. / Journal of Ethnophamucology IW (2N7) 331-337 o Bacterial susceptible index, BSI: 3. Results and discussion

100 x no. of extracts effective against each Results clearly demonstrated that the plants investigated exhibited signif,cant anti-microbial activity (Tables 1-3). The bacterial strain BSI : results showed that out of all the three solvents used for extrac- no. of total §amples tion, the methanol exffacts displayed a broadér spectrum of anti-bacterial activity (Table 3). The DCM extracts also showed BSI is used to compare the relative susceptibility among relatively good anti-bacterial acüvities, most particularly against the bacterial strains. BSI values ranges from '0' (resistant to Salmonella enteritidis, Enterubacter cloacae and Bacillus sub- all samples) to '100' (susceptible to all sample). tilisbü very little to no activity was noted against Escheichia coli and Pseudomonas aeruginosa (Table 2). The crude hex- 2. 5. 4. B io auto g raphy u s in g C lado spo ri um c uc urne rinum ane extracts were of no interest as none showed marked activity The protocol was as described by Homans and Fuchs ( 1970). (Table 1). Bioautography involved the development of chromatogram of In the present investigation, the extracts were prepared seri- the different crude extracts under investigation as described ally with the same dried and grounded plant material, the above (Section 2.3). After separation, the TLC plates were thor- first extractant being hexane followed by the other solvents of oughly dried to remove all traces of solvents. The bioautography increasing polarity. This serial extraction led to some fractiona- tests were done using Cladosportum cucumertnum, a parasitical tion of the anti-microbial compounds of theplants studied.It was fungus, which attacks different plants most particularly those observed that in most of the medicinal plants analy'sed, the anti- of the Cucurbitaceae family. A suspension of the fungus was bacterial potency resided in the most polar extractant used, i.e., made half an hourbefore the bioautography was to be set. Loops methanol. This corroborate well with the ethnobotanical claims full of the spore were transferred aseptically to sabouraud malt on the differcnt plants, as in the local folklore, the benef,cial broth (20 ml) to which was then added 8 mg of chloramphenicol, medicinal properties of these plants are derived in most cases which is a standard anti-bacterial agent. from decoctions and infusions in aqueous medium. The results The inoculum was thoroughly shaken every 10 min for 30 min revealed that the methanol extracts hold the most promise for to make a homogenised suspension of the Cladosporium cuc- e further work. umerinum, which was then sprayed thinly over the developed Invitro anti-microbial screening via the micro-dilution tech- TLC plates and then incubated for 48 h at room temperâture in a nique provided the iequired preliminary observation to select humid atmosphere. Nystatine (10 pl of a 1 mÿml aqueous solu- among the crude methanoüc plant extracts those with poten- tion) and chloramphenicol (10 pl of a I mg/ml ethanol solution) tially useful properties for further chemical and pharmaceu- spotted separately were used as control. Growth of the fungus tical investigations. According to NCCLS standards, a break- was denoted by a greyish green coloration of the TLC plates point for a pure antibiotic susceptibility is 8 -g/1. The low- with zones of inhibition, which appears as clear spots (Begue est minimum inhibitory concentration recorded for the differ- and Kline, 1972). ent crude meüanolic extracts against Staphylococcas aureus

Table 1 Results of preliminary anti-micrcbial screening of crude hexane extsact using the disk diffusion methgd

Scientific name SA" ECb SEc pAd EC BSf CGc RSh oFk CA

An tide s ma mada g a s c ari e n s e -+ Aphloia theifurnh -+ Erythrcrylum lauifolium -+ Mangifem indica Melia azedarach -: Michelia champaca Mortnga oleifera -+ (+) Positive result with conected diameter of inhibition with respect to blank less than 4 mm; (++) positive result with corrected diameter of inhibition with respect to blank more than 4 mm; (-) negative, no distinct zone of inhibition. r Staphylococcus aureus. b Eschertchiacoli. c Salmone ll.a ente ritidis. d P s e udomonas ae ru g ina sa. e Ente robacter c loacae. r Bacillus subtilis. E Colletotrichum glocospotttides. h Rhizoctoniasolani, I Sclemtinia sclerotium. t Guigrurd.iasp. k Fusarium orysporurn. I Candidaalbicans. F" O. Rangasany et al. / Joumal of Ethnopharmacology 109 (2N7) 331-337

Table 2 Resulls of preliminary anti-microbial screening of crude DCM extract using the disk difftrsion method

Scientific name sAa ECb sEc PAd EC" BSf CGC RSb ss' G9 oFk CAI

An ti d e sma mada g a s c a i e ns e Aphloia theifonnis Erythruxylum lauifolium Mangiftra indica . Melia azedarach Michelia chornpaca Moringa oleifem

(+) Positive result üth correcæd diameter of inhibition with respect to blank less than 4 mm; (++) positive result with conected dianeter of inhibition with respect to blank more than 4 mm; (-) negative, no distinct zone of inhibition. a Staphylococcus aureus. b Escherichiacoli. c Salmonella enteritidis. d P se udo monas ae ru ginos a. e Enterbactercloacae. r Bacillus subtilis. E Colle tot richum g locosporuide s. n Rhizoctonia solani. I Sclerotinia sclerutium. i Gui7rurdia sp. k Fusarium orysporum. I Candidaalbicans.

(500 pg/ml), Escherichia coli (1000 pg/rnl), §almqnella enter- activity, as the anti-bacterial component(s) from this plant can itidis (125 FClml), Enterubacter claacoe (250 pÿml), Bacillus be dihræd in 1043 ml of solvent and still inhibits growth of subtilis (1000 p"ÿml) and the mould tungus Canàida albicans Salmonella enteritidis (total activity =l0/,3mVg), followed by (I25 y,elnl) (Table 4) were all above the breakpoint recom'- methanol extract of Erythrorylum laurifulium agairlist Can- mended. Nevertheless, these MIC values, according to Fabry dida albicans (730mÿg), methanol extract of Melia azedarach et al. (1998) are demonstative ofthe potential clinical use and against Enterubacter cloacae (138 mUg) and methanol extract interest of these exEacts as they are crude exftacts of uncertain of Erythrorylam laurifolium against Bacillus subtilis (91 mVg). composition and with components that can have synergistic or BSI values (Table 5) were useful in evaluating the suscep- antagonistic effects. tibility of the different strains of bacteria towards the plant Total activity values (Table 4) revealed that methanol extract extracts investigated. Bacillus subtilis, Enterobacter cloacae and of Michelia chnmpaca has a high magnitude of anti-bacterial Salmonella entertfidis were the three test organisms found to be

Table 3 Results of preliminary anti-microbial screening of crudc methanol exbact using the disk diff.rsion pethod

Scientific name sAâ ECb sEc pAd Ece Bsf CGs RSh sst GSJ oFk CAI Antidesmamadtgsscariense++++++++- -+ Aphloiatheifurmis + - ++ + + + - + Erythmrylumlnurifulium+++++- -+ Mangiftraindica+-++- -+ Meliaazedarach - + + - ++ + - ++ Micheliachartpaca - + ++ + + + - -+ Moringa oleifera - + ++ - ++ ++ -+ (+) Positive result with corected diarreter of inhibition with respect to blank less than 4 mm; (++) positive result with corected diameær of inhibition with respect to blank more than 4 mm; (-) neg4tive, no distinct zone of inhibition. a S lap hy loc occu s au@u s - b Eschcrichiotcoli. c Salmonella ente ritidis. d P s e udomonas oe ru ginosa. e Enteroba.cte r claacae. r Bacillus subtilis- e C oile tot r ichum g locos po mide s. h Rhizoctonia solani. ' Sclerutinia sclerutium. ) Guignardia sp. k Fusaiurnoxysporum. I Cardidaalbicaw. F 336 O. Rangasanry et al. / Joumal of Ethnopharmacology 109 (2N7) 331-337

Table 4 Cladospoium cucumerinum is often used to detect the pres- Minimum inhibitory concentration values ofthe most active methanol extracts ence of anti-fungal compounds in plant extracts. Rahalison et al. Test organisms MIC Total activity (1993) investigated the activity of several plant extracts against (peltnl) (nÿg) the phytopathogenic fungus Cladospoium cucilmerinum arrd yeast Candida albicans. Out of 20 plant-derived Anti.desma Salmonelln enteitidis 125 326 the motld madagascariense Staphylococcus aureus s00 81 compounds they analysed, they found that 15 gave a positive response with Cladosporium cucumerinum, out of which 13 Aphloia thzifurmis S at ml, n e ltn ent e ri tid is 500 74 present Staphylococcus aureus 500 't4 were found active against Candida albicans. In the work, results of bioautography with Cladnsporium cucumer- Erythrcrylum Bacillus subtilis r000 91 the extracts exhibited by far the most lnurifolium Candida albicans 125 730 inumrcvealedthat DCM the S almone I la ente ritidi s 500 r82 appreciable activity, followed by the hexane extracts while methanol extracts demonstrated no activity. The results on the Mangifera inlica Candidc albicans 250 54 Enterobacter cloacae 250 54 bioautograms supported those obtained via the disk diffusion Salmnella enteritilis 250 54 assay, as in both case, the DCM exuacts illustrated the most out- of the different plants pos- Melia azedarach Candida albicans 250 n11 standing activity. The DCM extracts Enterobacter cloacae 500 138 sessed broador spectrum of activity than extracts of methanol, Salmone I la ente ritid.is 500 138 ranging from Colletotrichum gloc osporoides, S clerotinia scle - (Table 2). The Michelia champaca S almonell.a e nte ritidis 125 ro43 rotium, Guignardia sp. arrd Fusarium orysporum serial extraction method used here can account forthe low anti- Moinga oleifera Salmonella enteitidis 500 169 fungal activity of the methanol extracts, as the previous solvent, DCM, had most probably removed the active components from Table 5 the plant matÊrials. Manifest activity against Colletotrichum glo- Bacterial susceptibility index, BSI, calculated for the different strain ofbacteria cosporoides was noted for the DCM extract of Antidesma mada- used for screening of the seven different methanola extractsl ! gascariense and Michelia champaca in particular (Table 2). Test organisms No. of active extracts BSI values The thorough literature search that was achieved prior to the beginning of the experiments had yielded a wide choice Bacillus subtilis 7 100 ' quantifying the activity plant Entembarter cloacae 7 100 of protocols that aimed at of Eschzrichia coli 4 ,ts7.l extracts against frlarhentous fungi. However, as our ambition Pseudomonas aeruginosa 4 57.t in the present study was only to demonsfiate the broad spec- Sa lmone lla ente ritidi s 7 100 trum of activity of the different plants chosen, the disk diffusion Staphylococcus aureus ) 28.6 assay employed was found suitable and easily manageable for u BSI values were sought only forthe methanolic extacts, on which emphasis such preliminary screening. The anti-fungal activity recorded was laid due to their high activity. (Table 3) for some of the extracts were at levels that hint at a probable therapeutic worth. We intent to pursue the in vitro antt' susceptible to the crude methanol extracts of the different plants fungal experiments with a broader collection of yeast, mould investigated (1007o susceptibility), followedby Escherichia coli and fllamentous fungi using established protocols to determine (57.l%o) and Pseudomonas aeruginosa (57.17o), and Staphylo- MIC values. coc cus aureus (28.6Vo). After a carèful examination of the results of bioautography, Percent activity values recorded, further rationalized the folk- the DCM extract of Michelia champaca, which possessed a loric use of these plants in the treatment of infectious diseases maximum of five clearly distinguishable Cladosporium cuc- as the values in general were found to be above 507o (Table 6). ume rinum growth inhibiting compounds (Rr = 0. 1 3, O.32, 0.43, Methanol extract of Antidesma madaga^scaiense in particular 0.61 and 0.97) was further fractionated on a Sephadex column, showed noticeable efflciency (1007o activiÿ) against the differ- using DClWmethanol 1/1 as eluant. This yielded six different ent bacterial strain used (Table 6). fractions, all of which exhibited growth of Cladosporium cuc- umerinum. The fraction eluted last being predominantly more growthof Table 6 active with at least two compounds inhibiting the Cla: Percent activiÿ values of the differont methanolic extracts, demonstrating the do sporiurn c ucumerinum (R1 = 0.54, 0.74). total anti-microbial poæncy oftbe extracts The results clearly indicate that the different plants screened which agrees with the Plants Number of susceptible Percent activity possess substantial anti-microbial activity, bacterial strains values (%) use of these plants in the traditional Mauritian pharmacopoeia as plants having therapeutic anti-infective potential. This'prelimi- An ti d e sm a mada g as c a r i e n s e 6 100 these extracts shows Aphloia theiformis 5 83.3 nary investigation of the activiÿ of crude Erythmrylum laurifolium 4 66.7 that it is important to continue screening medicinal plants as an Mangiftra inlica 3 50 alternative for finding new or better anti-microbials. Melia azedsrach 4 «.7 As far as we know, this study is the first in Mauritius to Michelia champaca 5 83.3 demonstrate the anti-infective properties of medicinal plants Moinga oleifera 4 6.7 using the referred micro-diluüon assays. Moreover, the wide !F O. Rangasamy et al. / Iounnl ofEthnophannacology 109 (2N7) 331-337 difference in polarity of the anti-microbial components detected Homans, 4.L., Fuchs, 4., 1970. Direct bioautography on thin layer chro- may suggest possible clinical application, Active components matograms as a method for detecting firngitoxic subsiances. Joumal of are being isolated for certain plant extracts for chemical Chromatography 51, 327 -329. Klepser, M.E., Banevicius, M.A., Qinüliani, R., Nightingale, C.H., 1996. Char- characterisation. acterization of bacæricidal actiüÿ of clindamycin agaitst Bactemides fragilis ia kill curve methods. Antimicrobial Agents and Chemotherapy, Acknowledge-"oO 1941-1944. . Martini, N., Eloff, J.N., 1998. The preliminary isolation of several ântibacterial compounds from Combretwn erythmphyllum (Combretaceae). Journal of The authors wish to thank the following institutions: Univer- Etlrnopharmacol ogy 62, 255'263. 'Université sity of Mauritius, the catholique de Louvain' and Matsen, J.M., lylg. Baclsir^l susceptibility testing and assays. In: Henry, J.B. the Commission universitaire porr le développement (CLJD) (Ed.), Clinical Diagnosis and taboratory Method§. \YB Saunders, Philadel- as well as the Tertiary Education Commission, Mauritius, for phia, pp. 1900-1938. financial support. The contribution of the Institut Malgache des Mitscher, L.A., Drake, S., Gollapudi, A.R., Okwute, S'K., 1987' A modem Joumal of NaÜral Products 50' Recherches Appliquées, IMRA, is deeply acknowledged for look at folklore use of anti-infective agents. lû25-1M0. providing laboratory facilities. Murray, C.J.L., Lopez, A.D., 1997. Mortality by cause for eight regions of lhe world: global burden of disease study. Lancet 349,1269-1276. Referenees National Committee for Clinical Laboratory Standards (NCCLS), 1990' Meth- ods for Dilution Antimicrobial Susceptibility Tests for Bacteria that Grows (M7-A2). Villanova, PA. Begue, W.J., Kline, R.M., 1972. The use of tetr^zolium salts in bioautographic Aerobically. Approved Standard NCCLS, procedures. Iournal of Chromatography 64, 182-184. National Committee for Clinical LS), 1992. Bonjar, G.H.S.,2@4. New approaches in screening for antibact€rials in plants. Approved StandedMl00-S4 (M forAntimi- Villanova, PA (Fourth Infor- Asian Joumal of Plant Sciences 3, 55-60. crcbial Disk §uscèptibility Testing. NCCLS, Eloff, J.N., 2000. On expressing the antibacterial acüvity of plant extracts-a mational §upplement). prescdbing: the cure small first step in applÿng scientiflc knowledge to rural primary health care. Peterson, L.R., Dalhotr, A.2ffJ/.. Towards targeted will through improved South African Joumal of Science 96, 16-l I 8. for antimicrobial resistance be specific, directed therapy diagnostic testing? Ioumal of Antimicrobial Chemotherapy 53, 9O2-9O5. Eloff, J.N.,2004. Quantifying the bioactivity of plant expcts during screening and the search for new jun- and bioassay-güded fractionation. Phytomedicine 71, 370-37 l. Plotkin, M.J., 1988. Conservation, ethnobotany Fabry, $[., Okema, P.O., Ansorg, R., 1998. Antibacterial activity of plant gle medicines: pharmacognosy comes of ageagain. Pharmacotherapy 8, extracts of East African medicina.l plants. Joumal of Ethnopharmacology 6Q 257-262. 79-84. Rahdison, L., Hamburger, M., Monod, M., Frenk, E., Hostettmann, K., 1993' bioauto- Gurib-Fakim, 4.,2û2. Mauritius through its medicinal plantstoÿards a betoer Antifungal tests ilr, phytochemical investigations: comparison of pathogenic fungi. Planta understanding of medicinal plants of the Indian Ocean Islands. Edition Le graphic methods using phytopathogenic and human Printemps, Mauritius. Medica 60,4l-.44. Art Press, Rochester, Hoareau, L., Da Silve, E.J-, 1999. Medicinal plants: a re-emerging health aid. Shultes, R.E., Hofmzrn, A., 1992. Plants of Gods. Healing Journal of Biotechnology 2. Vermont.

F P.16: Y. Yvon, EG. Raoelison, R. Razafindrazaka, A. Randriantsoa, M. Romdhane, N. Chabir, M.G. Mkaddem, J. Bouajila. Relation between chemical composition or antioxidant activity and antihypertensive activity of six essential oils. Journal of Food Science 2012, 77(8), H184-H191.

R M EO EO , μ Laurus 0.98). Plasmodium L. nobilis = EO induced Melaleuca leaves, 14 , 2 Hyptis fruticosa 0.05 mg/L) and R Lauris nobilis 0.76). A good ± , and were also evaluated = Eucalyptus gracilis Melaleuca armillaris 2 EO from Portugal has (1.24 H. fruticosa R J. phoenicea EO was able to antagonize T.capitatus Lauris nobilis , M). EO was attenuated significantly 27 mg/L) and a good value has 2012 Institute of Food Technologists L. nobilis M. armillaris 6 Further reproduction without permission is prohibited C − T. capitatus = uniperus phoenicea -myrcene ( M. villosa β 12 mg/L). Mentha x villosa ,J 50 = IC 50 . In the 1,1-diphenyl-2-picrylhydrazyl M. villosa IC Thymus capitatus 0.90), and = E. gracilis , 32 compounds (representing 98.7% of total oil) 2 FcB1 strain ( Eucalyptus gracilis R EO (Barla and others 2007), for 2 mg/L, respectively). Correlations between chemical ± The cardiovascular activity of some EOs was evaluated. L. nobilis been reported (Ferreiracytotoxic and activities others 2006). ofelsewhere The EO (Chabir antimalarial of and and EO others had 2011). mild activity They againstfalciparum have the chloroquine-resistant found that the and others (2004)increasing have concentrations shown of that in isolated rat aortic rings, Cardiovascular effects of EOs have beenothers reported (Guedes 2007). andeffects Results others for 2004; experiments have carried Santosrings shown out and of on noticeable rat rats.concentration-dependent superior In and relaxations intact mesenteric of and tonus beneficial artery, isolated inducedphenylephrine by 10 (Santos and others 2007). In addition, Guedes Bounatirou and others 2007; Amartinobilis and others 2008), for (Hayouni and others(AChE) 2008). inhibitory Study capacity of on the acetylcholinesterase been found concerning cytotoxicbreast cancer activity cells against ( MCF7 human the effects of phenylephrine (1and mM), prostaglandin F2a KCl (10activity mM), (80 induced by mM)-induced contractions. The vasorelaxant (Ben Marzoug and others 2010), and for berries, 11 compounds (representing 99.6% of total oil) for -elemene ( β (leaves and berries), 3and59 ´ ePaul- ´ e Chim- ± Thymus 0.86), J. phoenicea = ´ eactivit 2 R ), but were inactive e´ Razafindrazaka, Adolphe Randriantsoa, Mehrez Romdhane, Naziha Chabir, leaves and berries EOs Juniperus phoenicea ´ eculaires et R ´ ´ es, Tunisia. Direct inquiries to author elisation, Analyse et commande des leaves and berries EOs have -azinobis-3-ethylbenzothiazoline-6-sulfonate assay. Antihypertensive activity was evaluated , were screened for their antioxidant and antihypertensive activity as well as their chemical Vol. 77, Nr. 8, 2012 doi: 10.1111/j.1750-3841.2012.02812.x r J. phoenicea ´ edeMod Mucor ramamnianus ´ enieur, 6029 Gab Thymus capitatus , , and 26 compounds (representing 99.3% of total oil) for Univ. de Toulouse, UMR CNRS 5623, Universit Eucalyptus gracilis , Juniperus phoenicea Six essential oils (EOs), antihypertensive activity, antioxidant activity, essential oil, , 32 compounds (representing 98.9% of total oil) for EO strongly inhibited the growth of gram-positive ,and EO (Kandil and others 1994; Daferera and others 2003; Journal of Food Science M. armillaris Antioxidant activity can contributeno to study the has prevention of beenEOs the reported can increase until find of now its the of interest blood correlation and pressure. between application According antihypertensive in to activity a the and medicinal literature, area. antioxidant activity. Natural were most active for an antihypertensivecomposition activity or (29 antioxidant activity and/orfor antihypertensive activity antihypertensive activity were studied. and Significant p-cymene correlation ( has been found correlation has been found between antihypertensive activity and antioxidant activity by DPPH assay ( by testing the vasorelaxing capacity of EOs on rat aorta precontracted by phenylephrine (10 Keywords: Abstract: Melaleuca armillaris (DPPH) assay, the antioxidant activity wasgave the in best the activity in range the of 2,2 0.59 to 2183.6 mg/L, whereas armillaris for compositions. We identified andcompounds quantified (representing 24 98.8% of compoundsT. total (representing capitatus oil) 99.8% for of total oil) for ` emes, Ecole Nationale d’Ing Antimicrobial and antibacterial activities were tested for Essential Oils (EOs) are mainly used for their medicinal Bouajila (E-mail: [email protected]). Sabatier, 118 routeRazafindrazaka, de and Narbonne, Randriantsoa F-31062 areisation, with Toulouse, Laboratoire France. IMRA, de Authors BP: Phytochimie Raoelison, Chabir, et 3833, and Standard- Antananarivo Mkaddem 101, aresyst Madagascar. with Authors Unit Romdhane, H184 MS 20120131 Submitted 1/26/2012,Bouajila Accepted are 5/12/2012. Authorsique with Yvon et Laboratoire and Photochimique des Interactions Mol Introduction Activity for Six Essential Oils Yan Yvon, Emmanuel Guy Raoelison, Ren Mounira Guedri Mkaddem, and Jalloul Bouajila Relation between Chemical Composition or Antioxidant Activity and Antihypertensive against gram-negative strains (Ennajar and othersactivities 2009). Cytotoxic were also shown(El-Sawi for and others 2007). capitatus properties (Burt 2004; Bakkaliof and others biological 2008). Determinations activitiesstudies have of beencarried chemical commonly out composition found for been of with described EOs. the (EnnajarJ. Antimicrobial and phoenicea tests othersmicroorganisms 2009). and fungi They ( showed that

H: Health, Nutrition & Food 07adcniin fE xrcinwr h aea cited as same the were extraction 2009). others EO and (Ennajar of above conditions (Gab Tunisia and of south 2007 Concerning from 2009). collected were others parts and aerial same (Ennajar the above were extraction cited EO as of Conditions 2009. April in Tunisia nobilis, L. and 2010), et,adcreain eecekdbtenantihypertensive between potential. antioxidative checked and/or composition chemical were with activity correlations antihypertensive and and antioxidant tests, on obtained results presents ticle Enjradohr 2009), others and (Ennajar eeprhsdfo im,Adih lk (Saint-Quentin, Fluka Aldrich, Sigma, France). from purchased were theEOsaredescribedelsewhere: eprtr a rdal asdfo 0t 260 to 60 from raised gradually was temperature nltcltechniques Analytical used Chemicals methods distillation and material Plant Methods and Material oils: essentials 6 berries, these for literature in mined of receptors. opiate mechanism of have participation a the involves test) with probably that effect, formalin action antinociceptive and dose-dependent test; administered a plate orally (acetic-acid- that hot out shown test; carried tests writhing different the induced From mice. on effects Ara 2009). others plate (De and hot work (Amorim Moreover, effect in hypothermic time from a showed latency and isolated test, the EO increased acetic-acid-induced constrictions, The the abdominal inhibited 2009). significantly others leaves pitungueira and antinociceptive (Amorim of an evaluated effect Moreover, hypothermic removal. and endothelium either by . . . activity antihypertensive and activity antioxidant or composition chemical between Relation nerto fec ekwscekdadcretdmnal if manually corrected and checked was necessary. peak each The (Varian). of software au- Workstation integration performed 2100 were Saturn quantification with tomatically correction and the of integration use in the Peak injected without factors. data sample was percentage electron- area obtained mL diluted were FID data 1.0 from A Quantitative ically of 10). of : min. rate 1 v/v) (ratio flow 57 ether, mode split petroleum a was in at time 100 gas, 340 : analysis (1 to Total carrier the raised mL/min. was finally 1 99.999%) and (purity min, Helium 15 for held aeinzto eetr(I)adaD-M ailr column capillary a DB-5MS with a m equipped and (30 (FID) chromatograph detector gas ionization 3400Cx flame Star France) Ulis, are u on out carried aunG/SM Dms-eetv eetri h electron the GC in Varian detector for a mass-selective with as 4D equipped GC/MS/MS chromatograph conditions using gas Saturn gas) 3400 same carrier Star the Varian of the the rate flow under and temperature, oven performed (column, was EOs eprtrswr e t20ad270 and 200 at set were temperatures a chromatography. Gas reagents All grade. reagent analytical of were used chemicals All of isolation for conditions and material plant of Collection oorkolde oathpresv ciiyhsbe deter- been has activity antihypertensive no knowledge, our To a hoaorpyms spectrometry. chromatography-mass Gas × .capitatus T. .5m;fimtikes 0.25 thickness, film mm; 0.25 rs evswr olce rmrgo fMannouba of region from collected were leaves fresh .armillaris M. lii zerumbet Alpinia , .nobilis L. Cai n tes21) Concerning 2011). others and (Chabir j ih n tes20)hsbeen has 2005) others and Pinho ujo ` O eeaaye sn ain(Les Varian a using analyzed were EOs , .uniflora E. .capitatus T. .gracilis E. Ot vlaeantinociceptive evaluate to EO lii zerumbet Alpinia .phoenicea J. ,and ◦ μ ,rsetvl.Teoven The respectively. C, ) netraddetector and Injector m). Oo iehsbeen has mice on EO Madmadothers and (Mkaddem .armillaris M. .phoenicea J. ◦ Cat40 evsadberries and leaves ◦ Cat5 Opromotes EO s nMarch in es) ´ nlssof Analysis evsand leaves .gracilis, E. .Thisar- ◦ ◦ C/min. C/min, as aaiyo rerdclsaegn a xrse by expressed was scavenging The standard. radical a free as used of measured was capacity acid was Ascorbic absorbance mixing. initial the after then min fresh 6 and with mL), react (900 to allowed solution was ABTS mL) diluted (100 EO the nm of sample, 734 solution at each methanol units For diluted 0.02 spectrophotometer. was to Helios 0.70 mixture the of using The absorbance an use. give before to at ethanol h dark with 16 the for in temperature storage by room followed concentration) (final persulfate 5mMNa ss( yses ocuea5%dces niiilDP ocnrto.The concentration. DPPH initial in decrease 50% a cause to O)(. L eemxdwt . Lo . Mmethanolic 25 mM at min 0.2 30 a of period of incubation mL an 1.5 After solution. with DPPH mixed or were antioxidant (pure mL) (1.5 materials test EOs) modifi- the some with of (1958) dilutions Various Blois by cations. described as radical free DPPH assay. ing aeil n h bobnea eoddas recorded test the as without absorbance solution the a to and procedure material, same carried also the was applying experiment blank out A UK). Cambridge, (Unicam, CM nlssrslsaegvni al .Aldeterminations and All GC averaged. 1. and KI. Table duplicate of in in calculation given performed were are the results C24) in analysis to points GC-MS (C5 reference Alkanes as or 2.62). used commercial version were in (NIST02 2.62). contained libraries those version literature with 02 values contained MS (NIST comparing experimental those by libraries obtained was with compounds literature of data identification Tentative or experimental commercial coinjection MS values, in and (KI) Index standards, Kovats based their of identified of were comparison components the The on (amu). units mass atomic ecnaeihbto codn otefloigequation: as following calculated the then to was according inhibition solution percentage each of activity scavenging radical nixdn activity Antioxidant a eemnduigtesm qainpeiul sdfrthe for used previously equation method. same DPPH the using determined was coefficient correlation line a (least-squares of analyses calculation regression with linear method using identified was eeae ymxn MAT tp . 5m NaH mM (5 7.4 pH was at ABTS ABTS 1999). mM deter- others 7 was and mixing action by (Re generated radical elsewhere ABTS described the as mined for assay. antioxidants of scavenging capacity radical (ABTS) triplicate. 0 fAT aias h aaiyo rerdclscavenging scavenge radical free to of capacity required The radicals. concentration ABTS of the 50% represents which value, PH a eoddas of recorded absorbance maximum was of wavelength DPPH, the nm, 520 at absorbance ag fstsato ewe .9 n .9) l measurements triplicate. All in 0.998). performed and were 0.994 between satisfaction of range f1 Aadeeto utpirvlaea 50V h trap The V. 1500 current at emission voltage multiplier an 250 electron for was and temperature adjusted mA was 10 of spectrometer mass The ietmeaue htwr e t20ad300 and transfer 200 spectrometer at mass set were and that Injector temperatures eV). line (70 mode impact nixdn ciiyo etcmonso xrcswsexpressed was extracts or compounds test of activity Antioxidant ,-ihnl2pcyhday DP)rdclscaveng- radical (DPPH) 1,1-Diphenyl-2-picrylhydrazyl 2,2 IC 50 R  -Azinobis-3-ethylbenzothiazoline-6-sulfonate 2 enda h ocnrto ftets aeilrequired material test the of concentration the as defined , = 2 HPO .9 o099.Almaueet eepromdin performed were measurements All 0.999). to 0.992 nixdn cvnigatvt a tde sn the using studied was activity scavenging Antioxidant % o.7,N.8 2012 8, Nr. 77, Vol. inhibition 4 IC n 5 MNC)wt . Mpotassium mM 2.5 with NaCl) mM 154 and , 50 ◦ a dnie sn ierrgeso anal- regression linear using identified was ,adms cnigwsfo 0t 650 to 40 from was scanning mass and C, A = sample 100 sn eisspectrophotometer Helios a using , × r ora fFo Science Food of Journal ( A blank A h aia scavenging radical The − blank A sample A ◦ ,respectively. C, blank ) IC R .Thefree 2 50 ihthe with (mg/L) ◦ 2 ,the C, H185 PO IC IC (1) 50 50 4 ,

H: Health, Nutrition & Food ) 80 03 07 12 06 26 36 95 04 02 05 56 09 05 04 02 38 33 30 05 44 16 25 02 44 12 13 05 13 ...... Continued ( 10 71 60 30 50 80 80 50 ...... 95 5–– 780 9–0 92 71 3–– 7–0 01 8–– 7–– 1–0 80 11 5–– 8–– 74 8–0 2–– 0–0 7–0 6–– 6–– 7–– 7–– 8–0 5–– ...... 2– – 85 4– – – 2– – – 2– – – 1– – 0 5– – – 31 1– – – 555 0– – – 11 1– – – 41 30 2– – – 4– – – 5– – 0 1– – – 5– – 5 30 4– – 0 7– – 0 31 1– – – 1– – – 3– – – 7– 1 10 6– – 0 6– – – 12 2– – 0 20 2– – – 62 2– – – 3– – 0 0– – – 6– – – 4– – – 1– – – ...... 22 – – – – 68 0 19 0 10 1 05 1 40 – 0 08 7 72 – – – – 04 0 06 0 09 0 ...... 2– – – – 0 5– – – – – 3– 6 23 2– – – – – 1– 0 1– – 0 1– 0 2– 6 5– –20 – – – 4– 0 1– 0 7– – 2 2– –3– – – – – – – – 55 8– – 0 5– – 0 3– – – – – 6– – – – – 0–20 20 ...... ––0 ––0 E. gracilis T. capitatus L. nobilis J. phoenicea leaves J. phoenicea berries M. armillaris Vol. 77, Nr. 8, 2012 r ∗ ∗ -Cadinene – – – 2 -Terpinene 0 -Elemene – – – 1 -Muurolene 0 -Phellandrene 0 -Elemene – – 0 -Caryophyllene – 0 -Terpineol – – 0 -Gurjunene – – 0 -Pinene – – 0 -Caryophyllene oxide 0 -Myrcene – – 0 -Selinene – – 2 -Thujene – – – – – 0 -Pinene 2 -Cadinene – – 0 -Fenchene 0 -Terpinolene 0 -Calacorene – – – 0 -Himachalene – – 0 -Campholenal -Humulene – 0 -Terpineol – 0 -Terpinene – 0 -Amorphene – – – 0 -Chamigrene – – – 0 -Phellandrene – – – 0 -Cubebene – – – 1 -3-Carene – – – 10 β γ α β α γ β α α α α β γ β β β α α β α α γ δ α β α α α α Chemical composition (%) of the 6 studied EOs. Journal of Food Science – 1376 Neryl acetate – – 0 1033 1513 1137 4-Acetyl-1-methyl-Cyclohexene 0 930 KI866 m-Xylene Compounds – – – – – 0 1389 1038 Eucalyptol – – 0 1514 Cubenol – – 0 1140 Cis-verbenol – – 0 936 1406 Methyl eugenol – – 2 1057 1524 d-Cadinene – – 0 11411142 Cis-sabinol Cis-verbenol 4 – – 0 1415 1534 1143 Camphor – – 34 951 1086 1540 1159 954 Camphene 0 1428 1432 1094 Linalool – – 1 15501554 Elemol Germacrene B – – – – – – 2 – – 0 1166 Borneol 0 974 1435 Terpinyl propionate – – 0 1108 Fenchol – – 0 15761580 Spathulenol 1 1168 Pinocarvone 1 976 Sabinene – – 2 1450 1128 1178 Terpinen-4-ol – 0 985 1454 1185 14601480 Allo-aromadendrene – – – 0 1185 p-Cymen-8-ol 0 1011 1480 Germacrene D – – 0 12021208 Myrtenol Verbenone 0 0 1025 p-Cymene 4 1018 1482 1210 Verbenone – – 0 1488 1237 Cuminaldehyde 1 1028 Limonene 0 1502 12721280 Isopulegyl1285 acetate Piperitone Bornyl acetate – – 0 – – – 5 0 1030 1512 Calamenene 1288 p-Cymen-7-ol 0 1030 1,8-Cineole 78 1290 Thymol – 89 1302 Carvacrol 0 1345 1350 Eugenol – – 0 1372 Carvacrol acetate – 0 H186 Relation between chemical composition or antioxidant activity and antihypertensive activity . .Table . 1

H: Health, Nutrition & Food odtrieterrespective their determine to CO n uuaiecnetain nari ig rcnrce with precontracted 10 rings phenylephrine aortic increasing on with concentrations tested cumulative were and extracts active the studies, detailed For ocnrto rdcn 0 ftemxmmrsos:for 50% response: producing concentration maximum the represents the it of product, relaxing 50% producing concentration sltdadctit ig f2m nlnt.Ternswere Krebs rings containing The baths NaHCO length. organ mM, 5.9 in in KCl was g mM, mm 122 2 aorta (NaCl 2 of solution thoracic tension of the a rings were under killed; into mounted g were cut 300 Animals and and study. 200 isolated between the weighing in sex used either of rats Wistar ie n loe oeulbaei rb ouinfr3 min. 30 for solution 3 Krebs rinsed in (10 was equilibrate phenylephrine organ to Then, the allowed obtained, and contraction times of maximum the wasaddedtotheorganbath(20mL)tosensitizetheorgan.Once spretg ftesed-tt otato nue ythe by induced contraction steady-state (10 the phenylephrine of percentage as regression. linear using calculated ftetse Owsaddit h ra ah(0m)to mL) (20 bath organ the the as into expressed was added that was activity vasorelaxant EO the evaluate tested mg/mL 0.5 the to steady-state 0.05 of the from At concentrations in cumulative used. product level was contraction, the manner single of concentration addition a cumulative product, using a tested the of of Instead concentration of undertaken. was above described ot.Atrteeulbainpro,peyehie(10 phenylephrine period, precontracted equilibration phenylephrine test, the on preliminary After tested For aorta. was committee. activity international ethic relaxant local with the the accordance and in All guidelines transducer. performed force were were isometric relaxations 7003 experiments and period, Ugobasile Contractions equilibration an g. this with 2 of recorded end to the readjusted At was min. tension 30 every rinsed and nfnto fconcentration, of function in ouinwsmitie t37 at maintained was solution al 1–Continued. Table CaCl mM, 1.25 activity Vasorelaxant . . . activity antihypertensive and activity antioxidant or composition chemical between Relation O(.5m/L a netdit h ahslto orelax and to EO solution bath calculated. was the value tested relaxation into percentage the injected The contraction, was aorta. mg/mL) steady-state (0.25 the EO At contraction. induce dHpae24dmtya–––0 0 – – – – – – p-Menth-4(8)-en-9-ol 2,4-dimethyla Heptane Dihydroumbellulone nd Trans-ferruginol nd nd 2328 68Sahlnl–––––0 0 – – – – – – – – – 13-Isopimaradiene 1960 1654 Spathulenol 1651 1648 1638 ∗ Compounds 1596 KI ettvl dnie upre ygo ac fM.nd MS. of match good by supported identified Tentatively eaainpretgsidcdb h xrcswr calculated. were extracts the by induced percentages Relaxation aoeaatatvt a tde yuigioae a aorta. rat isolated using by studied was activity Vasorelaxant h eaatefc ftetse rdcswsexpressed was products tested the of effect relaxant The h otatv O o ecnaerelaxation, percentage for EOs active most 2 The 2 n 5 O 95% and tes8 Others eqiepnsOyeae 2 Oxygenated Sesquiterpenes eqiepnshdoabn 0 hydrocarbons Sesquiterpenes ooepnsOyeae 84 Oxygenated Monoterpenes ooepn yrcros3 hydrocarbons Monoterpene α β γ α hnlc 0 Phenolics .nobilis L. Cdnl–––––0 – – – – – -Cadinol Cdo 0 – – -Cedrol Edso 0 -Eudesmol Edso 0 -Eudesmol 2 2 .Arawr qiirtdi h eimfr2h 2 for medium the in equilibrated were Aorta ). − − 6 O eetse o hi aoeaigeffect vasorelaxing their for tested were EO, .5m,adguoe1 M.Tebath The mM). 11 glucose and mM, 1.25 6 .Fo h uv hwn nue effect induced showing curve the From M. ) The M). − ∗ 6 ∗ M)wasaddedtotheorganbathto ◦ EC n asdwt abgn(5% carbogen with gassed and C EC EC 50 50 50 ausfrrlxto were relaxation for values .gaii .cpttsL oii .poncalae .poncabrisM armillaris M. berries phoenicea J. leaves phoenicea J. nobilis L. capitatus T. gracilis E. h aeexperience same The . whichisdefinedasthe 0 – 0 – – ––– ...... 15 80 11 60 ––––– – 53 – – 1– ––––– – – – – – – 6– – 5– 289 = o eemnd – o detected. not “–" determined. Not 3 5m,MgCl mM, 15 .capitatus T...... 67 26 02 10 84 08 373 13 110 91 51 15 EC − 5 50 M) . 2 nixdn rbooia ciiywr acltduigM Excel MS function). using statistical and calculated (CORREL software were composition activity chemical biological or between antioxidant relationship the determine h otato rdcn otato f5%o h maximal the of 50% aorta. is of the it of contraction and contraction a agonist, producing contractile for contraction and the contraction the of relaxation BSatoiattssadvsrlxn ciiytest. chemical activity vasorelaxant EO and and of tested DPPH tests tests: Each antioxidant pharmacological EO. Identification ABTS different tested to each submitted EOs. then for was made in also was content content composition compounds 2009), terpenoids phenolics others total and and the (Khanavi quantified established separately we well is activity tioxidant ( coefficients Correlation values. obtained the of confi- majority at The set measurements. were triplicate limits dence of (SD) deviations standard considered were and when test significant Student’s be a to using mean analyzed the were Values as represented are results eelwrta nlae O.Madmadohr 21)have (2010) others and Mkaddem that EOs. cited leaves in than lower were of contents the whereas con- the instance, for of EOs, tents Berries different. was from berries obtained and EOs leaves of composition chemical The (2.9%). cadinene EOs, leaves in hydrocarbons are (5.1%), EOs cadinene that of berries EOs in explained found of compounds constituents have Major main the (2009) were monoterpenes others 1). (Table elsewhere and discussed and Ennajar studied been have 2011) others cilis hmclcmoiino EOs of composition Chemical Discussion and Results analysis Statistical ihol 1cmoet.Peoisrpeetd8.5 ftotal of 89.15% represented Phenolics (89.06%) components. content 11 thymol only high with and (0.09%) content carvacrol low 2009), others and hmclcmoiinof composition Chemical an- and compounds phenolics between correlation the As means as expressed were activity antioxidant of data All The experiments. interday 3 in repeated was test relaxant Each ...... BnMrogadohr 00,and 2010), others and Marzoug (Ben 82 21 418 13 872 ––– – 3– ––– – 1– α -pinene, .capitatus T. o.7,N.8 2012 8, Nr. 77, Vol. α δ pnn (55.7%), -pinene .capitatus T. 3crn 45) and (4.5%), -3-carene γ cdnn and -cadinene ...... Oi eetdceoyecaatrzdby characterized chemotype selected a is EO 00 500 5101 4086 588 6–– 6–– P δ P -3-carene, < < 0.05. .phoenicea J. Madmadohr 2010), others and (Mkaddem .5 Dddntece %frthe for 5% exceed not did SD 0.05. r ora fFo Science Food of Journal β α δ crohlee eehigher, were -caryophyllene, ± trio,adgrarn B germacrene and -terpinol, 3crn 1.%,and (10.7%), -3-carene evsadbris(Ennajar berries and leaves .. of S.E.M .armillaris M. β . . . 50 310 5– crohlee(2.6%); -caryophyllene α pnn (80.7%), -pinene n observations. Cai and (Chabir .phoenicea J. R .gra- E...... H187 03 77 05 51 33 13 02 22 2 )to γ γ ± - - .

H: Health, Nutrition & Food - - - 1) β β γ and = n 3and 0.85), 0.80), ,leaves All EOs = activities = L. nobilis ± -blockers, 2 in mg/L) 2 J. phoenicea . 0.98) have β R R 50 0.89), and 50 = in vitro = 2 EC IC . T. capitatus 2 -terpinene, R 2) 139 ( -himachalene, and test what can give R γ showed a moderate α = EO (80%) has a high (35%) and (96.9%) and n -blockers, M. armillaris EO 30, 306, and 139 mg/L, α -terpinene ( 1( 1mg/L γ -muurolene, and ± in vivo ± γ ± E. gracilis E. gracilis -muurolene ( T. capitatus γ 1, 240 -Muurolene has the highest value M. armillaris and -himachalene ( γ α 0.98, 0.94, and 0.96, respectively). ± -caryophyllene, β = 3) 350 3 and 350 EOs. Furthermore, more precise works 2 = 0.99), R ± n = 0.73, respectively). As reported in previous 0.77), and 2( J. phoenicea -himachalene, 2 0.99) have presented a good correlation. These ± α = = -myrcene, p-cymene, R β 2 2 L. nobilis = R 2 R R and showed the best antihypertensive activity (29 3) 59 in Table 3) have been carried out on all EOs. Values of -terpinene was previously reported (Ben Marzoug and are between 29 2 mg/L, respectively). In other side, = γ 50 -Pinene, n 50 ± Preliminary tests and quantification by Many treatments of different pharmacodynamic mechanism are α -terpinene ( -muurolene. These correlations are obtained for -caryophyllene ( 3( EC antihypertensive activity (350 been determined in literature for these 6 EOs. and berries EOs of respectively). To our knowledge, no antihypertensive activity has L. nobilis 59 angiotensin antagonism, angiotensinbition, converting nervous enzyme systemand inhi- others inhibitors, 2005).capacity and of For vasodilatators the the EOs,precontracted (Blacher rat vasorelaxing evaluation aorta activity was of test used. on the phenylephrine antihypertensive caryophyllene, caryophyllene oxide are thein these most 6 frequently studiedthese found EOs. compounds compounds Correlationactivity. has and been Significant both checked antioxidant between DPPH- activity and (low ABTS-scanvenging Only phenolicsDPPH presented assay a ( satisfactory correlation with the available to treat hypertension: diuretics, γ Antihypertensive activity of EOs sesquiterpenes, and phenolics presentedwith a the satisfactory ABTS correlation assay ( studies, the antioxidant activitynolic of contents EOs was (Ennajar related2010), and to hydrocarbons monoterpene others their (Chabir 2009; phe- andsesquiterpenes others Mkaddem compounds 2011), and (Mukazayire and and others others 2011). have been tested forrat their aorta. vasorelaxing Figure activity 1 onobtained showed with precontracted the 250 relaxation mg/L percentageon of on rat the rat aorta EOs. aorta have Best been(82%) relaxation found EOs percentages for and the lowest were berries (47%) EOs. Itvalue appears of that relaxation percentageT. and capitatus close to the highest values for results are logical;might concerning be ABTS attributed assay,Concerning to antioxidant DPPH others assay, activity it chemical appears functions that p-cymene nonphenolic. ( EC β must be found withConcerning ABTS high essay, it percentage appears of that the studied compound. muurolene ( presented a good correlation. ( γ and compounds shoulddifferent be results. evaluated of correlation inand both assays. Antioxidant activity of p-cymene others 2010; Chabirour and others knowledge, 2011). correlationthe No following report between compounds: mentions, antioxidant to activity and ± T. capitatus L. nobilis M. armillaris E. gracilis ± EO. 9 05 9 . . . (mg/L) 3 2 0 4 3 4 -pinene L. nobilis 50 α ± ± ± ± ± ± EO. The 1 3 . . EO has the IC 24 M. armillaris . L. nobilis 1) 29 berries = EO has the best n EO. In addition, results M. armillaris M. armillaris J. phoenicea (73.1%), and -pinene (6.5%), and bornyl EO has the second amount T. capitatus 0.02 mg/L, DPPH assay and α was mainly constituted by 3 247 . (Chabir and others 2011) EO 02 1 981 ± . . -pinene (2.29%) were the main Vol. 77, Nr. 8, 2012 0 2 44 29 159 (mg/L) ABTS EO, mono- and sesquiterpene α -pinene (1.95%), and sabinene r 2) 306 ( ± ± ± ± 50 α T. capitatus 3 6 0.59 . . = 59 L. nobilis IC L. nobilis . leaves n 44.3 mg/L, DPPH assay and 247.3 0 = 1052 236 . Ben Marzoug and others (2010) have 2183 ± 50 30 ( T. capitatus M. armillaris E. gracilis ± IC (84.6%), J. phoenicea 2183.6 species (72.5% and 88.3% for leaves and berries, T. capitatus = values for the tested EOs. 50 50 leaves – 127 berries – 64 E. gracilis IC EC Antioxidative potential of the 6 studied EOs. 0.05 mg/L, ABTS assay) when Journal of Food Science – ± (mg/L) 240 J. phoenicea number of replicate. 50 In this work, correlation between chemical composition The capacity of EOs to give antioxidative effect was studied Chemical composition shows that oxygenated monoterpene -terpineol (7.2%), borneol (6.7%), = “–": not tested. EOsJ. phoenicea DPPH J. phoenicea L. nobilis T. capitatus M. armillaris E. gracilis EC n EOs showed a logical correlation between theantioxidant amount activity. of Indeed, phenolics and of phenolic and showed the second best antioxidant activity. and antioxidant potentialcompounds has families, been monoterpene checked. hydrocarbons, Concerning oxygenated the weakest ( was the major compounds of 3.9 mg/L, ABTS assay).to Significant phenolics antioxidant activity (Mkaddem is and related others 2010), and thymol (89.05%) 1.24 H188 Table 3– Table 2 Antioxidative potential of EOs (85.8%), camphene (5.05%), (1.3%) were the main compounds in composition, and theimportant oxygenated proportions (84.6%). monoterpenes 1,8-Cineole (77.97%), had(4.20%), cis-sabinol p-cymene the (4.55%), and most compounds. Mono- andslight sesquiterpenes constituents hydrocarbons for were(10.33% the and 1.05%, respectively). Thehad oxygenated monoterpenes the most important proportions (86.51%). 1,8-Cineole Relation between chemical composition or antioxidant activity and antihypertensive activity . . . chemical composition. Low content ofwere mono- found and sesquiterpenes in shown that forhydrocarbons represent the 3.5% and 2.8%, respectively, in chemical 2010; Mkaddem andFrom others each 2010;ABTS studied antioxidative Chabir potential EO assays, andantioxidative in activity ( others this 2011). article, in both DPPH and previously (Ennajar and others 2009; Ben Marzoug and others of the majorfrequently compounds in of the studied each EOs. EO,(Table 2) appearing most (86.5%)). Monoterpene hydrocarbonsthe were the majorrespectively). part Finally, for phenolics compounds (89.2%).and Table 1,8-cineole 1 were showed the that 2 chemical compounds, taking part oxygenated monoterpenes have beenEO found (subjected 47.4% article).α in Camphor (34.4%), 1,8-cineoleacetate (5.0%) (20.2%), were detected as major compounds in constituted the major partall of EOs chemical ( compounds for 3 out of

H: Health, Nutrition & Food o10 gL n ueo 1t 00m/)rlxdthe relaxed mg/L) 1000 (1 to EO (1 with the eugenol similarly the and that contraction and showed EO phenylephrine-induced mg/L) the Results 1000 rings with to eugenol. out aortic compound, carried isolated hypertensive been active in have (DOCA-salt) evidence Experiments rats. in deoxycorticosterone-acetate put from effect hypotensive a ciiy ncmaio,or6tse O,wihaeamxof ( mix EOs nobilis 2 a L. context, are an- similar the a which to In response activity. EOs, tihypertensive better have tested compounds, active 6 chemical our many antihypertensive comparison, must the to In Eugenol related activity. rats. compound pressure active aortic pentobarbital-anesthetized an the considered in 40% be to rate up Concentration reduce heart rats. can and eugenol normotensive of mg/kg on 10 eugenol of pure been has with (2004) effect Vasodilator tested others eugenol. of and activity Lahlou the respectively. confirmed mg/L, have 196.8 and 226.9 of ewe ueo otie in contained eugenol between eainbtenceia opsto ratoiatatvt n niyetnieatvt . . . activity antihypertensive and activity antioxidant or composition chemical between Relation eue pt 5.Tyo a ena ciecmon from compound active an rate been heart has this the Thymol In compounds, 35%. to tested administration. up the (i.p.) mg/kg reduced of intraperitoneal concentrations 2 10 of for between and range and mg/kg 0.1 administration 40 between (i.v.) and intravenous Doses in 2003). tested ischemia were myocardial others and and hypertension of (Garzon treatment in used EOs. were our in found compounds chemical (2004). others and Lahlou and (2007) others and Interaminense by neaies n tes(07 eotda observation an reported (2007) others and Interaminense nte td eotdbnfiilcrivsua s fsome of use cardiovascular beneficial reported study Another aemr neetn au of value interesting more have ) cmmgratissimum Ocimum EC α Pnn n derivatives and -Pinene 50 hntoeobtained those than .cpttsand capitatus T. EC 50 Oand EO values on ihteceia aiyo xgntdsesquiterpenes oxygenated of family chemical been ( has the correlation with Significant found checked. been has composition ciiyi oprsnwith comparison in activity where manner, ihpeyehiet 10 to phenylephrine with and the in observed effect cardiac Sprague, ammi Trachyspermum potential antioxidative and composition chemical parameters: 2 and activity antihypertensive between Relation hscs,pse xeiet aebe are u opoethe prove to the and out for activity antihypertensive carried on been effect have concentration experiments pushed case, this 2003). and others transient The and was (Lim pressure. phenomenon shown were and this (percentage) blood that tested data no in fact was the response concentration) related mg/kg authors dose-dependant 10 a rats. to normotensive produced (1 and hypertensive acetate known in Bornyl also pressure was blood acetate Bornyl reduce 1995). and to mg/kg, others pressure 10 and blood (Aftab to rate in (1 heart reductions thymol dose-dependent rats, produced anaesthetized i.v.) In 1995). others and R orlto ewe niyetnieatvt n chemical and activity antihypertensive between Correlation oersos o .cpttsadL oii EOs. nobilis L. and capitatus T. for response Dose 2 .nobilis L. = .nobilis L. .3.Mlclso hs hmclgop a aean have can groups chemical these of Molecules 0.73). a ot tmltdb hnlprn t10 at phenylephrine by stimulated aorta rat capitatus and T. relaxation of percentage between 2–Relation Figure at 10 phenylephrine by precontracted on aorta EOs rat studied thoracic 6 the of effect 1–Antihypertensive Figure O.Terslso iue2soe that showed 2 Figure on results The EOs. o.7,N.8 2012 8, Nr. 77, Vol. O eae h sltdrtaraprecontracted aorta rat isolated the relaxed EOs − 6 .capitatus T. .Tse O ocnrto:20mg/L. 250 concentration: EOs Tested M. and − 6 .nobilis L. .nobilis. L. hwdtebte antihypertensive better the showed hc a yoesv n brady- and hypotensive a has which ihaconcentration-dependent a with M nvivo in r ora fFo Science Food of Journal O ocnrto nisolated on concentration EOs tde Mea13;Aftab 1934; (Maeda studies − 6 M( .capitatus T. .capitatus T. n = H189 3). In

H: Health, Nutrition & Food - β ´ egies Gaertn. J Boil Sci -pinene and p-cymene α -elemene, and ´ erapie 6(6):342–7. β Monodora myristica ` ujo MEM. 2006. The in vitro screening for EOs have the best vasorelaxant 0.98). Further works envisaged to = ´ erielle: quelles recommandations et comment les appli- 2 ` aes PJC, Leal-Cardoso JH, Duarte GP, Lahlou S. 2007. R L. nobilis ¨ umen G, Kingston DGI. 2007. Identification of cytotoxic sesquiter- and ` a DM, Magalh ¨ uz S, T ´ edicale 34(18):1279–85. ¨ ¸a C, Serralheiro MLM, Ara Oks ¸u G, ` T. capitatus ujo Pinho FVS, Coelho-De-Souza AN, Morais SM, Santos CF, Leal-Cardoso JH. 2005. ´ erapeutiques dans l’hypertension art 7:937–42. vascular effects ofnormotensive eugenol, rats. a J Cardiol phenolic Pharmacol compound 43(2):250–7. present in many plant essential oils, in A, Oveisi MR. 2009.some Comparison Stachys of species. the Afr antioxidant J activity Biotechnol and 8:1143–7. total phenoliccomposition contents and in hypotensive effects of essential oil of 105(1):146–55. review. Int J Food Microbiol 94(3):223–53. M, Bouajila J. 2011. ChemicalMelaleuca study armillaris and antimalarial, (Sol antioxidant, Ex and Gateau) anticancer Sm activities essential of oil.growth J of Med Botrytis Food cinerea, 14(11):1383–8. FusariumCrop sp. Prot and 22(1):39–44. Clavibacter michiganensis subsp. michiganensis. Antinociceptive effects of12(6–7):482–6. the essential oil of Alpiniaactivity zerumbet of essential on oils of mice.Tradit Complememt leaves Phytomedicine Altern and berries Med 4(4):417–26. of Juniperus phoenicea L. grownmoquinonated in Nigella Egypt. sativa Afr volatile J in oil rats. and Saudi its Pharm major J components 11(3):104–10. M. 2009. The influenceantioxidant of and organ, antimicrobial season activities andAgric drying of 90(3):462–70. method Juniperus on phoenicea chemical L. composition essential and oils.acetylcholinesterase J inhibition Sci and Food antioxidantEthnopharmacol activity 108(1):31–7. of medicinal plants from Portugal.27. J Bicyclic CB2 canabinoid receptor ligand. U.S. patent 20,050,020,544. hypotensive and vasorelaxant effects ofin rats. the Phytomedicine essential 11(6):490–7. oil from aerial parts ofMelaleuca armillaris Mentha x (Sol. Ex villosa Gaertu)parametric Sm. flow cytometry essential and oil automated on microtiter-based six assay. Food LAB Chem strains 111(3):707–18. as assessedPharmacological by evidence multi- of calcium-channelmum blockade and by its main essential constituent, oilrats. eugenol, of Fund in isolated Ocimum Clin aortic Pharmacol gratissi- rings 21(5):497–506. from DOCA-salt hypertensive tracts and fractions44(1):19–24. of Thymus capitatus exhibit antimicrobial activities. J Ethnopharmacol from Trachyspermum ammi (L.) sprague. Phytomedicine 2(1):35–40. Chaouch A. 2008. Chemicalmoroccan Thymus composition capitatus and and antimicrobial Thymus bleicherianus. activity Phytoth of theand essential hypothermic oils of evaluationuniflora of L. the (Brazilian Pitanga). leaf Phytomedicine essential 16(10):923–8. oil andreview. isolated Food Chem terpenoids Toxicol 46(2):446–75. from Eugenia penes from Laurus nobilis L. Food Chem 104(4):1478–84. hane M. 2010. Eucalyptuschemical composition (gracilis, and antioxidant oleosa, and antimicrobial salubris, activities. J and Med Food salmonophloia) 13(4):1005–12. essentialth oils: their quer? La Presse M 181(4617):1199–200. Barroso JG, Pedro LG.of 2007. Chemical the composition, essential antioxidant and oils antibacterial isolated activities from Tunisian Thymus capitatus Hoff. et Link. Food Chem myrcene showed significant correlationto coefficient. determine Experiments antihypertensive activity have topure be carried molecule. out Antioxidant with antihypertensive activity activity ( could be correlated with Lahlou S, Interaminense LFL, Magalhaes PJC, Leal-Cardoso JH, Duarte GP. 2004. Cardio- Koudou J, Etou Ossibi AW, Aklikokou K, Abena AA, Gbeassor M, Bessiere JM. 2007. Chemical Burt S. 2004. Essential oils: their antibacterial propertiesChabir and N, potential Romdhane applications M, in Valentin A, foods—a Moukarzel B, Ben Marzoug HN, BenDaferera Brahim DJ, N, Mars Ziogas BN, Polissiou MG. 2003. The effectiveness ofDe plant Ara essential oils on the El-Sawi S, Motawae H, Alib A. 2007. Chemical composition, cytotoxic activity and antimicrobial El Tahir KEH, Al-Ajmi MF, Al-Bekairi AM. 2003. Some cardiovascular effectsEnnajar of M, the Bouajila dethy- J, Lebrihi A, Mathieu F, Savagnac A, Abderraba M, Raies A, Romdhane Ferreira A, Proenc Garzon A, Fink G, Dalit ED, Menashe N, NudelmanGuedes A, Greenberg DN, O, inventors. Silva 2003 January DF, Barbosa-Filho JM, De Medeiros IA.Hayouni EA, 2004. Bouix Endothelium-dependent M, Abedrabba M, Leveau JY, Hamdi M. 2008.Interaminense Mechanism LFL, Juc of action of Kandil O, Radwan NM, Hassan AB, Amer AMM, El-Banna HA,Khanavi Amer M, WMM. Hajimahmoodi 1994. M, Ex- Cheraghi-Niroomand M, Kargar Z, Ajani Y, Hadjiakhoondi that effect onantihypertensive rat activity aorta. anddifferent chemical Correlations EOs. composition It have from these appears been that made p-cymene, between References Aftab K, Atta Ur R, Usmanghani K. 1995.Amarti Blood F, pressure Satrani lowering B, action Aafi of A, active principle Ghanmi M, Farah A,Amorim Aberchane ACL, M, Lima CKF, El Hovell Ajjouri AMC, M, Miranda ALP, El Rezende Antry CM. S, 2009. Antinociceptive Bakkali F, Averbeck S, Averbeck D, IdaomarBarla A, M. Topc 2008. Biological effects of essentialBen Marzoug oils—a H, Bouajila J, Ennajar M, Lebrihi A, Mathieu F, Couderc F, Abderraba M, Romd- Blacher J, Baes M, Marchal A, Younes W,Blois Legedz MS. L, 1958. Safar Antioxidant M. determinations 2005. byBounatirou Nouvelles the S, Smiti S, strat use Miguel MG, of Faleiro L, a Rejeb MN, Neffati stable M, Costa free MM, Figueiredo radical AC, [10]. Nature be investigated to determine whichthe chemical antihypertensive activity. molecules implying 50 ± what test. EC 0.90), 56.31 = 7.4% and -elemene, − 2 β ± in vivo in vivo R 0.98). To our 11.7% the heart -pinene and p- = Monodora myristica α -phellandrene and L/kg of dethymo- ± 3.0% the heart rate. 2 α μ R ± (antioxidative assay) has -elemene ( -myrcene and β 50 β IC Vol. 77, Nr. 8, 2012 r 0.86), = 2 -pinene can reduce to 33.8 R 0.76). The chemical structure of these 3 α 2.15%, respectively, with a concentration = ± 2 R EO to 0.04%. At 0.06% concentration, the fre- L/kg of (the black seed) on the cardiovascular system of 17.28 μ − -pinene was 8 times more active than p-cymene, but α Journal of Food Science -myrcene ( L/kg of p-cymene can reduce to 33.7 β μ M. myristica Correlation between antihypertensive activity and antioxidative Another report showed the influence of Koudou and others (2007) have worked on Significant correlation has been found for the compounds Investigations on hypotensive and antioxidant activities have both of compounds may decrease the blood pressure quinonated volatile oil can reduceAs to a 53.6 result, 4 32 rate. So, potential has been checked. Correlation showed by a low composition. These concentrations are about 400 to 600 mg/L. cymene containedNigella in saliva theurethane-anesthetized dethymoquinonated rats (El volatile Tahir andshown oil others that 2003). these of They 2 have pressure compounds decreased and both the the arterial heart blood rate. A dose of 16 isolated heart byVariations of recording the arterialthe amplitude heart blood and isolated pressure2.45% frequency from and variations. of toadof showed contractions a of decreasequency to and amplitude wereantihypertensive reduced activity to 100%. were The not results correlated of with the the chemical p-cymene. Cardiovascular effects have been checked on frog no investigations in literaturewith have the been antihypertensive reported activitygives on to orientation correlation our to knowledge. testis This these most result compounds important. with tests EO in which the major compounds were been found in the case of DPPH assay ( H190 (antihypertensive assay) with low Conclusion effect on thereport antihypertensive presented activity. an Tochemical family our antihypertensive of knowledge, activity oxygenated sesquiterpenes. no attributed to the as follows: p-cymene ( Figure 3–Chemical structure of components that present athe correlation antihypertensive with activity. Relation between chemical composition or antioxidant activity and antihypertensive activity . . . knowledge, no reports haverelation been between found antioxidant activity in and antihypertensive literature activity. concerning been checked with 6 EOs on rat aorta. Experiments have shown and compounds is different (Figure 3). For

H: Health, Nutrition & Food kde ,Rmhn ,IrhmH naa ,LbiiA ahe ,Bujl .2010. J. Bouajila F, Mathieu A, Lebrihi M, Ennajar H, Ibrahim M, Romdhane of M, respira- on Mkaddem Influence series carvacrol 2003. and thymol SP. the Hong of compounds OM, various of Kim action The BC, 1934. T. Kim Maeda MJ, Kang GM, Na YW, Ki DY, Lim . . . activity antihypertensive and activity antioxidant or composition chemical between Relation assetoer nlssadatmcoiladatoiatatvte.JMdFo 13(6) Food Med J activities. antioxidant and chromatography- antimicrobial gas Tunisia: and Matmata, 1500–4. from analysis Link. et spectrometry Hoff. mass capitatus Thymus of of raising oil Essential its in adrenaline to 18(2–3):79–94. action Jpn antagonistic Pharmacol their Folia pressure. with Pharmacol blood together Appl pressure, blood J and rat. tion the of contractility strips aortic and pressure 11(2):119–25. blood on acetate bornyl atsMV avloA,Mdio A le B acir ,Atnol R 2007. AR. Antoniolli M, Marchioro PB, Alves IA, Medeiros Antioxidant 1999. AA, C. Carvalho MRV, Rice-Evans M, Santos Yang A, Pannala A, Proteggente N, Pellegrini R, Re St JC, Tomani M-J, Mukazayire adoaclrefcso ytsfuioaesniloli as ioeai 78(3):186– Fitoterapia rats. in oil essential fruticosa Hyptis of 91. effects Med Cardiovascular Biol Rad Free assay. decolorization cation radical 26(9–10):1231–7. ABTS improved an applying activity 129(3):753–60. Chem spectrom- Food chromatography-mass activities. antioxidant gas and herbs: hepatoprotective analysis Rwandese etry four of oils Essential o.7,N.8 2012 8, Nr. 77, Vol. vgyC hlhtJ,Cnot ,MnciiF uzP 2011. P. Duez F, Menichini F, Conforti JC, Chalchat C, evigny ´ r ora fFo Science Food of Journal H191

H: Health, Nutrition & Food

COMMUNICATION COMMUNICATIONS Certains résultats des travaux présentés dans le Volume 2, synthèse des travaux ont fait l’objet de communications notées C.01 à C.03 :

C.01 : Mamy H. Rafamantanana, Benjamin Debrus, Guy E. Raoelison, Suzanne Urverg- Ratsimamanga, Philippe Hubert, Joëlle Quetin-Leclercq. Application of experimental design for the HPLC-UV-MS separation of aporphine alkaloids from leaves of Spirospermum Penduliflorum Thouars. International Symposium on Drug Analysis. Anvers, Belgique, 09, 10 et 11 septembre 2010 (Communication affichée).

C.02: Mamy Rafamantanana, Eric Rozet, Guy E. Raoelison, Kiban Cheuk, Suzanne Urverg- Ratsimamanga, Phillipe Hubert, Joëlle Quetin-Leclercq. Quantification of triterpenic glycosides in leaves of Centella asiatica Urb (APIACEAE) by HPLC-UV. 14th Forum of Pharmaceutical Sciences, organisé par la Société belge des Sciences Pharmaceutiques. Blankenberge, Belgique, 14 et 15 mai 2009 (Communication orale).

C.03 : J-R. Ioset, G.E. Raoelison, K. Hostettmann. Méthode par LC/DAD-UV/MS pour une détection rapide de l’acide aristolochique dans les préparations à base de plantes. Sections Sciences Fondamentales et Appliquées, Académie Nationale des Arts, des Lettres et des Sciences. Antananarivo, 15 mai 2003 (Communication orale).

C.01 : Mamy H. Rafamantanana, Benjamin Debrus, Guy E. Raoelison, Suzanne Urverg- Ratsimamanga, Philippe Hubert, Joëlle Quetin-Leclercq. Application of experimental design for the HPLC-UV-MS separation of aporphine alkaloids from leaves of Spirospermum Penduliflorum Thouars. International Symposium on Drug Analysis. Anvers, Belgique, 09, 10 et 11 septembre 2010 (Communication affichée).

Communication affichée lors de l’ International Symposium on Drug Analysis. Anvers, Belgique 09, 10 et 11 septembre 2010

APPLICATION OF EXPERIMENTAL DESIGN FOR THE HPLC-UV-MS SEPARATION OF APORPHINE ALKALOIDS FROM LEAVES OF SPIROSPERMUM PENDULIFLORUM THOUARS

Mamy H. Rafamantananaa,b, Benjamin Debrus c, Guy E.Raoelisonb, Suzanne Uverg- Ratsimamangab ,Phillippe Hubertc, Joëlle Quetin- Leclercqa, aLouvain Drug Institute, Université catholique de Louvain, Av. Mounier 7230, 1200 Bruxelles, Belgium, bInstitut Malgache de Recherches Appliquées (IMRA), BP 3833, Itaosy, Antananarivo 102, Madagascar, cLaboratoire de Chimie Analytique, Département de Pharmacie, CIRM, Université de Liège, CHU, Av. de l’hôpital 1, B36, B-4000 Liège, Belgium

Spirospermum penduliflorum Thouars (Menispermaceae) is an endemic species of Madagascar and traditionally used for the treatment of hypertension. Recently, two aporphine alkaloids known to possess antihypertensive properties (dicentrine and neolitsine) were isolated and identified from this plant. Therefore, we decided to develop an analytical method to separate these molecules to control the quality and effectiveness of this plant. The optimization of chromatographic methods for plant extracts is often intricate and can be time consuming. In fact, it is a thorny problem to separate components due to the similarities between their chromatographic behaviours or to elute them all well separated when some components have widely distinct properties (e.g. polarities, pKa, Log P …). In order to optimize the separation and simultaneously evaluate the method robustness, experimental design and design space methodologies were applied [1]. Three common chromatographic parameters (mobile phase pH, initial proportion of methanol and gradient slope) were selected to construct a full factorial design of 36 experiments. The times at the beginning, the apex (i.e. retention time) and the end of each peak were recorded. The logarithm of the retention factors and the semi-widths were then modelled by multi-linear regressions and these were used to optimize the separation. The relationship between predicted and experimental retention times is depicted in Figure a. The corresponding residuals were normally distributed. Subsequently, the predictive error propagation was analyzed to evaluate the ability of the method to remain unaffected by small parameters variations. Figure b shows the probability surface (at pH 3.5) to reach a selected threshold of the separation criterion (S>0 min, i.e. baseline resolved peaks). The design space (DS) is defined as the zone where the probability to attain S>0, is higher than the quality level. The DS is surrounded in red for a quality level of 64% in Figure b.

The application of design of experiments and design space led to separate 13 molecules including dicentrine and neolitsine. An isomer of a dicentrine was also evidenced. Other molecules were identified by the comparison of literature data and our data bank. References: [1] P.Lebrun, B. Govaerts, B. Debrus, A. Ceccato, G. Caliaro, P. Hubert, B. Boulanger. Chemom. Intell. Lab. Sys.: 91 (2008) 4-16.

C.02: Mamy Rafamantanana, Eric Rozet, Guy E. Raoelison, Kiban Cheuk, Suzanne Urverg- Ratsimamanga, Phillipe Hubert, Joëlle Quetin-Leclercq. Quantification of triterpenic glycosides in leaves of Centella asiatica Urb (APIACEAE) by HPLC-UV. 14th Forum of Pharmaceutical Sciences, organisé par la Société belge des Sciences Pharmaceutiques. Blankenberge, Belgique, 14 et 15 mai 2009 (Communication orale).

Communication orale lors du 14th Forum of Pharmaceutical Sciences, organisé par la Société belge des Sciences Pharmaceutiques. Blankenberge, Belgique, 14 et 15 mai 2009.

SIMULTANEOUS QUANTIFICATION OF TRITERPENIC GLYCOSIDES AND AGLYCONES IN LEAVES OF CENTELLA ASIATICA URB. (APIACEAE)

Mamy Rafamantananaa, Eric Rozetc, Guy E. Raoelisonb, Kiban Cheukb, Suzanne Urverg-Ratsimamangab, Phillipe Hubertc, Joëlle Quetin-Leclercqa.

aLouvain Drug Institute, Université catholique de Louvain, Av. Mounier 7230, 1200 Bruxelles, Belgium, bInstitut Malgache de Recherches Appliquées (IMRA), BP 3833, Itaosy, Antananarivo 102, Madagascar, cLaboratoire de Chimie Analytique, Département de Pharmacie, CIRM, Université de Liège, CHU, Av. de l’hôpital 1, B36, B-4000 Liège, Belgium

The simultaneous quantification of madecassoside, asiaticoside, madecassic acid and asiatic acid in Centella asiatica by HPLC-UV is proposed. Asiaticoside was used as reference for the quantification of heterosides and asiatic acid for aglycones. The evaluation of the extraction efficiency of the four molecules led to use Soxhlet extraction for 8 hours. The method was validated and was found to be accurate in the concentration range of 1.0 to 3.0 mg/ml for asiaticoside and 0.5 to 2.0 mg/ml for asiatic acid with CV < 3% for all investigated compounds. LOD and LOQ were respectively 0.0113 and 1.0 mg/ml for asiaticoside and 0.0023 and 0.5 mg/ml for asiatic acid. This method was shown to be convenient for routine analysis of samples of Centella asiatica.

C.03 : J-R. Ioset, G.E. Raoelison, K. Hostettmann. Méthode par LC/DAD-UV/MS pour une détection rapide de l’acide aristolochique dans les préparations à base de plantes. Sections Sciences Fondamentales et Appliquées, Académie Nationale des Arts, des Lettres et des Sciences. Antananarivo, 15 mai 2003 (Communication orale).

Communication orale lors de la séance des Sections Sciences Fondamentales et Appliquées, Académie Nationale des Arts, des Lettres et des Sciences. Antananarivo, Madagascar, 15 mai 2003.

METHODE PAR LC/DAD-UV/MS POUR UNE DETECTION RAPIDE DE L’ACIDE ARISTOLOCHIQUE DANS LES PREPARATIONS A BASE DE PLANTES

Jean Robert IOSETa, Emmanuel Guy RAOELISONb, Kurt HOSTETTMANNa aInstitut de Pharmacognosie et Phytochimie, BEP, Université de Lausanne, Suisse ; bInstitut Malgache de Recherches Appliquées, BP : 3833, Avarabohitra, Itaosy-Antananarivo 103

Une nouvelle méthode utilisant le couplage HPLC avec les détecteurs DAD-UV/MS a été développée pour permettre la détection rapide de l’acide aristolochique dans les préparations à base de plantes médicinales L’utilisation des détecteurs DAD-UV et MS confère à cette méthode une plus grande sélectivité : avec ces deux types de détection, une limite de détection de l’ordre de ng a été obtenue. Ces deux méthodes sont également quantitatives. Parallèlement, une méthode par chromatographie sur couche mince a été développée afin de faire un dépistage préliminaire de l’acide aristolochique I dans des mélanges complexes de plantes.