sign in sign up
<<
Home , Befloxatone, Crassula falcata, Herbacetin, Kalanchoe brasiliensis, Petrosedum, Pratensein

Thesis

Phytochemical investigation of two : rosea L., the New "Herbal Stress Buster", and dasyphyllum L.

VAN DIERMEN, Daphné

Abstract

Rhodiola rosea L. (Crassulaceae), the most investigated species of the Rhodiola, grows at elevated altitudes in the and in mountainous regions throughout and Asia, where it is also knows as "Golden root" or "Arctic root". The roots have been used for centuries in to enhance physical and mental performance, improve resistance to high altitude sickness and to treat , psychological stress and depression. The present work aims at clarifying the pharmacological effects of R. rosea and identifying the active metabolites. The was tested on three targets : , acetylcholinesterase and oxidative stress. As the roots of R. rosea exhibited interesting activities against the three targets, a investigation was undertaken on the plant. An agronomical study was also realised on R. rosea. Another part of this work consisted in studying different Crassulaceae species in order to discover new potential herbal stress buster. Six new were isolated from Sedum dasyphyllum L.

Reference

VAN DIERMEN, Daphné. Phytochemical investigation of two Crassulaceae species: Rhodiola rosea L., the New "Herbal Stress Buster", and Sedum dasyphyllum L.. Thèse de doctorat : Univ. Genève, 2009, no. Sc. 4122

URN : urn:nbn:ch:unige-46200 DOI : 10.13097/archive-ouverte/unige:4620

Available at: http://archive-ouverte.unige.ch/unige:4620

Disclaimer: layout of this document may differ from the published version.

1 / 1 UNIVERSITE DE GENEVE FACULTE DES SCIENCES Section des Sciences Pharmaceutiques Laboratoire de Pharmacognosie et de Phytochimie Prof. Kurt Hostettmann

Phytochemical Investigation of two Crassulaceae Species: Rhodiola rosea L., the New "Herbal Stress Buster", and Sedum dasyphyllum L.

THESE

Présentée à la Faculté des Sciences de l’Université de Genève Pour obtenir le grade de Docteur ès Sciences, mention Sciences Pharmaceutiques

par Daphné van Diermen de Blonay (VD)

Thèse N o 4122

Atelier d’impression ReproMail - Uni Mail Genève 2009

REMERCIEMENTS

En premier lieu, je tiens à exprimer toute ma gratitude à mon directeur de thèse, Monsieur le Professeur Kurt Hostettmann, pour m’avoir accueilli au sein de son laboratoire de renommée internationale et permis d’effectuer mon travail de thèse dans d’excellentes conditions. Je remercie également le Professeur pour m’avoir donné l’occasion de présenter une partie de mes résultats de recherche lors de congrès nationaux et internationaux, notamment en Indonésie et en Grèce. Je tiens également à lui exprimer ma reconnaissance pour la confiance qu’il m’a accordée en m’incluant dans un projet de développement de médicaments traditionnels améliorés à Haïti.

Je remercie chaleureusement mon responsable technique, Monsieur le Docteur Andrew Marston pour ses précieux conseils et son soutien.

Mes sincères remerciements vont également aux membres de mon jury de thèse : Monsieur le Professeur Pascal Richome du Laboratoire : Substances d’Origine Naturelles et Analogues Structuraux de l’Université d’Angers, Monsieur le Docteur Bruno David des Laboratoires Pierre Fabre de Ramonville, et Madame le Docteur Pia Malnoe de l’Agroscope Changins-Wädenswill de Conthey, pour leur lecture approfondie de mon manuscrit et leur remarques pertinentes et pour avoir fait le déplacement lors de la soutenance de thèse. Je remercie tout particulièrement le membre interne du jury, Madame le Docteur Karine Ndjoko Ioset du Laboratoire de Pharmacognosie et Phytochimie de l’Université de Genève qui m’a guidée tout au long de ce travail. Merci également à Monsieur le Professeur Pierre-Alain Carrupt pour avoir accepté de présider ce jury.

Ce travail de thèse a également pu être réalisé grâce à différentes collaborations :

Le Laboratoire de Chimie Thérapeutique, groupe de Pharmacochimie de l’Université de Genève, en particulier Monsieur le Docteur Juan Bravo qui a testé mes extraits et produits purs sur la monoamine oxydase.

L’Agroscope de Changins-Wäadenswill, plus spécialement Madame le Docteur Pia Malnoe qui nous a fourni une partie du matériel végétal nécessaire pour ce projet.

Je tiens également à remercier Monsieur Egidio Anchisi pour avoir récolté les différentes plantes nécessaires à mes recherches.

I

Mes sincères remerciements vont également à tous mes collègues de laboratoire :

A Jean-Luc, Karine et Emerson pour leur aide et leurs conseils expérimentés en RMN et en spectrométrie de masse. Et à Philipe Christen pour, entre autres, ses conseils en extraction.

A Anne-Emmanuelle Hay pour ses conseils, son soutien et plus particulièrement pour son amitié.

A Monica, ma diplômante, pour son excellent travail qui a contribué à cette étude.

A Sandra et Fred pour nos nombreuses et précieuses discussions scientifiques ou non ainsi que pour les bons moments partagés à Bangkok et à Bali.

A Martine, Peihong, Trixie, Claudia, Gaëtan, Sylvian, Raimana, Caro et Anne-Laure avec qui j’ai travaillé pendant trois ans. Merci pour leur aide, leur motivation, leur encouragement et les bons moments sérieux et un peu moins sérieux partagés durant les congrès et les différentes sorties du Laboratoire.

Je souhaite encore remercier tout particulièrement mes parents pour leurs encouragements et pour leur soutien tout au long de mes études et également pour les corrections de l’anglais du présent manuscrit.

Finalement, je remercie du fond du cœur Cédric pour sa patience, sa compréhension et son précieux soutien.

II

SCIENTIFIC COMMUNICATIONS

The present work was carried out at the Laboratory of Pharmacognosy and Phytochemistry, School of Pharmaceutical Sciences, University of Geneva, University of Lausanne from January 2006 to Mai 2009, under the guidance of Prof. Dr. Kurt Hostettmann. Some aspects of the present research have been published and presented at scientific congresses as poster presentations.

Publications

Van Diermen D., Marston A., Bravo J., Reist M., Carrupt, J.-C. and Hostettmann K. Monoamine oxidase inhibition by Rhodiola rosea L. roots (2009). Journal of Ethnopharmacology 122 : 397-401.

Van Diermen D., Ndjoko Ioset K., Marston A., Malnoe P. and Hostettmann, K. Chemical profile dynamics in a wild population of Rhodiola rosea L. from the Swiss Alps. (Submitted to Journal of Agricultural Sciences ).

Van Diermen D., Pierreclos M., and Hostettmann, K. Antioxidant phenolic compounds from Sedum dasyphyllum L. (Submitted to Journal of Natural Products ).

Ndjoko Ioset K., Nyberg N. T., van Diermen D., Malnoe P. and Hostettmann K. Metabolic profiling of Rhodiola rosea by 1H NMR spectroscopy. (Submitted to Metabolomics ).

Hostettmann K. and van Diermen D. La plante du jour: Rhodiola rosea (2007). Phytotherapie Européenne 37 : 14-17.

Posters

Van Diermen D., Marston A., Bravo J., Reist M., Carrupt, P.-A. and Hostettmann K. Inhibition of monoamine oxidase and acetylcholinesterase by Rhodiola rosea L. root extract. 7th Joint Meeting of AFERP, ASP, GA, PSE & SIF, Athens, Greece, August 2008.

Van Diermen, D., Marston A., Ndjoko Ioset K. and K. Hostettmann. Analyses and bioactivities of wild populations of Rhodiola rosea L. (Crassulaceae) from Switzerland. Swiss Chemical Society, Lausanne 2007.

Van Diermen D., Ndjoko Ioset K., Marston A. and Hostettmann, K. Qualitative and quantitative analyses of wild populations of Rhodiola rosea L. (Crassulaceae) from Switzerland. International Symposium by IOCD, “Biology, Chemistry, Pharmacology and Clinical studies of Asian plants”, Surabaya, Indonesia. April 2007.

III

ABBREVIATIONS AND SYMBOLS

The units used in the present work are in agreement with the International System of Units (SI).

2D Bidimensional 4CL Hydroxycinnamate-CoA ligase 5-HT δ Chemical shift (NMR)

λmax Wavelength of the absorption maxima

[α]D Specific rotation at the wavelength D of sodium Abs Absorbance ACh Acetylcholine AChE Acetylcholinesterase AD Alzheimer’s disease ALDH Aldehyde dehydrogenase ATP Adenosine triphosphate br s Broad singlet (NMR) BuOH Butanol CAD Cinnamyl alcohol dehydrogenase CAM Crassulaceae acid CCR Cinnamyl-CoA reductase

CD 3OD Deuterated methanol CG Cyanogenic glycosides CoA Coenzyme A CNS Central nervous system COMT Catechol-O-methyltransferase COSY Correlation spectroscopy (NMR) CPC Centrifugal partition chromatography d Doublet (NMR) DA DAD Diode array detector (UV) DCM Dichloromethane dd Double doublet (NMR) DMSO Dimethylsulfoxide

DMSO-d6 Deuterated dimethylsulfoxide DNA Deoxyribonucleic acid DPPH 1,1-diphenyl-2-picrylhydrazyl

EC 50 Effective concentration of a compound required for 50% scavenging activity

V EI Electron impact (MS) ESI Electrospray ionization (MS interface) EtOAc Ethyl acetate EtOH FA Formic acid GSH Glutathione HMBC Heteronuclear multiple bond correlation (NMR) HPLC High performance liquid chromatography HPLC-UV/DAD HPLC coupled with ultraviolet photodiode array detector HPLC-MS HPLC coupled with mass spectrometry HR-ESI/TOF/MS High resolution ESI TOF MS HSQC Heteronuclear single quantum coherence (NMR) Hz Herz Inhibitory concentration of ligand to equal 50% of the effect produced by a IC 50 standard OCD Obsessive-compulsive disorder i.d. Internal diameter IT Ion trap (MS detector) J Coupling constant (NMR) LC Liquid chromatography LDL Low density lipoprotein LLE Liquid-liquid extraction LPP Laboratory of Pharmacognosy and Phytochemistry LPLC Low pressure liquid chromatography m Multiplet (NMR) m/z Mass per electronic charge MAO Monoamine oxidase MeCN Acetonitrile MeOH Methanol MPLC Medium pressure liquid chromatography MPP + 1-methyl-4-phenylpyridinium MPTP 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine MS Mass Spectrometry MW Molecular Weight NA Noradrenaline NADPH Nicotinamide adenine dinucleotide phosphate NMR Nuclear magnetic resonance NOESY Nuclear overhauser effect spectroscopy (NMR)

VI NST-PEG Naturstoff-Polyethylenglycol PAL ammonia lyase PD Parkinson’s disease PEA Phenylethylamine pH Power of hydrogen ppm Parts per million (NMR Unity) PTSD Post-traumatic stress disorder RIMAs Reversible inhibitors of MAO A ROS Reactive oxygen species RSD Relative standard deviation s singlet (NMR) Semi-prep. HPLC Semi-preparative HPLC SOD Superoxide dismutase SPE Solid phase extraction t triplet (NMR) TIC Total ion chromatogram TLC Thin layer chromatography TMS Tetramethylsilane (NMR) TOCSY Total correlation spectroscopy (NMR) TOF Time of flight (MS detector) UV Ultra violet

VII

TABLE OF CONTENTS

REMERCIEMENTS ...... I SCIENTIFIC COMMUNICATIONS ...... III ABBREVIATIONS AND SYMBOLS ...... V TABLE OF CONTENTS ...... IX RESUME DU TRAVAIL DE THESE ...... XIII

1. AIM OF THE PRESENT WORK ...... 1

2. INTRODUCTION ...... 5 2.1. THE CRASSULACEAE FAMILY ...... 7 2.1.1. Systematicclassificationofthefamily ...... 7 2.1.2. EvolutionofCrassulaceaesystematics...... 8 2.1.3. BotanicalaspectsofCrassulaceae...... 10 2.1.4. CharacteristicmetabolismpathwayoftheCrassulaceae ...... 10 2.1.5. Phytochemicalaspects ...... 11 2.1.6. Interestsofthefamily...... 22 2.2. RHODIOLA ROSEA L...... 23 2.2.1. Generalities ...... 23 2.2.2. Thegenus Rhodiola ...... 25 2.2.3. Morphologyandgeographicdistributionof R. rosea ...... 28 2.2.4. Traditionalusesof R. rosea ...... 31 2.2.5. R. rosea inmodernmedicine...... 32 2.2.6. Secondarymetabolitesfrom R. rosea ...... 33 2.2.7. Pharmacologicalactivitiesof R. rosea ...... 39 2.2.8. Marketpotentialandthreatstatusof R. rosea ...... 45 2.3. SEDUM DASYPHYLLUM L...... 46 2.3.1. Thegenus Sedum ...... 46 2.3.2. MorphologyandgeographicaldistributionofS. dasyphyllum ...... 51 2.4. BIOLOGICAL ACTIVITY...... 54 2.4.1. Monoamineoxidaseinhibitoryactivity ...... 54 2.4.2. Antioxidantactivity ...... 63

3. RESULTS ...... 71 3.1. CHARACTERISATION OF BIOACTIVE COMPOUNDS FROM RHODIOLA ROSEA L. ROOTS...... 73 3.1.1. Extraction...... 73 3.1.2. Biologicalandchemicalscreeningof R. rosea rootextracts ...... 73 3.1.3. BioguidedisolationofAChEinhibitorsfrom R. rosea DCMextract ...... 77 3.1.4. Monoamineoxidaseinhibitionby Rhodiola rosea L.roots ...... 81

IX

3.2. HPLC-UV QUANTITATIVE ANALYSIS OF SALIDROSIDE AND ROSAVINS IN RHODIOLA ROSEA L. PLANTS ...... 89 3.2.1. Chemicalprofiledynamicsinawildpopulationof Rhodiola rosea L.plantsfromthe SwissAlps ...... 91 3.3. CHEMICAL AND BIOLOGICAL SCREENING OF SEDUM SPECIES (CRASSULACEAE) ...... 105 3.3.1. Plantmaterials ...... 105 3.3.2. Screeningof Sedum speciesextracts ...... 106 3.3.3. Antioxidantphenoliccompoundsfrom Sedum dasyphyllum L...... 107

4. CONCLUSION AND PERSPECTIVES...... 129 4.1. CHARACTERISATION OF BIOACTIVE COMPOUNDS FROM RHODIOLA ROSEA L...... 131 4.2. HPLC-UV QUANTIFICATION OF SALIDROSIDE AND ROSAVINS IN R. ROSEA L. PLANTS ...... 133 4.3. CHEMICAL AND BIOLOGICAL SCREENING OF SEDUM SPECIES ...... 134

5. EXPERIMENTAL PART ...... 137 5.1. PLANT MATERIAL AND EXTRACTION ...... 139 5.1.1. Plantmaterial...... 139 5.1.2. Secondarymetabolitesextraction ...... 140 5.1.3. Liquidliquidextraction(LLE)...... 140 5.2. ANALYTICAL CHROMATOGRAPHIC METHODS ...... 141 5.2.1. Thinlayerchromatography(TLC) ...... 141 5.2.2. Highperformanceliquidchromatographycoupledtoultravioletphotodiode arraydetector(HPLCUVDAD)...... 142 5.2.3. Highperformanceliquidchromatographycoupledtomassspectrometry(HPLCMS) ...... 144 5.2.4. Ultraperformanceliquidchromatographycoupledwithmassspectrometry(UPLCMS)...... 144 5.3. PREPARATIVE SEPARATION TECHNIQUES...... 145 5.3.1. Columnchromatography ...... 145 5.3.2. Preparativepressureliquidchromatography...... 146 5.3.3. Centrifugalpartitionchromatography(CPC) ...... 148 5.3.4. Solidphaseextraction(SPE) ...... 148 5.4. PHYSICO-CHEMICALS METHODS ...... 149 5.4.1. Opticalrotation([α]D)...... 149 5.4.2. Ultravioletspectrophotometry ...... 149 5.4.3. Nuclearmagneticresonancespectrometry(NMR) ...... 150

X

5.5. CHEMICAL METHODS ...... 150 5.5.1. ReagentsforTLCdetection ...... 150 5.5.2. DPPHfreeradicalscavengingassayonTLC...... 151 5.5.3. DPPHfreeradicalscavengingmicroplateassay...... 151 5.5.4. Acidhydrolysisofandsugaridentification...... 152 5.6. BIOLOGICAL METHODS ...... 152 5.6.1. AcetylcholinesteraseinhibitorTLCbioassay...... 152 5.6.2. Acetylcholinesteraseinhibitorsmicrotitreplatebioassay ...... 153 5.6.3. Monoamineoxidaseinhibitormicrotitreplatebioassay ...... 154 5.7. PHYSICAL CONSTANTS AND SPECTRAL DATA FOR THE ISOLATED COMPOUNDS ...... 155 5.8.1. CompoundsisolatedfromtheDCMextractof Rhodiola rosea L...... 155 5.8.2. CompoundsisolatedfromtheMeOHextractofRhodiola rosea L...... 162 5.8.3. CompoundsisolatedfromtheMeOHextractofSedum dasyphallum L...... 173

6. REFERENCES ...... 199

XI

RESUME DU TRAVAIL DE THESE

Introduction

Rhodiola rosea L. (orpin rose) est une plante des régions montagneuses de l’hémisphère Nord. Sa distribution en Europe comprend les pays Scandinaves, l’Islande, les Alpes et la plupart des massifs d’Europe centrale. Elle fait partie de la famille des Crassulaceae qui englobe plus de 1500 espèces réparties en 35 genres. Les membres de cette famille ont la particularité de stocker l’eau dans leurs feuilles, tiges et racines ce qui explique leur nom de plantes « succulentes ». En 77 après J.-C., R. rosea était déjà citée par Dioscoride sous le nom de rodi riza , dans son ouvrage . Plus tard, Linné la renomma Rhodiola rosea en référence à l’odeur de rose que dégagent les racines fraîchement coupées. Durant des siècles, ces racines furent utilisées dans les médecines traditionnelles russe, scandinave et asiatique pour augmenter l’endurance physique et la résistance au mal d’altitude ainsi que pour le traitement de certains troubles du système nerveux et pour lutter contre la fatigue. Depuis les années 1960, les cosmonautes et les athlètes russes l’utilisent pour améliorer leurs performances physiques. En 1975, la plante a été introduite dans la Pharmacopée russe laquelle la préconise pour traiter les maladies somatiques ou infectieuses et, chez les personnes saines pour combattre la fatigue, stimuler la mémoire, l’attention et améliorer la condition physique. Bien que la plante ait été largement étudiée pour ses vertus antistress et antifatigue, ses propriétés sont restées longtemps ignorées des populations occidentales. Ceci étant certainement dû au fait que la majorité des études scientifiques menées auparavant étaient publiées en langues slaves ou scandinaves. Toutefois, suite à un article paru en 2003 dans le magazine américain Newsweek , intitulé « Herbal Stress Buster ?», R. rosea est devenue célèbre dans le monde entier. En effet, une telle publicité, à une époque où la plupart des gens sont soumis à un stress occasionnel ou permanent, a suscité l’intérêt des chercheurs et des fabricants de phytomédicaments. Dès lors, de nombreuses investigations phytochimiques et pharmacologiques ont été effectuées sur cette plante. Depuis 2002, une centaine d’études scientifiques ont été publiées à son propos. Plus d’une septantaine de métabolites secondaires ont été isolés à partir de ses racines, dont des flavonoïdes, des mono- et des tri-terpènes, des dérivés du phénylethanol et du phénylpropane. Parmi les dérivés du phénylpropane, des glycosides de l’alcool cinnamique, la rosine et la rosavine, se sont révélés être spécifiques à cette plante. D’après l’ensemble des études pharmacologiques, R. rosea répond aux trois critères définissant un adaptogène : perturbation minimale de la fonction physiologique normale, effet stimulant et effet stress-protecteur. Plusieurs études démontrent ses effets positifs sur le système nerveux central grâce à son influence sur les taux en monoamines biogéniques (sérotonine, dopamine et noradrenaline) dans le

XIII

cortex cérébral et dans l’hypothalamus. Plusieurs métabolites isolés se sont révélés être antioxydants et jouent donc un rôle protecteur contre les radicaux libres au niveau du système nerveux. D’autres études ont montré que R. rosea augmente la capacité physique et réduit le temps de récupération entre des périodes d’exercices physiques intenses. Une récente étude clinique en double aveugle contre placebo démontre une activité antidépressive significative de l’extrait hydro-alcoolique des racines. La plante ne présente pas de toxicité et très peu d’effets secondaires ont été rapportés jusqu’à ce jour. Cependant, malgré les nombreuses études effectuées, les effets physiologiques de la plante sur l’homme n’ont toujours pas pu être identifiés précisément, et certains effets pharmacologiques restent inexpliqués. A l’heure actuelle, de nombreuses préparations à base de R. rosea sont sur le marché mondial. En 2005, plus de 46 compagnies utilisaient l’orpin rose dans leurs produits et plus de 30 compagnies étaient désignées comme fournisseur de la plante. Suite à l’abondante récolte des plantes sauvages pour la préparation des phytomédicaments, les populations naturelles risquent de disparaître. La culture semble être la seule solution pour pouvoir satisfaire la demande générale. Plusieurs expériences de mise en culture ont été entreprises en Russie ainsi qu’en Scandinavie, Pologne et au Canada. Cependant, la culture présente des inconvénients au niveau des coûts engendrés et du temps nécessaire pour obtenir des rhizomes suffisamment développés.

Objectifs du présent travail

Ce travail est divisé en trois parties dont la première a consisté à clarifier les effets physiologiques de R. rosea au niveau du système nerveux central et à identifier les métabolites actifs. Dans ce but, la plante a été testée sur trois cibles différentes : la monoamine oxydase, l’acetylcholinesterase et le stress oxydatif. Ces trois cibles jouent un rôle important au niveau du système nerveux. La monoamine oxydase (MAO) est une enzyme clé dans la métabolisation des neurotransmetteurs tels que la sérotonine (5-HT), la dopamine (DA) et la noradrenaline (NA). Ceux-ci sont directement impliqués dans la modulation de l’humeur et des émotions ainsi que dans le contrôle des fonctions motrices, perceptuelles et cognitives. Deux types de MAO (A et B) sont distingués en fonction de leur affinité pour la dopamine, la sérotonine et la noradrenaline. La MAO A ayant une plus grande affinité pour la NA et la 5-HT, joue un rôle important dans la dépression tandis que la MAO B qui a une grande affinité pour la dopamine est impliquée dans certaines maladies neurodégéneratives telles que les maladies de Parkinson et d’Alzheimer. L’acetylcholinesterase (AChE) est l’enzyme responsable de la

XIV

métabolisation de l’acétylcholine, neurotransmetteur du système cholinergique qui est impliqué notamment dans les fonctions cognitives. L’inhibition de cette enzyme va engendrer une diminution du turn-over de l’acétylcholine et donc augmenter les effets cholinergiques. Les inhibiteurs de l’acetylcholinesterase sont utilisés pour diminuer les symptômes de la maladie d’Alzheimer. Le stress oxydatif, la troisième cible, peut être défini comme un déséquilibre entre les pro-oxydants et les antioxydants en faveur des premiers, qui engendre des dommages dans l’organisme, au niveau des lipides, des protéines et de l’ADN. Le stress oxydatif est impliqué dans de nombreuses affections telles que les maladies neurodégéneratives.

La seconde partie de ce travail avait pour objectif la mise en culture de R. rosea, en collaboration avec l’Agroscope de Changins-Wädenswil, et l’optimisation des conditions de culture et de récolte afin d’obtenir des plantes riches en métabolites secondaires actifs. Dans ce but, quatre plantes de la même population ont été sélectionnées. Une partie de leurs racines a été repiquée en pot, placée sous serre, puis mise en terre. Ces plantes seront récoltées après quatre ans (en 2010), temps nécessaire à la croissance du rhizome. En attendant, des analyses préliminaires ont été effectuées sur ces quatre plantes sauvages. Les racines ont été analysées par chromatographie liquide à haute performance couplée à l’UV afin de caractériser leur profil chimique. Celui-ci a été suivi durant une année afin d’établir une dynamique et d’identifier ainsi la période la plus propice à la récolte de rhizomes riches en principes actifs. De plus, les profils chimiques d’une quinzaine de plantes femelles et mâles ont été comparés, dans le but d’identifier le genre le plus favorable à la culture.

La dernière partie de ce travail a consisté à rechercher d’autres espèces suisses, de la famille des Crassulaceae, susceptibles de présenter des propriétés similaires à R. rosea . Parmi les espèces du genre Rhodiola , seule R. rosea croît en Suisse. Les autres espèces sont principalement distribuées en Asie, en Russie et en Amérique du Nord. Cependant, de nombreuses espèces du genre Sedum, sont recensées en Suisse. Le genre Sedum est proche de celui de Rhodiola. Auparavant, certains botanistes les avaient même regroupés en un seul genre. Le genre Sedum (orpins) regroupe entre 350 et 500 espèces selon les auteurs. Les orpins se développent presque exclusivement dans l’hémisphère Nord, particulièrement sur le continent américain et en Europe. En outre, les deux tiers des espèces européennes sont réparties dans les pays bordant la Méditerranée. Cependant, plus de 21 espèces sont recensées en Suisse. Parmi ces espèces, six ont été sélectionnées pour une investigation phytochimique. Le choix des plantes s’est fait en fonction de leur disponibilité en grande quantité. Ainsi, les espèces menacées ont été écartées. Les plantes sélectionnées ont été extraites puis soumises

XV

à un criblage biologique et chimique afin d’identifier de potentielles activités. L’espèce Sedum dasyphyllum L. a révélé les activités anti-oxydante et anti-acetylcholinesterase les plus intéressantes. Cette herbe rampante aux fleurs gris-vertes est distribuée en Europe centrale jusqu’aux côtes de la Méditerranée, entre 0 et 2500 m d’altitude. Elle est principalement utilisée en pour ses propriétés couvrantes. Aucune investigation phytochimique ou pharmacologique n’ayant été décrite dans la littérature, son investigation phytochimique bio-guidée a donc été entreprise afin d’identifier les composés responsables des activités.

Résultats

Afin de tester R. rosea sur les trois cibles mentionnées précédemment, les rhizomes ont été successivement extraits avec du dichlorométhane (DCM), du méthanol (MeOH) et de l’eau, de manière à obtenir trois extraits de polarité différente. Leur activité a été évaluée à l’aide de trois tests : un test bioautographique sur CCM pour détecter les composés inhibiteurs de l’acetylcholinesterase (AChE), un test chimique sur CCM avec le DPPH comme réactif afin d’identifier les capteurs de radicaux libres et un test en solution sur microplaque afin d’identifier les inhibiteurs de la MAO. L’extrait DCM a révélé sur CCM plusieurs taches d’inhibition de l’AChE. Les extraits MeOH et aqueux ont présenté une activité inhibitrice supérieure à 80% contre la MAO A et B à une concentration de 100 g/ml. L’investigation phytochimique de chaque extrait a alors été entreprise. L’extrait DCM a été fractionné afin d’isoler les métabolites secondaires responsables de l’activité inhibitrice de l’AChE. Une première séparation par extraction liquide-liquide a été effectuée afin d’éliminer les métabolites les plus lipophiles. La fraction polaire qui montrait la meilleure activité a été fractionnée par chromatographie sur colonne ouverte sur silice. Cette séparation a permis d’isoler le β-sitosterol (DCM3) . Parmi les autres fractions, deux (F4 et F8) présentaient une activité inhibitrice de l’AChE. La purification de F4 par chromatographie par partage centrifuge (CPC) a abouti à l’isolement de deux composés, l’alcool cinnamique (DCM2) et l’acide linoléique (DCM4) . La séparation par HPLC semi-préparative de F8 a fourni l’acétate de tyrosol (DCM1) . L’identité des composés isolés a pu être déterminée par spectrométrie de masse à haute résolution (SM/HR) et résonance magnétique nucléaire (RMN), puis confirmée par comparaison avec la littérature. Parmi ces composés, seuls l’alcool cinnamique et l’acide linoléique ont présenté une activité inhibitrice de l’AChE sur CCM. La quantité minimale d’inhibition mesurée s’élève à 5 g pour l’alcool cinnamique et 1 g pour l’acide linoléique tandis qu’elle est de 0.01 g pour la galanthamine,

XVI

substance de référence. L’activité inhibitrice de l’AChE de l’acide linoléique a déjà été décrite dans la littérature, tandis que l’activité de l’alcool cinnamique est démontrée ici pour la première fois. L’alcool cinnamique a été testé en solution selon la méthode décrite par Ellman afin de mesurer son -5 IC 50 . Cependant, il n’a montré aucune activité à une concentration de 10 M. Il ne présente donc qu’une activité inhibitrice très faible.

Les extraits MeOH et aqueux ont montré une activité inhibitrice des MAOs A et B remarquable dans le test in vitro en solution. Le fractionnement bio-guidé de ces deux extraits par différents moyens chromatographiques tels que la CPC et la Lobar® a donc été entrepris. Celui-ci a permis l’isolement de douze composés. La majorité présentait des spectres UV caractéristiques des dérivés du phénylpropane ( λmax 280-290 nm). Leur structure a été déterminée en comparant les données obtenues par SM/HR et RMN avec celles de la littérature. Sept dérivés du phénylpropane ont été identifiés : le salidroside ( RR1 ), la rosarine ( RR5 ), l’alcool cinnamique ( RR6 ), la triandrine ( RR8 ), la rosavine (RR9 ), le tyrosol ( RR10 ) et la rosine ( RR11 ). Le glucoside azoté rhodiocyanoside A ( RR 7 ), le monoterpène rosiridine ( RR12 ) et un dimère de l’épigallocatechine gallate ( RR2 ) ont également été caractérisés. De plus, un mélange des isomères rhodiolosides B et C ( RR3-RR4 ) a été obtenu. Tous les produits isolés ont été testés in vitro contre les MAOs A et B. La rosiridine et le mélange d’isomères ont présenté la plus forte activité anti-MAO. Le pIC50 (-logIC50) de la rosiridine a été mesuré à 5.38 ± 0.05 ce qui représente une activité modérée en comparaison avec celle du contrôle positif, la , qui possède un pIC 50 de 7.23 ± 0.04.

En seconde partie de ce travail, le profil chimique de quatre plantes sauvages de R. rosea récoltées en Valais (>2200 m) a été caractérisé. Le salidroside et trois dérivés de l’alcool cinnamique : la rosarine, la rosine et la rosavine ont été sélectionnés comme marqueurs. Ces dérivés du cinnamyle sont mentionnés dans la littérature comme étant spécifiques à la plante. En outre, la Pharmacopée russe recommande de standardiser les extraits de R. rosea en rosavines (rosavine, rosarine et rosine) 3% et en salidroside 1%. Pour les analyses, un échantillon des racines a été prélevé sur chaque plante dans leur habitat. Ils ont été soigneusement lavés, coupés, lyophilisés afin d’éliminer toute trace d’eau, puis cryo-broyés. Une méthode d’extraction spécifique a été développée afin d’optimiser l’extraction des marqueurs : elle consiste à extraire les échantillons (10 mg) aux ultrasons durant une heure dans 1ml de MeOH 60%. Après filtration et dilution, les extraits ont été analysés par HPLC-UV. Cette méthode d’analyse a été choisie en raison de son utilisation simple relativement peu coûteuse et de sa robustesse. La détection UV, reproductible, a été préférée à la MS car la méthode développée était

XVII

destinée à être implantée dans un autre laboratoire. Ce type de détection était d’autant plus justifié que les métabolites analysés absorbaient dans l’UV à 210 nm ou à 254 nm et qu’ils étaient en quantité largement supérieure à la limite de détection. La méthode HPLC a été optimisée afin d’obtenir une séparation des marqueurs optimum en un temps réduit. Celle-ci a ensuite été validée dans le but d’obtenir des résultats significatifs et comparables. Par la suite, la dynamique du profil chimique de chaque plante a été établie. Pour cela, des échantillons de rhizomes ont été prélevés à cinq dates différentes au cours de l’année 2006 : avant, pendant et après la floraison. Au total, vingt échantillons de plantes ont été récoltés. Chaque échantillon a été analysé en triplicat. Après traitement des résultats, les dynamiques du profil chimique de chaque plante ont pu être caractérisées puis comparées entre elles, faisant ressortir plusieurs éléments. Premièrement, le contenu des marqueurs fluctue considérablement entre les plantes même lorsque celles-ci appartiennent à la même population. Par exemple, la teneur en salidroside peut varier de moins de 0.05% à plus de 2.25 % d’une plante à l’autre. Deuxièmement, il a été constaté que les teneurs en marqueurs sont restées plus ou moins constantes durant toute la période étudiée. Les teneurs n’ont donc pas été influencées par les différents stades de croissance. Il est donc possible d’en conclure qu’il n’y a pas de période, durant l’année, plus propice à la récolte de rhizomes riches en principe actifs. En parallèle, une étude comparative sur quinze plantes sauvages a été entreprise afin de distinguer les possibles différences entre plantes males et femelles. Leur profil chimique a été établi selon la méthode analytique décrite précédemment. Après traitement des résultats, aucune différence significative n’a pu être observée entre les deux genres.

La dernière partie de ce travail a consisté à investiguer d’autres espèces de Crassulaceae récoltées en Suisse, afin de tenter de découvrir un nouvel adaptogène naturel. Six espèces, sélectionnées selon les critères cités précédemment, ont été récoltées : L., Sedum album L., Sedum atratum L., Sedum dasyphyllum L., Sedum sarmentosum Bunge et Sedum sexangulare L. Celles-ci ont ensuite été congelées puis lyophilisées dans le but d’éliminer par sublimation toutes traces d’eau. Après cryo- broyage, le matériel végétal a été successivement extrait par du DCM et du MeOH. Ceci a permis de faire une première séparation entre les métabolites secondaires lipophiles et ceux de polarité moyenne à élevée. Les 12 extraits bruts ainsi obtenus ont alors été soumis à un criblage chimique et biologique directement effectué sur des plaques CCM afin d’évaluer leur activité contre l’acétylcholinestérase et contre le radical DPPH. Toutefois, l’activité inhibitrice contre la MAO n’a pas pu être évaluée sur ces extraits pour cause d’indisponibilité du test à cette période.

XVIII

Concernant l’activité anti-radicalaire contre le DPPH, les extraits MeOH de S. dasyphyllum, S. atratum et S. acre se sont révélés être les plus actifs. Tandis que seuls les extraits MeOH de S. dasyphyllum et S. acre ont présenté une activité inhibitrice de l’AChE. Les extraits de S. acre et S. dasyphyllum présentaient donc les activités les plus intéressantes. S. acre ayant déjà fait l’objet de plusieurs investigations phytochimiques pour son contenu en alcaloïdes, S. dasyphyllum a donc été la seule espèce retenue pour une investigation phytochimique approfondie. Le fractionnement des extraits de S. dasyphyllum a été guidé par les activités anti-radicalaire et anti- acetylcholinesterase détectées sur les plaques CCM afin d’isoler les métabolites actifs. L’extrait

MeOH a été fractionné par extraction liquide-liquide. Trois fractions (EtOAc, BuOH et H 2O) de polarités différentes ont été récupérées. La fraction EtOAc a été séparée par chromatographie liquide à moyenne pression (CLMP). Ce premier fractionnement a permis l’isolement de sept produits ( RR1- RR7 ). La purification d’une des fractions par chromatographie sur gel LH-20 a permis d’isoler un composé supplémentaire (RR8 ). L’analyse HPLC-UV-DAD de ces composés a permis d’identifier, pour certains, la classe chimique à laquelle ils appartenaient. La majorité présentait des spectres UV caractéristiques des flavonols avec des maxima d’absorbance aux alentours de 260-280 nm et 350-370 nm. Cependant trois composés ( RR3, RR6 et RR7 ) présentaient des spectres UV correspondant aux ( λmax 245-275 nm). La séparation de la fraction BuOH par MPLC, suivie d’une purification sur gel LH 20 ou par chromatographie semi-préparative pour certaines fractions, a mené à l’isolement de trois composés (SD9-SD11 ) dont deux ( SD9-SD10 ) présentaient des spectres UV de flavonols. Le fractionnement de la phase aqueuse par MPLC a permis d’isoler huit composés supplémentaires (SD12-SD19 ) dont la majorité a également été identifiée, selon leur spectre UV, comme étant des flavonols. Au total, 19 composés ont été isolés. Leur structure a été déterminée à l’aide de la SM/HR et de la RMN. Les spectres 1H RMN des flavonoïdes présentent des déplacements chimiques ( δ) aux alentours de 6 ppm correspondant aux protons aromatiques. En outre, le proton du groupe hydroxyle en position 5 qui forme un pont hydrogène avec le carboxyle en position 4 présente un signal caractéristique à 12.5 ppm lorsque l’analyse est effectuée dans le DMSO-d6. De plus, les sucres présentent des signaux caractéristiques aux alentours de 3-4.5 ppm, plus particulièrement le proton lié au carbone anomérique qui présente un doublet autour de 4-4.5 ppm. Ainsi, 13 flavonoïdes ont été identifiés, dont six sont décrits ici pour la première fois : le kaempférol 3-O-α-rhamnoside-7-O-β- sophoroside ( SD9 ), la gossypétine 3,7-di-O-β-glucoside-8-O-β-glucuronide ( SD13 ), l’herbacétine 3,7- di-O-β-glucoside-8-O-β-glucuronide ( SD14 ), l’hibiscétine 3-O-β-glucoside-8-O-β-glucuronide

XIX

(SD16) , l’herbacétine 3-O-β-(3’’-acetylglucoside)-7-O-β-glucoside-8-O-β-glucuronide ( SD17 ), et l’herbacétine 3-O-β-(3’’-acetylglucoside)-8-O-β-glucuronide ( SD19 ). Parmi les autres composés, trois isoflavonoïdes déjà connus ont été identifiés : le 3’-O-methylorobol 7-O-β-D-glucoside ( SD3 ), la dalspinosine 7-O-β-glucoside ( SD7 ) et l’iristectorigenine B ( SD9 ). Ceux- ci présentent des spectres RMN proches des flavonoïdes. Cependant, le proton ainsi que le carbone en position 2 présentent respectivement des déplacements chimiques caractéristiques autour de 8 et 156 ppm, permettant de distinguer les isoflavonoïdes des flavonoïdes. Le lignane secoisolariciresinol 4-O- β-glucoside (SD11 ) a également été identifié grâce à la comparaison des ses données spectroscopiques avec celles de la littérature. Un glycoside cyanogénétique, la lotaustraline ( SD12 ), a également été isolé. La valeur impaire de sa masse obtenue par SM/HR suggérait la présence d’un nombre impair d’azote. Sa structure a ensuite été élucidée grâce aux analyses RMN et aux données de la littérature. L’acide caféique et un dérivé de l’acide férulique ont également été purifiés. Le glycoside cyanogénétique lotaustraline ( SD12 ) a été isolé ici pour la première fois dans le genre Sedum , tandis que le flavonol herbacétine 3-O-β-glucoside-8-O-β-glucuronide et les isoflavonoïdes dalspinosine 7-O-β-glucoside ( SD7 ) et l’iristectorigénine B ( SD9) sont mis en évidence pour la première fois dans la famille Crassulaceae ainsi que le lignane secoisolariciresinol 4-O-β-glucoside (SD11 ). Les composés purs ont été testés pour leur activité anti-oxydante et anti-AChE sur des plaques CCM. L’herbacétine 3-O-β-glucoside-7-O-α-rhamnoside ( SD10 ) et la lotaustraline ( SD12 ) ont révélé une activité inhibitrice de l’AChE, cependant lorsque ces composés ont été testés in vitro en solution, ils n’ont présenté aucune activité à une concentration de 10 -5M. En ce qui concerne l’activité anti- oxydante, sept composés ont présenté une activité remarquable : l’hibifoline ( SD2 ), la mélocorine (SD4 ), l’herbacétine 3-O-β-glucoside-7-O-α-rhamnoside ( SD10 ), le secoisolariciresinol 4-O-β- glucoside ( SD11 ), la lotaustraline ( SD12 ), la gossypétine 3,7-di-O-β-glucoside-8-O-β-glucuronide (SD13 ) et l’hibiscétine 3-O-β-glucoside-8-O-β-glucuronide ( SD16 ). Ces métabolites ont été testés en solution sur microplaque afin de mesurer leur IC 50 . Exceptée la lotaustraline, tous les composés ont présenté une activité anti-oxydante élevée, plus particulièrement l’hibifoline et la mélocorine dont l’IC 50 est comparable à celle de la quercétine (contrôle positif dont l’activité anti-oxydante a été largement décrite dans la littérature).

XX

Conclusions et perspectives

Ce travail a permis de clarifier une partie du mécanisme par lequel R. rosea agit dans l’organisme au niveau du système nerveux central. Plusieurs études cliniques ont démontré des activités pharmacologiques qui seraient dues à l’influence de la plante sur les taux de différents neurotransmetteurs tels que la sérotonine, la noradrénaline ou la dopamine. Dans la présente étude, nous avons identifié précisément un des effets physiologiques de la plante sur l’organisme pouvant expliquer ses activités. En effet, les extraits MeOH et aqueux de R. rosea ont été identifiés comme étant des inhibiteurs de deux isozymes responsables de leur dégradation: les MAOs A et B. En inhibant ces isozymes, R. rosea permet d’augmenter la concentration des neurotransmetteurs au niveau des synapses nerveuses et ainsi de stimuler les diverses voies dopaminergique, sérotoninergique et adrénergique responsables entre autres des fonctions cognitives, émotionnelles, perceptuelles et motrices. Lors de l’investigation phytochimique des extraits, dix composés ont été isolés dont trois monoterpènes glycosylés, la rosiridine et les isomères rhodiolosides B et C. Ces monoterpènes se sont révélés être les plus actifs contre les MAOs, plus particulièrement contre la MAO B. Le pIC 50 de la rosiridine a été évalué à 5.38 ± 0.05. Cette activité reste cependant relativement modérée par rapport au contrôle positif, la sélégiline (pIC 50 = 7.23 ± 0.04). Il semblerait que l’activité des extraits s’explique par un effet additif de différents composés ou par synergisme, parmi lesquels la rosiridine et les rhodiolosides B et C sembleraient être les plus actifs. L’activité peut également être renforcée par la présence de flavonoïdes, tels que la quercétrine, qui ont déjà été décrits comme étant des inhibiteurs modérés de la MAO. Cependant, des études in vivo sont nécessaires pour confirmer cette activité inhibitrice des MAOs. Des études supplémentaires doivent encore être effectuées afin d’identifier précisément tous les effets physiologiques exercés par R. rosea et ainsi déterminer exactement les effets pharmacologiques de celle-ci.

La seconde partie du travail a permis d’établir la dynamique du profil chimique de plantes de R. rosea sur une année, en utilisant comme marqueurs les rosavines et le salidroside. Deux éléments principaux sont ressortis de cette étude. Premièrement, la teneur en marqueurs varie considérablement d’une plante à l’autre malgré leur appartenance à la même population, deuxièmement les teneurs en principes actifs ne présentent que très peu de variations au cours de l’année. De plus, la comparaison du profil chimique de plantes femelles et mâles a également démontré que les deux genres contiennent des teneurs en marqueurs similaires.

XXI

Dès 2010, il sera possible d’analyser les plantes mise en culture en 2006. De nombreuses informations pourront être tirées de la comparaison des profils chimiques des plantes cultivées et des plantes sauvages. Il sera par exemple possible de déterminer si les variations des teneurs en principes actifs entre plantes sont dues à des facteurs génétiques ou a des facteurs extérieurs.

La troisième partie du travail a consisté à investiguer d’autres espèces suisses de la famille Crassulaceae. Après un criblage biologique et chimique de six espèces de Sedum , une investigation phytochimique a été entreprise sur l’espèce présentant les activités les plus intéressantes : Sedum dasyphyllum L. Au final, 19 composés ont été isolés dans cette plante, dont six sont décrits ici pour la première fois. Certains d’entre eux se sont révélés être de bons capteurs de radicaux libres. Des investigations supplémentaires pourraient encore être effectuées sur Sedum dasyphyllum . Entre autre sur l’extrait DCM. Il serait également intéressant de tester les extraits contre la MAO, afin d’identifier s’ils présentent des activités similaires à R. rosea. Cependant, les profils chimiques de ces deux espèces sont très différents. On peut donc supposer ne pas retrouver une activité similaire à R. rosea contre la MAO. De nombreuses espèces de Sedum suisses n’ont pas encore été étudiées. Leur investigation pourrait apporter une meilleure compréhension à la chimiotaxonomie très complexe du genre Sedum. De plus, de nouveaux composés actifs pourraient être isolés. En effet, l’extrait MeOH de S. atratum a présenté une activité anti-oxydante intéressante lors du criblage.

Les travaux menés durant ce projet ont conduit à la rédaction de trois publications scientifiques, dont une est déjà parue et deux ont été soumises.

XXII

1. AIM OF THE PRESENT WORK

1. AIM OF THE WORK

Since recent years, a plant with very attractive indications appeared on the Western market: Rhodiola rosea L., also named “Golden root” or “Roseroot”. This with a thick rose-perfumed rhizome belongs to the Crassulaceae family. In 77 AC, Dioscorides first described it in De Materia Medica . Since, Roseroot has been used in the traditional medicine of Russia and Scandinavian countries to increase physical endurance, work productivity and to treat fatigue and depression. In 1960’s the Soviet Union seriously began it investigation, in part to maximize the performance of the Olympic athletes. Despite the great success in these countries, the plant remained unknown from the Western populations since most reports were published in Slavic or Scandinavian languages. However, in the late 1980’s, the plant began to be known in the West. In February 2003, the publication of the front page article, written by Anne Underwood, untitled “ Rhodiola -Herbal Stress Buster?” in the consumer magazine Newsweek rose a great impact to a large public. Since many pharmaceutical industries are interested in developing R. rosea -based products. All the clinical studies undertaken on the hydro-alcoholic extract of R. rosea led to the same conclusions: R. rosea has anti- fatigue, anti-stress, antioxidant and physically and intellectually stimulant virtues. All these properties meet the present demand of Western populations. Though, the pharmacological activities have been verified by clinical studies, its mechanisms of action are still unclear and more investigations are necessary to order to identify the active metabolites. Considering the remaining work to be done on the biological activities of the plant, physiological effects of R. rosea were evaluated in the course of the present study. Tests were based on three different targets, i.e. monoamine oxidases (MAO), acetylcholinesterase (AChE) and oxidative stress which are all involved in central nervous system functions and/ or disorders. AChE plays an important role in the cholinergic system which contributes to the memory functions. MAO is one of the key enzymes in the metabolism of neurotransmitter such as serotonin and noradrenaline which are involved in the modulation of mood and emotion, as well as the control of motor, perceptual and cognitive functions. Finally, oxidative stress is known to be involved in several neurodegenerative diseases. While R. rosea extracts exhibited inhibitory activity against MAO and AChE and scavenging properties, bio-guided investigations were initiated in order to identify the active metabolites. Victims of its own success, demand for R. rosea raw material rose from the industries, threatening the natural wild populations. Cultivation of roseroot seems to be the solution for producing raw material in sufficient quantities. In collaboration with the Agroscope of Changins-Wädenswill, in Conthey, selected wild plants collected in Wallis were planted in the field under controlled conditions. While waiting for the plants growing (a four year-period), samples of the selected wild plants were analysed. The objective was to define the optimum harvesting period to obtain the highest active metabolites in

3 1. AIM OF THE WORK

the plants. Therefore, the dynamic of the chemical profile was monitored on four wild plants during a growing season. In a second part of this project, a comparison study between male and female plants was performed in order to define the most appropriate gender for further cultivation. Another part of this work consisted in finding new potential herbal stress buster growing in Switzerland. We choose to investigate plants from the same family as R. rosea . Since no other Rhodiola species than Rhodiola rosea L. is found in Switzerland, the plant selection was focussed on Sedum species considering that this genus is close to Rhodiola . Six available species selected among the 21 species found in Switzerland were biologically investigated. Among them, Sedum dasyphyllum L. exhibited the most interesting biological activities. Therefore, the methanol and aqueous extract of this species was investigated in order to identify the active compounds.

4

2. INTRODUCTION

2. INTRODUCTION

2.1. THE CRASSULACEAE FAMILY

Crassulaceae is a morphologically diverse and systematically complex angiosperm family containing 35 genera and 1500 species. Members of the family are succulent, usually herbaceous, and often have five-parted, radially symmetrical flowers with two whorls of five stamens each (Mort et al. , 2001). The family is distributed throughout the Northern hemisphere as well as in Southern Africa and Australia with conspicuous centres of diversity in Mexico, South Africa, the European and temperate

Asiatic mountains (Figure 2-1).

Figure 2-1 Geographical distribution of Crassulaceae (GBIF Data Portal, 2009)

2.1.1. Systematic classification of the family

The Crassulaceae are generally considered to be monophyletic and classified in the Rosales complex, in which they are closely affiliated with the Saxifragaceae and Penthoraceae (Cronquist, 1988; Thorne, 1992). Within the complex, they are relatively primitive in floral structure and embryological features (Dahlgreen, 1983; Takhtajan, 1997). The details of the position of the Crassulaceae in the different plant systems are shown in Table 2-1.

7

2. INTRODUCTION

Table 2-1 Systematic position of the Crassulaceae in different plant taxonomy systems

De Candolle Cronquist Takhtajan Dahlgreen Thorne APG (1828) (1988) (1997) (1983) (1992) (1998)

Kingdom Plantae Plantae Plantae Plantae Plantae Plantae

Division Magnoliophyta Magnoliophyta Angiospermae

Class Dicotyledoneae Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Eudicot

Subclass Calyciflorae Rosidae Rosidae Magnoliidae Magnoliidae Core-eudicot

Superorder Saxifraganae Rosanae

Suborder

Order Rosales Saxifragales Saxifragales

2.1.2. Evolution of Crassulaceae systematics

The family was erected by Augustin Pyramus De Candolle in 1801. His concept of this group of succulent has been a very sound and successful one. De Candolle’s delimitation of the Crassulaceae was instantly accepted and has never been seriously challenged, except for the position of two Penthorum species. In fact, recent studies confirm that the Crassulaceae are a “natural” group. In contrast to the distinctness of this large family (about 1500 species), the infrafamilial classification and the delimitation of a majority Crassulaceae genera are subjected to an ongoing debate since more than 200 years ago. As the infrafamilial classification is quite complex, only a brief overview of systematics will be discussed in the following paragraph. The most comprehensive classification is certainly that one of Berger (1930). He recognized six subfamilies and 33 genera in the family based on the number and arrangement of floral parts, the degree of sympetaly, and phyllotaxis. Because of its comprehensiveness and great practical value, Berger’s classification has been the most widely followed, despite the general conviction that it is highly artificial. The core of the systematic problem in the Crassulaceae lies in the generic and infrageneric classification in subfamily Sedoideae. The Sedoideae comprise Sedum , the largest genus with ca. 470 species and nearly cosmopolitan in distribution, and several smaller genera. Ideally, Sedum displays the herbaceous, predominately perennial, Crassulaceae with alternate and entire with a single abaxial subapical hydathode, and 5-merous, obdiplostemonous flowers with free ('t Hart and Bleij, 2005). The genus encompasses a broad range of species. As many appear transitional,

8

2. INTRODUCTION

the infrageneric taxonomy is difficult. Several cytological studies have been carried out on the family ('t Hart and Eggli, 1988; Uhl, 1992). However, opinions about the delimitations and infrageneric classification of Sedum still remain highly divergent. Various classifications of the subfamilies have been proposed and there are two opposing viewpoints (Ohba, 1978). One view retains Sedum as a catchall taxon and recognises only a few additional genera under the Sedoideae ('t Hart, 1982; Fröderström, 1930; Fröderström, 1931). The other view segregates some groups as genera ( e.g., Rhodiola, , Phedimus ) from Sedum on the basis of unique morphological character sets and recognises a number of genera within the Sedoideae (Berger, 1930; Ohba, 1978). These conflicting viewpoints need still to be clarified. Molecular phylogenetic studies ('t Hart, 1995; Van Ham, 1995; Van Ham and 't Hart, 1998) concluded that many of the subfamilies proposed by Berger (1930) are not monophyletic and detected seven major clades: , , Telephium, , Aeonium, Leucosedum, and Acre clades. In their analyses, Sedum genus is shared in five clades and is interpreted to be highly polyphyletic. Based on these results, ‘t Hart (1995) suggested a revised classification with only two subfamilies: Crassuloideae and Sedoideae (Figure 2-2). He restricted Sedum to the species of the Sempervivum, Aeonium, Leucosedum, and Acre clades. Later, Mayuzumi and Ohba (2004) divided Telephium clade into four subclades: Hylotelephium, Rhodiola, Phedimus, and Umbilicus .

Berger (1930) ‘t Hart (1995) Ham (1995) Mayuzumi (2004)

Crassuloideae Crassuloideae Crassula

Kalanchoideae Sedoideae Cotyledonoideae Tribe Kalanchoe Kalanchoe Echeverioideae Hylotelephium Tribe Sedoideae Phedimus Subtribe Telephiinae Telephium Sedoideae Sempervivum Rhodiola

Subtribe Sedinae Leucosedum Umbilicus Aeonium

Acre

Figure 2-2 Subfamilial classification of Crassulaceae by Berger (1930) and the revised classification by ‘t Hart (1995). The clades recognised by Ham (1995) and Mayuzumi and Ohba (2004) are also indicated.

9

2. INTRODUCTION

2.1.3. Botanical aspects of Crassulaceae

The main morphological characteristics of Crassulaceae plants are detailed in Table 2-2.

Table 2-2 Main morphological characters of Crassulaceae species (Stephenson, 1994; Thiede and Eggli, 2007)

Main morphological characters of Crassulaceae species

General morphology Perennial or rarely annual plant, (sub)shrub rarely aquatics, or treelike, epiphytic or scandent, with ± succulent leaves, sometimes with succulent stems or underground caudices or succulent roots

Leaves (Sub)sessile or rarely petiolate, usually alternate and spiral, or opposite-decussate or rarely whorled, frequently aggregate into rosettes, simple.

Inflorescences Usually terminal, bracteates, usually many-flowered spikes or panicles

Flowers Bisexual or unisexual, actinomorphic, frequently 5-merous but varying from 3- to 32-merous

Corolla Petals free or basally united to form a short long tube

Calyx Sepals free or basally united

Fruit Usually dehiscent follicles, capsular

Seed Smallish, to 1.5-3 mm, elongate, smooth, papillate to longitudinally ridged, mostly brownish

2.1.4. Characteristic metabolism pathway of the Crassulaceae

Most species of the family are Crassulacean Acid Metabolism (CAM) plants. CAM plants have evolved a special way of avoiding transpiration. Non-CAM plants, which breathe through their stomata during the day, use the energy of the sun to convert absorbed carbon dioxide into sugars in the process known as . They therefore suffer from tremendous water loss during the heat of the day, or of high evaporation, if grown on windward slopes. CAM plants avoid this water loss through a very specialized form of photosynthesis in which their stomata are open only during the night to take in carbon dioxide. By keeping their stomata closed during the day, they cut down transpiration and therefore can survive arid conditions. The carbon dioxide they take in at night is stored overnight in the form of malic acid which makes the plant bitter- tasting in the morning. The next day, CAM plants use light energy to convert malic acid to sugar

(Figure 2-3). The trigger to complete this process of photosynthesis is not light but a marked temperature change. Because this odd kind of photosynthesis was first noticed in the Crassulaceae family, it has been named with the acronym CAM.

10

2. INTRODUCTION

CO 2

Mesophyll cell Night

Step 1: CO 2 incorporated into Organic acid four-carbon organic acids

Day

Step 2: Organic Calvin cycle acids release CO 2 to Calvin cycle

Sugars

Figure 2-3 Crassulaceae Acid Metabolism

2.1.5. Phytochemical aspects

Several phytochemical investigations were undertaken on Crassulaceae plants, mainly on Sedum and Rhodiola species. The most representative secondary metabolites are listed in the next Paragraphs.

2.1.5.1.

Alkaloids have always fascinated mankind due to their wide range of pharmacological activities. The alkaloids of Sedum are no exception in this respect. The first Sedum was isolated by Kolesnikow and Scharzmann in 1993 in Sedum acre L., who named their substance “sedamine” (Franck, 1958). Since then, many alkaloids have been isolated in this plant (Kooy, 1976). According to a chemotaxonomy study effected by Stevens et al. (1995), only Sedum species of the Acre clade were found to contain pyrrolidines and piperidines alkaloids types.

11

2. INTRODUCTION

H N CO H H2N O H N 2 H2N 2 lysine 5-aminopentanal ∆1-piperidine

-HSCoA O O -CO2 O H Mannich C C CH3 N N CH3 N CH3 COSCoA H H COSCoA H acetoacteylCoA pelletierine

Figure 2-4 Biosynthesis of piperidine alkaloid pelletierine (Mann, 1987)

The pyrrolidine and piperidine nuclei originate respectively from ornithine and lysine. In the case of piperidine alkaloids, decarboxylation and transamination of lysine leads to the formation of 5- aminopentanal, which is in equilibrium with the cyclic imine ∆1-piperidine (Mann, 1987). This active intermediate may react with β-keto acids via a Mannich type condensation, and subsequent decarboxylation yields pelletierine (Figure 2-4) (Gupta and Spenser, 1969; Mann, 1987). After formation of the skeleton, modifications may occur on the side-chain such as N-methylation and hydrogenation (Mann, 1987). The side-chain of sedamine has been found to originate from phenylalanine (Gupta and Spenser, 1967).

S. acre contains the widest variety of piperidine alkaloids with about 20 alkaloids. Figure 2-5 showed the major alkaloids isolated from it: pelletierine (1) , sedridine (2) , sedamine (3) , sedinone (4) , sedinine (5) and sedacrine (6) .

12

2. INTRODUCTION

O OH OH

N CH3 N CH3 N H H H

pelletierine (1) sedridine (2) sedamine (3)

O OH OH OH O OH

H3C N H3C N H3C N CH3 CH3 CH3

sedinone (4) sedinine (5) sedacrine (6)

Figure 2-5 Alkaloids isolated from S. acre

Beside their occurrence in the Crassulaceae, Sedum alkaloids have been found in a few other families mainly in higher plants. However, the typical S. acre alkaloids, i.e. sedamine (3) , sedinone (4), sedinine (5) and sedacrine (6) have not been reported from other families (Stevens, 1995).

2.1.5.2. Tannins

Tannins are defined as water-soluble phenolic compounds having molecular weights between 500 and 3’000 Da (Bate-Smith and Swain, 1963). Vegetable tannins can be divided into two major chemical groups: condensed tannins and hydrolysable tannins. The condensed tannins also named are flavonol polymers. The structural unit is the flavan-3-ol nucleus. Proanthocyanidins exist as oligomers, containing two to five or six catechin units, and polymers. The flavan-3-ol units are linked principally through the 4 and 8 positions. Procyanidins (7) and prodelphinidins (8) (Figure 2-6) are the most commonly found types of condensed proanthocyanidins and they are generally associated with plants growing in a woody habitat.

13

2. INTRODUCTION

R OH

HO O OH R OH OH n = 2-50 8 HO 7 O 2 OH 6 4 3 R 5 OH OH HO O OH R = H procyanidin (7) R = OH prodelphinidin (8)

OH

Figure 2-6 Common condensed proanthocyanidins

The hydrolysable tannins are oligo- or polyesters of sugar unit and variable number of phenol acid molecules. The sugar is generally D-glucose. The phenol acid corresponds to gallic acid in gallic tannins or to hexahydroxydiphenic acid in the ellagitannins (Bruneton, 2005). Vegetable tannins have been described in over 80 dicotyledonous plant families. The condensed tannins are the most widely found. They are also present in many monocotyledons, gymnosperms and pteridophytes. Tannins present various pharmacological activities. Several tannins, particularly the hydrolysable, act as antioxidants in cellular pro-oxidant states. They also inhibit different enzymatic activities by precipitating proteins and peptides (Bruneton, 2005). Proanthocyanidins, in majority prodelphinidins, have been found in all clades of Crassulaceae family, except for the Acre clade where they are rare or absent and replaced by alkaloids (Stevens et al. , 1995). Proanthocyanidins are widespread both in woody and herbaceous species (Stevens et al. , 1995). Galloyl esters are common within the family, but ellagitannins are absent (Thiede and Eggli, 2007).

2.1.5.3. Flavonoids

Various flavonoids were isolated from Crassulaceae species; i.e. derivatives of 14 aglycones (, herbacetin, , , , and their 8-hydroxy and/or 8-methoxy derivatives) were identified by Stevens et al. (1996) in 63 species of Sedoideae and Sempervivoideae .

14

2. INTRODUCTION

The flavonoids are a large group of natural products (more than 4000) which are widespread in higher plants. All have a common biosynthetic origin: the condensation of three molecules of malonyl-CoA with a hydroxycinnamic acid CoA ester (Figure 2-7).

O OH

3x O

CoAs CoAs

malonyl-CoA O 4-coumaroyl CoA

OH

HO OH

4,2',4',6'-tetrahydroxychalcone OH O

OH

HO O

OH O (flavanone)

OH OH HO O B HO O HO O A C OH O OH OH OH O OH O () dihydrokaempferol (flavone) (dihydroflavonol)

OH OH OH

HO O HO O HO O

OH OH OH OH OH O OH

afzelechin kaempferol pelargonidin (catechin) (anthocyanidin) (flavon-3-ol)

Figure 2-7 Biosynthesis of flavonoids

15

2. INTRODUCTION

The first central step is catalysed by the chalcone synthase and lead to the formation of a chalcone intermediate (the 4,2’,4’,6’-tetrahydroxychalcone). Transformation by stereospecific action of the chalcone isomerase will produce the first flavonoid, a flavanone (naringenin). The introduction of a double bond between C-2 and C-3 by a dioxygenase will lead to the abundant class of (apigenin), whereas an oxidative rearrangement by the isoflavone synthase will yield an isoflavone (genistein), and a direct hydroxylation in C-3 will lead to the formation of dihydroflavonols (dihydrokaempferol). These ones are the intermediates in the formation of (afzelechin), anthocyanidins (pelargonidin) and flavonols (kaempferol), formed by the introduction of a double bond catalyzed by the flavonol synthase. Hydroxylation, in particular of ring B, and methylation of hydroxyl functions lead to the various aglycones structures within each flavonoid class. Further frequent modifications of the flavonoid skeleton are due to the glycosylation of hydroxyl groups and to acylation of these sugar moieties (Davies and Schwimm, 2006).

Anthocyanidins represent one of the most important classes of flavonoid. These are intensely coloured and widely distributed in floral tissues. The six common pigments reported in Crassulaceae family, pelargonidin (9) , cyanidin (10) , peonidin (11) , delphinidin (12) , petunidin (13) , and malvidin (14)

(Nielsen Allan et al. , 2005; Van Wyk and Winter, 1995), are represented in Figure 2-8.

R1 R =R =H pelargonidin (9) OH 1 2 R1=OH, R2=H cyanidin (10) R =OCH , R =H peonidin (11) HO O 1 3 2 R =R =OH delphinidin (12) R2 1 2 R1=OCH3, R2=OH petunidin (13) OH R1=R2=OCH3 malvidin (14) OH

Figure 2-8 Anthocyanidins isolated from Crassulaceae species

Flavonols and flavones are also among the most important groups of flavonoids. Their colour is generally pale yellow. They can be found both in flower and in leaf tissues. Figure 2-9 illustrates the three common flavonols, kaempferol (15), quercetin (16) and myricetin (17) widely distributed in the family. The less frequent 8-OH substituted flavonols (18) and herbacetin (19) are also

16

2. INTRODUCTION

present, particularly in the genus Sedum . Many flavonoids occur also as glycosides, they are more commonly substituted on the C-7, C-8 or C-3 hydroxyl with β-D-glucose, α-L- or β-D- glucuronide.

R3 3' OH 2' R1 = R2 = R3 = H kaempferol (15) R2 B 4' R1 = OH, R2 = R3 = H quercetin (16) 8 7 2 HO O R1 = R3 = OH, R2 = H myricetin (17) R1 R1 = R2 = OH, R3 = H gossypetin (18) 6 AC 3 R1 = R3 = H, R2 = OH herbacetin (19) 5 4 OH OH O

Figure 2-9 Common flavonols present in Crassulaceae species (common position of glycoside substitutions is indicated with arrows)

Anthocyanidins, flavonols and flavones play an important role in the coloration of flowers and fruits. As pigments (anthocyanidins) or co-pigments (flavonones and flavonols) in floral structures, they play a pollinator attractant role. The flavonoids are also present in the foliar cuticle and in the epidermal cells of the leaves to ensure the tissue protection against the harmful effect of UV radiation (Bruneton, 2005).

Isoflavonoids differentiate themselves from flavones by the linkage, in position 3, of the C-ring to the B-ring. They are almost completely confined in the Fabaceae family but their structural diversity is important. They occur in a free state or as glycoside, and are obtained from root, wood, bark or seed rather than leaf or flower. Majority of isoflavones are phytoalexins, i.e. substances produced by the plant resulting from a pathogen infection. Few pharmacological properties are attributed to these metabolites. However two main activities may be distinguished: insecticidal activity of (20) and the phytoestrogenic activity of isoflavones (genistein (21) ) (Figure 2-10).

17

2. INTRODUCTION

8 O O HO O 2 7 O A C 6 3 5 B O OH O O OH O rotenon (20) genistein (21)

Figure 2-10 Structures of the two different types of isoflavonoids

Only two species of Crassulaceae (Sedum alfredii Hance and Sedum lineare Thunb.) have been described to contain isoflavones (22-26) (Li and Zuo, 1991; Mackova et al. , 2006; Men, 1986). They are illustrated in Figure 2-11.

R3O O

OH O OR1 OR2

R1 = CH3, R2 = R3 = H pratensein (22) R1 = CH3, R2 = H, R3 = Glc pratensein 7-O-β-D-glucoside (23) R1 = R2 = CH3, R3 = Glc 3'-O-methylpratensein 7-O-β-D-glucoside (24) R1 = R2 = H, R3 = Glc 7-O-β-D-glucoside (25) β− R1 = H, R2 = CH3, R3 = Glc 3'-O-methylorobol 7-O- D-glucoside (26)

Figure 2-11 Isoflavonoids isolated from Crassulaceae species

The flavonoids display a remarkable array of biochemical and pharmacological actions. They have been shown to affect a large variety of enzymes ( e.g . catechol-O-methyltransferase, protein kinase C, amylase, etc.), to possess important enzyme-inducing activities ( e.g . epoxyde hydrolase), to possess free-radical scavenging activity, to chelate certain metal cations, to have antioxidant properties, to

18

2. INTRODUCTION

affect the immune and inflammatory cell functions, to have antitoxic and hepatoprotective effects and antiviral effects (Clifford and Brown, 2006).

2.1.5.4. Cyanogenic and nitrile glycosides

The cyanogenic glycosides (CG) are amino acid-derived plants constituents, present in more than 2500 plant species (Vetter, 2000). They are composed of an α-hydroxynitrile type aglycone and of a sugar moiety. The aglycones can be grouped into aliphatic and aromatic compounds; the sugar is mostly D- glucose, but gentibiose or primeverose can also be found (Vetter, 2000).

O OH O OH

R' R' OH 2-oximino acid α-amino acid NH2 N R R

R' CN R OH nitrile aldoxime N R' R

OH HO OH 2-hydroxynitrile 2-hydroxyaldoxime N R' CN R R' R

O-Glucose Glucose-O OH Cyanogenic glucoside N R' CN R R' R

Figure 2-12 Pathways of cyanogenic glycoside biosynthesis (Tapper and Reay, 1973, dotted arrows indicate alternative pathways)

Figure 2-12 illustrates the general pathway of cyanogenic glycosides biosynthesis described by Tapper and Reay (1973). The hydroxylation of α-amino acids gives N-hydroxylamino acid, which is converted to an aldoxime, and finally into a nitrile. This latter is hydroxylated to form an α- hydroxynitrile, which is glycosylated to form the corresponding CG.

19

2. INTRODUCTION

The cyanogenesis is the generation of cyanide from CG. It is a two-step process involving the initial deglycosilation of CG and the cleavage of CG to acetone cyanohydrin to form acetone and cyanide

(Figure 2-13). These reactions are catalysed by a β-glucosidase and by α-hydroxynitrile lyase. The tissue level compartmentalisation of CG and their hydrolysing enzymes prevents large-scale hydrolysis in intact plant tissue.

CCN C CN H2O O Glucose OH Glucose

pH 3.5-6.0 temp < 65°C hydroxynitrile lyase CCN HCN C O OH

Figure 2-13 Cyanogenesis from lotaustralin (according to McMahon et al. , 1995)

Several well known plants with economical value contain CG, i.e. linamarin in Manihot esculenta Crantz , Linum usitatissimum L. , Trifolium repens L., dhurrin in Sorghum species, amgydalin in rosaceous plants, lotaustralin in Lotus corniculatus L., etc. Linamarin and lotaustralin have a relatively broad distribution in the plant kingdom. They have been detected in several plant families, i.e . Compositae, Euphorbiaceae, Linaceae, Papaveraceae and Fabaceae.

As cyanogenic glucosides, nitrile glucosides are β-glucosides of hydroxy nitriles. However, in nitrile glucosides, the hydroxyl and nitrile groups are not linked to the same carbon atom of the aglycone. Accordingly, hydrolysis of nitrile glucosides by β-glucosidases does not result in HCN release. While the biosynthetic pathway for cyanogenic glucosides has been elucidated and some biological roles have been identified, no experimental data are available on how nitrile glucosides are synthesised and what their biological function might be. Cyanogenic and nitrile glucosides have been found in some Crassulaceae species. Ahmad et al. (1994) and Varma et al. (1986) highlighted the presence of CG in Crassula falcata J.C. Wendl. and Kalanchoe integra (Medik.) Kuntze. The presence of sarmentosin (27) was reported in Sedum

20

2. INTRODUCTION

sarmentosun Bunge (Fang et al. , 1979) , Sedum cepeae L. (Narhrstedt et al. , 1982) , Rhodiola rosea L. , Rhodiola kirilowii (Regel) Maxim. , Rhodiola semenovii (Regel & Herder) Boriss. and Rhodiola yunnanensis (Franch.) S.H. Fu (Bjarnholt et al. , 2008). Different studies reported the presence of lotaustralin (28) and rhodiocyanosides A, B and D (29-31) (Figure 2-14) in different species of Rhodiola, including Rhodiola rosea L. (Akgul et al. , 2004; Bjarnholt et al. , 2008; Yoshikawa et al. , 1995; Yousef et al. , 2006). Rhodiocyanosides A (29) and D (31) are not restrained to the genus Rhodiola ; their presence has also been identified in Lotus japonicus (Regel) K. Larsen (Fabaceae). Yoshikawa et al. (1995) reported an antiallergic effect of rhodiocyanosides A (29) and B (39) in Rhodiola quadrifida (Pall.) Fisch. et Mey.

CN HO HO OH O O CN O (E) O OH OH OH OH OH OH sarmentosin (27) lotaustralin (28)

OH OH CN CN HO HO HO O O O OH O O (Z) O (E) O OH OH O OH CN OH OH OH OH OH OH rhodiocyanoside A (29) rhodiocyanoside B (30) rhodiocyanoside D (31)

Figure 2-14 Cyanogenic (28) and nitrile (27, 29-31) glucosides isolated from Crassulaceae species

21

2. INTRODUCTION

2.1.6. Interests of the family

2.1.6.1. Horticulture

No species of this family is an important crop plant, but many are popular for horticulture. Several members have an intriguing appearance and are quite hardy, needing thus only minimal care. Familiar species include the Jade plant or "friendship tree" (Crassula ovata (Mill.) Druce) and the "Florists Kalanchoe" (Kalanchoe blossfeldiana Poelln.) which are annually produced in large quantities as popular pot plants (Figure 2-15).

Figure 2-15 Crassula ovata and Kalanchoe blossfeldiana (Photos Courtesy Missouri Botanical Garden PlantFinder)

Sedum and Sempervivum species are also well known by gardeners. They are attractive and useful plants for rock garden, “green roofs”, and scree gardens. Because of their succulent nature, Sedum species are particularly renowned for their ability to grow in the driest part of the garden. Several species, as Kalanchoe pinnata (Lam.) Pers., are aggressive invaders in the tropics (Thiede and Eggli, 2007). Nowadays, Crassulaceae are not anymore used in food. In Medieval time, Petrosedum rupestre L. (“Trip-Madame”) was recommended for salad and was locally used as pot-herb. The fleshy leaves may appear appealing in arid environements, but they are completely tasteless or bitter and are generally avoided, even by cattle ('t Hart, 1997).

22

2. INTRODUCTION

2.1.6.2. Traditional medicinal use

Several genus of Crassulaceae are used in traditional medicine for various purposes. Some examples of the medicinal uses are detailed in Table 2-3.

Table 2-3 Medicinal uses of Crassulaceae species

Species Indications Region References

Sedum kamtschaticum Inflammatory disorders North-East Asia (Kim Dong et al. , 2004)

Sedum sarmentosum Chronic inflammatory diseases China, Korea (Jung et al. , 2008a)

Rhodiola sacra (Prain ex (Yoshikawa et al. , 1997) Hamet), Burns, contusions, cough, (Nakamura et al. , 2007) Rhodiola sachalinensis Boriss., China bleedind Rhodiola quadrifida (Pall.) (Yoshikawa et al. , 1996) Fisch. et Mey.

Rhodiola sacra (Prain ex Circulation disorders and Tibet (Shih et al. , 2008) Hamet), hypertension

Orostachys japonicus Gastric ulcer or gastric Korea (Jung et al. , 2007) disease

Kalanchoe crenata (Andrews) Otitis, headache, inflammations, Africa (Nguelefack et al. , 2006) Haworth convulsions and general debility

Kalanchoe petitiana Skin disorders Ethiopia (Tadeg et al. , 2005)

Sedum dendroideum Moc & Gastric and inflammatory Brazil (De Melo et al. , 2005) Sesse disorders

Kalanchoe brasiliensis Inflammatory and infectious Brazil (de Paiva et al. , 2008) diseases

Rhodiola rosea L. represents the species with the most pharmacological activities, as described in the next Chapter.

2.2. RHODIOLA ROSEA L.

2.2.1. Generalities

Rhodiola rosea L., also known as “Golden root”, “Roseroot” or “Arctic root”, is the most investigated species of the genus Rhodiola (Brown et al. , 2002). The plant grows at high altitude and is distributed in the arctic areas of Europe and Asia. For centuries, it has been used in the traditional medicine of Russia and . Between 1748 and 1961, various medicinal applications of R. rosea appeared

23

2. INTRODUCTION

in the scientific literature of these countries (Brown et al. , 2002). Since 1961, the plant has been extensively studied as adaptogen with various health-promoting effects; however its properties remained largely unknown in the West. This may be explained by the fact that the bulk of research was published in Slavic and Scandinavian languages. However, since the late 1980’s the plant began to be known in the West and since, the demand for R. rosea -based phytomedicines largely increased.

As shown in Figure 2-16, the number of publications about R. rosea largely increased since 2005. In 2006, there were over 46 companies worldwide using R. rosea in their products and over 30 companies listed as ingredients suppliers (Ampong-Nyarko et al. , 2006). Furthermore, an increasing number of articles about R. rosea are published in consumer magazine which will amplify even more the demand for R. rosea based-products. The elevated number of Internet Web pages (> 300’000) containing the terms Rhodiola rosea point out the importance of the phenomenon. The pharmacological activities attributed to the plant may explain its success; the indications correspond to a large part of syndromes which the Western populations suffer from, i.e. fatigue, stress and depression. Although many clinical and phytochemical studies have been undertaken on this plant, the physiological effects are still unclear. An overview of the traditional uses, phytochemistry and pharmacological effects are presented in the next Chapters.

140

120

100

80

60

40 Numberof publications

20

0 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 Year

Figure 2-16 Evolution of R. rosea publication number for the ten last years (SciFinder Scholar, 2007)

24

2. INTRODUCTION

2.2.2. The genus Rhodiola

2.2.2.1. Generalities

The genus Rhodiola is distributed in Europe, , East Asia, and in , especially in subarctic and subalpine zones (Figure 2-17).

Figure 2-17 Distribution of Rhodiola species (GBIF Data Portal, 2009)

The genus name originates from the Greek rhodon , meaning rose, an allusion to the odour of the rootstock of R. rosea. The genus Rhodiola L., based on the species R. rosea L., was described in the first edition of “Species Plantarum ” by Linnaeus (1753). He distinguished the genus Rhodiola from Sedum because Rhodiola species are dioecious plants unlike Sedum . Later, Rhodiola was reduced to a synonym within Sedum genus by Scopoli (1777), De Candolle (1828), Schönland (1891), Berger (1930), Fröderström (1930) until Nakai (1938). Borissova (1939) clarified the differences between the two genera. He distinguished Rhodiola from Sedum by the well-developed rhizomes and by the annual flowering

25

2. INTRODUCTION

stems arising from the axils of the scaly radical leaves. In their molecular phylogenetic study, Mayuzumi and Ohba (2004) confirmed that Rhodiola is well-separated from Sedum . There are still different opinions among specialists about new species which should or should not be included in the genus Rhodiola . The rational and defined criteria for the boundaries of the genus remain somehow controversial. This problem is illustrated by a wide disparity in the number of estimated Rhodiola species. Ohba (2005) accounted for ca. 60 species in the whole genus, while Fu et al. (2001) reported ca . 90 species only in China.

The main morphological characters of the genus are detailed in Table 2-4 (Thiede and Eggli, 2007).

Table 2-4 Main morphological characters of Rhodiola species (Ohba, 2005)

Main morphological characters of Rhodiola species

General morphology Dioecious or rarely monoecious, perennial plant with well developed rhizomes

Rhizomes Monopodial, massive or slender, apical part with foliage and/or scaly leaves

Leaves Alternate, simple, herbaceous (to ± fleshy), mostly flat

Inflorescences Usually cymose or reduced to a solitary flowers or rarely racemose, mostly bracteate

Flowers 4-5(-6)-merous, pedicellate, when diocecious: petals and ovaries opposite in male but alternate in female plants, when monoecious (very rare): petals and ovaries always opposite

Corolla Free petals, white, reddish, deep purple-red or pale yellow to greenish, petals always longer than sepals in female plants

Calyx Fleshy, in the female plant forming a tube divided into 4-5 (-6) ± equal lobes

Fruit Dry follicles

Seed < 3mm, ± fusiform, brownish, longitudinally striate

2.2.2.2. Traditional uses

In China, the roots and rhizomes of various Rhodiola plants have been used as adaptogen, hemostatic, and tonic in traditional medicines for thousands years. R. kirilowii, R. yunnanensis, R. crenulata (Hook. f. & Thomson) H. Ohba, R. fastigata (Hook. f. & Thomson) S.H. Fu, and R. quadrifida were the mainly used species. The rhizomes and roots of R. crenulata are still commonly use as a traditional Chinese medicine named ‘‘Suo-Luo-Ma-Bu,’’ which has been accepted by the Chinese Pharmacopoeia 2005. It is used for activating blood circulation, thoracic obstruction, apoplexy, hemiplegia, lassitude, and asthma. Although the other four species are not recorded in the Chinese

26

2. INTRODUCTION

Pharmacopoeia, they are sold under the same name as a mixed form. These species are also used individually for their medicinal properties (Tao Li, 2008). For example, R. kirilowii also known as the ‘‘Li-Ga-Du-Er’’ is used for hemostasis, alleviating pain, trauma, irregular menstruation, and dysentery and R. yunnanensis is used to treat rheumatism, mastitis, furuncle, and open fracture (Tao Li, 2008).

2.2.2.3. Phytochemical and pharmacological aspects

Numerous phytochemical and pharmacological studies have been undertaken on R. rosea as described in Chapter 2.2.6 and 2.2.7. Other species have been investigated to a lesser extent, particularly plants included in the Traditional Chinese Medicine. Different types of secondary metabolites were identified i.e. flavonoids, terpens, nitrile glucosides, and phenylethanolic derivatives (Tao Li, 2008) . Furthermore, specific constituents were characterised such as crenulatin (32) in R. crenulata (Wang and Wang, 1992), sachalol (33) in R. sachalinensis (Li et al. , 2008), rhodiocyanosides (29-31) in R. sacra (Yoshikawa et al. , 1997) and R. sachalinensis (Nakamura et al., 2007) and fastigitin ( 34)

A in R. fastigata (Yang et al. , 2002) (Figure 2-18).

HO O O OH OH OH

crenulatin (32)

OH HO OH O O OH O OH OH sachalol (33) fastigitin A (34)

Figure 2-18 Specific compounds isolated from Rhodiola species

27

2. INTRODUCTION

In recent years, several studies dealing with biological activities of Rhodiola species have been published. They were carried out with plant extracts as well as with pure compounds. Table 2-5 reviews the research done in this field and the pharmacological activities detected.

Table 2-5 Biological activities of Rhodiola species

Species Tested part Pharmacological activity Reference

R. crenulata Phenolic-enriched extract Breast cancer prevention (Tu et al. , 2008) Crenulatin Antioxidant (Su et al. , 2007)

R. imbricata Aqueous rhizome extract Adaptogen (Gupta et al. , 2008) Ethanol rhizome extract Wound healing (Gupta et al. , 2007) Hydro-alcoholic rhizome extract Radioprotective (Arora et al. , 2005) Aqueous and alcohol rhizome extract Cytoprotective/ antioxidant (Kanupriya et al. , 2005) Aqueous rhizome extract Immunostimulant (Mishra et al. , 2006)

R. quadrifida Rhodiocyanosides A/B Anti-allergic (Yoshikawa et al. , 1995)

R. sachalinensis Salidroside Sedative, hypnotic (Li et al. , 2007) Sachalosides III/IV and rhodiosin Hepatoprotective (Nakamura et al. , 2007) Flavonoids Antioxidant (Zhang et al. , 2004) Phenolic compounds Prolyl endopeptidase inhibitors (Fan et al. , 2001) (memory and learning function)

R. sacra Water and methanol extracts, and Antioxidant (Ohsugi et al. , 1999) phenolic pure compounds

Rhodiocyanoside D, sacranosides A/B Anti-allergic (Yoshikawa et al. , 1997)

Water-soluble fraction Cardio-vascular effect (Shih et al. , 2008)

Caffeic phenyl ester Anti-inflammatory (Jung et al. , 2008b)

2.2.3. Morphology and geographic distribution of R. rosea

R. rosea was first described by Linnaeus in 1749. It belongs to the subgenus Rhodiola section Rhodiola according to Ohba (1978). R. rosea possesses more than 50 synonyms enumerated by Ohba (2005), the most frequent used are probably Sedum rosea (Linné) Scopoli (1771) and Sedum rhodiola De Candolle (1805).

The main morphological characters of the species are summarised in Table 2-6 (Ohba, 2005). R. rosea roots and flower are illustrated in Figure 2-19 and Figure 2-20.

28

2. INTRODUCTION

Table 2-6 Main morphological characters of R. rosea

Main morphological characters of R. rosea

General morphology Dioecious plant occasionally hermaphrodite, 5-50 cm tall

Rhizomes Cylindrical to long obconical, branched when well developed, diameters of 1-2 cm, with scaly reddish-brown leaves, with a rose-like sent

Stems Flowered, glabrous, smooth, pale green, sometimes glaucous

Leaves Widely spreading, 7-40mm long, sessile, fleshy, oblong or obovate to oblanceolate or ovate, pale green or sometimes reddish, ± glaucous beneath, glabrous, smooth, tip around to acute

Inflorescences Terminal corymbous to umbellate cymes, 25- to 50 (-70) flowered

Flowers 3-7 mm of diameters

Calyx 2-2.5 mm, glabrous, lobes linear-subulate to narrowly oblong triangular, ascending

Petals yellow, greenish, reddish or purple, linear to narrowly oblong (male) or linear subulate (female)

Fruits Dry follicles

Seeds Narrowly oblong, 1.2-1.4 mm, tip round, brownish, longitudinally lowly ridged

Figure 2-19 R. rosea aerial parts and rhizomes, Wallis, Switzerland (Photo D. van Diermen)

29

2. INTRODUCTION

Figure 2-20 R. rosea flower, Wallis, Switzerland (Photo D. van Diermen)

R. rosea is a variable circumpolar species present in cool temperate and subarctic areas of the Northern hemisphere, including Greenland, North America, Europe and Asia. The European distribution includes Scandinavian countries and most of the mountains of central Europe, southwards to the Pyrenees, central Italy, and Bulgaria. The Asian distribution includes Arctic and Alpine regions in the Altai Mountains Eastern Siberia, Tien-Shan, the Far East, and south to the Himalayan Mountains. Several varieties of R. rosea have also been identified across Alaska, Canada, and the

Northern Mountains of the continental United States (Figure 2-21). This perennial plant grows in areas up to 2500 meters elevation.

30

2. INTRODUCTION

Figure 2-21 Geographical distribution of R. rosea L. (GBIF Data Portal, 2009)

2.2.4. Traditional uses of R. rosea

Several reports relating traditional uses of R. rosea have been found in Asia, Eastern Europe and Scandinavia. The Greek physician, Dioscorides, first recorded medicinal applications of rodi riza in 77 AC in De Materia Medica (Mell, 1938). Later, Linnaei renamed it Rhodiola rosea referring to the rose-like fragrance of the fresh cut roots. He described R. rosea as an astringent and useful for the treatment of hernia, leucorrhoea, hysteria, and headache (Linnaei, 1749). The oldest report on its use in Scandinavia concerns a journey made by King Christian IV from Norway who travelled to Finnmark and Kola in 1599. In that area, the plant was used by both the Laps and the Russians (Hansen and Schmidt, 1985). According to several Norwegian reports, in the 18 th century the plant was known as an effective remedy for scurvy (Alm, 1996). The Alaska Native people used roseroot as a vegetable and it may still be used in that way. They cooked or mixed it in various meals and ate the leaves in salad. The ground leaves were used to make bread. The children ate the raw leaves (Galambosi, 2006). The Inuit in Greenland have eaten roseroot and in 1762 Herman Ruge, priest from Valdres in Norway, wrote: “I have myself eaten it, both fried and roasted as well as boiled. And I have found it disagreeable either in taste or in effect” (Lagerberg et al. , 1955). The plant has been much

31

2. INTRODUCTION

used medicinally, especially for treating burns. It was also taken internally against lung inflammation and as remedy for urination disorders. In addition, roseroot has been used as cosmetics to wash hair since it gave a pleasant scent and was supposed to have a benefit effect (Galambosi, 2006). Russian and Asiatic traditional folk medicine used it to increase physical endurance, work productivity, longevity, resistance to high altitude sickness, and to treat fatigue, depression, anaemia, impotence, infections, gastrointestinal and nervous system disorders (Brown et al. , 2002). In the 13 th century, the legendary Ukrainian Prince Danila Galitsky, whose reputation rivalled that of Casanova, was believed to use roseroot as an aphrodisiac (Small and Catling, 1999). In , R. rosea tea was believed to be the most effective treatment for cold and flu during severe Asian winters. Mongolian doctors prescribed it for tuberculosis and cancer (Khaidaev and Menshikkova, 1978). For centuries, only family members knew where to harvest the wild golden root and the methods of extraction (Saratikov and Krasnov, 1987). Chinese emperors sent expeditions to Siberia to bring back roseroot for medicinal preparations. Siberian secretly transported the herb down ancient trails to the Caucasian Mountains where it was traded for Georgian wines, fruits, garlic, and honey. In mountain villages of Republic of Georgia, a bouquet of R. rosea roots is still given to couples prior to marriage to enhance fertility and assure the birth of healthy children (Brown et al. , 2002; Saratikov and Krasnov, 1987). It is known nowadays that during the war in Afghanistan, Russian soldiers received R. rosea roots from their mother in order to better resist to stress, fatigue and . Russian athletes also use the plant in order to improve their performances since the late 1960’s.

2.2.5. R. rosea in modern medicine

Since 1969, R. rosea has been officially included in Russian medicine. The Pharmacological and Pharmacopoeia Committee of the Soviet Ministry of Health recommended medicinal use and industrial production of liquid R. rosea extract. In 1975, the Soviet Ministry of Health approved and registered a preparation as medicine and tonic, allowing large-scale production under the name Rhodiola Extract Liquid, an alcohol-based extract. Medical and pharmacological texts recommend its use as a stimulant for asthenia (fatigue), for somatic and infectious illnesses, in psychiatric and neurological conditions, and in healthy individuals to relieve fatigue and to increase attention, memory, and work productivity. The common dose is 5–10 drops 2–3 times a day, 15–30 minutes before eating for a period of 10–20 days. In psychiatric disorders with fatigue, a starting dose of 10 drops 2–3 times a day is gradually increased up to 30–40 drops for 1–2 months. In Sweden, R. rosea was recognized as an Herbal Medicinal Product in 1985 and has been described as an anti-fatigue

32

2. INTRODUCTION

agent in the Textbook of Phytomedicine for Pharmacists. In the Textbook of Pharmacology for dispenser training in Sweden, R. rosea is mentioned as a plant with a stimulant action. In Denmark, R. rosea is registered as a medicinal product in the category of botanical drugs. These registered preparations are extensively used in Sweden and other Scandinavian countries to increase mental work capacity during stress, as a psychostimulant, and as a general strengtheners (Brown et al. , 2002).

2.2.6. Secondary metabolites from R. rosea

Phytochemistry investigations of R. rosea reported the identification of more than 70 metabolites mainly from the chemical classes detailed in Table 2-7.

Table 2-7 Different types of secondary metabolites isolated from R. rosea

Chemical class Reference

Phenylpropanoids (Kurkin and Zapesochnaya, 1986; Zapesochnaya and Kurkin, 1982)

Phenylethanol derivatives (Kurkin and Zapesochnaya, 1986; Troshchenko and Kutikova, 1967)

Flavonoids (Hillhouse et al. , 2004; Kurkin et al. , 1984b; Petsalo et al. , 2006; Zapesochnaya and Kurkin, 1983b; Zapesochnaya et al. , 1985)

Phenolic acids (Furmanowa et al. , 1998; Kurkin et al. , 1988; Kurkin et al. , 1984b; Yousef et al. , 2006; Zapesochnaya et al. , 1987)

Monoterpene glycosides (Ali et al. , 2008; Kurkin and Zapesochnaya, 1986; Kurkin et al. , 1985; Ma et al. , 2006; Wiedenfeld et al. , 2007; Zapesochnaya et al. , 1987)

Triterpenes (Krasnov et al. , 1966; Kurkin et al. , 1985)

Cyanogenic and nitrile glucosides (Akgul et al. , 2004; Byung et al. , 1999)

Essential oil (Kurkin et al. , 1985; Rohloff, 2002)

Coumarins (Furmanowa et al. , 1995; Revina et al. , 1976)

Lactones (Furmanowa et al. , 1995)

In 1967, Troshchenko and Kutikova isolated tyrosol and its glycoside rhodioloside from the roots of R. rosea. Rhodioloside was later identified as salidroside (35) (Thieme et al. , 1969) (Figure 2-22), which was first found in Salix triandra L., even the term salidroside is derived from the Salix name (Briegel and Beguin, 1926).

33

2. INTRODUCTION

OH OH O HO HO O OH

salidroside (35) Figure 2-22 Structure of salidroside

Zapesochnaya and Kurkin (1984a; 1983b) isolated from the rhizomes several flavonoids i.e. (36) , rhodiosin (37) , rhodiolin (38) , acetylrhodalgin (39) , 8-methylherbacetin (40) , kaempferol 7-O-α- L-rhamnopyranoside and methyl gallate. They identified additional phenolic compounds (rhodionidin (41) , rhodalin (42) , rhodalidin (43) and caffeic acid) from the aerial parts of R. rosea (Kurkin et al. ,

1984b) (Figure 2-23) . Petsalo et al. (2006) identified 10 supplementary flavonoids (gossypetin, herbacetin, quercetin and kaempferol derivatives) from the aerial parts.

HO HO O HO OH OH HO OH O O OH O O O O

OH OH OH O OH O rhodionin: R = α-rhamnose (36) rhodiosin: R = β-glucose (1->3) α-rhamnose (37) rhodiolin (38) O

O OH HO OH OR3

R2O O OH O O HO O OR1 OH O OH OH O 8-methylherbacetin R1= H, R2 = H, R3 = CH3 (40) rhodionidin R1 = OH, R2 = rhamnose, R3 = glucose (41) rhodalin R = OH, R = H, R = xylose (42) 1 2 3 acetylrhodalgin (39) rhodalidin R1 = glucose, R2 = H, R3 = xylose (43)

Figure 2-23 Phenolic compounds isolated from R. rosea

34

2. INTRODUCTION

Kurkin et al. (1985) while studying the flavonoids of roseroot, obtained terpenoid-like fractions, which were identified as rosiridol (44) , rosiridin (45), daucosterol, and β-sitosterol (Figure 2-24). According to Rohloff (2002), the dried rhizomes contain 0.05% essential oil. He identified 75 compounds, mainly from the following chemical classes: monoterpene hydrocarbons (25%), monoterpene alcohols (23%) and straight chain aliphatic alcohols (37%). The most abundant compounds were found to be decanol (30%), geraniol (12%) being the most important rose-like odour compound, and 1,4-p-menthadien-7-ol (5%).

HO OH OH O OH HO O HO OH

(-)-rosiridol (44) (-)-rosiridin (45)

Figure 2-24 Monoterpenes found in R. rosea

Zapesochnaya and Kurkin (1983a) isolated phenylpropanoid derivatives from the rhizomes. They identified them as glycosides of cinnamyl alcohol and named them rosarin (46) , rosavin (47) and rosin (48) . Tolonen et al. (2003) isolated further cinnamyl alcohol glycosides: cinnamyl-(6´-O-β- xylopyranosyl)-O-β-glucoside (49) , 4-methoxy-cinnamyl-(6´-O-α-arabinopyranosyl)-O-β-glucoside (50) , triandrin (51) , and picein (52) from the rhizomes (Figure 2-25).

35

2. INTRODUCTION

O HO HO O OH O HO HO (E) HO O O O HO OH O HO (E) O HO HO cinnamyl-(6'-O-ß-xylopyranosyl) OH -O-ß-glucopyranoside (49)

rosarin (46) OH O HO O OMe OH OH O HO O HO O OH HO O OH O HO (E) 4-methoxy-cinnamyl- HO O (6´-O-α-arabinopyranosyl) OH -O-β-glucopyranoside (50) rosavin (47) HO OH O HO (E) HO O OH HO triandrin (51) O HO (E) HO O OH HO rosin (48) O HO HO O OH

O picein (52) Figure 2-25 Phenylpropane glycosides isolated from R. rosea

Besides the above mentioned compounds, nitrile glucoside rhodiocyanoside A (29) and cyanogenic glucoside lotaustralin (28) (Figure 2-14 ) were also isolated from R. rosea roots by Byung et al. (1999) and Akgul et al. (2004), respectively. Phenolic acids (chlorogenic-, hydroxycinnamic- and gallic acid) were also identified in R. rosea (Brown et al. , 2002). Kurkin et al. (1986) found that the cinnamyl alcohol glycosides (rosavins (46-48) ) occurred only in R. rosea , distinguishing it from other species. Today cinnamyl alcohol glycosides and salidroside are considered to be the most important components of R. rosea , all demonstrating adaptogenic activity

36

2. INTRODUCTION

(Furmanowa et al. , 1995; Panossian and Wagner, 2005). Extracts used in most clinical trials are standardized to minimum 3% cinnamyl alcohol glycosides and 0.8-1% salidroside as the naturally occurring ratio of these compounds in the plant rhizomes is approximately 3:1 (Brown et al. , 2002).

2.2.6.1. Biosynthetic pathway of cinnamyl alcohol glycosides specific to R. rosea

Cinnamyl alcohol glycosides are products of phenylpropanoid metabolism, derived from phenylalanine, which is a derivative of the shikimic-chorismic acid pathway. The enzyme that directs carbon to the synthesis of phenylpropanoid metabolites is known as phenylalanine ammonia lyase (PAL ). It converts phenylalanine to cinnamic acid. At this level, the pathway leaves the main phenylpropanoid biosynthesis way, which leads to coumarins, flavonoids or lignins and lignans. However, the same types of enzymes take part in the further biosynthesis of the cinnamyl alcohol glycosides. Cinnamyl-CoA ester is formed from cinnamic acid through hydroxycinnamate-CoA ligase (4CL ). This CoA ester is reduced to cinnamaldehyde by cinnamyl-CoA reductase ( CCR ). The cinnamaldehyde is further reduced by cinnamyl alcohol dehydrogenase ( CAD ) to cinnamyl alcohol. The enzymes that take part in the formation of the cinnamyl alcohol glycosides are not yet described. Rosin, the simplest glycoside of roseroot, is formed by the transfer of one glucose. Rosavin is formed by the connection of an on rosin and rosarin by the connection of an arabinofuranose (Figure 2-26). Depending on the sugar type and the site it is connected to, further glycosides may be formed (György, 2006).

37

2. INTRODUCTION

O phenylalanine H2N OH

PAL

HO trans-cinnamate O

4CL

cinnamyl-CoA CoA

CCR

(E) O cinnamaldehyde

CAD

(E) OH cinnamyl alcohol

+ glucose OH O HO (E) rosin HO O OH

+ arabinofuranose + arabinose OH HO O O O O HO HO O OH O HO (E) HO (E) O HO O HO HO OH OH rosarin rosavin Figure 2-26 Biosynthetic pathway of the cinnamyl alcohol glycosides specific to R. rosea

38

2. INTRODUCTION

2.2.7. Pharmacological activities of R. rosea

Numerous scientific studies have been undertaken on R. rosea , especially in Russia and China . Brown (2002) in his overview reported over 180 pharmacological, phytochemical and clinical studies published since 1961. Between 2002 and nowadays, more than 100 supplementary reports have been published in English. Considering the great number of in vitro or in vivo pharmacological effects reported in the literature for R. rosea , only the most documented activities will be detailed in the next Paragraphs.

2.2.7.1. Effect on memory and cognitive functions

Βrown (2002) reported that R. rosea stimulated noradrenaline (NA), dopamine (DA), serotonin (5- HT), and nicotinic cholinergic effects in the central nervous system (CNS) and also enhanced the blood brain barrier permeability to precursors of DA and 5-HT. He discussed the possible effects of the plant on neurotransmitters in different neuronal pathways. Briefly, R. rosea promotes release of NA, 5-HT, and DA in ascending pathways that activate the cerebral cortex and the limbic system. Consequently, the cognitive functions of the cerebral cortex and the attention, memory, and learning functions of the prefrontal and fontal cortex are enhanced. R. rosea plays also a role in the cholinergic system. By using acetylcholine as neurotransmitter, the system contributes to memory function via pathways ascending from the memory storage systems of the limbic system to various areas of the cerebral cortex. Agents that block acetylcholine (ACh) suppress the activity of these ascending pathways and interfere with memory. R. rosea metabolites (hydroquinone (53) , rhodioflavonoside (54) and rhodiolgin (55) ) reverse this blockade by inhibiting acetylcholinesterase (Hillhouse et al. , 2004; Wang et al. , 2007). The deterioration of these systems with age results in age associated memory loss. R. rosea may prevent or ameliorate some age related dysfunction in these neuronal systems.

OH OH OH OR OH

OH OH OH O

α hydroquinone (53) rhodiolgin R = -rhamnose (54) rhodioflavonoside R = β-glucose(1->3) α-rhamnose (55)

Figure 2-27 Acetylcholinesterase inhibitors isolated from R. rosea

39

2. INTRODUCTION

R. rosea is rich in phenolic compounds (triandrin (51) , salidroside (35) , caffeic acid and flavonoids), known to have strong antioxidant properties (Chen et al. , 2008; Furmanowa et al. , 1998; Petsalo et al. , 2006). Furthermore, Zhang (2007a) et al. reported that salidroside (35) , a major component of R. rosea , has protective effects against oxidative stress-induced cell apoptosis, which might be a potential therapeutic agent for treating or preventing neurodegenerative diseases implicated with oxidative stress. Thus, in addition to enhancing cognitive functions, learning, and memory by stimulating NE, DA, 5-HT and ACh neuronal systems, R. rosea may exert positive effects on memory and cognition by improving resistance to oxidative stress.

2.2.7.2. Anti-depressive effect

Perfumi et al. (2007) firstly pointed out the anti-depressive and anxiolytic activities of R. rosea in mice. Darbinyan et al. (2007) undertook the first randomized double-blind placebo controlled study with 89 patients suffering from a current episode of mild to moderate depression. The study demonstrated that standardised extracts of R. rosea possess a clear and significant anti-depressive activity. Furthermore, no serious side effects were reported. Panossian et al. (2008a) identified salidroside (35) and tyrosol (56) (Figure 2-28) as the active compounds in the behavioural despair test on mice. Moreover, they demonstrated their synergistic action.

HO

OH

tyrosol (56)

Figure 2-28 Tyrosol, active compound in behavioural despair test on mice

It is known that depression is caused by a deficiency in the function of biogenic amines, e.g. NA, DA, 5-HT (Priest et al. , 1995). Consequently, antidepressants may act in many ways, through increasing the availability of these amines. Extracts of R. rosea have been reported to influence the levels and activities of biogenic monoamines in the cerebral cortex, brain stem and hypothalamus (Kurkin and Zapesochnaya, 1986). Stancheva et al. (1987) believed that these changes in monoamines levels were

40

2. INTRODUCTION

due to the inhibition of the enzymes responsible for their degradation, and also to the facilitation of neurotransmitter transport within the brain. However, no study allows determining the exact mechanisms of the anti-depressive effect of R. rosea and therefore further appropriated studies remain necessary.

2.2.7.3. Adaptogen

The term adaptogen was introduced in the middle of the 20 th century by the Russian scientist Lazarev (Brekhman and Dardymov, 1969) to describe medicinal plants that were able to enhance the so-called “state of non-specific resistance” of an organism to stress. It is now accepted that true adaptogens must (Brekhman and Dardymov, 1969): • possess stress-protective effects such as anti-fatigue, anti-infection and restorative activities • present stimulating effects, following both single and multiple administration, that give rise to an increase in work capacity and mental performance against a background of fatigue and stress • be innocuous and not disturb the normal level of body functions, but rather present normalizing influence on the pathologic states, independent of the nature of that state

In order to avoid confusion, it is necessary to differentiate the term ‘adaptogen’ from traditional herbal medicinal products of related action: Tonics are substances, which mitigate conditions of weakness or lack of tone within the entire organism, or in particular organs. The term is typical for traditional medicine, where tonics are used in conditions of “asthenia”. The tonic effect may be characterised by multiple doses that increase general well being and work capacity (Brekhman and Dardymov, 1969). Stimulants cause a temporary increase in work capacity, which is followed by a period of decreased work capacity. The stimulating effect may be characterised by a single dose that increases work capacity (Brekhman, 1968a). The term is used in modern and in traditional medicine. In contrast to stimulants, adaptogens are reputed to cause an increased work-capacity that is not followed by a decrease. According to Panossian and Wagner (2005), only Schizandra chinensis (Turcz.) Baill., Eleutherococcus senticosus (Rupr. & Maxim.) Maxim., Bryonia alba L. and R. rosea have been found to be fully compliant with this specific definition of adaptogen. More recently, plant adaptogens have been defined as products that increase the ability of an organism to adapt to environmental factors and to avoid damage from such factors (Panossian, 2005).

41

2. INTRODUCTION

The European Medicines Agency for the evaluation of medicines for human uses considers the term “adaptogen” not appropriate for marketing authorisations. They judge that more clinical studies and data on the efficacy in well defined clinical conditions should be necessary. However, they consider the concept adaptogen sufficient for the assessment of traditional herbal medicine products e.g . monograph on R. rosea (Committee on herbal medicinal products, 2008). The pharmacological assessment of adaptogens typically includes evaluation of their stimulating, tonic and stress-protective activities in model systems in which animals are subjected to various stress conditions. Several pharmacological and clinical studies of R. rosea reported its capacity to improve physical and mental performance and its anti-fatigue activity (Darbinyan et al. , 2000; Perfumi and Mattioli, 2007; Shevtsov et al. , 2003; Spasov et al. , 2000). Brown et al. (2002) reported different neurological mechanisms (acting on serotonin, catecholamines and endorphins) how R. rosea could enhance the non specific resistance related to different kinds of stress. For example, the serotonin system implicated in the stress response reaction and in the adaptation to new environmental conditions could be decreased in the hypothalamus by numerous stressors. Thus, the ability of R. rosea to increase non specific resistances may be theoretically related to its capacity to increase serotonin in the hypothalamus and midbrain. Another mechanism discussed by Brown is the reduction of the release of stress response factors i.e. endorphins, opioid peptides and corticotrophin. Despite considerable research efforts, it still remains unclear which mediators or stress response are predominantly involved in the mechanism of adaptogen action. However, Panossian et al. (2007) demonstrated that nitric acid and cortisol were appropriate stress markers that can be employed in the evaluation of the anti-stress effects of adaptogens. In fact, cortisol produced by the adrenal cortex is linked to the stress suppression response in the immune system. In normal release, cortisol has widespread actions that help to restore homeostasis after stress. In chronic stress, prolonged cortisol secretion causes muscle wastage, hyperglycaemia, and suppresses immune/inflammatory responses. However, short-term exposure to cortisol helps to create memory. Nitric oxide, a short-lived free radical, is released during psychological or physiological stress which may modulate stress-induced activation of the hypothalamic-pituitary axis (HPA) and the sympatho-adrenal medullary system. Olsson et al. (2009) have demonstrated clinically that R. rosea exerts its beneficial health effects on stress-induced disorders by cortisol modulation. Recently, Panossian et al. (2008b) suggested that the increased tolerance to stress induced by adaptogen is associated to the stimulation of the molecular chaperone Hsp72, a mediator of stress response involved in reparation of proteins during physical load. The molecular chaperones are a

42

2. INTRODUCTION

group of proteins that promote correct three-dimensional folding of other proteins, prevent their aggregation and assist in refolding of missfolded proteins which are the main contributors to many devastating human diseases. The protein Hsp72 plays a central role in the mechanism that rids the cell of stress-induced missfolding or incompletely synthesised polypeptides that otherwise would interfere with normal cellular function, thereby playing a critical roles in maintaining cellular homeostasis and in protecting the cell from stressful conditions and increase cell survival. The authors have demonstrated that adaptogens significantly enhance endurance of mice and the time to exhaustion by increasing the Hsp72 levels in systemic circulation.

2.2.7.4. Effect on physical work capacity

Several studies have shown that R. rosea increased physical work capacity and considerably shortened the recovery time after high intensity exercise. These studies included normal individuals exposed to maximal work (De Bock et al. , 2004; Walker and Robergs, 2006). Brown (2002) reported a placebo- controlled study with 42 competitive skiers (20-25 years old) who took either R. rosea extract or placebo 30-60 minutes before a biathlon. Athletes who were given R. rosea had statistically significant increased shooting accuracy, less arm tremor and better coordination. Thirty minutes after work performance, the heart rate in the R. rosea group was 104-106 percent of baseline, versus 128 percent in the placebo group. He concluded that R. rosea improved recovery time, strength, endurance, cardiovascular measures, and coordination. Animal studies suggest that R. rosea may stimulate essential energy metabolites, adenosine triphosphate (ATP) synthesis or resynthesis in mitochondria of skeletal muscles during exercise. The accumulation of lactic acid and the presence of ammonia in muscles are the metabolic factors causing fatigue during exercise and decreasing the rate of ATP and cyclophosphate synthesis in mitochondria. Abidov et al. (2003) showed that treatment with R. rosea promoted the decrease of ammonia concentration in mouse muscles and thus increase their physical work capacity and endurance.

43

2. INTRODUCTION

2.2.7.5. Diverse pharmacological effects

Besides the activities described previously, numerous pharmacological effects have been reported since 1961, Table 2-8 resumed these less documented activities (reviewed by György (2006).

Table 2-8 Pharmacological effects of the alcohol-aqueous extract of R. rosea

Type of Pharmacological effect Reference experiment

Antitumour and antimetastatic effects in vivo (Dementieva and Laremenko, 1987)

in vivo (Bocharova et al. , 1995)

Inhibits the growth of tumours in liver by 39% in vivo (Udintsev and Shakhov, 1991)

Anti-arrhythmia effect in vivo (Lishmanov et al. , 1993)

Prevents stress-induced cardiac damage in vivo (Maslova et al. , 1994)

Improves coronary flow in vitro (Lishmanov et al. , 1997)

Anti-hyperglycaemic and insulin stimulating activities in vivo (Molokovskij et al. , 2002)

Stimulates bone marrow erythropoiesis during paradoxical sleep in vivo (Provalova et al. , 2002) deprivation

Prevents the ischemic brain damage development in vivo (Pogorelyi and Makarova, 2002)

Hepatoprotective effect in vivo (Yaremii and Grigoreva, 2002)

in vivo (Shevtsov et al. , 2003)

Anti-inflammatory effect, protects muscle tissue in vivo (Abidov et al. , 2004)

Protects against hypochlorous acid induced oxidative damage in vitro (De Sanctis et al. , 2004)

Expedites the recovery after acute non-specific pneumonia in vivo (Narimanian et al. , 2005)

2.2.7.6. Toxicity and interactions

R. rosea has a very low level of toxicity. In rat toxicity studies, the LD 50 (lethal dose at which 50 percent of animals die) was estimated to be 3,4 g/kg of extract (Brown et al. , 2002). .Since the usual clinical doses are 200–600 mg/day, there is a huge margin of safety. Moreover, R. rosea has few side effects. Most users find that it improves their mood, energy level, and mental clarity. Some individuals, particularly those who tend to be anxious, may feel agitated. If this occurs, then a smaller dose with very gradual increases may be needed. R. rosea should be taken early in the day because it can interfere with sleep or cause vivid dreams during the first weeks. It is contraindicated in excited

44

2. INTRODUCTION

states. Because the plant has an activating antidepressant effect, it should not be used in individuals with bipolar disorder who are vulnerable to becoming maniac when given antidepressants or stimulants (Brown et al. , 2002). Recently, Panossian et al. (2009) investigated the possible interactions of R. rosea extracts with theophylline and warfarin in rats. They conclude that concomitant treatment with the extracts and theophylline or warfarin does not affect significantly their pharmacokinetics. Thus, R. rosea might be a useful antidepressant agent for the treatment of patient with mild or moderate depression since its application would not necessitate stringent precautions in order to avoid possible interactions. Unlike Hypericum perforatum L. extracts , the most popular anti-depressive herbal medicines, which present interactions with warfarin, theophylline, HIV drugs and cyclosporine.

2.2.8. Market potential and threat status of R. rosea

Based on the documented pharmacological effects and on its safe use, commercial interest in roseroot- based products has rapidly increased worldwide. Preparations are marketed in various forms, like pure alcoholic extracts or tablets or in combinations with other medicinal plants. As said before there were over 46 companies worldwide using R. rosea in their products and 30 companies listed as ingredients suppliers in 2005 (Ampong-Nyarko et al. , 2006). In Switzerland, only one preparation based on R. rosea extract is available, but it will not take a long time before the arrival of many other R. rosea based-preparations on the market. Cosmetic sector has also interests e.g. a cosmetic product containing R. rosea , used as skin protector against the electromagnetic waves is already sold in Switzerland. Presently one of the biggest problems is to meet the raw-material requirement for the increasing industrial demand. Nearly all the raw material for industrial processing is obtained by collection from natural populations (Galambosi, 2006). The largest populations are situated in the Altai area of Southern Siberia. The estimated quantity of dry roots exported from Russia was around 20-30 tons/year in 2005 (Galambosi, 2006). Due to intensive collection, the natural populations were seriously threatened. For this reason, roseroot has been added on the Red List in Russia and its collection is nowadays strictly regulated. Natural populations of the plant exist in the countries of the Alps and Carpathians, as well as in the Scandinavian countries, but collection in these European has less economic importance because of the high labour costs and the difficulties of transport in the high mountains areas. Although collection in European countries is not as extensive as in the Altai area, the species has been reported as a threatened medicinal plant in several countries, and is listed as an endangered plant in Czech Republic and Bosnia-Herzegovina and as vulnerable in Slovakia (Lange, 1998). The rapidly growing industrial demand and high prices for raw material could cause increased pressure on natural habitats.

45

2. INTRODUCTION

Cultivation of roseroot seems to be the only solution for producing raw material in sufficient quantities. Cultivation experiments have been carried out in several parts of the former Soviet Union (Kucinskaite et al. , 2007), in Russia, as well as in Sweden, Poland, Germany, Canada and Finland (Galambosi, 2006; Galambosi et al. , 2007; Kucinskaite et al. , 2007). However, roseroot field cultivation meets intrinsic problems. The cultivation costs are high because the fields have to be established by seedling transplantations. The cultivation period from planting to harvesting requires a period of five years. Post-harvesting of the roots are labour-intensive. In addition, for a continuous supply in raw material, new plantations have to be established every year. It seems that only stable financial project for long-term sustainable agronomical research and introduction of R. rosea could create market possibilities for the growers. Another way to produce material is the in vitro culture of plant cells. Different groups studied the production of R. rosea glycosides in compact callus aggregates cultures (Furmanowa et al. , 1999; Gyoergy et al. , 2004; Tolonen et al. , 2004).

2.3. SEDUM DASYPHYLLUM L.

2.3.1. The genus Sedum

2.3.1.1. Morphology and geographic distribution of the genus Sedum

The genus Sedum was first described by Linnaeus in “Species Plantarum” (1753). It is the largest genus in the Crassulaceae, and Berger placed it with a few other very small genera in the subfamily Sedoideae. The estimated number of species amounts to ca. 500 (Berger, 1930) or ca . 350 (Fröderström, 1930). When considering the morphology, it is the most diverse of all genera of the Crassulaceae. Theoretically, Sedum comprises the herbaceous, predominantly perennial Crassulaceae with alternate and entire leaves with a single abaxial subapical hydathode, and 5-merous, obdiplostemonous flowers with free petals. Table 2-9 summarised the main morphological characters of Sedum species. However, most Sedum species do not fully agree with the genus characteristics description. Although there is a kind of a general consensus about the classification of the bulk of the species that belong to Sedum , there is still much controversy about the proper delimitation of the genus ('t Hart and Eggli, 1995).

46

2. INTRODUCTION

Table 2-9 Main morphological characteristics of Sedum species ('t Hart and Bleij, 2005)

Main morphological characters of the genus Sedum

General morphology Glabrous or pubescent, perennial or annual herbs with erect, ascending, procumbent or creeping, often rooting, usually much branched non flowering shoots

Rhizomes Usually fibrous, rarely tuberous

Leaves Alternate, sometimes opposite or in whorls of 3 or 4, sessile or rarely semi-petiolate or petiolate, succulent, rarely flat, with a single abaxial apical hydathode

Inflorescences Many-flowered cymes or corymbs

Flowers Usually 5-merous, rarely 3- or 4-merous or 6- to 12-merous, generally obdiplostemonous.

Corolla Petals yellow, white, pink, purple or reddish

Calyx Sepals green, broadly sessile or free and spurred

Fruit Erect, suberect or stellate-patent without lips along the ventral suture or stellate-patent with distinct lips bordering the ventral suture

Seeds Usually ± 1 mm, ovoid to ellipsoid, testa costate or reticulate

The genus Sedum is almost completely restricted to the temperate and subtropical regions of the Northern hemisphere though a few species occur in Central East Africa, Australia and South America

(Figure 2-29). Species richness is the highest on the American continent where ± 170 species occur, i.e. ± 30 species in Canada and the USA, ± 120 species in Mexico, and ± 20 species in Central and South America. Asia is second in species richness, with 130-140 species, whereas the Eurasian region is third with ± 100 species ('t Hart and Bleij, 2005).

47

2. INTRODUCTION

Figure 2-29 Geographical occurrence of Sedum species (GBIF Data Portal, 2009)

2.3.1.2. Traditional uses of Sedum species

Sedum plants have been included in herbalist’s lists of cures for a long time. The leaves have been used externally to bring relief to wounds. Before the Middle Age, they were taken internally as emetic or anthelmintic (Stephenson, 1994). Various Sedum species are still used in Chinese Traditional Medicine, generally mixed with other plant extracts. These Sedum -based preparations are used as antitumour, anti-inflammatory, radioprotective, vasodilating, liver protector or wound-healing. The mainly used species in these preparations is S. sarmentosum (Su and Hua, 2005). Sedum species are also found in Brazil folk medicine, e.g. S. dendroidum for gastric and inflammatory diseases. However, Sedum species are mainly used in horticulture as mentioned in Chapter 2.1.6.

2.3.1.3. Phytochemical and pharmacological aspects

Numerous phytochemical and pharmacological studies have been undertaken on Sedum species. Mainly flavonoids, alkaloids and phenolic acids were identified in Sedum species. The genus Sedum is characterised by the widespread occurrence of 3’-O-methylated as well as 8-hydroxy and 8-O- methylated flavonols (see section 2.1.5.3). These derivatives might be considered as specialised

48

2. INTRODUCTION

compounds relative to kaempferol (15), quercetin (16), and myricetin (17). The flavonoids chemistry of Sedum genus is quite diverse and showed the enormous morphological and cytological variation present in this taxon. Alkaloids were only identified in Sedum species of the “ Acre clade” as described in section 2.1.5.1. Other less frequent compounds were identified in Sedum species. Generally, they were found only in specific species, e.g. lactones (telephinones A (57) and B (58) and isosiphonodin

(59) ) in S. telephium L. (Fung et al. , 1990) (Figure 2-30) , coumarins (esculetin and p-coumaric acid) in S. acre, S. album L. and S. maximum Suter (Wolbis, 1987), isoflavones (pratensein, orobolside and tritriacontane) in S. alfredii Hance and Sedum lineare Thunb. (Li and Zuo, 1991; Mackova et al. , 2006; Men, 1986), nitrile glucoside sarmentosin (Fang et al. , 1979; Su and Hua, 2005), megastigmanes ( sedum osides A4, A5, A6 and myrsinionosides) (Ninomiya et al. , 2007) and triterpenes ( δ-amyrone, 3-epi-δ-amyrin, δ−amyrin and sarmentolin) (He et al. , 1998) in S. sarmentosum Bunge .

O O O O

O O

telephinone A (57) telephinone B (58) O O

OH

isosiphonodin (59)

Figure 2-30 Lactones isolated from S. telephium

Figure 2-31 illustrates hepatoprotective metabolites (60-66) isolated from S. sarmentosum (Fang et al. , 1982; He et al. , 1998; Ninomiya et al. , 2007; Zhang et al. , 2007b). Antioxidant and anti-inflammatory activities were also described in this plant by Heo et al. (2007) and Jung et al. (2008a), respectively.

49

2. INTRODUCTION

CN HO OH OGlc OH O O (E) OH R HO OH myrsinionoside A R = O (60) OH sarmentosin (63) myrsinionoside D R = OH (61) (3S,5R,6S,9R)-megastigman-3,9-diol (62)

OH (E) OH OGlc OH O O HO O

sedumoside F1 (64) O OH OH OH O OH OH O O HO OH (E) O HO O O O OH O sarmentolin (65) sarmenoside III (66) OH OH

Figure 2-31 Hepatoprotective metabolites isolated from S. sarmentosum

Thuong et al. (2007) isolated two new antioxidant phenolic compounds from S. takesimense Nakai:

2,6-di-O-galoylarbutin (67) and gossypetin-8-O-β-D-xylopyranoside (68) (Figure 2-32).

OH OH OH xyl-O galloyl-O O HO O O OH OH OH O-galloyl OH O

2,6-di-O-galloylarbutin (67) gossypetin-8-O-β-D-xylopyranoside (68)

Figure 2-32 Antioxidant compounds isolated from S. takesimense

50

2. INTRODUCTION

A recent study focused on the fresh leave juice from S. dendroideum used in Brazilian traditional medicine for gastric and inflammatory disorders. Four antinociceptive and anti-inflammatory kaempferol glycosides (69-72) (Figure 2-33) have been isolated (De Melo et al. , 2005).

OH

R2O O

OR1 OH O

R1 = rhamnose, R2 = rhamnose (69) R1 = glucose, R2 = rhamnose (70) R1 = neohesperidoside, R2 = rhamnose (71) R1 = neohesperidoside, R2 = glucose (72)

Figure 2-33 Antinociceptive and anti-inflammatory kaempferol glycosides isolated from S. dendroideum

Altavilla et al. (2008) demonstrated that the methanolic extract of S. telephium might be useful as a potential anti-inflammatory agent. Furthermore, Bonina et al. (2000) indicated that the leave juice of this species has strong antioxidant properties and photoprotective effects against UV B-induced skin damage.

2.3.2. Morphology and geographical distribution of S. dasyphyllum

Sedum dasyphyllum L. was first described by Linneaus in Species Plantarum (1753). This species is part of Leucosedum clade according to Mort et al. (2001). It is an orthocarpic Sedum species without yellow flowers (Stephenson, 1994). The main morphological characters of S. dasyphyllum are presented in Table 2-10. A chart is displayed in Figure 2-34.

51

2. INTRODUCTION

Table 2-10 Main morphological characters of S. dasyphyllum ('t Hart and Bleij, 2005)

Main morphological characters of S. dasyphyllum

General morphology Glandular-pubescent glaucous often tufted perennial herbs with short branching and rooting stems

Leaves Densely imbricate, ovoid, sessile, obtuse or acute, flattened on the upper face, glaucous, glabrous to densely and coarsely glandular-pubescent, sometimes sticky, 3.5-7 mm

Inflorescences Cymes with (1-) 2 (3-) cincinni

Bracht Small, ovate

Flowers 5-merous

Corolla Basally free sepals, oblong-elliptic to lanceolate, acute, white, often with a red keel or pinkish, outside glaucous, purplish-green, 3-5 mm

Fruit Erect, brown or bluish

Seeds Ovoid, pale brown, costate

Figure 2-34 Chart of S. dasyphyllum (Fitch, 1924)

52

2. INTRODUCTION

S. dasyphyllum is distributed from Central Europe to the Mediterranean coastlines from the sea-level to over 2500 m (Figure 2-35). Several distinct forms are endemic to the Atlas Mountains region of North Africa, and adjacent isles. Normally found in dry, rocky areas, it also grows on stone walls (Stephenson, 1994).

Figure 2-35 European distribution of S. dasyphyllum ('t Hart, 2003)

S. dasyphyllum is mainly known in horticulture to cover soil. The only phytochemical study on the plant has been done by ‘t Hart (1999) who described the presence of triterpenes ( β-amyrenyl acetate and germanicyl acetate), flavonols (herbacetin and quercetin) and gallic acid. No pharmacological study has been reported on this plant.

53

2. INTRODUCTION

2.4. BIOLOGICAL ACTIVITY

During this work, plant extracts and isolated pure compounds have been tested against several targets. Extracts have been tested against the radical DPPH (which reveals radical scavenging activity) and acetylcholinesterase on TLC assays while pure compounds have been tested on TLC and microplate assays to determine their IC 50 . Furthermore, extracts and pure compounds have been tested on a monoamine oxidase inhibitory microplate assay. Only the biological activities which led to published results, i.e. monoamine oxidase inhibitory and antioxidant activities, are described in this Chapter.

2.4.1. Monoamine oxidase inhibitory activity

2.4.1.1. Introduction

Monoamine oxidase (MAO) is a mitochondrial bound enzyme which belongs to the class of flavin containing amine oxidoreductases. First described by Hare (1928) as oxidase, Zeller (1938) later called it monoamine oxidase to distinguish it as member of the class of enzymes responsible for the oxidative deamination of dietary amines, monoamine neurotransmitters and hormones. This broad array of substrates include several notable biogenic molecules: serotonin (5-HT), noradrenaline (NA), and dopamine (DA) and the neuromodulator phenylethylamine (PEA) (Shih and Thompson, 1999). The rapid degradation of these brain monoamines is essential for the correct functioning of synaptic neurotransmission (Figure 2-36).

54

2. INTRODUCTION

a) Release of dopamine ( DA ) b) DA receptor activation c) Reuptake by DA transporter (DAT) d) DA degraded by two enzymatic pathways:

(1) In the first pathway, MAO and aldehyde dehydrogenase (ALDH) convert DA into 3,4- dihydroxyphenylacetic acid (DOPAC); this compound is then processed by catechol-O- methyltransferase (COMT) into (HVA)

(2) In the second pathway, COMT metabolizes DA into 3-methoxytyramine (3-MT), which is then converted into HVA by MAO and ALDH

Figure 2-36 Example of monoamine neurotransmitter (dopamine) degradation in synapse by MAO and COMT (Bortolato et al. , 2008)

Monoaminergic signalling is regarded as one of the key mechanisms for the modulation of mood and emotion, as well as the control of motor, perceptual and cognitive functions (Bortolato et al. , 2008).

The chemical reaction catalysed by MAO, exemplified in Figure 2-37, consists in the degradation of monoamines into the corresponding aldehydes, which are then oxidized into acids by aldehyde dehydrogenase or converted into alcohols or glycols by aldehyde reductase. The by-products of these reactions include a number of potentially neurotoxic species, such as hydrogen peroxide and ammonia. In particular, hydrogen peroxide can trigger the production of reactive oxygen species (ROS) and induce mitochondrial damage and neuronal apoptosis (Bortolato et al. , 2008).

R 1 O R1 R N O2 H2O N H2O2 R2 R H H R2

Figure 2-37 Overall catalytic reaction of MAO

55

2. INTRODUCTION

2.4.1.2. Molecular characteristics of MAO

Two forms of MAO, designated as MAO A and MAO B have been first identified on the basis of substrate and inhibitor sensitivity, before their molecular characterisation. Of the two, MAO A exhibits a higher affinity for NA (73) and 5-HT (74) and for the inhibitor clorgyline, whereas MAO B has a higher affinity for PEA (75) , benzylamine (76) , and the inhibitor deprenyl. Tyramine (77) and

DA (78) are substrates for both MAO A and MAO B (Table 2-11). This classical grouping of MAO A, MAO B and MAO A- and MAO B-substrates must be taken with caution, since most of the natural substrates can be metabolised by both isozymes. Although most tissues express both isozymes, human placenta and fibroblasts express predominantly MAO A, and platelets and lymphocytes express only MAO B (Shih and Thompson, 1999). Studies have established that in the brain, MAO A is predominantly localised in catecholaminergic neurones, whereas MAO B is mainly expressed in serotoninergic and histaminergic neurons, as well as in astrocytes (Saura et al. , 1994; Westlund et al. , 1988). This finding, though well documented, is in apparent contrast with the pharmacological evidence that serotonin levels are enhanced only following MAO A, but not MAO B, inhibition. The reasons of this mismatch are still unclear.

Table 2-11 Endogenous substrates of MAO in humans

Preferred isozyme Endogenous substrate H MAO A OH N

OH

NH2 OH H N OH 2

serotonin ( 74 ) noradrenaline ( 73 )

NH2 MAO B NH2

phenylethylamine ( 75 ) benzylamine ( 76 ) NH NH MAO A + MAO B 2 2

HO HO

OH tyramine ( 77 ) dopamine ( 78 )

56

2. INTRODUCTION

2.4.1.3. Pharmacological inhibition of MAO

The development of MAO inhibitors started with the finding that , initially an anti- tuberculosis drug, had psychostimulant side effects by inhibiting MAO. This discovery, led to the design and production of other MAO inhibitors such as . However, this first generation of irreversible non-selective MAO inhibitors used as drugs for mood disorders caused various side- effects including liver toxicity. Another undesirable side effect of MAO non-selective inhibitors was the so called “cheese-effect”, consisting in severe, potentially lethal hypertensive crisis with cerebral haemorrhages, following the consumption of tyramine-rich food (wine, cheese and other fermented foods). Due to the lack of intestinal metabolism by MAO B, these compounds are absorbed and enter in circulation, to induce increased NA release in the medulla, which in turn activates the sympathetic system and, in the absence of MAO A-mediated metabolism, causes the sudden increment in blood pressure. The quest for MAO inhibitors devoid of side effects prompted research to characterise selective MAO A and MAO B inhibitors. Subsequently, the discovery that the neurotoxin 1-methyl-4- phenyl-1,2,3,6-tetrahydropyridine (MPTP (79) ) is metabolised by MAO B to the 1-methyl-4- + phenylpyridinium cation (MPP (80) ) (Figure 2-38), which causes death of dopaminergic neurons and induces symptoms very similar to Parkinson’s disease in humans, revived the interest for monoamine oxidase-catalysed reaction. In fact, MAO B inhibitors such as selegiline were revealed to be efficacious in the therapeutic management of Parkinson’s disease (Bortolato et al. , 2008). Selective MAO inhibitors have attracted much attention and are now of considerable interest in the management of a variety of MAO-related disorders as detailed in next Paragraphs.

MAO B

N N

+ MPTP (79) MPP (80)

Figure 2-38 MAO B catalysed oxidation of neurotoxic MPTP

57

2. INTRODUCTION

Depression and mood disorders. MAO inhibitors show high mood-enhancing ef fi cacy (Bortolato et al. , 2008). The mechanism of antidepressant action is generally interpreted as based on MAO A inhibition and the consequent ability to counter the reduction in 5-HT and NA (and, to a more limited extent, DA) that characterises depression. The current guidelines for treatment of major depression disorder, issued by the American Psychiatric Association and the British Association for Psychopharmacology, suggest that MAO inhibitors should be considered as second-choice agents, after serotonin selective reuptake inhibitors and tricyclic agents, due to the numerous side effects of these compounds. Indeed, the introduction of selective agents, which display a better tolerability profile, has gradually supplanted the use of MAO inhibitors in the clinical practice. However, the development of reversible inhibitors of MAO A (RIMAs) has renewed interest for this category of compounds (Amrein et al. , 1993), particularly in view of their ef fi cacy in treatment-resistant depression and atypical depression (Nierenberg et al. , 1998). One MAO A inhibitor (, Aurorix®) is found on the Swiss market.

Anxiety disorders. MAO inhibitors are indicated for some anxiety-spectrum disorders, namely social phobia, panic disorder, post-traumatic stress disorder (PTSD) and obsessive –compulsive disorder (OCD). Again, moclobemide and other RIMAs are a promising category of drugs to address these disorders, and may be an optimal choice particularly for PTSD and panic disorder (Cyr and Farrar, 2000).

Parkinson's disease. The use of MAO B inhibitors in Parkinson's disease (PD) was originally based on the concept that DA is preferentially deaminated by this isozyme in the human nigrostriatal dopaminergic system. Thus, the increase in DA levels caused by MAO B inhibitors should compensate for the nigrostriatal deficits in this neurotransmitter (Knoll, 2000). However, studies on the MAO B inhibitor, deprenyl, revealed that MAO inhibition elicits also neuroprotective actions. As stated previously, one of the by-products of MAO-mediated reaction is hydrogen peroxide, which contributes to the formation of other ROS and can trigger mitochondrial damage and neuronal death. Thus, MAO inhibition plays also a neuroprotective role by decreasing this ROS formation. This has been pointed out by recent evidences which show that moclobemide, the reversible inhibitor of MAO A (RIMA) has also anti-parkinsonian effects (Bortolato et al. , 2008). Furthermore, emerging evidence show that some of the neuroprotective actions of deprenyl do not depend on MAO B inhibition, but on other mechanisms (Bortolato et al. , 2008).

58

2. INTRODUCTION

Alzheimer's disease. Age-related increases in MAO B activity, as well as the neuroprotective effects of its inhibitors, have been considered as rational bases to use MAO B inhibitors in Alzheimer's disease (AD) management. However, the therapeutic effects of MAO B inhibitors for AD have been challenged by recent meta-analyses (Birks and Flicker, 2003).

2.4.1.4. Synthetic and natural monoamine oxidase inhibitors

The discovery of the two isoforms of the enzyme led to considerable gain in the understanding of MAO function and the implications of its selective inhibition. Nowadays, a large range of irreversible and reversible selective MAO inhibitors are available. While irreversible MAO B inhibitors are used as adjunctive agents in L-DOPA therapy of Parkinson’s disease, reversible MAO A inhibitors are mainly administered for the treatment of depression. The MAO inhibitors can be characterised by their specificity for one of the two forms. Moreover,

MAO A inhibitors are differentiated according to their reversibility. Table 2-12 reviewed the different classes of MAO inhibitors.

Table 2-12 MAO inhibitors from non-natural sources (Bortolato et al. , 2008)

Group Class Compound MAO selectivity

Non-selective-irreversible Hydrazines A and B

Phenelzine A and B

Iproniazid A and B

Non-selective-reversible Amphetamine derivates A and B

Selective-irreversible Propargylamines Clorgyline A

Selegiline B

Rasagiline B

Selective-reversible (RIMAs) Piperidylbenzofurans A

Morpholinobenzamides Moclobemide A

Oxazolidinones A

Linezolid A

Befloxatone A

Cimoxatone A

Mixed MAO-cholinesterase inhibitors Propylamines A and B

59

2. INTRODUCTION

The reversible inhibitors correspond to the inhibitors which form a complex with the enzyme which can be readily reversed by any procedure that reduces the inhibitor concentration. There is no covalent bond formation between the reversible inhibitor and the enzyme. If the inhibitor concentration is lowered, enzymatic activity rises again. Numerous compounds of natural and synthetic origin with great structure variety have been shown to reversibly inhibit MAO. Reversible MAO A inhibitors are mainly administered for the treatment of depression (Gnerre, 2000). The irreversible inhibitors decrease the amount of enzyme without affecting the activity of the remainder. Many irreversible inhibitors form non-covalent complex with the enzymes, with subsequent reaction leading to covalent-bond formation within this complex. The best known selective irreversible inhibitors are clorgyline and deprenyl which inhibit MAO A and B, respectively (Gnerre, 2000).

Several MAO inhibitors have been isolated from plants. These active metabolites can be grouped in five different classes, as detailed in the following section. Xanthones : In 1981, Suzuki et al. (1981) related the MAO inhibitory activity of naturally occurring xanthones. Since, many reports described the activity of xanthones isolated from diverse plants (Thull, 1995), especially from the Gentianaceae family e.g. bellidifolin (81) and bellidin (82) in Gentiana lactea Phil. (Schaufelberger and Hostettmann, 1988), bellidin and swertianolin (83) from Gentianella amarelle ssp. acuta (Urbain et al. , 2008) (Figure 2-39) .

OR O OH OH OOH

O O O OH OH OH bellidifolin R = OH (81) bellidin (82) swertianolin R = β-Glc (83)

Figure 2-39 MAO inhibitory xanthones isolated from Gentianaceae species

60

2. INTRODUCTION

β-carbolines and quinolines. Glover et al. (1982) proved that β-carbolines alkaloids and their derivatives inhibit human and rat MAO. The South American psychotropic beverage ayahuasca obtained by boiling the bark of the liana Banisteriopsis caapi (Spruce ex Griseb.) C.V. Morton (Malpighiaceae) together with the leaves of various plants, principally Psychotria viridis Ruiz & Pav. (Rubiaceae) is known to contain β-carbolines alkaloids ( (84) and (85) ) which inhibit MAO A (Gambelunghe et al. , 2008) (Figure 2-40).

N N

O N O N H H

harmaline (84) harmine (85)

Figure 2-40 MAO inhibitory β-carbolines alkaloids contained in the psychotropic ayahuasca beverage

Mitsui et al. (1989) described the MAO B inhibitory activity of three quinoline derivatives (quinine, cinchonicinol, cinchonaminone). Later, Kong et al. (2001) demonstrated the MAO inhibitory activity of the isoquinoline alkaloids berberine (86) and palmatine (87) (Figure 2-41) isolated from Coptis chinensis Franch. (Ranunculaceae). Recently, Han et al. (2007b) isolated MAO inhibitory quinolines from the fruits of Evodia rutaecarpa (Rutaceae).

O N+ O N+ O O O O O O

berberine (86) palmatine (87)

Figure 2-41 MAO inhibitory isoquinolines from Coptis chinensis

Vinca alkaloids . The neurotoxin MPTP and vinca alkaloids (vinblastine and vincristine) possess a common structural feature: vindoline (88) (Figure 2-42). Son et al (1990) described the MAO B inhibitory activity of these vinca alkaloids.

61

2. INTRODUCTION

N

OCOCH3 OH O N COOCH3 CH3

vindoline (88)

Figure 2-42 Structures of vindoline (common portion of the vinca alkaloids); the structural common feature with MPTP is marked in bold

Flavonoids . Investigation of Melastoma candidum D. Don (Melastomataceae) leaves revealed the presence of MAO inhibitory flavonoids: , isoquercitrin, and (Lee et al. , 2001). Hwang et al. (2005) found that the methanol extract of Sophora flavescens Aiton (Fabaceae) showed inhibitory effect on mouse brain monoamine oxidase. Bioactive-guided isolation yielded two flavonoids, (89) and kushenol F (90) (Figure 2-43) which showed significant inhibitory effects on MAO in a dose-dependant manner. Han et al. (2007a) described the MAO inhibitory activity of apigenin, and quercetin isolated from Cayratia japonica (Thunb.) Gagnep. (Vitaceae). The three flavonoids had a more potent effect on MAO A than MAO B.

OH HO O

HO O

OH O O OH O formononetin (90) kushenol F (89)

Figure 2-43 MAO inhibitory flavonoids isolated from Sophora flavescens

Stilbene derivatives . Ryu et al. (1988) isolated from Reynoutriae radix and Rhei undulati rhizomes (Polygonaceae) various stilbene derivatives displaying MAO A inhibitory activities, with resveratrol being to most active.

62

2. INTRODUCTION

2.4.2. Antioxidant activity

2.4.2.1. Oxidative stress

The oxidative stress can be defined as a imbalance between reactive oxygen species (ROS) and antioxidants in favour of the former, leading to potentially damage (Sies, 1985). The ROS encompass • - a variety of diverse chemical species including oxygen radicals (superoxide anions (O 2 ) and hydroxyl • radicals (HO )) and non-radical derivatives of O 2 (hydrogen peroxide H 2O2, hypochlorous acid

(HOCl), and ozone (O 3)). The ROS are the result of normal intracellular metabolism in mitochondria and peroxisomes, as well as cytosolic enzyme systems. In addition, a number of external agents can trigger ROS production. However, they can be produced at elevated rates under pathophysiological conditions which can result in significant damage on several targets in biological systems such as DNA, proteins, and lipids (Finkel and Holbrook, 2000). The generation of ROS requires the activation • - of molecular O 2 in O 2 . This superoxide is produced by different pathways involving enzymes, haem proteins, auto-oxidation reactions or mitochondrial electron transport. The most studied enzymes in • - • - O 2 production are NADPH oxidases which produce deliberately O 2 in phagocytic cells to digest • - pathogens and metallo-enzymes such as xanthine oxidase which reduce O 2 in O 2 (Halliwell and Gutteridge, 2007). Exogenous sources, i.e. ultraviolet light, ionizing radiation, chemotherapeutics and environmental toxins are also ROS generators (Figure 2-44). The biomedical literature is full of claims that free radicals and other reactive species are involved in human diseases. They are implicated in over 150 disorders ranging from rheumatoid arthritis and intestinal ischemia through cardiomyopathy and Alzheimer’s and Parkinson’s disease to osteoporosis (Halliwell and Gutteridge, 2007). Some diseases are probably caused by oxidative stress e.g. excess radiation exposure causes many biological consequences. However in most human diseases, oxidative stress is a consequence and not a cause of the disease. Tissue damages lead to formation of increased amount of putative “injury mediators”, such as prostaglandins, leukotrienes, cytokines and, of course, ROS. All of these have, at various times, been suggested to play important roles in tissue injury. Thus, the essential question: whether ROS play a critical role as mediators of the pathology, or are simply produced in response to some other form of injury, still need to be clarified.

63

2. INTRODUCTION

Figure 2-44 Sources and cellular responses to reactive oxygen species (according to Finkel and Holbrook, 2000)

2.4.2.2. Antioxidant defences: endogenous and diet derived

As discussed previously, O 2 is poisonous; however aerobes organisms survive in its presence because they have evolved antioxidant defences. Antioxidants are synthesised in vivo or taken in from the diet. They can be defined as “any substance that, when present at low concentrations compared with that of an oxidisable substrate, significantly delays or inhibits oxidation of that substrate” (Halliwell and Gutteridge, 2007).

64

2. INTRODUCTION

There are numerous endogenous antioxidant defences designed by organisms to evade oxidative stress which comprise (Figure 2-45): • Enzymes that catalytically remove ROS , such as the superoxide dismutase (SOD), superoxide reductase, catalase and peroxidase enzymes. The SODs are well known enzymes as their

discovery led to the postulation of the superoxide theory of O 2 toxicity and the realisation that free radicals are important metabolic products. These proteins co-factored with copper and zinc, or manganese, iron, or nickel, remove ROS by catalyzing dismutation of superoxide into oxygen and hydrogen peroxide. • Agents that decrease ROS formation , e.g. mitochondrial uncoupling proteins, transferrins, albumin, haptoglobins and other proteins which minimise the availability of pro-oxidant such as iron and copper ions or haem. • Proteins that protect biomolecules against oxidative damage by other mechanism, e.g. chaperones • “Sacrificial agents” that are preferentially oxidized by ROS to preserve more important • - biomolecules, e.g . glutathione reacts with various radicals but not O 2 (or very slowly), • •, • bilirubin scavenges ONOOH, RO 2, and RO , urate is a powerful scavenger of O 3 and NO 2. Furthermore it protects proteins against nitration and chelate metal ions, and albumin bonds haem and copper ions.

GSH Glutathion GSSH peroxidase

O 2 H2O2 H2O Superoxyde dismutase Catalase Fenton reaction

OH H2O + 1/2 O2

Figure 2-45 Some endogenous antioxidant reactions

Many dietary constituents have been suggested to act as antioxidants. Some examples of low- molecular-mass antioxidants agents derived from the diet are discussed below:

65

2. INTRODUCTION

Ascorbic acid : this white crystalline solid, which L-enantiomer (Figure 2-46) corresponds to vitamin C (91) , is an essential nutriment. In fact, humans, primates, guinea primates and some fish have lost the enzyme required to synthesise it. Good dietary sources of vitamin C include citrus fruits, guava, berries, mango, kiwi, broccoli and peppers. Ascorbate in animals acts as cofactor for at least eight enzymes, of which the best known are the prolyl and lysyl hydroxylases involved in collagen • • - biosynthesis. Ascorbate scavenges various free radicals as RO 2, O 2 , and also HOCl, ONOOH and more (Halliwell and Gutteridge, 2007). Vitamin E : the nutritional term vitamin E is not a specific chemical; it covers eight naturally-occurring tocopherols and tocotrienols. The most effective form in animals is RRR-α-tocopherol (92) (Figure

2-46). Dietary sources of vitamin E include wheat-germ, vegetable oil, nuts, grains, and green leafy vegetables. Fat-soluble α-tocopherol concentrates in the interior of membranes and in lipoproteins. Tocopherols and tocotrienols inhibit lipid peroxidation by scavenging lipid peroxyl radicals much faster than these radical can react with adjacent fatty acid side chains or with membrane proteins. Nonetheless, supplementing well-nourished humans with α-tocopherol has only modest effects on lipid peroxidation. Carotenoids : These red or orange coloured pigments widespread in plant tissues are precursor of the fat-soluble vitamin A, sometimes called retinol (93) . Vitamin A (Figure 2-46) is essential for cell growth and differentiation, and in vision. The most important carotenoids are β-carotenes. In plants, carotenoids play a key antioxidant role, helping quenching O 2 and deter its formation during photosynthesis. Tebbe et al. (2001) showed that β-carotene administration protected porphyria patients against light-induced skin damage. Alaluf et al. (2002) demonstrated that sunlight depletes β-carotene in skin, consistent with a protective role even in healthy subjects. Carotenoids are capable of scavenging some ROS, but do not protect LDL from peroxidation.

HO O

O (E) (E) HO O OH H (E) (E) OH HO OH vitamin C (91) RRR-α-tocopherol (92) trans-retinol (93)

Figure 2-46 Low-molecular-mass antioxidant agents derived from the diet

66

2. INTRODUCTION

Plant phenols . Plants contain a huge range of mono-, di-, and . Most of them exert antioxidant effects in vitro , inhibiting lipid peroxidation by acting as chain-breaking peroxyl radical • • scavengers. In addition, phenols often scavenge other ROS, such as NO 2, HO , N 2O3, ONOOH and • - HOCl. Some can react with O 2 , mostly the di- and polyphenols (Taubert et al. , 2003). Compounds with catechol groups, or other chelating structures, can bind transitional metal ions (especially iron and copper), often in poorly reactive forms. The number of phenolic –OH groups and their relative positions are key determinants for the antioxidant activity. The human diet is rich in a variety of phenols, such as catechins in green tea, resveratrol (94) in wine, genistein in soy beans and rosmarinic acid (95) in rosemary (Figure 2-47). The USA and Netherlands total daily flavonoids consumption is estimated as at least 23 mg, of which about 16 mg is quercetin (Halliwell and Gutteridge, 2007). Phenols in red wine were found to inhibit LDL oxidation in vitro , and it was suggested that they could exert cardioprotective effects by limiting LDL oxidation in vivo. This was proposed as an explanation of the lower incidence of heart disease in certain areas of France named as the French Paradox, despite the high prevalence of risk factors, such as smoking and high fat intake (Frankel et al. , 1995). Furthermore an epidemiological study in the Netherlands (the Zutphen study) (Keli et al. , 1996) suggested an inverse correlation between the incidence of coronary heart disease and stroke and the dietary intake of flavonoids, which originated mainly from tea, fruits and vegetables in the population examined.

HO

HO CO2HO OH HO HO O OH resveratrol (94) rosmarinic acid (95) OH

Figure 2-47 Antioxidant plant phenols

67

2. INTRODUCTION

2.4.2.3. Neurodegenerative diseases and oxidative stress

Stress oxidative is involved in a broad array of diseases as mentioned previously, particular attention will be taken in this section for its role in the neurodegenerative diseases since Crassulaceae species, particularly R. rosea are known to have potentially influence on central nervous system disorders. The nervous system, including the brain, spinal cord, and peripheral nerves, is rich in both unsaturated fats (which are prone to oxidation) and iron. The high lipid content of nervous tissue, coupled with its high metabolic (aerobic) activity, makes it particularly susceptible to oxidant damage. The high level of brain iron may be essential, particularly during development, but its presence also means that injury to brain cells may release iron ions that can lead to oxidative stress via the iron-catalyzed formation of reactive oxygen species (Emerit et al. , 2004). There are substantial evidences ( e.g. increased lipid peroxidation and oxidation of DNA in the substantia nigra) that oxidative stress is a causative or at least ancillary factor in the pathogenesis of major neurodegenerative diseases, including Parkinson's and Alzheimer's diseases (Barnham et al. , 2004). A number of in vitro and in vivo studies have shown that antioxidants, both endogenous and dietary, can protect nervous tissue from damage induced by oxidative stress. For example, both vitamin E and β-carotene was found to prevent neuronal damage and cell death (apoptosis) in rat neurons subjected to hypoxia followed by oxygen reperfusion (Mitchell et al. , 1999; Tagami et al. , 1998). In a Dutch study, it was found that the risk for Parkinson's disease was lower for subjects who had higher dietary intakes of antioxidants, particularly vitamin E (de Rijk et al. , 1997). In another study, it was found that patients suffering from Parkinson's disease had consumed less antioxidants β-carotene and vitamin C than did non-sufferers of the disease, implying that dietary antioxidants do play a protective role in this disease (Hellenbrand et al. , 1996). There is a large amount of evidence indicating that oxidative stress plays a crucial role in aging as well as in neurodegenerative but it is still unclear whether oxidative stress is the primary initiating event associated with neurodegeneration or a secondary effect related to other pathological pathways but a growing body of evidence implicates it as being involved in the propagation of cellular injury leading to different damages observed in neurodegenerative diseases (Mariani et al. , 2005). Nevertheless, these studies demonstrated the antioxidant influence on neurodegenerative diseases and strengthen the idea that to prevent their occurrence it is important to avoid causes of oxidative stress such as radiation exposure, large alcohol consumption, smoking, obesity and reinforce the antioxidant defences with food and plants rich in antioxidants molecules.

68

2. INTRODUCTION

2.4.2.4. Antioxidant and radical scavenging activity detection

Several in vitro methods are proposed to detect antioxidants and measure their power: for example ferric reducing antioxidant power assay, β-carotene/linoleic acid assay, inhibition of LDL protein oxidation assay, and finally DPPH assay (Szabo et al. , 2007). The latter was suggested in 1958 by Blois (Blois, 1958) to provide a convenient method for detecting and measuring the concentration of antioxidants in biological materials. DPPH is a stable radical (2,2-diphenyl-1-picrylhydrazyl), thanks to its spare electron delocalisation over the whole molecule. The delocalisation causes a deep violet colour with λmax around 520 nm. When a solution of DPPH is mixed with a substrate acting as a hydrogen atom donor, a stable non-radical form of DPPH is obtained with simultaneous change of the purple colour to pale yellow (Szabo et al. , 2007) (Figure 2-48).

N N

N• NH + RH + R• O2N NO2 O2N NO2

NO2 NO2

DPPH• DPPH pale yellow purple

Figure 2-48 Reaction of DPPH with a radical scavenging molecule

69

3. RESULTS

3. RESULTS

3.1. CHARACTERISATION OF BIOACTIVE COMPOUNDS FROM RHODIOLA ROSEA L. ROOTS

3.1.1. Extraction

R. rosea roots were collected in Val d’Aoste (Italy) in November 2005. The roots were washed and cut. As the succulent roots did not dry at ambient temperature and moreover, there was a high risk of mould contamination, the cut roots were frozen and placed in vacuum in order to remove water by sublimation with a lyophiliser. Once dried, the roots were ground in liquid nitrogen to prevent the deterioration of thermolabile compounds which may occur during the grinding process. The milling helps to maximize the exposition of the plant cells to the solvent, increasing therefore the surface of extraction and improving the solubilisation of the secondary metabolites. The powdered roots were then successively extracted with an apolar solvent (dichloromethane, DCM), and a polar solvent (methanol, MeOH), 3 times during 24 hours. This led to a first separation between lipophilic secondary metabolites in the DCM extract and the medium and high polarity secondary metabolites in the MeOH extract. After filtration, the enriched solvents were evaporated to dryness under vacuum. The extraction of one kilo of powdered roots yielded 27 g of crude DCM extract (2.7%, w/w) and 160 g of MeOH crude extract (16%, w/w). Moreover, a crude water extract (1.5 g, 30%, w/w) was obtained by extraction of 5 g of powdered roots at room temperature during 24 hours.

3.1.2. Biological and chemical screening of R. rosea root extracts

The obtained extracts were submitted to three chemical and biological assays in order to determine their activities. Two types of assays may be distinguished, on TLC plates or in solution on microtitre plates. The traditional medicinal uses as well as the availability at the LPP or through collaboration, directed the choice of the assays.

3.1.2.1. TLC assays

As R. rosea exhibits pharmacological effects that may be linked to antioxidant activity, the radical scavenging activity of the extracts was tested on TLC (Cuendet et al. , 1997), using the radical DPPH as detailed in Chapter 2.4.2.4.

73

3. RESULTS

In the central nervous system, the cholinergic system is involved in the memory functions. Its key neurotransmitter, acetylcholine, is catabolised by AChE. The inhibition of this enzyme slows down the turn-over of the neurotransmitter in the synapses (Scarpini et al. , 2003), and therefore increases cholinergic functions. The identification of AChE inhibitors in R. rosea would furnish information about the way this plant acts on cognitive functions. Thus, the AChE inhibitory activity of roots extracts were evaluated by the method described by Marston et al. (2002) on TLC. The TLC assays allow to test several extracts at the same time and to assess quickly the interest of the extracts with regard to the target. The activities are visually estimated, and compared with pure reference compounds. However, this technique has a major inconvenient, the results are purely qualitative and the activities can not be numerically assessed. It is worth emphasising that when extracts are tested on TLC, the observed inhibitory or activity spots depend upon two parameters; the intrinsic activity of the compounds, and their relative amount within the extracts. The spots obtained may thus either be due to minor compounds highly active or to one or more major compounds slightly active.

Table 3-1 summarises R. rosea extracts activities detected on TLC assays. The amounts of extract spotted were 100 g for the assay with DPPH and 10 g for the AChE test.

Table 3-1 Chemical and biological activities of R. rosea extracts

Type of extract DPPH AChE (100 µg) (10 µg)

DCM + ++

MeOH ++ +

Aqueous ++ + (-: inactive; +: moderate activity; ++: high activity)

Regarding the DPPH assay, the DCM extract exhibited weak radical scavenging activities. Only two small inhibition spots were observed on the TLC with low R f indicating the presence of rather polar active compounds. The MeOH and aqueous extracts presented more important radical scavenging activities. Since many antioxidant metabolites were previously described in R. rosea hydro-alcoholic extracts e.g. triandrin, caffeic acid (Furmanowa et al. , 1998), salidroside (Zhang et al. , 2007a) and

74

3. RESULTS

flavonoids (Petsalo et al. , 2006), the bio-guided isolation of active compounds from these extracts was not based on this activity. Another target was chosen as detailed in the Chapter 3.1.2.2. Concerning the AChE inhibition assay, the MeOH and aqueous extracts showed only weak inhibition spots. Hillouse et al. (2004) recently described the isolation of AChE inhibitor flavonoids (gossypetin- 7-O-α-rhamnopyranoside and rhodioflavonoside) from R. rosea roots. On the other hand, the DCM extract exhibited remarkable inhibitions spots as shown in Figure 3-1. This activity had not been reported previously. Therefore, the DCM extract was selected for an extensive phytochemical investigation, presented in Chapter 3.1.3.

Rf

1

0 DCM ext. Galanth.

Figure 3-1 Bioautographic detection of AChE inhibitory activities in the DCM extract DCM ext: DCM extract of R. rosea roots (10 g): Galanth.: galanthamine (1 g); solvent system: hexane: EtOAc (1:1); support: Merck Silicagel 60 F254 aluminium sheets; detection: 1-naphtyl acetate and Fast Blue B salt after incubation with acetylcholinesterase (Marston et al. , 2002)

3.1.2.2. Monoamine oxidase inhibition assay in solution

As described in Chapter 2.3.3, MAOs are key enzymes in the metabolism of neurotransmitter such as serotonin, dopamine or noradrenaline. These bioamines are involved in the modulation of mood and emotions as well as the control of motor, perceptual and cognitive functions. The increase of the neurotransmitter levels induced by MAOs inhibition will stimulate this monoaminesynergic signalling. MAO A inhibitors have proven to be effective in the pharmacological treatment of depression (Priest et al. , 1995) while MAO B is implicated in age-related neurodegenerative diseases i.e. Parkinson’s and

75

3. RESULTS

Alzheimer’s diseases (Magyar and Szende, 2004). R. rosea root extracts were tested on this target as its pharmacological effects seemed to be the results of MAOs inhibition. MAOs A and B inhibitory activity was tested on a 96-well microtitre plate assay (Novaroli et al. , 2005) in collaboration with the Pharmacochemistry Laboratory, School of Pharmaceutical Sciences, University of Geneva. The three extracts of R. rosea roots were tested at a final concentration of 100 g/ml in dimethyl sulfoxide (DMSO). The DCM extract was found to be inactive while the methanol and water extracts exhibited prominent inhibitory activity against both isozymes. The separation and identification of the active secondary metabolites from these two extracts led to the publication “Monoamine oxidase inhibition by Rhodiola rosea L. roots”, presented in Chapter 3.1.4.

3.1.2.3. General methodology for the isolation of active compounds

Since extracts of R. rosea exhibited MAO and AChE inhibitory activities and radical scavenging property, their separation was undertaken in the aim of identifying the active compounds. The fractionation, guided by the activity, was performed by using various chromatographic methods i.e . liquid-liquid chromatography, column chromatography and preparative pressure liquid chromatography. The separation method was chosen according to the fraction or extract complexity and the size of the sample that could be fractionated. Centrifugal partition chromatography (CPC), medium pressure liquid chromatography (MPLC), low pressure liquid chromatography (LPLC) and open column methods were generally chosen for the fractionation of crude extracts or complex fractions as long as their amounts were greater than 500 mg. The purification of small quantity of fraction was preferentially performed on semi-preparative chromatography or on gel filtration (Sephadex LH-20). High performance liquid chromatography (HPLC) was used to “pilot” the preparative isolation of the secondary metabolites (optimisation of the experimental conditions, checking of the different fractions throughout the separation) and to control the final purity of the isolated compounds. Once the compounds were purified, their structures were determined by accumulating data from various analytical methods such as UV/VIS spectroscopy, HR-MS spectrometry and 1D and 2D NMR spectroscopy. Their activity was tested on both TLC test and on microtitre plate assay to measure their

IC 50 . The same methodology was used for the isolation of active compounds from S. dasyphyllum active extracts.

76

3. RESULTS

3.1.3. Bio-guided isolation of AChE inhibitors from R. rosea DCM extract

3.1.3.1. HPLC-UV-DAD analysis of the DCM extract of R. rosea roots

In order to obtain preliminary information on the main UV-visible constituents of the DCM extract of R. rosea roots, HPLC-UV-DAD analysis was performed.

Figure 3-2 HPLC-UV-DAD analysis of the DCM extract of R. rosea roots. Chromatographic conditions: Symmetry C 18 column (3 x 150 mm i.d., 5 m); gradient of MeOH / H 20 (2/98 -> 100/0 in 30 min), flow rate 1 ml/min; detection at 210 nm, UV spectra (DAD) were recorded between 200 and 500 nm.

The HPLC-UV chromatogram detected at 210 nm, as shown in Figure 3-2, indicated a rather simple profile with two major UV-visible compounds. The UV spectrum of the compound DCM1 exhibited maximum absorbance at 222 and 276 nm. The more lipophilic compound DCM2 showed maximum absorbance at 208 and 252 nm. These UV spectrum corresponded respectively to phenolic compounds and phenylpropanoid derivatives (Van Sumere, 1989). In order to identify the main constituents and to find out the active metabolites, the fractionation of the DCM extract of R. rosea roots was undertaken.

77

3. RESULTS

3.1.3.2. Fractionation of the DCM extract of R. rosea roots and isolation of compounds DCM1-DCM4

The DCM extract (20g) was first fractionated by liquid-liquid extraction (LLE) between MeOH and hexane. The hexane fraction contained the most lipophilic compounds. The extraction yielded 6.6 g of MeOH fraction and 12.1 g of hexane fraction. Only the MeOH fraction exhibited AChE inhibitory activity. Its separation on an open silica column with hexane-EtOAc-MeOH step gradient yielded 16 fractions. Compound DCM3 (207 mg) was obtained pure in fraction 2. The AChE activity was detected in fractions 4 (808 mg) and 8 (198 mg). Mixture 4 was purified by CPC with a biphasic solvent system heptane:MeCN:MeOH (6:3:1, v:v:v). The lower phase was used as mobile phase, giving 8 fractions (A-1 - A-8). Fractions were pooled together according to their similarity on thin-layer chromatography. This separation led to the isolation of DCM2 (126 mg, fraction A-1) and DCM4 (11 mg, fraction A-6). Fraction 8 was suspended in MeOH and centrifuged. The supernatant exhibited an activity. The MeOH-soluble compounds of fraction 8 were purified by semi-preparative chromatography on a

Symmetry Preparative C 18 column (19 x 150mm i.d., 7 m) using the solvent system MeOH:H2O (30:70) in isocratic mode. Tyrosol acetate ( DCM1, 1 mg) was present in fraction B-4. However, the active compound(s) of this fraction could not be identified as it was present in too small quantity.

The fractionation of the DCM extract is summarised in Scheme 3-1.

78

3. RESULTS

Rhodiola rosea L. roots DCM extract, 20 g (AChE +)

LLE (Hexane-MeOH)

MeOH fraction Hexane fraction 6.6 g (fatty acids)

Open column on Silicagel (4 x 56 cm i.d.; Hexane-EtOAc-MeOH step gradient 95:5:0 ->0:5:95)

16 fractions

Fraction 2 Fraction 4 Fraction 8 (AChE -) (AChE +) (AChE +) 207 mg 808 mg 198 mg

CPC (heptane-MeCN-MeOH (6:3:1) Centrifugation mobile ph.=lower ph.; rotation rate 1000 rpm; UV detection 254 nm) Supernatant (AChE +) 8 fractions 123 mg

Semi-prep HPLC (Symmetry C1 8 column (19x150mm, 7 µm; MeOH-H2O 30-100; flow rate 10ml/min; UV detection 210 nm) Fraction A-1 Fraction A-6 126 mg 11 mg Fraction B-4 1 mg

HPLC-UV-MS HPLC-UV-MS HPLC-UV-MS HPLC-UV-MS 1 H, 13 C RMN 1 H, 13 C RMN 1H, 1 3 C RMN 1H, 13 C RMN

βββ-sitosterol Cinnamyl alcohol Linoleic acid Tyrosol acetat (DCM3 ) (DCM2 ) (DCM4 ) (DCM1) AChE - AChE + AChE + AChE -

Scheme 3-1 Fractionation scheme of the DCM extract of R. rosea roots (+ : active; -: inactive)

The comparison of the spectral data (1H and 13 C NMR and MS) of the purified compounds (Figure

3-3) with the literature allowed the structure elucidation. Their structure was confirmed as tyrosol acetate ( DCM1 ) (Grasso et al. , 2007), cinnamyl alcohol ( DCM2 ) (Wiedenfeld et al. , 2007), β- sitosterol ( DCM3 ) (Goad, 1991) and linoleic acid ( DCM4 ) (Batchelor et al. , 1974).

79

3. RESULTS

O HO O HO (E)

tyrosol acetat cinnamyl alcohol (DCM1) (DCM2)

O HO OH (−)−β-sitosterol linoleic acid (DCM3) (DCM4)

Figure 3-3 Compounds isolated from R. rosea DCM extract

3.1.3.3. AChE inhibitory activity of the compounds DCM2 and DCM4

The AChE inhibitory activity of the four isolated compounds was tested on TLC assay (10 g spotted) with the Fast Blue B method. At this level, only cinnamyl alcohol and linoleic acid exhibited inhibitory activity. Cinnamyl alcohol was found to be active in the bioautographic test with a minimal inhibition amount of 5 g while the reference standard galanthamine, an alkaloid largely used in the treatment of Alzheimer’s disease’s symptoms, gave inhibition down to 0.01 g. This is the first time that cinnamyl alcohol is described as an AChE inhibitor. In order to determine the IC 50 value, the metabolite was tested in solution using Ellman’s method (1961). However, cinnamyl alcohol did not exhibit AChE inhibitory activity at a concentration of 10 -5 M. It was thus considerate as not sufficiently active to be of interest as hit compounds for the development of a novel AChE inhibitor. Kissling et al. (2005) already described the AChE inhibitory activity of linoleic acid on the TLC test with a minimal inhibition amount of 1 g. However, as cinnamyl alcohol, linoleic acid did not exhibit inhibitory activity at concentration of 10 -5 M in the Ellman’s solution assay.

80

3. RESULTS

3.1.4. Monoamine oxidase inhibition by Rhodiola rosea L. roots

81

Journal of Ethnopharmacology 122 (2009) 397–401

Contents lists available at ScienceDirect

Journal of Ethnopharmacology

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

Ethnopharmacological communication Monoamine oxidase inhibition by Rhodiola rosea L. roots

Daphne van Diermen a, Andrew Marston a, Juan Bravo b, Marianne Reist b, Pierre-Alain Carrupt b, Kurt Hostettmann a,∗

a Laboratory of Pharmacognosy and Phytochemistry, School of Pharmaceutical Sciences, University of Geneva, University of Lausanne, Quai Ernest-Ansermet 30, CH-1211 Geneva 4, Switzerland b LCT-Pharmacochemistry, School of Pharmaceutical Sciences, University of Geneva, University of Lausanne, Quai Ernest-Ansermet 30, CH-1211 Geneva 4, Switzerland

article info abstract

Article history: Aim of the study: Rhodiola rosea L. (Crassulaceae) is traditionally used in Eastern Europe and Asia to stim- Received 22 July 2008 ulate the nervous system, enhance physical and mental performance, treat fatigue, psychological stress Received in revised form 17 December 2008 and depression. In order to investigate the influence of Rhodiola rosea L. roots on mood disorders, three Accepted 3 January 2009 extracts were tested against monoamine oxidases (MAOs A and B) in a microtitre plate bioassay. Available online 9 January 2009 Materials and methods: Methanol and water extracts gave the highest inhibitory activity against MAOs. Twelve compounds were then isolated by bioassay-guided fractionation using chromatographic methods. Keywords: The structures were determined by 1H, 13 C NMR and HR-MS. Rhodiola rosea L. Crassulaceae Results: The methanol and water extracts exhibited respectively inhibitions of 92.5% and 84.3% on MAO ␮ and 81.8% and 88.9% on MAO B, at a concentration of 100 g/ml. The most active compound (rosiridin) −5 Depression presented an inhibition over 80% on MAO B at a concentration of 10 M (pIC50 = 5.38 ± 0.05). Senile dementia Conclusions: The present investigation demonstrates that Rhodiola rosea L. roots have potent anti- depressant activity by inhibiting MAO A and may also find application in the control of senile dementia by their inhibition of MAO B. © 2009 Elsevier Ireland Ltd. All rights reserved.

1. Introduction nervous system (CNS) activity (Stancheva and Mosharrof, 1987; Brown et al., 2002). These findings may explain the influence of Rhodiola rosea L. (Crassulaceae), the most investigated species Rhodiola rosea on mental disorders such as depression and senile of the genus Rhodiola, grows at elevated altitudes in the Arctic dementia. and in mountainous regions throughout Europe and Asia, where Although many studies have provided evidence that administra- it is also known as “golden root” or “arctic root” (Saratikov and tion of Rhodiola rosea extract elicits antidepressant activity (Kurkin Krasnov, 1987). The perennial plant reaches a height of 30–70 cm et al., 2006; Darbinyan et al., 2007; Perfumi and Mattioli, 2007; and produces yellow flowers; its thick rhizome has a rose-like fra- Panossian et al., 2008), the mechanism of action of Rhodiola rosea grance when cut. The roots have been used for centuries in the in the treatment of nervous system disorders still remains unclear. traditional medicine of Asia, Scandinavia and Eastern Europe to Monoamine oxidases regulate the metabolic degradation of cat- stimulate the nervous system, enhance physical and mental per- echolamines and serotonin by oxidative deamination in the central formance, improve resistance to high altitude sickness and to treat nervous system or peripheral tissues. Monoamine oxidase (MAO) fatigue, psychological stress and depression (Saratikov and Krasnov, A plays a pivotal role in the degradation of biogenic amines such as 1987; Wagner et al., 1994; Panossian et al., 1999; Spasov et al., 2000; epinephrine, , and serotonin (Shih and Thompson, Panossian, 2003; Shevtsov et al., 2003; Panossian and Wagner, 1999). MAO A inhibitors have proven to be effective in the pharma- 2005). Rhodiola rosea contains flavonoids, monoterpenes, triter- cological treatment of depression (Priest et al., 1995). MAO B is the penes, phenolic acids, phenylethanol derivatives (salidroside and main enzyme implicated in the metabolism of dopamine (Novaroli tyrosol) and phenylpropanoid glycosides such as rosin, rosavin and et al., 2005). Several studies have shown that MAO B is implicated rosarin specific to this plant (Ganzera et al., 2001) in aging-related neurodegenerative diseases such as Parkinson’s Investigation by Russian researchers has revealed that Rhodiola disease (Castagnoli et al., 2003; Magyar and Szende, 2004) and in rosea root extracts produce favorable changes in a variety of phys- the formation of plaque-associated astrocytes present in brains of iological functions, including neurotransmitter levels and central patients suffering from Alzheimer’s disease (Saura et al., 1994). The present report aims at explaining the influence of Rhodiola rosea root extracts on mood disorders by studying its effect on the ∗ Corresponding author. Tel.: +41 22 379 34 01; fax: +41 22 379 33 99. regulation of neurotransmitters by monoamine oxidase. To this end, E-mail address: [email protected] (K. Hostettmann). three extracts, dichloromethane, methanol and water, were tested

0378-8741/$ – see front matter © 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.jep.2009.01.007 398 D. van Diermen et al. / Journal of Ethnopharmacology 122 (2009) 397–401

against both MAOs A and B in a microtitre plate assay. Since the Table 1 ␮ latter two extracts presented MAO inhibitory activity a bio-guided MAO A and B inhibitory activities of Rhodiola rosea L. root extracts (at 100 g/ml) and the bio-guided isolated compounds (at 10−5 M). fractionation of the extracts was undertaken in order to identify the active compounds. Twelve compounds were isolated and identi- Sample Inhibition (%)b fied by means of spectroscopic and chemical methods, including 1D MAO A MAO B and 2D NMR experiments and HR-MS analysis. The MAO inhibitory DCM extract 50.5 ± 0.1 66.9 ± 0.3 activity is reported here, together with the main components of MeOH extract 92.5 ± 0.1 81.8 ± 0.3 each active extract that account for the demonstrated activity. Water extract 84.3 ± 0.8 88.9 ± 0.3

± ± 2. Material and methods Fraction G-2 96.8 0.2 81.4 0.6 Fraction G-8 21.6 ± 0.2 88.5 ± 0.4 Salidroside (1) – 35.8 ± 2.5 2.1. General EGCG dimer (2) 43.1 ± 0.4 37.7 ± 0.5 Rhodioloside B and C mixture (3, 4)– 61.9± 3.0 1 13 Rosarin (5)–– H and C NMR spectra were recorded in CD3OD at 500 and Cinnamyl alcohol (6)27.7± 0.6 43.2 ± 1.5 125 MHz, respectively, on a Varian Unity Inova NMR instrument. Rhodiocyanoside A (7)–27.7± 4.8 TMS was used as internal standard. HR-MS spectra were acquired Triandrin (8) – 40.8 ± 3.5 on a Micromass LCT Premier instrument. Thin layer chromatogra- Rosavin (9)–– phy (TLC) was carried out on silica gel 60 F254 Al sheets (Merck) Tyrosol (10) – 26.3 ± 0.7 using CHCl –MeOH–H O (65:35:5) as eluent. Low pressure liquid Rosin (11)–– 3 2 Rosiridin (12) 16.2 ± 2.3 83.8 ± 1.1 chromatography (LPLC) was performed on a Lobar RP-18 col- l-Deprenyla 36.0 ± 1.0 99.5 ± 0.2 umn (LiChroprep 40–63 ␮m, 310 mm × 25 mm i.d., Merck). Medium Clorgylinea 100.0 ± 0.2 80.2 ± 0.9 pressure liquid chromatography (MPLC) was carried out on a RP- a Reference compound. ␮ × 18 LiChroprep column (40–63 m; 460 mm 50 mm i.d., Merck). b Inhibition lower than 15% was considered as inactive. Centrifugal partition chromatography (CPC) was performed on a CCC-1000 instrument (Pharma-Tech Research Corp., Baltimore, MD, USA). Total volume of the three coils was 320 ml and the rotation DMSO were used in place of the inhibitor solution. Diluted MAO ␮ speed was 1000 rpm. The CPC solvent was pumped at a flow rate of (50 l) was then delivered to obtain a final protein concentration 3 ml/min by a 600A pump (Waters Assiociates, Inc., Milford, USA). of 0.015 mg/ml in the assay mixture. Incubation was carried out at ◦ ␮ Elution was monitored at 254 nm with a Knauer (Berlin, Germany) 37 C and the reaction was stopped after 20 min by addition of 75 l UV–vis. detector and a LKB (Bromma, Sweden) model 2210 inte- of NaOH (2N). Fluorescence emission at 400 nm was measured with grator. HPLC–UV/DAD was carried out on a HP1100 (Agilent) with a 96-well microplate fluorescent reader (FLx 800, Bio-Tek Instru- a Symmetry RP-18 column (5 ␮m; 150 mm × 3.9 mm i.d., Waters) ments, Inc., Winooski, VT, USA). Inhibition measure of the extracts and fractions were done in duplicate since only an approximate using a MeOH–H2O gradient (2:98–100:0) in 30 min. The detection was performed at 210, 254 and 360 nm. measure of the inhibition was necessary for the bio-guided frac- tionation. 2.2. Plant material Extracts and fractions were tested at a final concentration of 100 ␮g/ml while the purified compounds were at 10−5 M, in DMSO. The roots of Rhodiola rosea L. (Crassulaceae) as authenticated Data analysis was performed with Prism 4.0 (GraphPad Software, by Egidio Anchisi (Orsières, VS, Switzerland) were collected in Inc., CA, USA). The degree of inhibition IC50 was assessed by a Val d’Aoste, Italy, in November 2005. A voucher specimen (no. sigmoidal dose–response curve. The standard deviation was cal- 2005006) is deposited in the Laboratory of Pharmacognosy and culated for the sigmoidal regression. Phytochemistry, Section of Pharmaceutical Sciences, University of Geneva. The roots were washed, cut, freeze-dried and powdered. 2.4. Extraction and isolation

2.3. Monoamine oxidase inhibition assay Dried and powdered roots of Rhodiola rosea (1 kg) extracted sequentially with CH2Cl2 (3 × 24 h) and MeOH (3 × 24 h) at room Human MAOs A and B Supersomes TM, purchased from BD temperature yielded, after removing the solvent under vacuum, Gentest (Woburn, MA, USA), are mitochondrial membrane frac- 27 g of crude DCM extract (2.7%, w/w) and 160 g of MeOH crude tions of insect cells containing human recombinant MAOs A and extract (16%, w/w). A crude water extract (1.5 g, 30%, w/w) was B. MAO inhibition assays were carried out with a fluorescence- obtained by extraction of 5 g of roots at room temperature during based method (end-point reading) adapted from a standard BD 24 h. Gentest protocol. The substrate used for the assay was kynuramine, MeOH extract (5 g) was fractionated by CPC with which is non-fluorescent until it undergoes oxidative deamination CHCl3:MeOH:n-BuOH:H2O (7:6:3:4) as solvent system. The by MAO resulting in the fluorescent metabolite 4-hydroxyquinoline lower phase was first used as mobile phase, giving 14 fractions (Novaroli et al., 2005). Product formation was quantified by com- (G-1–14). Seven further fractions (G-15–21) were subsequently paring the fluorescence emission of the samples to that of known obtained by elution in the reversed-phase mode (upper phase amounts of authentic metabolite 4-hydroxyquinoline. as mobile phase). Fractions were pooled together according their Reactions were performed in black, flat bottom polystyrene similarity on thin-layer chromatography. This separation led to the 96-well microtitre plates with enhanced assay surface (Fluoro- isolation of 130 mg of salidroside (fraction G-10, 1) and 600 mg of Nunc/LumiNunc, MaxiSorpTM Surface, NUNCtM, Roskild, Den- epigallocatechin gallate dimer (fraction G-15, 2). mark) using a final volume of 200 ␮l. The wells containing 140 ␮l Fraction G-2 was separated by LPLC with a MeOH–H2Ostepgra- of potassium phosphate buffer (0.1 M, pH 7.4, made isotonic with dient yielding 10.0 mg of a mixture of rhodioloside B and C (3, 4), KCl), 8 ␮l of an aqueous stock solution of kynuramine (0.75 M to 1.4 mg of rosarin (5), 5 mg of salidroside (1), 0.7 mg of cinnamyl get a final concentration corresponding to its km value), and 2 ␮l alcohol (6). of the sample solution (final concentration of 1%, v/v), were prein- Fraction G-8 was chromatographed by LPLC with a MeOH–H2O cubated at pH 7.4, 37 ◦C for 10 min. As positive control, 2 ␮l of pure step gradient to afford 400 fractions. Fractions were pooled D. van Diermen et al. / Journal of Ethnopharmacology 122 (2009) 397–401 399

Fig. 1. Structures of the compounds isolated from the methanol and water extracts of the roots of Rhodiola rosea L. according to the HPLC/UV trace at 366 nm yielding 25 mg of rhodi- Structures of purified compounds were elucidated by direct ocyanoside A (7), 20 mg of salidroside (1), 3 mg of triandrin (8) and comparisons of their spectral data (1H NMR, 13 C NMR and HR-MS) 6 mg of rosavin (9). with those found in literature (LaLonde et al., 1976; Zapesochnaya Tannins in the water extract were removed on a polyamide car- and Kurkin, 1982; Yoshikawa et al., 1995; Fan et al., 2001; Tolonen tridge by solid phase extraction. The purified extract (350 mg) was et al., 2003; Lin and Chen, 2004; Kishida and Akita, 2005; Ma et separated by MPLC with MeOH–H2O step gradient (10:90–70:30 in al., 2006; Yousef et al., 2006; Takaya et al., 2007; Wiedenfeld et al., 5% steps) to afford four fractions. This purification led to the isola- 2007). tion of 3.0 mg of tyrosol (fraction 1, 10), 2 mg of rosin (fraction 2, 11), 2 mg of rosarin (fraction 3, 5) and 20 mg of fraction 4. Separation 3. Results and discussion of fraction 4 by CPC with CHCl3:MeOH:isopropanol:H2O (5:6:1:4; lower phase as mobile phase) yielded 3 mg of rosavin (9) and 8 mg Dichloromethane, methanol and water extracts of Rhodiola rosea of rosiridin (12). L. were tested against two enzymes: MAOs A and B, which are 400 D. van Diermen et al. / Journal of Ethnopharmacology 122 (2009) 397–401 targets in the search for new neuroprotective agents. Prominent pounds or possibly by synergism, the most active compounds being inhibitory activity against both MAOs was found in the methanol rosiridin (12), rhodioloside B and C isomers (3, 4), cinnamyl alco- and water extracts (Table 1). When the concentration of the extracts hol (6), triandrin (8) and EGCG dimer (2). The activity may also be was 100 ␮g/ml, the activities relative to the positive controls (l- reinforced by the presence of known flavonoids such as quercetrin deprenyl and clorgyline) were for MAO A: 92.5% and 84.3% and for which have already been described as moderate MAO inhibitors MAO B: 81.8% and 88.9%, respectively. (Lee et al., 2001; Petsalo et al., 2006). Bio-guided fractionation of both extracts was undertaken in The phenylpropanoid glycosides rosin (11), rosarin (5), and order to identify the active compounds. rosavin (9) have been described in the literature as the molecules First fractionation of the MeOH extract by CPC yielded eleven responsible for the antidepressant activity (Zapesochnaya et al., fractions with two active fractions and one active pure compound 1995; Kurkin et al., 2006). Furthermore, Russian researchers have (epigallocatechin gallate (EGCG) dimer, 2). The active fractions chosen rosavin as a marker compound. However, in this study, these showed an inhibitory activity of over 80% against MAOs A and/or phenylpropanoid glycosides presented no MAO inhibitory activity B at a concentration of 100 ␮g/ml. The major metabolites of the in the in vitro test. active fractions were isolated by various preparative chromatog- This is the first report providing direct evidence that the roots raphy methods. Six compounds were purified and identified as of Rhodiola rosea have an influence on the levels of serotonin and salidroside (1), cinnamyl alcohol (6), rosarin (5), rhodiocyanoside norepinephrine in the nerve terminals by inhibiting MAOs A and B. A(7), triandrin (8) and rosavin (9). A mixture of rhodioloside B and In conclusion, the present investigation demonstrates that C(3, 4) isomers was also obtained. Bio-guided fractionation of the extracts of Rhodiola rosea L. roots have potent anti-depressant activ- water extract by different chromatographic methods afforded five ity by inhibiting MAO A. At the same time, these extracts can compounds including three phenylpropanoid glycosides: rosin (11), influence the progress of problems associated with Parkinsonism or rosarin (5) and rosavin (9), the phenylethanol derivative tyrosol (10) Alzheimer’s disease by inhibiting MAO B. These findings reinforce and the monoterpene glycoside rosiridin (12). the claims made in ethnomedicine that Rhodiola rosea L. can be used The identities of the isolated compounds were established by as a remedy for depression and other nervous system disorders. HR-MS and comparison of their 1H and 13 C NMR spectra with those reported in the literature. Their structures are shown in Fig. 1. All the metabolites isolated from the active fractions were tested Acknowledgement against MAOs A and B (Table 1). Rosiridin (12) and the mixture of rhodioloside B and C isomers The authors thank the Swiss National Science Foundation for (3, 4) exhibited the highest inhibitory activity (over 60% at concen- financial support of this work (grant no. 200020-107775 to Prof. K. tration of 10−5 M) against MAO B. Hostettmann). The pIC50 (−log IC50) value of rosiridin for MAO B inhibition is ± estimated to 5.38 0.05. The pIC50 value for the positive controls References was 7.23 ± 0.04 for MAO B inhibition (l-deprenyl). The pIC50 values of rhodioloside B and C were not measured as these compounds Brown, R.P., Gerbarg, P.L., Ramazanov, Z., 2002. Rhodiola rosea. A phytomedicinal were obtained as a mixture. overview. HerbalGram 56, 40–52. Castagnoli Jr., N., Petzer, J.P., Steyn, S., Castagnoli, K., Chen, J.F., Schwarzschild, M.A., As the MAO B inhibitory activity of EGCG has already been Van der Schyf, C.J., 2003. Monoamine oxidase B inhibition and neuroprotection: described by Mazzio et al. (1998), its pIC50 for MAO B inhibition studies on selective adenosine A2A receptor antagonists. Neurology 61, 62–68. was measured although it only presented a moderate inhibition Darbinyan, V., Aslanyan, G., Amroyan, E., Gabrielyan, E., Malmstrom, C., Panossian, ± A., 2007. of Rhodiola rosea L. extract SHR-5 in the treatment of mild percentage. EGCG dimer gave a pIC50 of 4.82 0.04. In the sigmoidal to moderate depression. Nordic Journal of Psychiatry 61, 343–348. dose–response curve of EGCG dimer for MAO B inhibition, the Fan, W., Tezuka, Y., Ni, K.M., Kadota, S., 2001. Prolyl endopeptidase inhibitors from the Hill coefficient in the concentration–response equation is greater underground part of Rhodiola sachalinensis. Chemical & Pharmaceutical Bulletin than unity. The Hill coefficient is related to the stoichiometry of 49, 396–401. Ganzera, M., Yayla, Y., Khan, I.A., 2001. Analysis of the marker compounds of Rhodiola inhibitor–enzyme interactions. It also represents the steepness of rosea L. (golden root) by reversed phase high performance liquid chromatogra- the concentration–response relationship (Robert, 2005). A high phy. Chemical & Pharmaceutical Bulletin 49, 465–467. Hill coefficient can be diagnostic of non-ideal behavior. Notably, Kishida, M., Akita, H., 2005. Synthesis of rosavin and its analogues based on a Mizoroki-Heck type reaction. Tetrahedron Asymmetry 16, 2625–2630. compounds that cause an abrupt inhibition above a critical con- Kurkin, V., Dubishchev, A., Ezhkov, V., Titova, I., Avdeeva, E., 2006. Antidepressant centration, hence producing concentration–response relationships activity of some phytopharmaceuticals and phenylpropanoids. Pharmaceutical with the Hill coefficient much greater than unity, usually reflect Chemistry Journal 40, 614–619. LaLonde, R.T., Wong, C., Tsai, A.I.M., 1976. Polygluosidic metabolites of Oleaceae. The a nonspecific mechanism of inhibition. This can result for com- chain sequence of oleoside aglucon, tyrosol, and glucose units in three metabo- pounds that act as general protein denaturants. Such compounds lites from Fraxinus americana. Journal of the American Chemical Society 98, do not effect inhibition by a specific interaction with a defined 3007–3013. Lee, M.H., Lin, R.D., Shen, L.Y., Yang, L.L., Yen, K.Y., Hou, W.C., 2001. Monoamine binding pocket on the enzyme molecule and are therefore gener- oxidase B and free radical scavenging activities of natural flavonoids in ally not tractable as drug leads. High Hill coefficients can also result Melastoma candidum D. Don. Journal of Agricultural Food and Chemistry 49, from very tight binding of inhibitors to enzyme targets and from 5551–5555. Lin, L.-C., Chen, K.-T., 2004. New phenylpropanoid glycoside from Boschniakia rossica. irreversible inhibition of enzymes. Thus, the inhibition activity of Chinese Pharmaceutical Journal (Taipei, Taiwan) 56, 77–85. EGCG dimer against MAOs A and B is rather attributed to its denat- Ma, G., Li, W., Dou, D., Chang, X., Bai, H., Satou, T., Li, J., Sun, D., Kang, T., Nikaido, T., urant effect on proteins than to a specific mechanism of inhibition. Koike, K., 2006. Rhodiolosides A–E, monoterpene glycosides from Rhodiola rosea. Since the activity is not specific and thus not of great interest, the Chemical & Pharmaceutical Bulletin 54, 1229–1233. Magyar, K., Szende, B., 2004. (−)-Deprenyl, a selective MAO-B inhibitor, with apop- stereochemistry of the EGCG dimer has not been clarified. totic and anti-apoptotic properties. Neurotoxicology 25, 233–242. The MeOH and water extracts gave over 80% inhibition of MAOs Mazzio, E.A., Harris, N., Soliman, K.F., 1998. Food constituents attenuate monoamine A and B at a concentration of 100 ␮g/ml. The inhibitory activity of oxidase activity and peroxide levels in C6 astrocyte cells. Planta Medica 64, 603–606. the water extract may be explained by the presence of rosiridin, Novaroli, L., Reist, M., Favre, E., Carotti, A., Catto, M., Carrupt, P.A., 2005. Human −5 which gave an inhibition of 83.8% against MAO B at 10 M (pIC50 recombinant monoamine oxidase B as reliable and efficient enzyme source for 5.38 ± 0.05). After fractionation of the methanol extract, three frac- inhibitor screening. Bioorganic & Medicinal Chemistry 13, 6212–6217. Panossian, A., Nikoyan, N., Ohanyan, N., Hovhannisyan, A., Abrahamyan, H., tions showed elevated inhibitory activity. The activity of these three Gabrielyan, E., Wikman, G., 2008. Comparative study of Rhodiola preparations fractions may be explained by the additive effect of different com- on behavioral despair of rats. Phytomedicine 15, 84–91. D. van Diermen et al. / Journal of Ethnopharmacology 122 (2009) 397–401 401

Panossian, A., Wagner, H., 2005. Stimulating effect of adaptogens: an overview with Spasov, A.A., Wikman, G.K., Mandrikov, V.B., Mironova, I.A., Neumoin, V.V., 2000. particular reference to their efficacy following single dose administration. Phy- A double-blind, placebo-controlled pilot study of the stimulating and adapto- totherapy Research 19, 819–838. genic effect of Rhodiola rosea SHR-5 extract on the fatigue of students caused Panossian, A., Wikman, G., Wagner, H., 1999. Plant adaptogens III. Earlier and more by stress during an examination period with a repeated low-dose regimen. Phy- recent aspects and concepts on their mode of action. Phytomedicine 6, 287–300. tomedicine 7, 85–89. Panossian, A.G., 2003. Adaptogens: tonic herbs for fatigue and stress. Alternative & Stancheva, S.L., Mosharrof, A., 1987. Effect of the extract of Rhodiola rosea L. on the Complementary Therapies 9, 327–331. content of the brain biogenic monoamines. Dokladi Na Bolgarskata Akademiya Perfumi, M., Mattioli, L., 2007. Adaptogenic and central nervous system effects of Na Naukite 40, 85–87. single doses of 3% rosavin and 1% salidroside Rhodiola rosea L. extract in mice. Takaya, Y., Furukawa, T., Miura, S., Akutagawa, T., Hotta, Y., Ishikawa, N., Niwa, M., Phytotherapy Research 21, 37–43. 2007. Antioxidant constituents in distillation residue of Awamori spirits. Journal Petsalo, A., Jalonen, J., Tolonen, A., 2006. Identification of flavonoids of Rhodiola rosea of Agricultural Food and Chemistry 55, 75–79. by liquid chromatography-tandem mass spectrometry. Journal of Chromatogra- Tolonen, A., Pakonen, M., Hohtola, A., Jalonen, J., 2003. Phenylpropanoid glycosides phy A 1112, 224–231. from Rhodiola rosea. Chemical & Pharmaceutical Bulletin 51, 467–470. Priest, R.G., Gimbrett, R., Roberts, M., Steinert, J., 1995. Reversible and selective Wagner, H., Norr, H., Winterhoff, H., 1994. Plant adaptogens. Phytomedicine 1, 63–76. inhibitors of monoamine oxidase A in mental and other disorders. Acta Psy- Wiedenfeld, H., Dumaa, M., Malinowski, M., Furmanowa, M., Narantuya, S., 2007. chiatrica Scandinavica 386, 40–43, Supplementum. Phytochemical and analytical studies of extracts from Rhodiola rosea and Rhodi- Robert, A.C., 2005. Evaluation of Enzyme Inhibitors in Drug Discovery: A Guide for ola quadrifida. Pharmazie 62, 308–311. Medicinal Chemists and Pharmacologists. Wiley, New York, p. 296. Yoshikawa, M., Shimada, H., Shimoda, H., Matsuda, H., Yamahara, J., Murakami, N., Saratikov, S.A., Krasnov, E.A., 1987. Rhodiola rosea is a Valuable Medicinal Plant 1995. Rhodiocyanosides A and B, new antiallergic cyanoglycosides from Chi- (Golden root). Monograph Tomsk State University Press, Tomsk, p. 252. nese natural medicine “si lie hong jing tian”, the underground part of Rhodiola Saura, J., Luque, J.M., Cesura, A.M., Da Prada, M., Chan-Palay, V., Huber, G., Loffler, quadrifida (Pall ) Fisch. et Mey. Chemical and Pharmaceutical Bulletin (Tokyo) J., Richards, J.G., 1994. Increased monoamine oxidase B activity in plaque- 43, 1245–1247. associated astrocytes of Alzheimer brains revealed by quantitative enzyme Yousef, G.G., Grace, M.H., Cheng, D.M., Belolipov, I.V., Raskin, I., Lila, M.A., 2006. radioautography. Neuroscience 62, 15–30. Comparative phytochemical characterization of three Rhodiola species. Phyto- Shevtsov, V.A., Zholus, B.I., Shervarly, V.I., Vol’skij, V.B., Korovin, Y.P., Khristich, M.P., chemistry 67, 2380–2391. Roslyakova, N.A., Wikman, G., 2003. A randomized trial of two different doses of Zapesochnaya, G.G., Kurkin, V.A., 1982. Cinnamic glycosides of Rhodiola rosea rhi- a SHR-5 Rhodiola rosea extract versus placebo and control of capacity for mental zomes. Khimiya Prirodnykh Soedinenii, 723–727. work. Phytomedicine 10, 95–105. Zapesochnaya, G.G., Kurkin, V.A., Boiko, V.P., Kolkhir, V.K., 1995. Phenylpropanoids as Shih, J.C., Thompson, R.F., 1999. Monoamine oxidase in neuropsychiatry and behav- promising biologically active substances from medicinal plants. Pharmaceutical ior. American Journal of Human Genetics 65, 593–598. Chemistry Journal 29, 277–280.

3. RESULTS

3.2. HPLC-UV QUANTITATIVE ANALYSIS OF SALIDROSIDE AND ROSAVINS IN RHODIOLA ROSEA L. PLANTS

According to the 1989 Soviet Pharmacopeia, R. rosea root extracts should be standardised in total rosavins (rosavin, rosin and rosarin) and salidroside, respectively 3% and 1%. These metabolites were thus chosen as the main makers for the following analyses. The evolution of the marker amounts were monitored in four wild plants collected in the Swiss Alps during a growing season in order to define the period when the metabolites are in the highest concentrations in the rhizomes. Furthermore, a comparison study between the marker profiles of male and female plants was performed in order to select the most appropriate gender for further cultivation. In total, twenty R. rosea root samples were collected for the marker dynamics and fifteen for the comparison between male and female plants. The samples were extracted according to an improved process and then analysed by an optimised HPLC-UV method. The analysis of an extract is displayed in Figure 3-4. Salidroside was detected at 210 nm and rosavins at 254 nm.

210 nm mAU salidroside 160 140 120 100 80 60 40 7.5 10 12.5 15 17.5 20 22.5 25 27.5 min

rosavin mAU 254 nm

40

30

20 rosarin

10 rosin

0 7.5 10 12.5 15 17.5 20 22.5 25 27.5 min

Figure 3-4 HPLC-UV analysis of an hydro-alcoholic extract of R. rosea Chromatograpic conditions: Atlantis dC 18 column (3 x 150 mm i. d., 3 µm) equipped with a dC 18 precolumn; gradient of MeOH:MeCN + FA 0.1% (solvent A) / H 2O + FA 0.01% (solvent B) (0-7 min, 99-80% B; 7-26 min, 80-63% B; 26-30 min, 63-0% B; 35-45min, 0% B), flow rate 500 µl/min; analysis temperature 37°C; detection at 210 nm and 254 nm.

89

3. RESULTS

The HPLC technique coupled with UV detection is an easy-in-use, reproducible method, with few maintenance charges. In this work, the UV detection was preferred to MS as the developed method was intend to be implement in another laboratory. Furthermore, this method lends itself well to the analysis of rosarins and salidroside as they are UV-visible and are present in sufficient quantity in the extracts to be detected by UV. The analyses of the R. rosea samples and the comparison among them led to the publication “Chemical profile dynamics in a wild population of Rhodiola rosea L. from the Swiss Alps”, presented in Chapter 3.2.1.

90

3. RESULTS

3.2.1. Chemical profile dynamics in a wild population of Rhodiola rosea L. plants from the Swiss Alps

Daphne van Diermen 1, Karine Ndjoko Ioset 1, Andrew Marston 1, Pia Malnoe 2, Kurt Hostettmann 1*

1 Laboratory of Pharmacognosy and Phytochemistry, School of Pharmaceutical Sciences, University of Geneva, University of Lausanne, Quai Ernest-Ansermet 30, CH-1211 Geneva 4, Switzerland

2 Station de Recherche Agroscope Changins-Wädenswil ACW, Centre de Recherche Conthey, 1964 Conthey, Switzerland

Submitted to Journal of Agricultural Sciences

91

3. RESULTS

3.2.1.1. Abstract

Rhodiola rosea L. (Crassulaceae) has for a long time been used in traditional medicine. The rhizome has been associated with many beneficial biological activities; effects on memory, learning, immune response, and stress, cancer therapy and antidepressant. Phytochemical investigations of R. rosea have revealed the presence of phenylpropane glycosides specific to the plant, phenylethanoids, cyanosides, terpenes and flavonoids. Precise identification of the compounds responsible for the numerous health benefits remains to be confirmed. Based on the revised 1989 Soviet Pharmacopeia, the extracts of R. rosea are now standardized in both rosavins (rosavin, rosin and rosarin) and salidroside. As interest in R. rosea has considerably increased in the western world, cultivation is necessary to obtain sufficient raw material for the production of high quality phytopharmaceuticals. Thus, harvest and destruction of wild populations may be avoided. The present study aimed to establish the dynamic chemical profiling of the main markers: rosavin, rosin, rosarin and salidroside in the rhizomes of wild plants of R. rosea . The results showed that seasonal variations have no impact on the proportions of the markers in each individual plant. Moreover, the content comparison of the markers in female and male plants showed that there is no significant difference between the two gender. These results provide valuable information for the cultivation and harvesting of R. rosea .

Key Words -Rhodiola rosea L., Profile dynamics, Rosavins, Salidroside, Seasonal variation, Sex differentiation.

92

3. RESULTS

3.2.1.2. Introduction

Over the last few years, interest in Rhodiola rosea L. (Crassulaceae), also known as “Golden root” and “Arctic root”, has considerably increased in the western world due to its biological properties, particularly for the adaptogenic activity. This perennial herbaceous plant has been traditionally used in China and Russia to maintain human health. However, the plant has remained largely unknown in the West, because the bulk of the research has been published in Slavic and Scandinavian languages. In recent years, many biological activities related to R. rosea have been reported; adaptogenic, anti-fatigue, antioxidant, anti-depressive and enhancement in learning and memory processes (Darbinyan et al. , 2007; Darbinyan et al. , 2000; Khanum et al. , 2005; Panossian and Wagner, 2005; Perfumi and Mattioli, 2007). Phytochemical investigations have revealed the presence of specific phenylpropane glycosides (rosavin, rosin, and rosarin), phenylethanoid (salidroside), cyanosides (rhodiocyanosides), terpenes (rosiridol, rosiridin, rhodiolosides) and flavonoids (rhodioline, rhodionine, rhodionine) (Ali et al. , 2008; Ma et al. , 2006; Petsalo et al. , 2006; Tolonen et al. , 2003b; Yoshikawa et al. , 1995; Yousef et al. , 2006; Zapesochnaya and Kurkin, 1982). However, the compounds responsible for the numerous health benefits remain to be identified. According to the revised 1989 Soviet Pharmacopeia, the extracts of R. rosea should be standardized in both rosavins (rosavin, rosin and rosarin) and salidroside. Even though rosavins are now accepted as markers for the genetic identification of R. rosea , they are not necessarily the pharmacologically active ingredients responsible for the efficacy observed in clinical studies. R. rosea extracts used in most clinical studies are standardized to a minimum of 3% and 0.8-1% of rosavins and salidroside, respectively (Brown et al. , 2002). Based on the documented pharmacological effects and safe traditional usage of the plant, the commercial interest for Golden root-based products has quickly increased worldwide. Presently, one of the most important problems facing industry is meeting raw-material production requirements in order to satisfy market demand. R. rosea cultivation seems to be the only solution towards producing raw material in sufficient quantities and defined quality (Galambosi, 2006). Cultivation will equally avoid the depletion of the natural wild populations through uncontrolled harvesting. Several studies compared the chemical profiles of cultivated and naturally growing R. rosea plants (Galambosi et al. , 2007; Kovaleva et al. , 2003; Kucinskaite et al. , 2007). Similar amounts of salidroside were found in cultivated and wild plants while the results for the contents of rosavins were contradictory among the

93

3. RESULTS

studies. In vitro culture of R. rosea callus has also been studied; the proportion of phenylpropanoids were different between cultivated and wild plants: the rosin content was three to six time higher in the in vitro cultivated plants while rosavin was produced only in traces (Gyoergy et al. , 2004). The present study aimed to characterize the amount of the main markers: salidroside ( 1), rosarin ( 2), rosavin ( 3), and rosin ( 5) ( Figure 1 ) in wild plants for the optimization of R. rosea cultivation.

HO O HO HO O OH

OH

salidroside (1) MW = 300 HO O O HO O HO (E) O HO HO OH

rosarin (2) MW = 428 OH O HO O OH O HO (E) HO O OH

rosavin (3) MW = 428 O HO HO O OH O HO (E) HO O OH

cinnamyl-(6'-O-ß-xylopyranosyl) -O-ß-glucopyranoside (4) MW = 428

HO O HO (E) HO O OH

rosin (5) MW = 296

Figure 1 Structures of the analysed markers

94

3. RESULTS

The evolution of the marker contents was studied during a growing season. The analysis was undertaken on four wild plants from the same population collected in the Swiss Alps. The collection of 5 samples of the rhizome in its habitat was undertaken from May to September 2006. As the study of the chemical evolution was undertaken within the same plants, the chemical variability of individual organisms was minimized. Thus, comparison of the different chemical profile appears more significant. Moreover, the marker profiles of the wild male and female plants were compared in order to select the most appropriate plant for further cultivation. We report here the first study on the chemical profile evolution of R. rosea rhizomes carried out on the same plant throughout a growing season period.

3.2.1.3. Methods and materials

Chemicals

Salidroside ( 1), rosarin ( 2), rosavin ( 3), and rosin ( 5) were purchased from ChromaDex (Santa Ana, CA, USA). Methanol and acetonitrile for HPLC analysis (VWR, Leuwen, Belgium) were of HPLC grade. Formic acid (Sigma Aldrich, Steinheim, Germany) was of analytical grade.

Samples R. rosea rhizomes were collected in the Saas valley in the Wallis Alps, altitude about 2200 m. For the dynamic study, 4 sizeable wild plants ( R1-4) were selected and marked in their natural habitat in order to be able to follow them throughout the seasons. Parts of the rhizomes of R1-4 were collected at five different time points (25/05, 07/06, 07/07, 10/08, 29/09/2006). For the comparison of female and male plants, 15 wild plants ( P1-15 ) were collected in June 16 th 2006. As R. rosea is a dioecious plant, female and male plants could easily be distinguished from each other at the flowering stage. Voucher specimens (No 2006008-2006023) were deposited at the Laboratory of Pharmacognosy and Phytochemistry, University of Geneva. The raw material was stocked at –5°C before being freeze- dried and powdered.

Extraction The extraction procedure was optimized after a range of trial extractions, carried out using various methanol concentrations (0, 50, 60, and 100%) in water and different extraction times (30, 60 and 90 min) in an ultra-sonic bath. The procedure was monitored by HPLC analysis. The extraction of freeze- dried powdered roots (10 mg) during 60 minutes with 1 ml of 60% methanol was finally selected as

95

3. RESULTS

the extraction method. After centrifugation at 10’000 rpm during 5 minutes, 100 µl of the supernatant was diluted with 900 µl methanol 6%. The extractions were performed in duplicate.

HPLC analysis All samples were analyzed by HPLC-UV-DAD immediately after the extraction process. The chromatographic system consisted of a Hewlett-Packard 1100 (Waldbronn, Germany) instrument with a quaternary pump, an autosampler (HP 1100), and a photodiode array detector (HP 1100).

Analytical separation of the markers was carried out using a 3.0 x 150 mm i.d., 3 µm, Atlantis dC 18 column equipped with a dC 18 guard column. The solvents used were methanol and acetonitrile (1:1) with formic acid 0.1% (solvent A) and water with formic acid 0.1% (solvent B). The gradient program was as follows: 0-7 min, 99-80% B; 7-26 min, 80-63% B; 26-30 min, 63-0% B; 35-45 min, 0% B. The analysis temperature was set at 37°C. The injection volume was 10 µl, and the equilibration time between the runs was 15 min. The flow rate was set at 500 µl/min. Rosavins were detected at 254 nm and salidroside at 210 nm.

Identification and quantification Identification of the markers was based on retention time, UV and MS spectra, compared with commercial standards. All components were quantified by the external standard method. Calibration curves were constructed over six different concentrations of the standards dissolved in methanol 6%. Results of regression analysis on calibration curves and detection limits are presented in Table 1 . The limits of detection and quantification (LOD and LOQ) were evaluated on the basis of a signal-to-noise ratio of 3 and 10 respectively. The regression coefficients of the calibration curves were all greater than 0.99.

Table 1 Statistical data of the HPLC-UV analysis of the markers in R. rosea extracts

LOD LOQ Linear range RSD Analyte Regression equation a (µg/ml) r2 (ng) (ng) (%) Recovery (%)

(1) y= 30.745 x + 46.7605 0.5- 62.5 0.999 0.5 1 0.54 94.8-104.6

(2) y = 36.379 x + 34.984 0.1-100 0.999 0.1 1 2.79 100.1-103.2

(3) y= 57.723 x + 25.759 0.1-100 0.999 0.1 1 1.30 93.9-98.8

(5) y = 54.145 x + 17.218 0.1-100 0.999 0.1 1 5.00 97.9-102.4 a y and x are peak areas (mAu) and concentration of analytes, respectively

96

3. RESULTS

The precision of the method, expressed as the relative standard deviation (RSD), was studied. The intra-day and inter-day RSD values of the four analytes were in the range of 0.5-5.0%. Accuracy was evaluated by recovery experiments. The samples were spiked with the reference standard solution at concentrations corresponding to 70, 100 and 130% of the content in the sample. For all standards, the recovery ranged between 93.9% and 104.69%. All the analyses were done in triplicate.

3.2.1.4. Results and discussions

Chromatographic analyses The separation of the rosavins presented difficulties because of their closely related structures. The LC method described by Tolonen et al. (2003a) for the separation of the main constituents of R. rosea extracts was adapted. The solvents were acidified (pH = 2) and a longer column (150 mm vs. 50 mm) was chosen to improve the separation. The solvent gradient was adjusted for the column. The optimized RP-HPLC method provided a repeatable and baseline separation of salidroside (1), rosin ( 5) and the three isomers: rosarin ( 2), rosavin ( 3) and cinnamyl-(6’-O-ß-xylopyranosyl)-O-ß- glucopyranoside ( 4). The method was validated in terms of linearity, precision, and accuracy. The average recoveries of the four analytes were 93.9% and 104.69% and the RSDs ranged from 0.5 to 5.0%. The specificity of the method was checked by LC-MS-ESI in positive mode ( Figure 2 ). The isomer cinnamyl-(6’-O-ß-xylopyranosyl)-O-ß-glucopyranoside ( 4) was not quantified as its content was under the limit of quantification in each sample.

97

3. RESULTS

3 MAu UV 254 nm 80 2 5 40 1 0 100 1 - m/z 299 [M-H] % 0 100 3 2 + % m/z 451 [M+Na] 4 0 100 5 + % m/z 319 [M+Na] 0 10 12 14 16 18 20 22 24 26 28 Min

Figure 2 Specificity of the HPLC-UV method checked by HPLC-MS of R. rosea extract Chromatograpic conditions: Atlantis dC 18 column (3 x 150 mm i. d., 3 µm) equipped with a dC 18 precolumn; gradient of MeOH:MeCN (1:1) + FA 0.1% (solvent A) / H 2O + FA 0.1% (solvent B) (0-7 min, 99-80% B; 7-26 min, 80-63% B; 26-30 min, 63-0% B; 35-45min, 0% B), flow rate 500 µl/min; analysis temperature 37°C; UV detection at 210 nm and 254 nm; MS- ESI detection in positive mode.

Seasonal variation in rosavins and salidroside content Marker contents considerably fluctuate between plants, as shown in Figure 3 . For example, plant R-3 contained a very low amount of salidroside ( 1), less than the limit of detection (0.05%), throughout the year, while the content of the other three plants was quantified to at least 0.5%. The level of this marker varied from less than 0.05% to 2.25% from one plant to another. The variation is equivalent to 98%. The total rosavins content also changed from 1.00% to 4.25% among the four studied plants, which represents a variation of 76%.

98

3. RESULTS

%

4.50 % Salidroside % Total rosavins 4.00

3.50

3.00

2.50

2.00

1.50

1.00

0.50

0.00 MayJune JulyAug. Sept. MayJune JulyAug. Sept. MayJune JulyAug. Sept. MayJune JulyAug. Sept. R-1 R-2 R-3 R-4

Figure 3 Dynamic profiling of total rosavins and salidroside in four wild plants ( R1-R4 ) of the same population of R. rosea during 2006

The dynamics of rosavins and salidroside during the growing season in the four selected plants are shown in Figure 3 . To determine the dynamics, parts of the rhizome of the individual plant were collected at five different dates during the year 2006. Each selected date corresponded to a growth stage of the plant: 1) beginning of May: the start of the vegetative period, 2) beginning of June: the stem stage, 3) end of June: the flowering stage, 4) beginning of August: the seed stage, 5) end of September: the end of the vegetative period. The analysis of individual markers showed that contents of rosavin and rosarin were the highest compared to other derivatives. The variations of total rosavins and salidroside within the same plant during the growing season were less than 1%. By comparing the marker content variation within the four plants throughout the year, no correlation could be drawn between the collecting date and the content of rosavins and salidroside even when the rosavins were analyzed separately. Thus, seasonal variation has no significant impact on the rosavins and salidroside content in R. rosea rhizomes. These results indicate that the harvest of cultivated R. rosea can be done in the autumn in order to optimize the phytochemical composition as well as the biomass.

Chemical profile comparison between female and male plants Fifteen wild plants ( P1-15 ) were collected in the same location in the Saas valley. In this batch, there were seven female and eight male plants. R. rosea is a dioecious plant with female and male plants which can easily be distinguished from each other at the flowering stage. The marker analysis of the 15 plants is summarized in Figure 4 . The chemical profile comparison of the gender demonstrates no

99

3. RESULTS

significant links between the sex of the plant and the content of the markers. However, as mentioned before, there is an important variation of the constituent levels between plants from the same population without distinction of their gender.

% % Salidroside 4.50 % Total rosavins 4.00

3.50

3.00

2.50

2.00

1.50

1.00

0.50

0.00 P1 P7 P5 P8 P15 P10 P13 P2 P3 P4 P6 P9 P11 P12 P14 Females Males

Figure 4 Chemical profile comparison of female and male R. rosea plants ( P1 -P15 )

The developed HPLC-UV method has been validated and used for the quantitative determination of rosavins and salidroside content in wild R. rosea rhizomes harvested in different seasons. This is the first time that the dynamics of the main markers have been studied within the same plant. By this method, the variations due to the difference between individual plants were eliminated, allowing a reliable interpretation of the results. It showed that the seasonal variations have no significant impact on the chemical profile of the rhizomes. Moreover, the marker analyses of a population composed of female and male plants showed no significant differences between the two sexes. As a clear indication of the relationship between a specific secondary metabolite and its biological activity has not been demonstrated, the main criteria for the selection of high quality R. rosea rhizomes could rely on the elevated content in rosavins and salidroside. This study showed a large variation of the markers among individual plants from the same population. Plant R3 contains almost no salidroside while the other plants have an amount of this metabolite over 0.5%.To determine the importance of external and genetic factors on the production of salidroside and rosavins in the rhizomes, vegetative clones of plants R1-4 and P1-15 have been propagated in the greenhouse. These clones have been planted in the field side by side under identical conditions in the Bagnes valley at 1050m altitude. These plants will

100

3. RESULTS

be harvested and analyzed after 4 years of cultivation. If the same internal relationship between salidroside and rosavins is found in wild grown and cultivated clones, it would indicate the importance of the genetic factors.

Acknowledgements The authors thank the Swiss National Science Foundation for financial support of this work (Grant No. 200020-107775 to Prof. K. Hostettmann).

101

3. RESULTS

3.2.1.5. References

Ali, Z., Fronczek, F.R., and Khan, I.A. (2008). Phenylalkanoids and monoterpene analogues from the roots of Rhodiola rosea . Planta Medica 74 : 178-181.

Brown, R.P., Gerbarg, P.L., and Ramazanov, Z. (2002). Rhodiola rosea . A phytomedicinal overview. HerbalGram 56 : 40-52.

Darbinyan, V., Aslanyan, G., Amroyan, E., Gabrielyan, E., Malmstrom, C., and Panossian, A. (2007). Clinical trial of Rhodiola rosea L. extract SHR-5 in the treatment of mild to moderate depression. Nordic Journal of Psychiatry 61 : 343-348.

Darbinyan, V., Kteyan, A., Panossian, A., Gabrielian, E., Wikman, G., and Wagner, H. (2000). Rhodiola rosea in stress induced fatigue - A double blind cross-over study of a standardized extract SHR-5 with a repeated low-dose regimen on the mental performance of healthy physicians during night duty. Phytomedicine 7: 365-371.

Galambosi, B. (2006). Demand and availibility of Rhodiola rosea L. raw material, pp. 223-226, in: Bogers, R.J., Craker, L.E., and Lange, D. (eds). Medicinal and Aromatic Plants . Springer: Netherlands.

Galambosi, B., Galambosi, Z., and Slacanin, I. (2007). Comparison of natural and cultivated roseroot (Rhodiola rosea L.) roots in Finland. Zeitschrift Fur Arznei- & Gewurzpflanzen 12 : 141-147.

Gyoergy, Z., Tolonen, A., Pakonen, M., Neubauer, P., and Hohtola, A. (2004). Enhancing the production of cinnamyl glycosides in compact callus aggregate cultures of Rhodiola rosea by biotransformation of cinnamyl alcohol. Plant Science 166 : 229-236.

Khanum, F., Bawa, A.S., and Singh, B. (2005). Rhodiola rosea : a versatile adaptogen. Comprehensive Reviews in Food Science and Food Safety 4: 55-62.

Kovaleva, N.P., Tikhomirov, A.A., and Dolgushev, V.A. (2003). Specific characteristics of Rhodiola rosea growth and development under photoculture conditions. Russian Journal of Plant 50 : 527-531.

Kucinskaite, A., Poblocka-Olech, L., Krauze-Baranowska, M., Sznitowska, M., Savickas, A., and Briedis, V. (2007). Evaluation of biologically active compounds in roots and rhizomes of Rhodiold rosea L. cultivated in Lithuania. Medicina-Lithuania 43 : 487-494.

Ma, G., Li, W., Dou, D., Chang, X., Bai, H., Satou, T., Li, J., Sun, D., Kang, T., Nikaido, T. and others. (2006). Rhodiolosides A-E, monoterpene glycosides from Rhodiola rosea . Chemical & Pharmaceutical Bulletin 54 : 1229-1233.

Panossian, A. and Wagner, H. (2005). Stimulating effect of adaptogens: an overview with particular reference to their efficacy following single dose administration. Phytotherapy Research 19 : 819-838.

102

3. RESULTS

Perfumi, M. and Mattioli, L. (2007). Adaptogenic and central nervous system effects of single doses of 3% rosavin and 1% salidroside Rhodiola rosea L. extract in mice. Phytotherapy Research 21 : 37-43.

Petsalo, A., Jalonen, J., and Tolonen, A. (2006). Identification of flavonoids of Rhodiola rosea by liquid chromatography-tandem mass spectrometry. Journal of Chromatography A 1112 : 224- 231.

Tolonen, A., Hohtola, A., and Jalonen, J. (2003a). Comparison of electrospray ionization and atmospheric pressure chemical ionization techniques in the analysis of the main constituents from Rhodiola rosea extracts by liquid chromatography/mass spectrometry. Journal of Mass Spectrometry 38 : 845-853.

Tolonen, A., Pakonen, M., Hohtola, A., and Jalonen, J. (2003b). Phenylpropanoid glycosides from Rhodiola rosea . Chemical & Pharmaceutical Bulletin 51 : 467-470.

Yoshikawa, M., Shimada, H., Shimoda, H., Matsuda, H., Yamahara, J., and Murakami, N. (1995). Rhodiocyanosides A and B, new antiallergic cyanoglycosides from Chinese natural medicine "si lie hong jing tian", the underground part of Rhodiola quadrifida (Pall.) Fisch. et Mey. Chemical & Pharmaceutical Bulletin 43 : 1245-1247.

Yousef, G.G., Grace, M.H., Cheng, D.M., Belolipov, I.V., Raskin, I., and Lila, M.A. (2006). Comparative phytochemical characterization of three Rhodiola species. Phytochemistry 67 : 2380-2391.

Zapesochnaya, G.G. and Kurkin, V.A. (1982). Cinnamic glycosides of Rhodiola rosea rhizomes. Khimiya Prirodnykh Soedinenii : 723-727.

103

3. RESULTS

3.3. CHEMICAL AND BIOLOGICAL SCREENING OF SEDUM SPECIES (CRASSULACEAE)

This part of the work aimed to phytochemically investigate a selection of Sedum species from the Crassulaceae family in order to find new potential herbal stress buster originated from Switzerland. Since no other Rhodiola species than Rhodiola rosea L. are found in Switzerland, the plant selection was focussed on Sedum species considering that this genus is close to Rhodiola . About 21 species of Sedum grow in Switzerland. A first selection of plants has been done according to their availability knowing that several species are endangered (Sedum sediforme (Jacq.) PAU , Sedum rubens L. and Sedum villosum L.) or grows on steep rocky walls difficult to access ( Sedum anacampseros L. and Sedum atratum L.). Furthermore, for chemotaxonomy purpose, the preference was put on species for which no intensive phytochemical studies were done. Based on the criteria described above, 6 Sedum species were selected for the chemical and biological screening.

3.3.1. Plant materials

The different Sedum species collected in different places in Switzerland (from 450 to 2000 m) are enumerated in Table 3-2. Firstly, small quantities of plant material were collected for the screening. Afterwards, the selected plant for the investigation was collected in larger quantity to furnish higher amounts of extracts. Immediately after collecting, the plants were washed to take off the mud and cut. They were frozen and lyophilised in order to avoid mould contamination which could occur when drying at ambient temperature. The dried plants were ground in liquid nitrogen and extracted following the same protocol as used for R. rosea in Chapter 3.1.1. The amounts of extract obtained are displayed in Table 3-2.

105

3. RESULTS

Table 3-2 Sedum species selected for the screening

Type of Botanical species Collected parts Plant weight Extract quantity extract

Sedumacre L. whole plant 58 g MeOH 6.2 g DCM 2.0 g

Sedumalbum L. whole plant 5 g MeOH 620 mg DCM 280 mg

Sedumatratum L. whole plant 115 g MeOH 6.9 g DCM 4.7 g

Sedumdasyphyllum L. whole plant 14 g MeOH 800 mg DCM 570 mg

Sedumsarmentosum Bunge whole plant 7.5 g MeOH 990 mg DCM 300 mg

Sedumsexangulare L. whole plant 102 g MeOH 7.4 g DCM 5.4 g

3.3.2. Screening of Sedum species extracts

The extracts were submitted to TLC assays described in Chapter 3.1.2.: free radical scavenging, and AChE inhibitory. Since the test in solution on MAOs was not anymore available, this activity could not be tested on these extracts. Several extracts showed radical scavenging and AChE inhibitory properties. The results of the screening and the phytochemical investigation of the most active species, Sedum dasyphyllum L., led to the publication “Antioxidant phenolic compounds from Sedum dasyphyllum L.“ presented in Chapter 3.2.3.

106

3. RESULTS

3.3.3. Antioxidant phenolic compounds from Sedum dasyphyllum L.

Daphne van Diermen, Monica Pierreclos, Kurt Hostettmann *

Laboratory of Pharmacognosy and Phytochemistry, School of Pharmaceutical Sciences, University of Geneva, University of Lausanne, Quai Ernest-Ansermet 30, CH-1211 Geneva 4, Switzerland

Submitted to Journal of Natural Products

107

3. RESULTS

3.3.3.1. Abstract

Twelve crude extracts from six plants of the genus Sedum (Crassulaceae) collected in Switzerland were submitted to rapid TLC assays against 1,1-diphenyl-2-picrylhydrazyl (DPPH) radical and acetylcholinesterase. Sedum dasyphallum L., which showed interesting activities against these two targets, has been studied. The chemical investigation of the methanol extract from the whole plant afforded the new flavonols kaempferol 3-O-α-rhamnoside-7-O-β-sophoroside ( 9), gossypetin 3,7-di- O-β-glucoside-8-O-β-glucuronide ( 13 ), herbacetin 3,7-di -O-β-glucoside-8-O-β-glucuronide ( 14 ), herbacetin 3-O-β-(3’’-acetylglucoside)-7-O-β-glucoside-8-O-β-glucuronide ( 17 ), herbacetin 3-O-β- (3’’-acetylglucoside)-8-O-β-glucuronide ( 19 ), and hibiscetin 3-O-β-glucoside-8-O-β-glucuronide ( 16 ), along with thirteen known flavonols, isoflavones, cyanogenic glycoside, caffeic acid and ferrulic acid derivatives. The structures of the isolated compounds were established by means of spectroscopic data analysis. Among the isolates, seven exhibited strong scavenging activity against DPPH (EC50 from 20 to 75 M).

108

3. RESULTS

3.3.3.2. Introduction

Crassulaceae is a medium-sized family among angiosperms, with estimated numbers of species ranging from 900 to 1500, distributed in about 40 genera ('t Hart, 1997). Most species are herbaceous leaf succulents. The family has a nearly cosmopolitan distribution and occurs most commonly in (semi-) arid and mountainous habitats of the temperate and subtropical zones (Van Ham and 't Hart, 1998). The genus Sedum, including about 500 species, is almost restricted to the temperate and subtropical regions of the Northern hemisphere ('t Hart and Bleij, 2005). Previous phytochemical studies on this genus reported the presence of alkaloids, flavonoids, and cyanogenic compounds (Van Ham and 't Hart, 1998). Some Sedum plants have been documented as either vegetables or folk medicines for treatment of many diseases. Sedum dasyphyllum L. is found in areas from Central Europe to the Mediterranean coastlines. It is a very low, compact plant, only a few centimeters high, producing a dense mat of glaucous-gray foliage and white flowers. In our continuous program search for new potential antioxidants and acetylcholinesterase inhibitors from natural sources, a preliminary screening of twelve extracts (DCM and MeOH) from six plants of the genus Sedum collected in Switzerland has been undertaken. The extracts were submitted to rapid TLC assays against 1,1- diphenyl-2-picrylhydrazyl radical (DPPH) and acetylcholinesterase (AChE). The MeOH extract of the whole plant of S. dasyphyllum was selected due to its high activities and moreover it was not investigated previously. This paper describes the isolation and characterisation of nineteen compounds including twelve flavonols ( 2, 4, 5, 8-10 , 13 , 14 , and 16 -19 ) of which six are new derivatives of herbacetin ( 14, 17 and 19 ), gossypetin ( 13 ), hibiscetin ( 16 ), and kaempferol ( 9). Three isoflavones ( 3, 6 and 7), one lignan ( 11 ), one cyanogenic glycoside ( 12 ), caffeic acid ( 1) and one ferrulic acid glucoside ( 15 ) were also identified. The radical scavenging and the AChE inhibitory activities of these compounds were determined. Compounds 2, 4, 8, 10 , 11 , 13 and 16 exhibited significant scavenging activity similar to quercetin which is known to be a strong antioxidant.

3.3.3.3. Results and discussions

Twelve MeOH or DCM extracts were tested against two targets: DPPH which reveals radical scavenging activity, and AChE, which is a key enzyme in the pathology of Alzheimer’s disease. The results are presented in Table 1.

109

3. RESULTS

Table 1 TLC autographical screening of extracts from Sedum species

Botanical species Extract DPPH assay AChE assay

Sedumacre L. whole plant MeOH + + DCM - -

Sedumalbum L. whole plant MeOH + - DCM - -

Sedumatratum L. whole plant MeOH + - DCM - -

Sedumdasyphyllum L. whole plant MeOH + + DCM - -

Sedumsarmentosum Bunge whole plant MeOH + - DCM - -

Sedumsexangulare L. whole plant MeOH + - DCM - - (-: inactive and +: active)

MeOH extract from S. acre presented activities on both TLC assays, however the plant has already been largely studied for its alkaloid content (Anilakumar et al. , 2006; Hillhouse et al. , 2004; Vanderwal et al. , 1981). On the other hand, S. dasyphyllum L. had not been investigated before and the crude MeOH extract exhibited interesting radical scavenging and AChE inhibitor activities. Thus, its phytochemical bio-guided investigation was undertaken. The MeOH extract was first partitioned between water and EtOAc, and then water and n-BuOH. The three extracts showed significant activities against DPPH and/or AChE. Repeated column chromatography of these extracts resulted in the isolation of 19 compounds ( 1-19 ) including 6 new flavonols ( 9, 13 , 14 , 16 , 17 and 19 ).

110

3. RESULTS

OH OH 3' OH OR3 4 R1= R2= H, R3= Gluc OR 3 4' 10 R1= Glc, R2= Rha, R3= H B R2O O R O 8 O 14 R1= R2= Glc, R3= Gluc 2 2 1' 17 R1= Glc-3''-Ac, R2= Glc, R3= Gluc 7 A C 18 R1= Glc, R2= H, R3= Gluc OR 6 3 1 19 R = Glc-3''-Ac, R =H, R =Gluc 5 OR1 1 2 3 OH O OH O

2 R1= R2= H, R3= Gluc 13 R1= R2= Glc, R3= Gluc OH OH 6''' OH O HO 1''' O O OH HO O O HO HO O 1'''' 6'''' O OH OH O 1'' HO OH HO HO O 1'' OH OH OH O OH OH O O 9 CH 6'' 6'' 3 8 OH CO H 6''' 2 OH O OCH3 HO OH OH 3' OH HO 1''' O 4' 9' OH HO O 7 H3CO OH 1 8 8' 1' 3 7' HO OH 9 O 1'' OH HO 1'' 4 O OH OH HO OH O O O 6'' 6'' 11 OH OH 16

3.3.3.4. Structural elucidation of isolated compounds

The isolated compounds could be classified into four classes: flavonols, isoflavones, cyanogenic glycosides, and lignans. Caffeic acid ( 1) (Lu and Foo, 1997) and (Z)-ferulic acid 4-O-β-glucoside ( 15 ) (Baderschneider and Winterhalter, 2001) were also identified. Compounds 2, 4, 5, 8-10 , 13 , 14 and 16 -19 exhibited the similar UV spectra characteristic of 3-OH substituted flavonols (Markham, 1982).

The HR-ESI-MS data of compound 9 indicated a pseudomolecular formula of C 33 H40 O20 . Moreover, the ESI-IT-MS analysis in the negative-ion mode showed a molecular ion at m/z 755 ([M-H] -). The MS/MS analysis displayed fragments at m/ z 609 corresponding to the loss of a deoxy hexose ([M- 146] -) and at m/ z 285 corresponding to the aglycone. The difference between the two fragments corresponded to the loss of two hexoses. The 1H NMR spectrum (Table 2) showed signals of aromatic protons at δ 6.52 (1H, d, J = 2.0 Hz, H-6), δ 6.79 (1H, d , J = 2.0 Hz, H-8), δ 6.95 (2H, d, J = 8.3 Hz, H-2’), and δ 7.84 (2H, d, J = 8.3 Hz, H-3’). The coupling constants and integrations of the two latter

111

3. RESULTS

protons suggested that they were in ortho position from each other on the B-ring which was substituted in position 4’. Furthermore, the downfield shift of C-4’ ( δ 161.8) confirmed a 4’-OH substitution (Table 3). The coupling constant of H-6 and H-8 (2.0 Hz) corresponded to benzoyl protons in meta positions. The downfield signals of C-5 ( δ 162.8) and C-7 ( δ 164.4) confirmed a 5,7- OH substitution. From these elements, the aglycone was identified as kaemfperol. Anomeric proton signals at δ 5.44 (1H, d, J = 1.5 Hz, H-1’’), δ 5.29 (1H, d, J = 7.3 Hz, H-1’’’), and δ 4.67 (1H, d, J = 7.8 Hz, H-1’’’’) confirmed the presence of three sugars as suggested by mass analysis. The short-range correlations observed on the HSQC spectrum corroborated this information with the attachment of the three anomeric protons with carbons at δ 103.5 (C-1’’), δ 100.2 (C-1’’’) and δ 105.5 (C-1’’’’), respectively. The protons and the carbons of the sugars were attributed thanks to 1H-1H COSY and HSQC spectra suggesting the presence of two glucosides and a rhamnoside. Compound 9 was hydrolysed under acidic conditions and the mixture extracted by LLE with EtOAC. The aqueous fraction was then analysed by TLC analysis confirming that the two hexoses were glucoses. According to the long-range correlations observed on the HMBC spectrum between the anomeric H-1’’ and C-3 (δ 136.4), the rhamnoside was linked to the aglycone at position 3. The long-range correlations for the signal of H-1’’’’ (δ 4.67) with C-2’’’ (δ 83.6) indicated that the glucosides are linked by position 1’’’’ → 2’’’, therefore corresponding to sophoroside (Wolbis, 1989). The linkage of the sophoroside to C-7 on the aglycone was determined by the HMBC correlation from the anomeric proton at δ 5.29 (H- 1’’’) with C-7 ( δ164.4). Moreover, a comparison with a similar compound, isorhamnetin 3-O-α-L- rhamnoside-7-O-β-D-sophorosid (Wolbis, 1989) validated the structure and position of the sugars. Therefore, 9 was elucidated as a new compound named kaempferol 3-O-α-rhamnoside-7-O-β- sophoroside.

The HR-ESI-MS data of compound 13 displayed a molecular formula of C 33 H38 O24 . The ESI-IT-MS analysis in negative mode presented a molecular ion at m/z 817 ([M-H] -). The MS/MS analysis presented fragments at m/z 641 and m/z 479 compatible with a loss of a glucuronic acid ( m/z [M-176] -) and a hexose ( m/z [M-162] -), respectively. Another fragment at m/z 317 corresponding to the aglycone was observed. The difference between the fragments at m/z 479 and m/z 317 corresponded to the loss of another hexose. The 13 C NMR spectrum (Table 3) showed the presence of 15 carbon signals due to the flavonol skeleton and a set of signals arising from the sugars [anomeric carbons: δ 100.7 (C-1’’), δ 101.4 (C-1’’’), and δ 105.5 (C-1’’’’)]. The 1H NMR spectrum (Table 2) exhibited a characteristic proton signal at δ 12.55 corresponding to a chelated hydroxyl at C-5, a singlet proton signal at δ 6.72 arising from H-6 and an ABX coupling system [ δ 7.84 (1H, d, J = 2.2 Hz), δ 6.85 (1H, d, J = 8.5 Hz),

112

3. RESULTS

δ 7.66 (1H, dd , J = 2.2, 8.5 Hz)] referring to the 3 ′,4 ′-dihydroxylated ring B. These 1H and 13 C NMR spectral features suggested the presence of a 3,7,8-tri-O-substitued gossypetin. Anomeric protons signals at δ 5.55 (1H, d, J = 7.3 Hz, H-1’’), δ 4.98 (1H, d, J = 7.3 Hz, H-1’’’) and δ 4.82 (1H, d, J = 7.8 Hz, H-1’’’’) confirmed the presence of three sugars. The 13 C NMR spectrum displayed a signal at δ 170.1 (C-6’’’’) corresponding to the carboxylic carbon of the glucuronic acid confirming its presence. The protons and the carbons of the sugars were attributed thanks to the 1H-1H COSY and HSQC spectra verifying the presence of two glucosides and glucuronic acid. Acid hydrolysis of 13 was undertaken. The mixture was extracted by LLE with EtOAc. The aqueous fraction was analyzed by TLC with standards, which revealed that the hexoses corresponded to glucoses. The HMBC spectrum displayed long-range correlations from the anomeric protons of the glucosides δ 5.55 (H-1’’) and δ 4.98 (H-1’’’) to C-3 ( δ 135.8) and C-7 ( δ155.3), respectively. Thus, the glucosides are linked to the aglycone at positions 3 and 7 and the glucuronic acid is linked at position 8 as verified by the long- range correlation of H-1’’’’ ( δ 4.82) with C-8 ( δ125.6). Hence, the structure of 13 was elucidated as gossypetin-3,7-di-O-β-glucoside-8-O-β-glucuronide. This is the first time this compound is isolated. The ESI-IT-MS analyses, in negative mode, for compounds 14 , 17 , 18 and 19 showed a fragment at m/z 301 corresponding to the aglycone moiety, herbacetin. The HR-ESI-MS of 14 displayed a molecular formula of C 33 H38 O23 differing from 13 by an hydroxyl residue. Moreover, the only differences in the NMR spectra (Tables 2 and 3) of 14 when compared to 13 concerned the patterns of substitution in the B-ring. For 14 , protons AB–type signals were observed at δ 8.20 (2H, d, J = 8.8 Hz, H-2’) and δ 6.88 (2H, d, J = 8.7 Hz, H-3’), suggesting that the B-ring was only substituted in position 4’. The downfield signal of C-4’ ( δ 160.1) corroborated this hypothesis. Therefore, 14 was elucidated as a new natural compound, herbacetin 3,7-di -O-β-glucoside-8-O-β-glucuronide.

The molecular formula of 17 , deduced from HR-ESI-MS was C 35 H40 O24 , with an additional acetyl residue when compared to 14 . The 1H and 13 C NMR spectra (Tables 2 and 3) of compound 17 showed supplementary signals to 14 , at δH 2.06 (3H, s, 3’’-OCOCH 3), δC 21.1 (3’’-OCOCH 3) and δC 169.7 1 13 (3’’-OCOCH 3) corresponding to the acetyl group. The H and C spectra of the aglycone moieties of the two compounds were identical, showing that the acetyl group was linked to one of the sugars. The 1H NMR spectra indicated that the signal of H-3’’ was deshielded ( δ 4.87 vs 3.61 in 14 ). This shift suggested that the glucose in position 3 was acetylated in position 3’’. The HMBC spectrum confirmed this hypothesis with a correlation between H-3’’ ( δ 4.87, m) and the carbon at δ 169.7 (3’’-

OCOCH 3). Thus, compound 17 was identified as herbacetin 3-O-β-(3’’-acetylglucoside)-7-O-β- glucoside-8-O-β-glucuronide, a new natural product.

113

3. RESULTS

The HR-ESI-MS of compound 19 indicated a molecular formula of C 27 H28 O18 . Compound 19 appeared to be a derivative of herbacetin 3 -O-β-glucoside-8-O-β-glucuronide ( 18 ) with an additional 1 13 C2H2O residue, corresponding to an acetyl group. Direct comparison of both H and C NMR spectra (Tables 2 and 3) of 18 and 19 showed additional signals for 19 : one more proton signal at δ 2.05 (3H, s, 3’’-OCOCH 3) and two more carbon signals at δ 169.6 (3’’-OCOCH 3) and δ 21.1 (3’’-OCOCH 3) corresponding to an acetyl group. All other shifts were comparable, except for the position 3’’ for which 1H signal was deshielded ( δ 4.86 vs 3.10 in 18 ). This shift suggested that the glucose was acetylated in position 3’’. The long-range correlation between 3’’-OCOCH 3 ( δ 169.6) and H-3’’ ( δ 4.86) observed on the HMBC spectrum corroborated this hypothesis. Thus, the new structure 19 was identified as herbacetin 3-O-β-(3’’-acetylglucoside)-8-O-β-glucuronide, a new natural product.

The HR-ESI-MS of 16 indicated a molecular formula of C 27 H28 O20 , differing from herbacetin 3 -O-β- glucoside-8-O-β-glucuronide ( 18 ) by two hydroxyl residues. The NMR spectra (Tables 2 and 3) of 16 and 18 were similar. Significant differences could be observed for the aromatic signals. The 1H NMR spectrum of 16 displayed only two aromatic signals at δ 6.28 (1H, s, H-6) and δ 7.38 (2H, s, H-2’). As the proton signal at δ 7.38 integrated for two protons, the B-ring was supposed to be symmetric. The 13 C NMR indicated the signals of the B-ring carbons at δ 119.9 (C-1’), 109.1 (C-2’), 145.1 (C-3’) and 136.7 (C-4’). The downfield signals of C-3’, C-4’, suggested that the B-ring was 3’,4’,5’-OH substituted. These elements corroborated the presence of two additional hydroxyl substituents compared to 18 , as indicated by the HR-ESI-MS. Hence, the structure of 16 was elucidated as a new compound, hibiscetin 3-O-β-glucoside-8-O-β-glucuronide. Six known flavonols were also isolated and identified as hibifolin ( 2) (Lai et al. , 2007), melocorin ( 4) (Wolbis, 1989), afzeloside ( 5) (Fossen et al. , 1999), hirsutrin ( 8) (Markham and Chari, 1982), herbacetin 3-O-β-glucoside-7-O-α-rhamnoside ( 10 ) (Yoshikawa, 2006), herbacetin 3 -O-β-glucoside-8- O-β-glucuronide ( 18 ) (Sikorska et al. , 2004) by comparison of their physicochemical and spectroscopic data with those reported in the literature. Compounds 3, 6 and 7 presented characteristic UV spectra of isoflavones (Markham, 1982). Thanks to the HR-ESI-MS, 1D and 2D NMR data and the literature information, they were identified as 3’-O- methylorobol 7-O-β-D-glucoside ( 3) (Adinaray and Rajasekh, 1974; Viscardi et al. , 1984), formerly isolated from Sedum alfredii and lineare (Li and Zuo, 1991; Men, 1986) , dalspinosin 7-O-β-glucoside (6) previously isolated from Juniperus macropoda (Cupressaceae) (Sethi et al. , 1983), and iristectorigenin B ( 7) identified in Iris pseudacorus and spuria (Iridaceae) (Hanawa et al. , 1991; Shawl

114

3. RESULTS

et al. , 1984). As the NMR data of these isoflavones were not complete in the literature, they are reported in Tables 2 and/or 3.

The molecular formula of 11 , deduced from HR-ESI-MS corresponded to C 11 H19 O6. The NMR spectra indicated the presence of aromatic and sugar signals. By direct comparison with the literature data, compound 11 was identified as the lignan secoisolariciresinol 4-O-β-glucoside, previously isolated from Glehnia littoralis (Umbelliferae) (Yuan et al. , 2002). The HR-ESI-MS analysis of 12 indicated pseudomolecular mass of 306.1187 [M+CHOO] -, suggesting the presence of an odd number of nitrogen atoms. Compound 12 was thus identified as lotaustralin by comparison of the physicochemical and spectroscopic data with those reported in the literature (Akgul et al. , 2004). Lotaustralin has been reported in Rhodiola rosea L. (Akgul et al. , 2004), however it has never been described in the genus Sedum . Furthermore, dalspinosin 7-O-β-glucoside ( 6), iristectorigenin B ( 7), secoisolariciresinol 4-O-β-glucoside ( 11 ), (Z)-ferulic acid 4-O-β-glucoside ( 15 ) and herbacetin 3 -O-β-glucoside-8-O-β-glucuronide ( 18 ) have never been reported in the Crassulaceae family.

115

1 a b Table 2 . H NMR spectroscopic data of isolated compounds (500 MHz, J in Hz), in CD 3OD or in DMSO-d6

1H 3a 9b 13 b 14 b 16 b 17 b 19 b 2 8.18 s 5-OH 12.55 brs 12.55 brs 12.55 brs 12.55 brs 12.55 brs 12.55 brs 6 6.71 d ( J = 2.0) 6.52 d ( J= 2.0) 6.72 s 6.73 s 6.28 s 6.74 s 6.29 s 8 6.53 d (J = 2.0) 6.79 d ( J= 2.0) 2’ 7.19 d (J = 1.5) 6.95 d ( J= 8.3) 7.84 d ( J= 2.2) 8.20 d ( J = 8.8) 7.38, 2H, s 8.20 d ( J = 9.1) 8.21 d ( J = 9.1) 3’ 7.84 d ( J = 8.3) 6.88 d ( J = 8.7) 6.88 d ( J = 8.7) 6.88 d ( J = 9.1)

3’-OCH 3 3.91, 3H, s 5’ 6.86 d ( J = 8.3) 7.84 d ( J = 8.3) 6.85 d ( J = 8.5) 6.88 d ( J = 8.7) 6.88 d ( J = 8.7) 6.88 d ( J = 9.1) 6’ 7.01 dd ( J = 1.5, 8.3) 6.95 d ( J = 8.3) 7.66 dd ( J = 2.2, 8.5) 8.20 d ( J = 8.8) 7.38, 2H, s 8.20 d ( J = 9.1) 8.21 d ( J = 9.1) 1’’ 5.05 d ( J = 6.8) 5.44 d ( J = 1.5) 5.55 d ( J = 7.3) 5.55 d ( J = 7.3) 5.50 d ( J = 7.5) 5.64 d ( J = 7.3) 5.60 d ( J = 7.5) 2’’ 3.50 m 4.24 m 3.38 m 3.38 m 3.34 m 3.38 d (J =7.8) 3.38 m 3’’ 3.51 m 3.72 t (J = 7.6) 3.62 d (J = 9.5) 3.61 d (J = 11.7) 3.23 d (J = 8.3) 4.87 m 4.86 m 4’’ 3.43 m 3.34 m 3.34 m 3.34 m 3.17 d (J = 3.1) 3.38 d (J = 7.8) 3.35 m 5’’ 3.52 m 3.35 d (J = 2.4) 3.27 m 3.27 m 3.13 m 3.27 m 3.25 m 6’’ 3.74 m 3.52 m 3.52 m 3.30 m 3.61 m 3.42 d (J = 7.9) 0.94, 3H, d ( J = 5.4) 3.92 m 3.75 d (J = 11.7) 3.75 d (J = 10.7) 3.67 d (J = 11.9) 3.44 d (J = 5.4) 3.60 d (J = 10.1)

-OCOCH 3 2.06, 3H, s 1’’’ 5.29 d ( J = 7.3) 4.98 d ( J = 7.3) 4.99 d ( J = 7.3) 4.82 d ( J = 7.5) 5.00 d ( J = 7.3) 4.76 d ( J = 7.5) 2’’’ 3.78 d (J = 8.8) 3.40 m 3.40 m 3.45 d (J = 8.5) 3.40 m 3.44 d (J = 7.9) 3’’’ 3.70 d (J = 8.8) 3.31 m 3.31 m 3.45 d (J = 8.5) 3.31 m 3.33 m 4’’’ 3.39 m 3.15 dd (J = 1.8, 11.7) 3.15 m 3.31 m 3.18 d (J = 4.9) 3.50 t (J = 9.1) 5’’’ 3.29 m 3.26 m 3.27 m 3.69 d (J = 9.5) 3.44 d (J = 5.4) 3.78 d (J = 9.5) 6’’’ 3.73 m 3.75 d (J = 11.7) 3.75 d (J = 10.7) 3.74 d (J = 9.5)

3.93 m 3.50 d (J = 9.1) 3.51 m 3.50 d (J = 9.6)

-OCOCH 3 2.05, 3H, s 1’’’’ 4.67 d ( J = 7.8) 4.82 d ( J = 7.8) 4.81 d ( J = 7.8) 4.82 d ( J = 7.8) 2’’’’ 3.24 m 3.40 m 3.40 m 3.38 m 3’’’’ 3.41 m 3.65 d (J =11.9) 3.65 m 3.65 dd (J = 9.9, 2.8) 4’’’’ 3.45 d (J = 2.9) 3.50 d (J = 9.1) 3.50 d (J = 9.8) 3.50 t (J = 9.8) 5’’’’ 3.55 m 3.32 m 3.32 d (J = 9.8) 3.32 m

6’’’’ 3.59 m

3.65 d (J = 4.4)

3. RESULTS

Table 3. 13 a b C NMR data of isolated compounds (125 MHz ), in CD 3OD or in DMSO-d6

13 C 3a 6a 7a 9b 13 b 14 b 16 b 17 b 19 b

2 155.5 155.7 155.1 159.8 160.9 157.3 156.6 158.7 156.9 3 125.0 124.6 125.0 136.4 135.8 132.9 133.3 133.7 132.9 4 181.0 182.5 182.2 179.8 177.6 177.6 177.2 177.5 177.3 5 165.1 151.3 151.0 162.8 156.5 156.7 156.2 157.4 156.3 6 95.9 130.7 129.3 100.9 98.9 98.9 98.7 99.04 99.0

6-OCH 3 62.3 61.9 7 164.0 157.7 158.8 164.4 155.3 155.4 156.2 156.5 156.3 8 101.1 100.2 100.3 95.9 125.6 125.6 124.3 126.5 124.9 9 159.2 158.7 157.2 158.1 148.6 148.4 148.4 148.4 148.4 10 108.1 107.7 106.1 107.7 105.5 105.4 103.7 105.4 103.6 1’ 123.5 124.9 124.4 122.4 125.6 120.6 119.9 120.6 120.5 2’ 113.9 114.2 114.2 116.6 120.9 131.7 109.1 131.7 131.3 3’ 148.8 150.2 150.2 132.1 144.5 115.1 145.1 115.1 115.3 4’ 148.0 150.8 150.7 161.8 148.4 160.1 136.7 160.2 160.0 5’ 116.2 112.8 112.8 132.1 117.4 115.1 145.1 115.1 115.3 6’ 122.9 122.8 122.8 116.6 122.1 131.7 109.1 131.7 131.3

3’-OCH 3 56.5 56.5

4’-OCH 3 56.5 56.4 1’’ 101.6 102.0 103.5 100.7 100.6 100.8 100.4 100.6 2’’ 74.7 74.8 71.9 74.0 74.1 73.8 72.0 72.0 3’’ 77.0 78.4 77.1 75.6 75.7 76.5 77.4 77.5 4’’ 71.2 71.1 73.2 69.9 69.6 69.9 67.6 67.7 5’’ 77.4 78.0 72.1 76.5 76.3 77.7 77.2 77.2 6’’ 62.4 62.3 17.1 61.1 60.6 61.1 60.4 60.4 169.7 169.6 3’’-OCOCH 3 21.1 21.1 1’’’ 100.2 101.4 101.3 105.0 101.4 106.1 2’’’ 83.6 73.2 73.1 71.3 73.2 73.7 3’’’ 77.5 75.6 75.7 73.8 74.0 75.3 4’’’ 71.2 69.7 69.8 75.5 70.0 71.4 5’’’ 78.0 76.5 76.3 77.5 77.3 76.0 6’’’ 62.4 60.7 60.8 170.9 60.7 169.9 1’’’’ 105.5 105.5 104.9 104.8 2’’’’ 76.0 74.0 73.9 74.0 3’’’’ 77.7 77.7 77.6 75.7 4’’’’ 70.9 71.3 71.3 71.7 5’’’’ 78.1 77.3 77.2 75.4 6’’’’ 62.1 170.1 169.9 169.9

117

3. RESULTS

3.3.3.5. Radical scavenging properties of isolates

Among the isolates, caffeic acid and ferrulic acid derivatives which are widely distributed compounds in the plant kingdom are extensively known to have anti-oxidant activity (Chen and Ho, 1997). The other isolated compounds were tested for their scavenging properties against DPPH on TLC assay. Hibifolin ( 2), melocorin (4), herbacetin 3-O-β-glucoside-7-O-α-rhamnoside ( 10 ), secoisolariciresinol 4-O-β-glucoside ( 11 ), lotaustralin ( 12 ), gossypetin 3,7-di-O-β-glucoside-8-O-β-glucuronide ( 13 ), hibiscetin 3-O-β-glucoside-8-O-β-glucuronide ( 16 ) presented inhibitory spots on the TLC assay. Thus, their IC 50 values were evaluated on a microtitre plate assay (Table 4). Lotaustralin ( 12 ) showed only low activity while the lignan ( 11 ) and the flavonols ( 2, 4, 10, 13 and 16 ) exhibited high activity, particularly hibifolin ( 2) (EC50 = 21.0 M) which presented a similar activity to quercetin (EC50 = 10.1 M). It is the first time that the radical scavenging activity of melocorin ( 4) and herbacetin 3-O-β- glucoside-7-O-α-rhamnoside ( 10 ) is reported. There is general agreement that flavonoids possess both excellent iron chelating and radical scavenging properties (Havsteen, 2002; Van Acker et al. , 1996). Furthermore, all the flavonoids isolated were 3-OH substituted flavonols. It has been extensively described that the double bond between carbons 2 and 3 in conjugation with the 4-oxo function, which is responsible for electron dislocation in the B ring, and the presence of both 3- and 5-hydroxyl groups were responsible for maximal radical scavenging capacity and strongest radical absorption (Van Acker et al. , 1996).

Table 4 Radical scavening activities (DPPH) of the isolated compounds from S. dasyphyllum

Compounds EC 50 (µM) Hibifolin ( 2) 21.0 ± 0.1 Melocorin ( 4) 28.9 ± 0.1 Hirsutrin ( 8) 30.3 ± 0.1 Herbacetin 3-O-β-glucoside-8-O-α-rhmanoside ( 10 ) 74.1 ± 0.1 Secoisolariciresinol 4-O-β-glucoside ( 11 ) 56.3 ± 0.1 Lotaustralin ( 12 ) 419.9 ± 0.6 Gossypetin 3,7-di-Oβ-glucoside-8-O-β-glucuronide ( 13 ) 30.2 ± 0.1 Hibiscetin 3-O-β-glucoside-8-O-β-glucuronide ( 16 ) 35.8 ± 0.1 Herbacetin 30.4 ± 0.1 Kaempferol 30.1 ± 0.1 Gossypetin 20.8 ± 0.1 Hibiscetin 14.1 ± 0.1 Quercetin 10.1 ± 0.1

118

3. RESULTS

In order to establish a structure–antioxidant activity relationship, the aglycone of the flavonols, obtained by acid hydrolysis, were tested in the same way. By comparing the activities of the aglycones, it appeared that quercetin, hibiscetin and gossypetin presented similar activity while kaempferol and herbacetin showed lower activity. This indicated that the dihydroxy (catechol) structure in the B-ring, which confers great stability to the aroxyl radicals and participates in electron dislocation, played an important role in antioxidant activity. It showed also that the 8-OH-substitution had no influence on the scavenging activity. This element was reinforced by comparing the activity of herbacetin and its 8-O-glycoside ( 4) derivative: their activities were similar. It was also observed that a pyrogalloyl group in ring B instead of a catechol, as in hibiscetin, did not change the scavenging activity.

3.3.3.6. Acetylcholinesterase inhibitory activity of isolates

As the MeOH extract exhibited AChE inhibitory properties, the isolates were tested on this inhibitory activity. Herbacetin 3-O-β-glucoside-7-O-α-rhamnoside ( 10 ) and lotaustralin ( 12 ) presented inhibitory spot on the TLC assay. However, on the microtitre plate assay, these compounds exhibited very low inhibitory activity (IC 50 >100 M).

3.3.3.7. Concluding remarks

In this study, S. dasyphyllum was demonstrated to contain many phenolic compounds. Thirteen flavonols of which six new derivatives of herbacetin, gossypetin, hibiscetin and kaempferol, three isoflavones and one lignan were isolated. A cyanogenic glycoside was also purified. The genus Sedum is characterised by the widespread occurrence of 3’-O-methylated as well as 8- hydroxylated and 8-O-methylated flavonols. This is explained by the fact that hydroxylations at C-8 of the flavonol skeleton are common biosynthetic features throughout this genus (Stevens et al. , 1996). Thus, the isolation of several derivatives of gossypetin and herbacetin, which are rather rare flavonoids, is not surprising in S. dasyphyllum. Moreover, we also indicated that this plant and its constituents possess significant radical scavenging activity. Among the six new isolates, compounds 13 and 16 exhibited strong scavenging activity (EC50 = 30.2 and 35.8 M, respectively). The strong radical scavenging activities of hibifolin ( 2) and melocorin ( 4) (EC50 = 21.0 and 28.9 M, respectively) are reported here for the first time, while the antioxidant activities of caffeic acid ( 1), ferulic acid glucoside ( 15 ), and hisutrin ( 8) are already well known (Chen and Ho, 1997; Wang et al. , 2006).

119

3. RESULTS

3.3.3.8. Experimental

Generalexperimentalprocedure

25 [α] D: Perkin-Elmer-241 polarimeter. UV: Perkin-Elmer-Lambda-20 UV–VIS spectrophotometer. UV spectra were recorded in MeOH. 1H and 13 C NMR: Varian Unity Inova NMR instrument, Palo Alto, 1 13 CA, USA. H and C NMR spectra were recorded in DMSO-d6 or CD 3OD at 500 and 125 MHz, respectively. TMS: int. standard. ESI-MS: Finnigan MAT LCQ ion trap instrument, in negative mode. HRMS: Micromass LCT Premier, in negative mode. TLC: silica gel 60 F254 Al sheets (Merck) using hexane–EtOAc (1-1) or CH 2Cl 2–MeOH–H2O (13-7-1). MPLC: RP-18 LiChroprep column (40–63 m; 450 x 50 mm i.d.; Merck). HPLC-UV-DAD was carried out on a HP1100 (Agilent) with a Symmetry RP-18 column (5 m; 250 x 4.6 mm i.d.; Waters) using a MeCN+ FA 0.1%:H2O + FA 0.1% gradient (2:98–50:50) in 35 min. The detection was performed at 210, 254, 280 and 360 nm.

Plantmaterial The whole plant of Sedum dasyphyllum L. (Crassulaceae) as authenticated by Edigio Anchisi (Orsières, VS, Switzerland) was collected in Wallis, Switzerland, in August 2008. A voucher specimen (no 2008-005) is deposited in the Laboratory of Pharmacognosy and Phytochemistry, Section of Pharmaceutical Sciences, University of Geneva. The plant was washed, freeze dried and powdered.

Extraction,fractionation,andisolation

Dried and powdered plants of Sedum dasyphyllum (315 g) were sequentially extracted with CH 2Cl 2 (3 × 24 h) and MeOH (3 × 24 h) at room temperature under agitation. After removing the solvent under vacuum, 12 g of crude CH 2Cl 2 extract (3.8% w/w) and 15 g of MeOH crude extract (4.8% w/w) were obtained. The MeOH extract was suspended into H 2O (300 mL) to give a suspension that was partitioned consecutively with EtOAc and n-BuOH (1L each), then exhaustively concentrated to yield an EtOAc fraction (4.2 g), a BuOH fraction (4.8 g), and a H 2O fraction (5 g), respectively. The EtOAc fraction was subjected to MPLC eluted with a MeCN: H 2O+ FA 0.1 % step gradient (5:95 to 70:39 in 5% steps) to give 40 fractions (A1-40). This separation afforded compounds 1 (A16, 10 mg), 2 (A21, 52 mg), 3 (A23, 28 mg), 4 (A26, 406 mg), 5 (A27, 33 mg), 6 (A31, 59 mg) and 7 (A37, 86 mg). Fraction A22 was purified on a Sephadex LH-20 column (1 ×30 cm) using MeOH as eluent to give compound 8 (6.5 mg). The BuOH fraction (4.8 g) was separated by preparative MPLC eluted with

120

3. RESULTS

MeCN:H 2O+FA 0.1% step gradient (0:100 to 70:30 in 1-5% steps) to give 48 fractions (B1-48). Fraction B23 (60 mg) was applied to a Sephadex LH-20 column (2 ×70 cm) eluted with MeOH to give compound 9 (14 mg). The fraction B26 (40 mg) was purified on an analytical column ( SS 250 / 0.5"/

10 Nucleosil 7-C8) eluted with MeCN:H 2O+FA 0.1% gradient (15:85 to 30:70, 3 mL/min) to give compounds 10 (4.3 mg) and 11 (4 mg). The H 2O fraction (5 g) was separated by preparative MPLC eluted with MeCN-H2O+FA 0.1% step gradient (5:95 to 40:60 in 1-5% steps) to give 53 fractions (C1- 53). This separation afforded compounds 12 (C8, 7 mg), 13 (C11, 15mg), 14 (C15, 8.1 mg), 15 (C17, 8.3 mg), 16 (C21, 17.3 mg), 17 (C28, 18.8 mg), 18 (C33, 56.8 mg), and 19 (C43, 11.4 mg).

Kaempferol3Oαrhamnoside7Oβsophoroside( 9)

25 Green-yellowish amorphous solid. [ α] D -114.7° (c 1.0, MeOH); UV (MeOH) λmax nm (log ε): 266 (4.27), 322 (4.05), 366 (4.28); 1H and 13 C NMR: see Tables 2 and 3, HR-ESI-MS: m/z 755.2037 - (C 33 H39 O20 : [M-H] requires 755.2035).

Gossypetin3,7diOβglucoside8Oβglucuronide(13 )

25 Yellow amorphous solid. [ α] D +29.3° ( c 1.0, MeOH); UV (MeOH) λmax nm (log ε): 261 (4.19), 270 1 13 (sh) (4.17), 366 (3.97); H and C NMR: see Tables 2 and 3, HR-ESI-MS: m/z 817.1631 (C 33 H37 O24 : [M-H] - requires 817.1675).

Herbacetin3,7diOβglucoside8Oβglucuronide(14 )

25 Yellow amorphous solid. [ α] D +13.4° ( c 1.0, MeOH); UV (MeOH) λmax nm (log ε): 271 (4.51), 328 1 13 (sh) (4.36), 358 (4.31); H and C NMR: see Tables 2 and 3, HR-ESI-MS: m/z 801.1667 (C 33 H37 O23 : [M-H] - requires 801.1726).

Hibisccetin3Oβglucoside8Oβglucuronide( 16 )

25 Yellow amorphous solid. [ α] D +57.5° ( c 1.0, MeOH); UV (MeOH) λmax nm (log ε): 270 (3.99), 366 1 13 - (3.83); H and C NMR: see Tables 2 and 3, HR-ESI-MS: m/z 671.1068 (C 27 H27 O20 : [M-H] requires

671.1096).

Herbacetin3Oβ(3’’acetylglucoside)7Oβglucoside8Oβglucuronide( 17 )

25 Yellow amorphous solid. [ α] D – 18.8° ( c 1.0, MeOH); UV (MeOH) λmax nm (log ε): 270 (4.10), 356 1 13 - (3.91); H and C NMR: see Tables 2 and 3, HR-ESI-MS: m/z 843.1755 (C 35 H39 O24 : [M-H] requires 843.1831).

121

3. RESULTS

Herbacetin3Oβ(3’’acetylglucoside)8Oβglucuronide( 19 )

25 Brownish amorphous solid. [ α] D – 139.6° ( c 1.0, MeOH); UV (MeOH) λmax nm (log ε): 271 (4.06), 1 13 - 355 (3.88); H and C NMR: see Tables 2 and 3, HR-ESI-MS: m/z 681.1305 (C 27 H27 O18 : [M-H] requires 681.1303).

Acidhydrolysisandsugarsidentificationofcompounds( 9, 13 , 14 and 16 ) Compounds 9, 13 , 14 and 16 (5 mg of each) were heated in 15 ml of HCl 2N at 100°C for 1 hour. After cooling, the mixtures were extracted by LLE with EtOAc (3 × 15 ml). The aglycones were obtained from the EtOAc fractions after removal of solvent under vacuum. The aqueous solutions were neutralized by adding 10% NaHC0 3. The sugars, after removal of water, were analysed by TLC with standards (rhamnose, galactose, glucose, mannose, and glucuronic acid, 1 mg/mL). The solvent system used for the TLC analysis corresponded to iso -PrOH:MeOH:H2O (7:2:1). The sugars were revealed with diphenylamine reagent.

3.3.3.9. Free radical scavenging assays

A TLC autographic assay of radical scavenging activity using the 1,1’-diphenyl-2-picrilhydrazyl (DPPH) radical was employed for extract screening and bio-guided investigations. After application of

100 g of each sample on a silica gel 60 F 254 Al sheet (Merck), the TLC plate was developed in hexane:EtOAc (1:1) for the CH 2Cl 2 extracts, or CH 2Cl 2:MeOH:H2O (13:7:1) for the MeOH extracts, and then thoroughly dried for complete removal of solvents. A solution of DPPH (2 mg/mL in MeOH) was then sprayed. Radical scavengers appeared as yellow spots against a purple background. Quercetin was used as positive control. DPPH radical scavenging activity of isolated compounds was measured according to the method reported by Lee et al. (2003) with some modifications. The samples dissolved in MeOH were added to EtOH containing 96-well microtitre plates to make up a total volume of 30 L in each well with final sample concentrations from 0.5 to 250 g/mL (10 dilutions). Then, 90 L of 200 M DPPH was added, the mixture was shaken vigorously and incubated for 30 minutes at 25°C. The scavenging activity of a tested compound was measured as the decrease in DPPH absorbance at 490 nm and expressed as a percentage of the absorbance of a control DPPH solution without the tested compound.

The EC50 stands for the concentration required for 50% scavenging activity.

122

3. RESULTS

3.3.3.10. Acetylcholinesterase inhibitory activity assays

A bioautographic assay on TLC developed by Marston (2002) was used in this work for the extract screening and the bio-guided isolation. After application of the samples (10 g for extracts and 1 g for pure compounds) on a TLC plate, this latter was developed in hexane:EtOAc (1:1) for the CH 2Cl 2 extracts, or CH 2Cl 2:MeOH:H2O (13:7:1) for the MeOH extracts, and then thoroughly dried for complete removal of solvents. The plate was then sprayed with a solution of acetylcholinesterase (1500 U) and bovine serum albumin (225 mg) dissolved in 500 mL of a Tris–hydrochloric acid buffer at pH 7.8 and incubated at 37 °C for 15 min in polystyrene boxes with a moist atmosphere. Then, a solution (1:4) of naphthyl acetate (250 mg in 100 mL EtOH) and Fast Blue B salt (50 mg in 20 mL

H2O) was sprayed onto the plate to give a purple coloration after 1–2 min. Inhibitors appeared as white spots against a purple background. Galanthamine was used as positive control. AChE inhibitory activity of the isolated compounds was measured on microtitre plate according to the method described by Ellman (1961). Briefly, the reaction mixture consisted of 100 L of 0.1M phosphate buffer (pH 7.8), 20 L of a solution of AChE (final concentration 0.05 U/mL in 0.1 Tris– hydrochloric acid buffer, pH 7.8), 20 L of test compound (samples were dissolved in 0.1 M phosphate buffer containing 45% MeOH). The controls contained the corresponding volume of buffer instead of test compound solutions. The mixture was shaken and incubated at 37°C during 15 min in a 96-well microplate. Then, 40 L of Ellman’s reageant (0.75 mM final concentration of 5,5’-dithiobis- (2-nitrobenzoic acid) (DTNB) in 0.1M phosphate buffer pH 7.8) followed by 20 µL of 1.5 mM acetylthiocholine iodide (ATCI) solution to initiate the reaction. The hydrolysis of ATCI was monitored by the formation of a yellow 5-thio-2-nitrobenzoate anion as a result of the enzyme- catalyzed reaction of DTNB with thiocholines, using a microplate-reader at a wavelength of 405 nm. The rates of reaction (Mean V) were obtained over 180 sec, with an 18 sec interval, shaking before every reading. The reaction system where the tested solution was replaced by equivalent MeOH-buffer solution was used as blank control. The percentage of inhibition was determined by comparison of rates of reaction of samples relative to the blank sample. Inhibition % = (1-Vtest /V blank ) ×100 %.

StatisticalAnalysis All assays were done in triplicate. All data are expressed as mean ± standard deviation for the number of experiments. The concentration of tested compounds required to inhibit 50% of the enzyme activity under the assay conditions was determined from dose-response curves and defined as the IC 50 value.

123

3. RESULTS

The dose-response curves were obtained and calculated with GraphPad Prism software (Version 5, graphPad software Inc., San Diego, CA, USA).

Acknowledgement The authors thank the Swiss National Science Foundation for financial support of this work (Grant No. 200020-107775 to Prof. K. Hostettmann).

124

3. RESULTS

3.3.3.11. References

'T Hart, H. (1997). Crassulaceae, pp. 20-22, in: Oldfield, S. (eds). Cactus and succulent plants - status survey and conservation action plan. IUCN/SSC Cactus and Succulent Specialist Group: Gland.

'T Hart, H. and Bleij, B. (2005). Sedum , pp. 234-260, in: Eggli, U. (eds). Illustrated Handbook of Succulent Plants: Crassulaceae . Springer: New York.

Adinaray, D. and Rajasekh, J. (1974). Synthesis and study of 3'-O-methylorobol. Indian Journal of Chemistry 12 : 911-913.

Akgul, Y., Ferreira, D., Abourashed, E.A., and Khan, I.A. (2004). Lotaustralin from Rhodiola rosea roots. Fitoterapia 75 : 612-614.

Anilakumar, P.K.R., Khanum, F., and Bawa, A.S. (2006). Phytoconstituents and antioxidant potency of Rhodiola rosea - A versatile adaptogen. Journal of Food Biochemistry 30 : 203-214.

Baderschneider, B. and Winterhalter, P. (2001). Isolation and characterization of novel benzoates, cinnamates, flavonoids, and lignans from riesling wine and screening for antioxidant activity. Journal of Agricultural and Food Chemistry 49 : 2788-2798.

Chen, J.H. and Ho, C.-T. (1997). Antioxidant activities of caffeic acid and its related hydroxycinnamic acid compounds. Journal of Agricultural and Food Chemistry 45 : 2374-2378.

Ellman, G.L., Courtney, K.D., Andres Jr, V., and Featherstone, R.M. (1961). A new and rapid colorimetric determination of acetylcholinesterase activity. Biochemical Pharmacology 7: 88- 90.

Fossen, T., Larsen, Å., Kiremire, B.T., and Andersen, Ø.M. (1999). Flavonoids from blue flowers of Nymphaea caerulea . Phytochemistry 51 : 1133-1137.

Hanawa, F., Tahara, S., and Mizutani, J. (1991). Isoflavonoids produced by Iris pseudacorus leaves treated with cupric chloride. Phytochemistry 30 : 157-163.

Havsteen, B.H. (2002). The biochemistry and medical significance of the flavonoids. Pharmacology & Therapeutics 96 : 67-202.

Hillhouse, B., Ming, D.S., French, C., and Towers, N.G.H. (2004). Acetylcholine esterase inhibitors in Rhodiola rosea . Pharmaceutical Biology 42 : 68-72.

Lai, X.Y., Zhao, Y.Y., and Liang, H. (2007). Flavonoid glucuronide from Abelmoschus manihot (L.) Medik. Biochemical Systematics and Ecology 35 : 891-893.

125

3. RESULTS

Lee, S.M., Na, M.K., An, R.B., Min, B.S., and Lee, H.K. (2003). Antioxidant activity of two phloroglucinol derivatives from Dryopteris crassirhizoma . Biological & Pharmaceutical Bulletin 26 : 1354-1356.

Li, S. and Zuo, C. (1991). Chemical constituents of linear stonecrop ( Sedum lineare ). Zhongcaoyao 22 : 438-440.

Lu, Y.R. and Foo, L.Y. (1997). Identification and quantification of major polyphenols in apple pomace. Food Chemistry 59 : 187-194.

Markham, K.R. (1982). Techniques of flavonoids identification. Treherne, J.E. and P.H., R. (eds). Academic Press INC.: London.

Markham, K.R. and Chari, V.M. (1982). Carbon-13 NMR spectroscopy of flavonoids, pp. 19-134, in: Harborne, J.B. and J., M.T. (eds). The Flavonoids: Advances in Research . Chapman and Hall: New York.

Marston, A., Kissling, J., and Hostettmann, K. (2002). A rapid TLC bioautographic method for the detection of acetylcholinesterase and butyrylcholinesterase inhibitors in plants. Phytochemical Analysis 13 : 51-54.

Men, Y.D. (1986). Studies on the constituents of Sedum alfredii . Taiwan Yaoxue Zazhi 38 : 52-66.

Sethi, M.L., Taneja, S.C., Dhar, K.L., and Atal, C.K. (1983). Three isoflavone glycosides from Juniperus macropoda . Phytochemistry 22 : 289-292.

Shawl, A.S., Vishwapaul, Zaman, A., and Kalla, A.K. (1984). Isoflavones of Iris spuria . Phytochemistry 23 : 2405-2406.

Sikorska, M., Matlawska, I., and Franski, R. (2004). 8-hydroxyflavonoid glucuronides of Malope trifida . Acta Physiologiae Plantarum 26 : 291-297.

Stevens, J.F., Hart, H., Elema, E.T., and Bolck, A. (1996). Flavonoid variation in eurasian Sedum and Sempervivum . Phytochemistry 41 : 503-512.

Van Acker, S.A.B.E., Van Den Berg, D.-J., Tromp, M.N.J.L., Griffioen, D.H., Van Bennekom, W.P., Van Der Vijgh, W.J.F., and Bast, A. (1996). Structural aspects of antioxidant activity of flavonoids. Free Radical Biology and Medicine 20 : 331-342.

Van Ham, R. and 'T Hart, H. (1998). Phylogenetic relationships in the Crassulaceae inferred from chloroplast DNA restriction-site variation. American Journal of Botany 85 : 123-134.

Vanderwal, R., Kooy, J.H., and Vaneijk, J.L. (1981). Phytochemical investigation of Sedum acre L. Planta Medica 43 : 97-99.

Viscardi, P., Reynaud, J., and Raynaud, J. (1984). A new isoflavone glucoside from the flowers of Cytisus scoparius Link (Leguminosae). Pharmazie 39 : 781-782.

126

3. RESULTS

Wang, K., Chen, L., Li, N., and Yu, X. (2006). Antioxidant and radical-scavenging activity of flavonoids from Solidago canadensis . Zhongguo Yaoxue Zazhi 41 : 493-497.

Wolbis, M. (1989). Flavonol glycosides from Sedum album . Phytochemistry 28 : 2187-2189.

Yoshikawa, M.; Nihon Yakuyo Shokuhin Kenkyusho K. K., Japan; Harima Kanpo Seiyaku Co., Ltd., assignee. 2006 20050302. Sinocrassula -derived components and usage thereof. Japan patent 2005-58051; 2006241054.

Yuan, Z., Tezuka, Y., Fan, W.Z., Kadota, S., and Li, X. (2002). Constituents of the underground parts of Glehnia littoralis . Chemical & Pharmaceutical Bulletin 50 : 73-77.

127

4. CONCLUSION AND PERSPECTIVES

4. CONCLUSION AND PERSPECTIVES

4.1. CHARACTERISATION OF BIOACTIVE COMPOUNDS FROM RHODIOLA ROSEA L.

The first part of the present work consisted in the clarification of the physiological actions of Rhodiola rosea L. metabolites on the organism, more especially on the cognitive functions and on the mood. R. rosea root extracts were tested on three different targets involved in nervous central system functions i.e . monoamine oxidases A and B, acetylcholinesterase and oxidative stress. The DCM extract exhibited significant AChE inhibitory activity on TLC assay. While the MeOH and water extracts exhibited considerable MAOs inhibitory activity. This led to the bioassay-guided investigation of the three extracts. Four compounds were isolated from the DCM extract: tyrosol acetate ( DMC1 ), cinnamyl alcohol (DCM2 ), β-sitosterol ( DCM3 ) and linoleic acid ( DCM4 ). Cinnamyl alcohol ( DCM2 ) and linoleic acid ( DCM4 ) showed inhibitory activity against AChE on TLC assay. Their minimal inhibition amount corresponded to 5 and 1 g, respectively. The reference standard galanthamine exhibits a minimal inhibition amount of 0.01 g. Kissling et al. (2005) already described the AChE inhibitory activity of linoleic acid while AChE inhibitory activity of cinnamyl alcohol is reported for the first time. The presence of AChE inhibitors in R. rosea roots may partly clarify the action pathway of the plant in the stimulation of the cholinergic system and furthermore of the memory functions.

HO O

OH cinnamyl alcohol linoleic acid (DCM2) (DCM4)

Figure 4-1 AChE inhibitors isolated from the DCM extract of R. rosea roots

131

4. CONCLUSION AND PERSPECTIVES

Remarkable MAOs inhibitory activity was found in the methanol and water extracts. At a concentration of 100 g/ml, the activity related to the positive controls (selegiline/ clorgyline) against MAO A was 92.5% and 84.3% and 81.8% and 88.9%, respectively, against MAO B. The bio-guided fractionation of both extracts led to the isolation of twelve compounds of different chemical classes. Seven phenylpropane derivatives were purified: salidroside ( RR1 ), rosarin ( RR5 ), cinnamyl alcohol ( RR6 ), triandrin ( RR8 ), tyrosol ( RR10 ), rosavin ( RR9 ), and rosin ( RR11 ). Nitrile glucoside rhodiocyanoside A ( RR7 ), monoterpen glucoside rosiridin ( RR12 ) and epigallocatechin gallate dimmer ( RR2 ) were also isolated. A mixture of rhodioloside B and C ( RR3-4) isomers was also obtained. Their structures were determined by means of spectroscopic methods, including 1D and 2D NMR experiments and HR-MS analysis.

Rosiridin ( RR12 ) exhibited the highest inhibitory activity against MAO B. The pIC 50 (-logIC 50 ) value was estimated to 5.38 ± 0.05 (positive control selegiline: 7.23 ± 0.04). The mixture of rhodioloside B and C isomers ( RR3 and RR4 ) demonstrated also moderate inhibitory activity. As the mixture was obtained in too small quantity, the isomers could not be separated for individually testing. The others isolated metabolites exhibited low inhibitory activity against MAO except the rosavins which did not show any inhibitory effect against both MAOs. The presence of MAO inhibitors in R. rosea roots explains partly how the plant can increase the levels of serotonin, dopamine and norepinephrine in the nerve terminals. Therefore, the antidepressant activity of this plant could be explained by the inhibition of MAO A, while its effects on symptoms associated with neurodegenerative diseases may be explain by the inhibition of MAO B.

HO O HO O HO OH OH

rosiridin (RR12)

HO O OH HO O HO OH O OH HO HO O O O HO O HO O OH HO HO OH OH

rhodioloside B (RR3) rhodioloside C (RR4)

Figure 4-2 Compounds with MAOs inhibitory activity isolated from R. rosea roots

132 4. CONCLUSION AND PERSPECTIVES

This part of the work contributes to explain the action mode of R. rosea extracts on the central nervous system functions. The precise physiological effects of R. rosea are far from being completely elucidated. As R. rosea metabolites act on many pathways and various targets in the brain, it will be difficult to identify precisely all the physiological effects.

4.2. HPLC-UV QUANTIFICATION OF SALIDROSIDE AND ROSAVINS IN R. ROSEA L. PLANTS

The main objective of this part was to assess the chemical profile dynamics in R. rosea wild plants during a growing season in order to determine the most favourable period to harvest rhizomes with high metabolite concentrations. Furthermore, the marker’s profiles of female and male plants were compared with the aim of selecting the most appropriate gender for further cultivation. The results of the study led to the following conclusions: • The marker contents fluctuate considerably between plants even when belonging to the same population. For example, salidroside content varied from less than 0.05% to 2.25% from one plant to another. • The marker evolution established on one growing season indicated that seasonal variation has no significant impact on the rosavins and salidroside contents. These results demonstrated there is no favourable harvesting time concerning the marker’s contents. • The chemical profile comparison between female and male plants showed no significant links between the gender of the plant and the content of the markers. To determine the importance of external and genetic factors on the production of salidroside and rosavins in the rhizomes, vegetative clones of the same plants used during the study have been planted in the field under identical conditions in the Bagnes valley at 1050m altitude. The plants will be harvested and analysed after 4 years of cultivation (in 2010). Therefore, the comparison between wild and cultivated plants will put in relation the interdependency of the markers content and external or genetic factors. Furthermore, the chemical profile dynamics should also be performed in large scale on cultivated plants in order to confirm the preliminary results obtained on wild plants.

133 4. CONCLUSION AND PERSPECTIVES

4.3. CHEMICAL AND BIOLOGICAL SCREENING OF SEDUM SPECIES

This part aimed to investigate Sedum species collected in Switzerland in order to find a new potential herbal stress buster. Therefore, 12 extracts from six selected species were submitted to TLC assays to evaluate their antioxidant and AChE inhibitory activities. Sedum acre L. and Sedum dasyphyllum L. revealed the more interesting activities against both targets. However, Sedum acre have already been largely studied for its alkaloids content, unlike S. dasyphyllum for which no chemical study was reported previously. The chemical investigation of the methanol extract of Sedum dasyphyllum L. led to the isolation of 19 compounds of four different chemical classes i.e . flavonoids, lignans, phenolic acids and cyanogenic glucosides. Among the isolated flavonoids, six new compounds, displayed in Figure 4-3, were identified: kaempferol 3-O-α-rhamnoside-7-O-β-sophoroside ( SD9 ), gossypetin-3,7-di-O-β-glucoside- 8-O-β-glucuronide ( SD13 ), herbacetin 3,7-di-O-β-glucoside-8-O-β-glucuronide ( SD14 ), hibiscetin 3- O-β-glucoside-8-O-β-glucuronide ( SD16) , herbacetin 3-O-β-(3’’-acetylglucoside)-7-O-β-glucoside-8- O-β-glucuronide ( SD17 ), herbacetin 3-O-β-(3’’-acetylglucoside)-8-O-β-glucuronide ( SD19 ). Furthermore, isoflavonoids dalspinosin 7-O-β-glucoside ( SD6 ) and iristectorigenin B ( SD7 ), flavonoid herbacetin 3 -O-β-glucoside-8-O-β-glucuronide ( SD18 ), and lignan secoisolariciresinol 4-O-β- glucoside ( SD11 ) were here isolated for the first time in the Crassulaceae family. The structures of the isolates were determined by spectrometric and chemical methods, including 1D and 2D NMR experiments and HR-MS analysis.

134 4. CONCLUSION AND PERSPECTIVES

CO2H O HO OH OH HO O OH OH OH OH O O HO HO HO O O HO O O O OH O HO HO O O OH OH OH OH O OH O O HO OH HO OH OH SD9 O OH SD13 CH3

CO2H O HO CO2H OH HO O O OH OH OH HO OH O HO O HO OH HO O O HO O OH OH HO HO O OH O OH OH OH OH O O OH O O OH OH SD14 SD16

CO2H O HO CO2H HO O O OH OH OH HO OH O HO O HO OH HO O O HO O OH HO HO O OCOCH3 O OCOCH3 OH OH OH O O OH O O SD17 OH SD19 OH

Figure 4-3 New flavonols isolated from Sedum dasyphyllum L. methanol extract

Hibifolin ( SD2), melocorin ( SD4), herbacetin 3-O-β-glucoside-7-O-α-rhamnoside ( SD10 ), secoisolariciresinol 4-O-β-glucoside ( SD11 ), lotaustralin ( SD12 ), gossypetin 3,7-di-O-β-glucoside-8- O-β-glucuronide ( SD13 ), hibiscetin 3-O-β-glucoside-8-O-β-glucuronide ( SD16 ) presented radical scavenging activity against the radical DPPH on TLC assay. These compounds were evaluated in a solution assay using the same reaction as on TLC. Except lotaustralin, all the metabolites presented

135 4. CONCLUSION AND PERSPECTIVES

high scavenging activity particularly hibifolin and melocorin which EC50 were comparable to the reference standard quercetin. Two metabolites active against AChE have been identified on TLC assay: herbacetin 3-O-β-glucoside- 7-O-α-rhamnoside ( SD10 ) and lotaustralin ( SD12 ). However, they only exhibited weak activity in the Ellman’s test in solution. S. dasyphyllum has been demonstrated to contain many phenolic compounds with strong radical scavenging properties. Supplementary analysis could be performed on the plant, indeed the DCM extract was not investigated in this work. It would be also interesting to test the extracts against the MAOs to know if they present similar activity to R. rosea . However, the chemical profiles of the two species considerable differ. Thus, it may be supposed that a similar activity against MAO will not be recovered. The majority of the Sedum species found in Switzerland, e.g. S. atratum, S. villosum, S. rubens, has not been investigated yet. Their phytochemical investigation could bring a better understanding of the Sedum chemotaxonomy. Furthermore, it could lead to isolation of potential new active compounds, indeed Sedum atratum methanol extract exhibited interesting radical scavenging properties in the screening carried out in this work.

136

5. EXPERIMENTAL PART

5. EXPERIMENTAL PART

5.1. PLANT MATERIAL AND EXTRACTION

5.1.1. Plant material

The details concerning the analysed plants are shown in Table 5-1. The botanical identification of the plants was carried out by Mr E. Anchisi (Orsières, Wallis, Switzerland), Dr. P. Malnoe (Agroscope Changins-Wädenswil, Switzerland) and Dr J. Kissling (Botany Institute, University of Neuchatel, Switzerland). Specimen vouchers are deposited in the Laboratory of Pharmacognosy and Phytochemistry, School of Pharmaceutical Sciences, University of Geneva, Switzerland.

Table 5-1 Origin of the analysed plants

Date of Botanical species Collected part Collection site Identifier Plant weight collection

Rhodiola rosea L. whole plant Val d’Aoste, 11.11.2005 E. Anchisi 1 kg Italy 2450 m

Rhodiola rosea L. whole plant Saas valley, 25.05.2006 P. Malnoe 10-200 g Switzerland 07.06.2006 2200 m 07.07.2006 10.08.2006 29.09.2006

Sedum acre L. whole plant Leysin, 09.08.2006 E. Anchisi 58 g Switzerland 1200 m

Sedum album L. whole plant Lausanne, 18.06.2006 J. Kissling 5 g Switzerland 550 m

Sedum atratum L. whole plant Sanetsch, 07.1996 E. Anchisi 115 g Switzerland 2000 m

Sedum dasyphyllum L. whole plant Fully, 21.06.2006 E. Anchisi 14 g (for screening) Switzerland 14.06.2008 E. Anchisi 315 g (for isolation) 450 m

Sedum sarmentosum whole plant Lausanne, 18.06.2006 J. Kissling 7.5 g Bunge Switzerland 600 m

Sedum sexangulare L. whole plant Orsières, 18.06.2006 E. Anchisi 14 g Switzerland 1100 m

139 5. EXPERIMENTAL PART

5.1.2. Secondary metabolites extraction

Immediately after collecting, the plants were washed to take off the mud and cut in pieces of maximum 2 cm. The cut plants were frozen and placed in vacuum in order to remove water by sublimation with a lyophiliser (Lyophiliser Alpha 2-4D, Martin Christ GmbH, Osterode an Harz, Germany) in order to avoid mould contamination which could occur if dried at ambient temperature. Once dried, the plants were ground in liquid nitrogen to prevent the deterioration of thermolabile compounds that may occur during the grinding process. The milling allows to maximise the exposition of the plant cells to the solvent, therefore increasing the surface of extraction and improving the solubilisation of the secondary metabolites. Successive macerations were then undertaken at room temperature with solvents of increasing polarity, leading to the dichloromethane (DCM) and methanol (MeOH) extracts. The lipophilic secondary metabolites were first extracted in DCM, with about 1 L per 200 g of dry material, three times during 24 hours, and the compounds of medium and high polarity were extracted in MeOH three times with the same proportion of solvent and during the same time. A direct water extract of R. rosea was obtained by maceration of 5 g of freeze-dried ground roots in water at room temperature during 24 hours. After filtration, the enriched solvents were evaporated to dryness by rotary evaporation (Büchi Rotavapor) at 40°C, leaving semisolid extracts. These crude extracts were then frozen and lyophilised.

Another extraction method was performed on the plant material used for the chemical profile dynamics of R. rosea plants. Different extraction procedures were carried out in order to determine the best extraction method for rosavins and salidroside. Cut and powdered R. rosea roots were extracted with different volumes (1, 1.5 and 2 ml) of various methanol concentrations in water (0, 50, 60 and 100%) during different extraction times (30, 60 and 90 min) in an ultra-sonic bath. The procedure was monitored by HPLC-UV method. The extraction of freeze-dried powdered root (10 mg) during 60 minutes with 1 ml of 60% methanol was finally selected as the extraction method. The extractions were performed in Eppendorf vials, followed by centrifugation at 10’000 rpm during 5 minutes. The supernatant diluted 10 times in methanol 6% was used for the analyses.

5.1.3. Liquid-liquid extraction (LLE)

In order to pre-purify raw extract, liquid-liquid extractions (LLE) were performed on DCM and MeOH extracts. Usually a small scale test was carried out with 1 g of extract. Fractions were injected

140 5. EXPERIMENTAL PART

on HPLC to be analysed. Once the small scale test was validated, a larger amount of extract was extracted. The DCM extract of R. rosea (20 g) was dissolved in 400 ml of MeOH, then hexane was added to 800 ml. Both phases were then shaken together, and the hexane was recovered after decantation. This step was repeated three times, and all the phases of the same solvent were collected together and evaporated to dryness by rotary evaporation. Therefore, 6.6 g of MeOH fraction and 13.2 g of hexane fraction were obtained.

The MeOH extract of S. dasyphyllum (15 g) was partitioned between EtOAc and H 2O (1 L EtOAc and

300 ml H 2O). After decantation and separation of the two phases, the aqueous fraction was then partitioned with water-saturated n-BuOH (1 L). Phases were dried-evaporated by rotary evaporator at

40°C. BuOH and H 2O phases were then suspended or dissolved in H 2O, frozen and lyophilised.

5.2. ANALYTICAL CHROMATOGRAPHIC METHODS

5.2.1. Thin layer chromatography (TLC)

Thin layer chromatography (TLC) has so far been applied more than any other chromatographic technique. This is due to three main reasons; the time required for the identification of most characteristic constituents of a drug by TLC is very short, this method gives semi-quantitative information on the main constituents of a drug or drug preparation, and the chromatographic drug fingerprint provided by TLC is suitable for monitoring the identity and purity of drugs, and for detecting adulterations and substitutions (Wagner and Bladt, 1996). Because of those advantages, TLC has become widely adopted for rapid analyses of drugs and drug preparations. Besides, it is the method of choice for routine phytochemical analyses of crude extracts, fractions and pure compounds. It allows chemical, biochemical, and biological screenings of extracts and pure compounds, giving thus additional information.

In this work, TLC was carried out on commercially available pre-coated silicagel 60 F 254 aluminium sheets (Merck, Darmstadt, Germany). They were treated with a fluorescent compound in order to make the UV detection of compound at 254 nm easier. The TLC plates were developed in twin trough Camag (Muttenz, Switzerland) chromatographic tanks saturated with the appropriate eluent. The solvent systems, often consisting of a binary or tertiary mixture, were adapted to the specific needs of the analysis. However, standard conditions (Table 5-2) were employed systematically to test biological activities on extracts. No acid was utilised, since it could interfere with the biological assay. TLC was also performed to identify sugars from flavonoids hydrolysis. They were analysed with standards

141 5. EXPERIMENTAL PART

(rhamnose, galactose, glucose, mannose, and glucuronic acid, 1 mg/ml) on the same plate with a solvent system iso -PrOH:MeOH:H2O (7:2:1).

Table 5-2 Standard TLC conditions used for the separation of extracts and fractions

Extract types Solvent system Proportions

DCM extracts Hexane : EtOAc 1:1

MeOH and H 2O extracts CHCl 3 : MeOH : H 2O 65:55:5

In order to obtain sharply resolved zones, the quantity of material applied to the chromatogram was usually 50 to 100 µg of extracts and 5 to 10 µg of pure compounds, except for the AChE inhibition assay where 10 to 20 µg of extracts and 1 µg of pure compounds were spotted. After migration, the detection of the main compounds of a drug was carried out by the observation of the extinction of the fluorescence at 254 nm induced by the added compound, and of the appearance of fluorescence at 366 nm. Thereafter, revelation was prompted by spraying an appropriate chemical, biochemical or biological reagent; the compounds appeared then as spots of striking colours.

5.2.2. High performance liquid chromatography coupled to ultraviolet photodiode array detector (HPLC-UV-DAD)

High performance liquid chromatography (HPLC) is one of the most appreciated techniques in modern phytochemical analysis. The separation of compounds from complex mixtures is based, as for the TLC method, on the selective distribution of analytes between a liquid mobile phase and a solid stationary phase. The considerable advantages of HPLC compared to TLC are the higher resolution of the separation, the sensitive decrease of the detection limit, and the possibility of automation of the procedure. HPLC was used routinely to “pilot” the preparative isolation of secondary metabolites (optimisation of the experimental conditions, checking of the different fractions throughout the separation) and to control the final purity of the isolated compounds. A conventional HPLC system consists of a pump that delivers the high pressure solvent, an injector that can be either manual or automatic (autosampler), a column where the separation takes place, a

142 5. EXPERIMENTAL PART

detector and an interface (integrator or computer software) to visualise the separation (Niessen, 1999). Currently, one of the most widely used detectors coupled to HPLC is the diode array detector (DAD). The development of DAD in the early 1980s made a significant change in relation to classical UV detectors. In fact, the DAD adds a third dimension (the wavelength) besides time and absorption. This allows access to much information, such as peak purity and identity (Huber and George, 1993). In this study, analytical HPLC was used to quantify rosavins and salidroside in R. rosea extracts, to analyse raw extracts and fractions, to guide and optimize the separation and isolation processes using medium pressure liquid chromatography (MPLC), low pressure liquid chromatography (LPLC) and semi-preparative HPLC, and to check the purity of the isolated compounds. The HPLC system used for the different analyses consisted in a HP-1100 system (Hewlett Packard, Palo Alto, CA, USA) equipped with a binary pump, a photodiode array high speed spectrophotometric detector (DAD) and an autosampler. The various components of these two HPLC systems were controlled by Agilent ChemStation Rev. 10.02 software. The HPLC conditions routinely used are reported in Table 5-3. In some cases, these conditions were adapted to the specific requirements of the analyses.

Table 5-3 General HPLC conditions for the analyses of R. rosea and S. dasyphyllum extracts

Rhodiola rosea Rhodiola rosea Sedum dasyphyllum

Hydro-alcoholic extract DCM extract MeOH extract

® ® ® Column Atlantis dC 18 Symmetry C 18 Symmetry C 18 (Waters, Milford, MA, USA) (Waters, Milford, MA, USA) (Waters, Milford, MA, USA)

Column size 3.0 x 150 mm i.d.; 3 µm 4.6 x 250 mm i.d.; 5 µm 4.6 x 250 mm i.d.; 5 µm

Pre-column size 4.6 x 20 mm i.d.; 3 µm 4.0 x 9 mm i.d.; 5 µm 4.6 x 20 mm i.d.; 5 µm

Solvent system MeCN:MeOH (1:1) + FA 0.1%- MeOH - water MeCN + FA 0.1% - water + FA water + FA 0.1% 0.1%

Gradient 1-99 to 100-0 in 45 min 2-98 to 100-0 in 30 min 2-98 to 50-50 in 35 min (4 elution steps)

pH 2 Neutral Neutral

Temperature 37°C Ambient Ambient

Flow rate 500 µl/min 1 ml/min 1 ml/min

Detection 210 and 254 nm 210, 254 and 366 nm 210, 254, 366 and 520 nm (DAD) (scan 200-500) (scan 200-500) (scan 200-500)

143 5. EXPERIMENTAL PART

5.2.3. High performance liquid chromatography coupled to mass spectrometry (HPLC-MS)

The combination of high performance liquid chromatography (HPLC or LC) and mass spectrometry (MS) offers the possibility of profiting from the joint of both methods, respectively the high selectivity and separation efficiency of HPLC, and the powerful and sensitive detection of MS, with additional structural information for the identification (Niessen, 1999). Nowadays, MS detection is generally used coupled to HPLC and in combination with DAD detection, resulting in an extremely powerful analytical tool. The large amount of spectral data obtained can make of this tool a hyphenated technique. In this study, LC-UV-MS was employed for the qualitative analysis of the different extracts. The mass spectrometer was coupled to an HP 1100 HPLC system (Hewlett-Packard, Palo Alto, CA, USA) equipped with a binary pump, a UV/DAD detector and an autosampler. A Finnigan MAT (San Jose, CA, USA) ion trap (IT) mass instrument equipped with a Finnigan ESI was used. The flow rate was set to 150 µl/min. MS analyses were performed with an ESI source. The transfer capillary temperature was set to 200°C and the capillary voltage to -47 V in negative mode or 38 V in positive mode; the pressure of the sheath gas (N 2) was 70 arbitrary units. The ESI/MS system operated in the negative mode at 2.5 kV or in positive mode at 3 kV. Full scan MS spectra were measured from m/z 120 to 1000. Source induced dissociation energy (SID) was used at 10 V. Collision induced fragmentation experiments were performed in the ion trap using helium as the collision gas. The collision energy was set at 35% (isolation width of 2 amu). The most abundant ion in full scan MS was selected to undergo fragmentation for MS/MS.

5.2.4. Ultra performance liquid chromatography coupled with mass spectrometry (UPLC-MS)

Ultra performance liquid chromatography (UPLC) has been developed in order to decrease both the time of analysis and the column volume. The system commercialised by Waters (Milford, MA, USA) is able to work up to 1000 bar. Small diameter columns are used to limit the frictional heating and the organic solvent consumption. It is worth noticing that this type of support needs small extra-column volumes due to detection, tubing, and injection volume. It allows to increase considerably the separation and the resolution of the chromatographic peaks compared with a classic HPLC system. Some criteria have to be fulfilled for this type of detection; the detection time constant should be fast

144 5. EXPERIMENTAL PART

enough because peak widths are very small, the injection cycle time must be fast, at least for a time of analysis under 1 or 2 min, and a small gradient delay volume is required (Nguyen et al. , 2006). The coupling of this technique with a high resolution mass spectrometer, such as a time-of-flight (TOF) analyser, is ideal to establish a correlation between fractions or extracts, or to perform dereplication. UPLC-TOF analyses were performed in this work to identify the HR-MS of pure isolated compounds. UPLC was carried out on an Acquity UPLC system (Water, Milford, USA) using an ACQUITY

® UPLC BEH C 18 column (50.0 x 2.1 mm i.d.; 1.7 µm) and a solvents gradient of MeCN+0.1 % formic acid:H 2O+0.1% formic acid (10:90 for 2.61 min, from 10:90 to 70:30 in 10.42 min, then 70:30 to 95:5 in 2.60 min). MS-TOF spectra were obtained on a Micromass LCT Premier (Waters) using electrospray as the ion source, in negative mode. The capillary voltage was set at 2.8 kV, the cone voltage at 40 V, the MCP detector voltage at 2650 V, the source temperature at 120°C, desolvatation temperature at 250°C, the con gas flow 10 L/h and the desolvation gas (N 2) flow at 550 L/h. For the lockmass, a 5 µg/ml solution of leucine-enkephalin (Sigma-Aldrich, Steinheim, Germany) was infused through the Lock Spray TM probe at a flow rate of 10 µl/min with the help on a second Shimadzu LC- 10ADvp pump (Duisburg, Germany). The Lock Spray TM frequency was set at 15 scans.

5.3. PREPARATIVE SEPARATION TECHNIQUES

5.3.1. Column chromatography

5.3.1.1. Adsorption chromatography

Conventional gravity-driven, open column chromatography is generally employed for the fractionation of crude extracts or complex fractions. In this case, the MeOH fraction of the DCM extract of R. rosea was first separated by adsorption chromatography. The size of the column, the granulometry of the solid phase, the flow rate of the mobile phase and the size of the fractions obtained were adapted to the quantity and the nature of the samples. A 56 x 4 cm i.d. column filled with silicagel 60 (40-63 µm, Merck, Darmstadt, Germany) was used. The choice of the elution conditions, the follow-up of the separation and the final gathering of the fractions were performed on the basis of the TLC analyses. The sample was introduced in solid form, having first been mixed with 2.5 or 4 times its weight of solid phase. A stepwise gradient eluent with hexane:EtOAct:MeOH from 95:5:0 (v:v:v) to 0:5:95 was used with a flow rate of about 5 ml/min. The fractions (100 ml) were collected manually.

145 5. EXPERIMENTAL PART

5.3.1.2. Gel filtration

Separation of some S. dasyphyllum fractions was carried out with exclusion chromatography (or gel filtration) on a Pharmacia Sephadex ® LH-20 phase with 30 x 1 or 70 x 2 cm i.d. columns. The size of the column depended of the sample weight. The sample was dissolved in MeOH and then introduced on the top of the column. The chromatography was performed in the gel filtration mode, using a single solvent (MeOH), therefore compounds were separated according to their size and those with molecular weight greater than 4000 Da were not retained. The flow rate was set at about 0.5 ml/min. The fractions (10 ml) were collected manually. The follow-up of the separation and the final gathering of the fractions were performed on the basis of the TLC analyses.

5.3.2. Preparative pressure liquid chromatography

The term covers those techniques of column chromatography in which pressure applied by a pump operating above 2 bar pressure. Preparative, in this context, refers to amounts ranging from micrograms to kilograms. The division between low- (up to 5 bar), medium- (5-20 bar), and high pressure (>20 bar) liquid chromatography is not simply arbitrary but reflects the use of different columns with different size packing material and size of sample that can be fractionated.

5.3.2.1. Low pressure liquid chromatograph (Lobar®)

The most popular preparative low-pressure LC system is the Lobar range (Merck, Germany). Lobar columns (ready filled columns made of glass) can accommodate separation of gram quantities with resolution sometimes approaching those of HPLC (Hostettmann et al. , 1998). Low pressure liquid chromatography (LPLC) was used to purify enriched CPC fractions G-2 and G-8 of the MeOH extract of R. rosea . The system used was composed of a ProMinent (Heidelberg, Germany) Duramat ® pump with a Pharmacia LKB 2238 Uvicord SII UV detector with fixed wavelength (210 or 254 nm) and an LKB 2210 recorder. The column used was a Lobar ® Lichroprep ® RP-18 (40-63 µm, 310 x 25 mm i.d., Merck, Darmstadt, Germany) with an adapted stepwise gradient eluent MeOH:H 2O. The samples were dissolved in methanol, and introduced after the elimination of the possible insoluble part by centrifugation. The fractions were collected by a Pharmacia fraction collector.

146 5. EXPERIMENTAL PART

5.3.2.2. Medium pressure liquid chromatography (MPLC)

This technique involves longer column with large internal diameters and requires higher pressure than LPLC to enable sufficiently high flow-rates. Higher resolution and shorter separations times are obtained (Hostettmann et al. , 1998). The equipment used consisted of a Büchi (Flawil, Switzerland) B-681 chromatography pump, a Knauer (Berlin, Germany) K 2001 UV detector, a Pharmacia (Bromma, Sweden) LKB Rec 1 recorder

® and an automatic Büchi B-684 fraction collector. A LiChroprep C 18 stationary phase (40-63 µm, Merck, Darmstadt, Germany) was packed in a pressure-resistant column (400 x 50 mm i.d.). The column size was chosen according to the amount of sample (about 5 g). Prior to loading, the column was washed with methanol and then equilibrated with the initial eluent. The samples were introduced in solid form. They were dissolved in a suitable solvent (MeOH) and were mixed with about 2 to 4 times their weight of stationary phase. The solvent was then removed under reduced pressure using a rotary evaporator. The dry powder was loaded into a small introduction cartridge that was connected to the main column. The choice of the solvent system was performed by analytical HPLC and transposed directly to MPLC. As the pump worked only in isocratic mode, the gradient mixture

(MeOH:H 2O or MeCN:H 2O + FA 0.1%) was adapted step by step, with an increase of 1-5 % of the organic solvent at each step, from 5:95 to 70:30. The maximal pressure was set to 20 bars, the flow rate was about 8-10 ml/min and the separation was monitored with a wavelength fixed at 254 nm. MPLC was used for the separation of the water extract of R. rosea and different fractions of the MeOH extract of S. dasyphyllum .

5.3.2.3. Semi-preparative high performance liquid chromatography

The term semi-preparative high performance liquid chromatography (semi-prep) is conventionally applied to columns with an internal diameter (i.d.) of 8 to 10 mm and often packed with 10 µm particles. This technique is useful for the separation of 1 mg to 100 mg of mixtures, and is thus widely applied for the final isolation steps (Hostettmann et al. , 1998). In the present work, semi-prep was used in order to purify compound DCM1 from a fraction obtained by open silica column. The tools consisted of an LC-8A pump, equipped with an SPD-10A VP system controller, an SIL 10AP autoinjector, and an SPD-10A VP UV-Vis detector, all obtained from

Shimadzu (Kyoto, Japan), in combination with a Symmetry C 18 column (19 x 150 mm i.d., 7 µm). The solvent system used was MeOH:H 2O (30:70) in isocratic mode. The flow rate was set to 10 ml/min and the injected amounts ranged from 5.0 to 10.0 mg.

147 5. EXPERIMENTAL PART

The fraction B26, obtained from MPLC of the BuOH fraction of S. dasyphyllum, revealed three close major peaks in HPLC-UV analysis. As the separation of these peaks was not possible with semi- preparative columns, they were purified on an analytical column ( SS 250 / 0.5"/ 10 Nucleosil 7-C8). The separation was performed on an HP-1100 system (described in Chapter 5.2.2) using the solvent system MeCN:H 2O + FA 0.1% in gradient 15:85 to 30:70 in 30 min. The flow rate was set to 3 ml/min. The injection volume was of 100 µl. The fractions were collected manually.

5.3.3. Centrifugal partition chromatography (CPC)

Centrifugal partition chromatography (CPC) is now widely accepted as a routine preparative technique in both industrial and university laboratories. In the field of natural products, crude extracts and semi- pure fractions can be successfully chromatographed, with sample sizes ranging from milligrams to grams. This technique relies on the partitioning of compounds between two immiscible liquid phases. As there is no solid support, the separation is determined only by the different partition coefficient of the solutes (Hostettmann et al. , 1998). CPC was performed on the MeOH extract and on a MPLC fraction of the water extract of R. rosea roots. It was realised on a High Speed Counter Current Chromatograph CCC-1000 apparatus (PTR, Baltimore, USA) of 300 ml of capacity, and with a rotation rate of 1000 rpm. The mobile phase was delivered by a Waters pump with a flow rate of 3 ml/min. The mobile and stationary phases chosen for the MeOH extract consisted of a quaternary biphasic system CHCl 3:MeOH: n-BuOH:H 2O (7:6:3:4). The lower phase of the system was first used as mobile phase; afterwards, during the fractionation, the sense of the rotation was inversed and the upper phase became the eluent. The biphasic solvent system for the MPLC fraction 4 of the water extract was CHCl 3:MeOH:isopropanol:H 2O (5:6:1:4). The lower phase was first the mobile phase. The separations were monitored at 254 nm by a Knauer UV detector, and recorded by a W+W 600 recorder (Tarkan). The fractions were collected by an automatic collector Ultrorac 2070 (LKB), and gathered according to the UV trace and TLC analyses.

5.3.4. Solid phase extraction (SPE)

Solid phase extraction is a solid-liquid chromatographic technique using a cartridge filled with stationary phase, which allows a preliminary purification of the extracts. Elution from the cartridge is performed by vacuum. The advantages of this technique are that it is very quick, and also cheap in spite of the unique use made of each cartridge. In the present work, this technique was used to remove tannins from the water extract of R. rosea.

148 5. EXPERIMENTAL PART

A Chromabond Polyamid SPE column (Machery Nagel) was first conditioned with 6 ml of MeOH, in order to free the chains of the stationary phase. The SPE column was then conditioned with 6 ml of MeOH 5% in water. A sample of 150 mg of water extract was deposited onto the cartridge. A first fraction with 6 ml of MeOH 5% was discarded. Then a fraction with 6 ml of MeOH 90% was eluted, dried and used for following separations.

5.4. PHYSICO-CHEMICALS METHODS

5.4.1. Optical rotation ([ααα]D)

The optical rotation [ α]D of the compounds containing one or more asymmetrical carbon atoms was determined with a Perkin-Elmer-241 MC polarimeter (Wellesley, MA, USA). The rotation α of the polarised light by the products dissolved in methanol was measured in a 10 cm long tank at room temperature. The yellow D line (589.3 nm) of a sodium lamp was used as source of incident light. The specific rotation [ α]D [°] was thus defined as follows: α ⋅1000 []α T = D c ⋅l

α = observed rotation [°] (mean on 30 measurements), T = temperature. l = cell length [dm]. c = concentration of the compound [g/l].

5.4.2. Ultraviolet spectrophotometry

The UV spectra of the purified compounds were recorded in MeOH on a Perkin-Elmer Lambda 20 spectrophotometer (Wellesley, MA, USA). The UV measurements were performed in quartz cells. The results were reported in λmax nm (log ε), where λmax represented the wavelength of the maxima, and (log ε) represented the extinction coefficient at this wavelength according to the Lambert-Beer law:

A = ε ⋅ c ⋅ l

A = absorbance c = molar concentration [mol/l] ε = extinction coefficient l = cell length [cm]

149 5. EXPERIMENTAL PART

5.4.3. Nuclear magnetic resonance spectrometry (NMR)

The nuclear magnetic resonance spectrometry (NMR), chosen as main analytical method for the structural elucidation of the isolated compounds, is based on the phenomenon that occurs when the nuclei of certain atoms are immersed in a static magnetic field and exposed to a second, oscillating, magnetic field. Each element with a permanent nuclear magnetic moment, like 1H, 13 C, 14 N, 17 O, 19 F, or 31 P is able to produce a signal. Under appropriate conditions, the analyte can absorb electromagnetic radio-waves radiations at frequencies depending on its characteristics (Vollhardt and Schore, 1995). The nuclei of greatest interest, in this work, were mainly protons ( 1H) and carbons ( 13 C), as their resonances are the most important for the identification of natural organic molecules. The 1H and 13 C NMR spectra were recorded on a Varian Inova 500 spectrometer (Varian, Palo Alto, CA, USA) at 500 and 125 MHz, respectively. The instrument was controlled by a Solaris VNMR software of the same manufacturer, installed on a Sun workstation (Santa Clara, CA, USA). All NMR measurements were performed at 26°C in deuterated methanol (CD 3OD-d4) or in deuterated chloroform (CDCl 3-d) or at 30°C in deuterated DMSO (DMSO-d6) (Armar Chemicals, Döttingen, Switzerland), depending on the solubility of the compound. Tetramethylsilane (TMS) was used as 1 internal standard for H spectra, while the CD 3OD -d4 shift (49.0 ppm), CDCl 3-d shift (79.2 ppm) or 13 DMSO-d6 shift (39.7 ppm) were used as reference for C spectra. In order to observe homo- and heteronuclear correlations between the proton and carbon atoms of the analyte, complementary bidimentional (2D) experiments were performed. These 2D experiments, including gDQF-COSY, gHSQC, gHMBC and gNOESY, were run using standard pulse sequences provided in the original VNMR software.

5.5. CHEMICAL METHODS

5.5.1. Reagents for TLC detection

The detection of the various compounds present on the TLC plates was performed by various means after development and evaporation of the mobile phase. Visual observations were first carried out in daylight in order to detect colored substances, and then under UV light at 254 nm (fluorescence extinction) and 366 nm (own fluorescence). The use of chemical reagents in solution sprayed on the TLC plates complements the visual observations. A specific reagent could point out a compound or a class of compounds, while a polyvalent reagent is particularly adapted to regroup fractions after preparative or semi-preparative separations. Two main reagents were used during this work:

150 5. EXPERIMENTAL PART

• Godin’s reagent (Godin, 1954): polyvalent reagent. The plate was first sprayed with a solution of equivalent volumes of 1 % vanillin in ethanol and 3 % perchloric acid in water. Then, the dried layer was covered with an ethanolic solution of 10 % sulfuric acid and heated to about 100°C during 5 min. Different colorations appeared in daylight. • NST-PEG (Naturstoff-Polyethylenglycol) (Wagner and Bladt, 1996): specific reagent (flavonoids). The TLC was sprayed successively with a solution of 1% of diphenylboryloxyethylamine in MeOH (NST) and a solution of PEG 4000 5% in EtOH. Flavonoids appeared as yellow, orange, or blue fluorescent spots at 366 nm. Another reagent with diphenylamine was used to reveal the sugars (Wagner and Bladt, 1996). The TLC was sprayed with a solution of diphenylamine (0.5 g) with aniline (0.5 ml) and o-phosphoric acid 85% (2.5 ml) in 25 ml of MeOH. The plate was then heated to about 100°C during 1 hour. The sugars appeared as blue or grayish spots.

5.5.2. DPPH free radical scavenging assay on TLC

A TLC autographic assay of radical scavenging activity using 1,1-diphenyl-2-picrylhydrazyle (DPPH) radical was employed for extract screening and bio-guided isolation of active metabolites.

DPPH is a stable radical, with purple colour when in solution ( λmax 517 nm) and will discolour when reduced by a free radical scavenger (Cuendet et al. , 1997). The test was performed on TLC plates, where 100 µg of crude extracts or fractions or 10 µg of pure compounds were spotted. After elution, the plates were dried and sprayed with 0.2 % DPPH (Sigma- Aldrich, Buchs, Switzerland) methanolic solution. After an optimal reaction time of about 30 minutes, the active compounds clearly appeared as yellow spots on a purple background.

5.5.3. DPPH free radical scavenging microplate assay

The capacity of tested compounds to scavenge the stable radical DPPH was determined spectophotometrically by measuring the decolourisation of DPPH upon reduction by tested compounds. The method reported by Lee et al . (2003) was applied with some modifications. The assay was carried out at room temperature in clear polystyrene flat bottom 96-well plates (Millian, Geneva, Switzerland) and all the spectrophotometric measurements were recorded with a Bio-Tek EL 808 microplate absorbance reader and KC4 v3.3 software (Bio-Tec Instruments Inc., Winooski, USA). The samples dissolved in MeOH were added to EtOH containing 96-well microtitre plates to make up a total volume of 30 µl in each well with final sample concentrations from 0.5 to 250 µg/ml (10

151 5. EXPERIMENTAL PART

dilutions). The reaction was initiated by addition of 90 µl of DPPH 200 µM, the mixture was shaken vigorously for 2 sec and incubated for 30 minutes at room temperature. The scavenging activity of a tested compound was measured as the decrease in DPPH absorbance at 490 nm and expressed as a percentage of the absorbance of a control DPPH solution without the tested compound. The EC50 stands for the effective concentration required for 50% scavenging activity.

5.5.4. Acid hydrolysis of flavonoids and sugar identification

Some isolated flavonoids were hydrolysed in order to analyse the aglycone and the sugars individually. A small quantity of flavonoids (5 mg) was dissolved in 15 ml of HCl 2N at 100°C for 1 hour. After cooling, the mixture was extracted three times with the same volume of EtOAc. The organic phase was washed with H 2O until neutralisation. The aglycone was obtained from this fraction after removal of solvent under vacuum. The aqueous solution was neutralised by adding 10% NaHC0 3. The sugars, after removal of water, were recuperated in pyridine and analysed by TLC with standards.

5.6. BIOLOGICAL METHODS

5.6.1. Acetylcholinesterase inhibitor TLC bioassay

All the crude extracts were submitted to a biautographic test on TLC which detects anti- acetylcholinesterase agents (Marston et al. , 2002). The bio-guided isolation of active compounds from the DCM extract of R. rosea was monitored by this assay. AChE from electric eel (1000 U) purchased from Sigma Chemical Co. (St. Louis, MO, USA) was dissolved in 150 ml of 0.05 M Tris-hydrochloric acid buffer at pH 7.8. Bovine serum albumin (150 mg) (Merck, Darmstadt, Germany) was mixed with this solution in order to stabilize the enzyme during the bioassay. The developed and well-dried TLC plates were sprayed with the solution and incubated in a humid atmosphere at 37°C for 20 min after being dried again. A freshly prepared mixture of 1 part 1-napthyl acetate (Merck, Darmstadt, Germany; 2.5 mg/ml in ethanol) and 4 parts Fast Blue B salt (Fluka, Buchs, Switzerland; 2.5 mg/ml in water) was then applied onto the incubated plates. A purple coloration appeared on the whole surface after 1 or 2 min, due to the formation of an azo dye between Fast Blue B salt and 1-naphthol obtained by hydrolysis of 1-naphthyl acetate by AChE. Active compounds (AChE inhibitors) were localized by colourless spots. In general, 10 to 20 µg of crude extract and 1 to 0.001 µg of pure compounds were spotted on the layer. Galanthamine obtained from Sigma (Buchs, Switzerland) was used as positive control.

152 5. EXPERIMENTAL PART

5.6.2. Acetylcholinesterase inhibitors microtitre plate bioassay

AChE inhibitory activity of the isolated compounds was evaluated on microtitre plate according to the method described by Ellman (1961). All measurements were performed with a microplate reader (EL 808, BioTek Instruments, Winooski, USA), in conventional flat-bottom polypropylene microtitre plates (Millian, Geneva, Switzerland). Briefly, the reaction mixture consisted of 100 µl of 0.1M phosphate buffer (pH 7.8), 20 µl of a solution of AChE (Sigma Chemical Co., St. Louis, MO, USA) (final concentration 0.05 U/ml in 0.1 Tris– hydrochloric acid buffer (Sigma, Buchs, Switzerland), pH 7.8), 20 µl of test compound (samples were dissolved in 0.1 M phosphate buffer containing 45% MeOH). The controls contained the corresponding volume of buffer instead of test compound solutions. The mixture was shaken and incubated at 37°C during 15 min in a 96-well microplate. Then, 40 µl of Ellman’s reagent (0.75 mM final concentration of 5,5’-dithiobis-(2-nitrobenzoic acid) (DTNB, Sigma) in 0.1 M phosphate buffer pH 7.8) followed by 20 µl of 1.5 mM acetylthiocholine iodide (ATCI, Sigma) solution to initiate the reaction. The hydrolysis of ATCI was monitored by the formation of a yellow 5-thio-2-nitrobenzoate anion as a result of the enzyme-catalyzed reaction of DTNB with thiocholine, using a microplate- reader at a wavelength of 405 nm. The rates of reaction (Mean V) were obtained over 180 sec, with an 18 sec interval, shaken before every reading. The reaction system where the tested solution was replaced by equivalent MeOH-buffer solution was used as blank control. The percentage of inhibition was determined by comparison of rates of reaction of samples relative to the blank sample. Inhibition

% = (1-Vtest /V blank ) ×100 %.

Statistical Analysis All assays were done in triplicate. All data are expressed as mean ± standard deviation for the number of experiments. The concentration of tested compounds, which is required to inhibit 50% of the enzyme activity under the assay conditions, was determined from dose-response curves and defined as the IC 50 value. The dose-response curves were obtained and calculated with GraphPad Prism software (Version 5, GraphPad software Inc., San Diego, CA, USA).

153 5. EXPERIMENTAL PART

5.6.3. Monoamine oxidase inhibitor microtitre plate bioassay

The microtitre plate assay was conducted in the LCT-Pharmacochemistry laboratory, School of Pharmaceutical Sciences, University of Geneva, Switzerland. The crude extracts of R. rosea were submitted to monoamine oxidase (MAO) microplate assay. Moreover, the bio-guided isolation of the active metabolites from the methanol and water extracts was monitored with this assay. Human MAO A and MAO B Supersomes TM , purchased from BD Gentest (Woburn, MA, USA), are mitochondrial membrane fractions of insect cells containing human recombinant MAO A and B. MAO inhibition assays were carried out with a fluorescence-based method (end-point reading) adapted from a standard BD Gentest protocol. The substrate used for the assay was kynuramine, which is non-fluorescent until it undergoes oxidative deamination by MAO resulting in the fluorescent metabolite 4-hydroxyquinoline (Novaroli et al. , 2005). Product formation was quantified by comparing the fluorescence emission of the samples to that of known amounts of authentic metabolite 4-hydroxyquinoline. Reactions were performed in black, flat bottom polystyrene 96-well microtitre plates with enhanced assay surface (FluoroNunc/LumiNunc, MaxiSorpTM Surface, NUNCtM, Roskild, Denmark) using a final volume of 200 µl. The wells containing 140 µl of potassium phosphate buffer (0.1 M, pH 7.4, made isotonic with KCl), 8 µl of an aqueous stock solution of kynuramine (0.75 M to get a final concentration corresponding to its Km value), and 2 µl of the sample solution (final concentration of 1% v/v), were preincubated at pH 7.4, 37°C for 10 min. As positive control, 2 µl of pure DMSO were used in place of the inhibitor solution. Diluted MAO (50 µl) was then delivered to obtain a final protein concentration of 0.015 mg/ml in the assay mixture. Incubation was carried out at 37°C and the reaction was stopped after 20 min by addition of 75 µl of NaOH (2 N). Fluorescence emission at 400 nm was measured with a 96-well microplate fluorescent reader (FLx 800, Bio-Tek Instruments, Inc., Winooski, VT, USA). Inhibition measure of the extracts and fractions were done in duplicate since only an approximate measure of the inhibition was necessary for the bio-guided fractionation. Extracts and fractions were tested at a final concentration of 100 µg/ml while the purified compounds were at 10 -5 M, in DMSO. Data analysis was performed with Prism 4.0 (GraphPad Software Inc., CA,

USA). The degree of inhibition IC 50 was assessed by a sigmoid dose-response curve. The standard deviation was calculated for the sigmoid regression.

154 5. EXPERIMENTAL PART

5.8. T PHYSICAL CONSTANTS AND SPECTRAL DATA FOR THE 1

ISOLATED COMPOUNDS TTP PTT

5.8.1. Compounds isolated from the DCM extract of Rhodiola rosea L.

Compound DCM1 Structure 2' 2'' 1' 2 O 3' 1'' 1

O 4' HO Chemicalname 2(4’Hydroxyphenyl)ethylacetate Trivialname Tyrosolacetate(Grasso etal. ,2007)

Formula C10B HB 12B OB 3B B ExactMass 180.0780Da Molecularweight 180.2039Da Aspect Whiteamorphoussolid λ UV maxB B 226,278nm 1 P HNMRdataP SeeTable54 13 P CNMRdataP SeeTable54

P HRESIMS m/z 179.0698(C10B HB 11B OB 3B [MH]B P requires179.0702)

1 TP PT For known compounds, UV maxima were measured from the DAD/UV detector, and [ α ] DB B not measured.

155 5. EXPERIMENTAL PART

Compound DCM2 Structure 3 2' 1

1' 3' 2 OH

4' Chemicalname (2 E)3phenylprop2en1ol Trivialname Cinnamylalcohol(Wiedenfeld etal. ,2007)

Formula C9B HB 10B OB ExactMass 134.0726Da Molecularweight 134.1783Da Aspect Deliquescentwhitecrystallinesolid λ UV maxB B 204,252nm 1 P HNMRdataP SeeTable55 13 P CNMRdataP SeeTable55

P HRESIMS m/z 133.0688(C 9B HB 9B O[MH]B P requires133.0691)

156 5. EXPERIMENTAL PART

Compound DCM3 Structure 29

21 28

22 18 20 23 24 27 12 17 25 11 13 19 16 14 26 15 1 9

2 10 8

3 7 4 5 6 HO Chemicalname Stigmast5en3ol,(3 β) Trivialname ()βsitosterol(Goad,1991)

Formula C29B HB 50B OB ExactMass 414.3856Da Molecularweight 414.7181Da Aspect Whiteamorphoussolid λ UV maxB B 198nm 1 P HNMRdataP SeeTable56 13 P CNMRdataP SeeTable56

P HRESIMS m/z 413.3788(C 29B HB 49B O[MH]B P requires413.3793)

157 5. EXPERIMENTAL PART

Compound DCM4 Structure O

12 1 9 HO Chemicalname 9,12Octadecadienoicacid(9Z,12Z) Trivialname αLinoleicacid(Batchelor etal. ,1974)

Formula C18B HB 32B OB 2B B ExactMass 280.2396Da Molecularweight 280.4525Da Aspect Transparentoil λ UV maxB B 236nm 1 P HNMRdataP SeeTable57 13 P CNMRdataP SeeTable57

P HRESIMS m/z 279.1050(C 18B HB 31B OB 2B [MH]B P requires279.1058)

158 5. EXPERIMENTAL PART

Table 5-4 1 13 P HP and P CP NMR spectral data of compound DCM1

DCM1 1 13 No P HP NMR P CP NMR 1 4.19(2H ,d , J=15.6) 65.5 2 2.81(2H, d, J=15.6) 34.2 1’ 129.9 2’ 6.71(1H, m) 115.1 3’ 7.03(1H, m,) 130.9 4’ 156.0 5’ 7.03(1H, m) 130.9 6’ 6.71(1H, m,) 115.1 1’’ 172.0 2’’ 1.99(3H ,s ) 20.0 1P 13 P H and C spectra recorded at 500 MHz and 125 MHz, respectively, in CD B3BOD ,B Busing TMS as internal standard, δ values given in ppm, J values in Hz

Table 5-5 1 13

P P P H and P C NMR spectral data of compound DCM2

DCM2 1 13 No P HP NMR P CP NMR 1 4.22(2H ,d , J=5.6) 62.1 2 6.36(1H, dt , J=5.6,16.1) 129.1 3 6.59(1H, d, J=16.1) 130.9 1’ 137.0 2’ 7.29(1H, m) 128.5 3’ 7.38(1H, m) 126.3 4’ 7.20(1H, m) 130.9 5’ 7.38(1H, m) 126.3 6’ 7.29(1H, m) 128.5

1 13 H and C spectra recorded at 500 MHz and 125 MHz, respectively, in CD 3B BOD ,B Busing TMS as internal standard, δ values given in ppm, J values in Hz

159 5. EXPERIMENTAL PART

Table 5-6 1 13 P HP and P CP NMR spectral data of compound DCM3

DCM3 1 13 No P HP NMR P CP NMR 1 1.85(1H ,m ) 37.2 1.07(1H ,m ) 2 1.45(1H, m) 31.6 1.78(1H, m) 3 3.51(1H, m) 71.7 4 2.22(1H,m) 42.2 2.77(1H, m) 5 140.7 6 5.35(1H, m) 121.6 7 1.96(1H, m) 31.8 1.99(1H, m) 8 1.11(1H, m) 31.9 9 0.92(1H, m) 50.1 10 36.5 11 1.50(1H, m) 21.0 1.42(1H, m) 12 2.01(1H, m) 39.7 1.15(1H, m) 13 42.0 14 1.01(1H, m) 57.2 15 1.57(1H, m) 24.2 1.05(1H, m) 16 1.83(1H, m) 28.2 1.27(1H, m) 17 1.22(1H, m) 56.0 18 0.68(3H, s) 11.8 19 1.00(3H, s) 19.8 20 1.35(1H, m) 36.1 21 0.92(3H, m) 19.0 22 1.32(1H, m) 33.9 1.01(1H, m) 23 1.15(1H, m) 26.1 1.16(1H, m) 24 0.93(1H, m) 45.8 25 1.66(1H, m) 29.2 26 0.80(3H, m) 18.9 27 0.84(3H, m) 20.2 28 1.27 23.1 29 0.84(3H, m) 11.9 1H and 13 C spectra recorded at 500 MHz and 125 MHz, respectively, in CD 3B BOD ,B Busing TMS as internal standard, δ values given in ppm, J values in Hz

160 5. EXPERIMENTAL PART

Table 5-7 1 13 P HP and P CP NMR spectral data of compound DCM4

DCM4 1 13 No P HP NMR P CP NMR 1 177.5 2 2.27(2H, t,J =7.6) 34.0 3 1.60(2H, m) 26.2 4 1.34(2H, m) 30.2 5 1.34(2H, m) 30.2 6 1.34(2H, m) 30.2 7 1.34(2H, m) 32.6 8a 2.06(2H, m) 28.2 9b 5.34(1H, m) 130.9 10 b 5.31(1H, m) 129.1 11 2.78(2H, t,J =6.6) 125.1 12 5.31(1H, m) 129.0 13 5.34(1H, m) 130.9 14 a 2.06(2H, m) 28.1 15 1.34(1H, m) 30.3 16 c 1.34(1H, m) 32.6 17 c 1.31(1H, m) 23.6 18 0.90(3H, m) 14.4 1H and 13 C spectra recorded at 500 MHz and 125 MHz, respectively, in CD 3B BOD ,B Busing TMS as internal standard, δ values given in ppm, J values in Hz, a, b, and c interchangeable attributions

161 5. EXPERIMENTAL PART

5.8.2. Compounds isolated from the MeOH extract of Rhodiola rosea L.

Compound RR1 Structure HO 6'

O HO HO O 7 1 1' 8 OH 4

OH Chemicalname 2(4hydroxyphenyl)ethylβDglucopyranoside Trivialname Salidroside(Wiedenfeld etal. ,2007)

Formula C14B HB 20B OB 7B B ExactMass 300.1203Da Molecularweight 300.3091Da Aspect Whiteamorphoussolid λ UV maxB 222,278 nmB 1 P HNMRdataP SeeTable58 13 P CNMRdataP SeeTable58 HRESIMS m/z 345.1181 (CP 15B H21B O9B [M+HCOO] P requires345.1186)

162 5. EXPERIMENTAL PART

Compound RR5 Structure

HO 5'' O O 6' HO 1'' 4 O HO 8 O HO 1 HO 1' 9 7 OH Chemicalname βDGlucopyranoside,(2E)3phenyl2propen1yl6OαL arabinofuranosyl Trivialname Rosarin(ZapesochnayaandKurkin,1982)

Formula C20B HB 28B OB 10B B ExactMass 428.1676Da Molecularweight 428.4371Da Aspect Whiteamorphoussolid λ UV maxB B 204,252nm 1 P HNMRdataP SeeTable59 13 P CNMRdataP SeeTable59 HRESIMS m/z 473.1643(C21 H29 O12 [M+HCOO] requires473.1653)

163 5. EXPERIMENTAL PART

Compound RR7 Structure HO 6'

O HO 3 HO O 5 2 4 OH 1' C 1 N Chemicalname 2Butenenitrile,4(βDglucopyranosyloxy)2methyl,(2Z) Trivialname RhodiocyanosideA(Fan etal. ,2001;Yoshikawa etal. ,1995)

Formula C11B HB 17B NOB 6B B ExactMass 259.1050Da Molecularweight 259.2595Da Aspect Yellowamorphoussolid λ UV maxB B 208nm 1 P HNMRdataP SeeTable510 13 P CNMRdataP SeeTable510

P HRESIMS m/z 304.1017(C12B HB 18B NOB 8B [M+HCOO]B P requires304.1032)

164 5. EXPERIMENTAL PART

Compound RR8 Structure HO 6'

4 OH O HO 8 O HO 1

1' 9 7 OH Chemicalname βDGlucopyranoside,3(4hydroxyphenyl)2propen1yl Trivialname Triandrin(LinandChen,2004)

Formula C15B HB 20B OB 7B B ExactMass 312.1203Da Molecularweight 312.3202Da Aspect Yellowamorphoussolid λ UV maxB 206,264 nmB 1 P HNMRdataP SeeTable59 13 P CNMRdataP SeeTable59 HRESIMS m/z 357.1216(C16B HB 21B OB 9B [M+HCOO] P requires357.1186)

165 5. EXPERIMENTAL PART

Compound RR9 Structure OH

5'' O

O 6' HO 1'' 4 OH O HO 8 O HO 1

1' 9 7 OH Chemicalname βDGlucopyranoside,(2E)3phenyl2propen1yl6OαL arabinopyranosyl Trivialname Rosavin(KishidaandAkita,2005)

Formula C20B HB 28B OB 10B B ExactMass 428.1676Da Molecularweight 428.4371Da Aspect Whiteamorphoussolid λ UV maxB B 204,252nm 1 P HNMRdataP SeeTable59 13 P CNMRdataP SeeTable59

P HRESIMS m/z 473.1648(C 21B HB 29B OB 12B B [M+HCOO] P requires473.1653)

166 5. EXPERIMENTAL PART

Compound RR10 Structure 2 1 7 OH 3 8

4 HO Chemicalname Benzeneethanol,4hydroxy Trivialname Tyrosol(Takaya etal. ,2007)

Formula C8B HB 10B OB 2B B ExactMass 138.0675Da Molecularweight 138.1666Da Aspect Whiteamorphoussolid λ UV maxB B 222,276nm 1 P HNMRdataP SeeTable58 13 P CNMRdataP SeeTable58

P HRESIMS m/z 137.0658(C8B HB 9B OB 2B [MH]B P requires137.0665)

167 5. EXPERIMENTAL PART

Compound RR11 Structure HO 6'

4 O

HO 8 O HO 1

1' 9 OH 7 Chemicalname βDGlucopyranoside,(2E)3phenyl2propen1yl Trivialname Rosin(ZapesochnayaandKurkin,1982)

Formula C15B HB 20B OB 6B B ExactMass 296.1254Da Molecularweight 296.3209Da Aspect Whiteamorphoussolid λ UV maxB B 204,252nm 1 P HNMRdataP SeeTable59 13 P CNMRdataP SeeTable59 HRESIMS m/z 295.1244(C15B HB 19B OB 6B [MH]B P requires295.1248)

168 5. EXPERIMENTAL PART

Compound RR12 Structure HO 6'

O HO HO O 1 OH OH 1' 2 4 6 9

3 7 5

8 Chemicalname βDGlucopyranoside,(2E,4S)4hydroxy3,7dimethyl2,6octadien 1yl Trivialname Rosiridin(Wiedenfeld etal. ,2007)

Formula C16B HB 28B OB 7B B ExactMass 332.1829Da Molecularweight 332.3950Da Aspect Yellowamorphoussolid λ UV maxB B 196nm 1 P HNMRdataP SeeTable511 13 P CNMRdataP SeeTable511

P HRESIMS m/z 377.1797 (C17B HB 29B OB 9B [MHCOO]P requires 377.1812)B

169

Table 5-8 1 13

P P P H and P C NMR spectral data of compounds RR1 and RR10

RR1 RR10

No P1HP P13 CP P1HP P13 CP 1 128.6 132.3 2 7.04(2H, d,J =8.3) 129.7 7.02(2H, d,J =8.3) 129.7 3 6.66(2H, d,J =8.3) 114.9 6.63(2H, d,J =8.3) 115.1 4 155.6 155.9 7 2.73(2H, m) 34.8 2.55(2H, d,J =6.6) 38.1 8 3.56(1H, m) 69.8 3.63(2H, d,J =6.6) 64.2 3.87(1H, m) 1’ 4.16(1H, d,J= 7.8) 102.8 2’ 2.94(1H, t, J= 8.3) 73.4 3’ 3.10(1H, m) 76.8 4’ 3.08(1H, m) 70.0 5’ 3.10(1H, m) 76.8 6’ 3.42(1H, m) 61.1 3.65(1H, d,J= 11.2) 1 13 HP and P CP NMR spectra recorded at 500 MHz and 125 MHz, respectively, in DMSO-d6B B using TMS as internal standard, δ values given in ppm, J values in Hz

Table 5-9 1 13

P P P H and P C NMR spectral data of compounds RR5 , RR8 , RR9 and RR11

RR5 RR8 RR9 RR11 No P1HP P13 CP P1HP P13 CP P1HP P1HP P13 CP 1 137.0 129.9 137.0 137.0 2/6 7.42(2H, d,J =7.8) 127.4 7.24(2H, d,J =8.8) 129.2 7.42(2H, d,J =7.8) 127.6 7.42(2H, d,J =7.8) 127.6 3/5 7.30(2H, d,J =7.8) 129.5 6.66(2H, d,J =8.8) 116.7 7.30(2H, d,J =7.8) 129.6 7.30(2H, d,J =7.8) 129.6 4 7.21(1H, d,J =7.3) 128.6 158.0 7.21(1H, d,J =7.3) 128.7 7.21(1H, d,J =7.3) 128.7 7 6.70(1H, d,J =15.6) 133.8 6.56(2H, d,J =16.1) 134.6 6.70(1H, d,J =15.6) 133.8 6.70(1H, d,J =15.6) 133.8 8 6.37(1H, m) 123.8 6.20(2H, m) 123.7 6.37(1H, m) 123.8 6.37(1H, m) 123.8 4.32(1H, m) 70.8 4.27(1H, d,J =12.4,6.3) 71.5 4.32(1H, m) 70.9 4.32(1H, m) 70.9 9 4.52(1H, dd,J =11.9,4.6) 4.48(2H, d,J =11.7,5.4) 4.52(1H, dd,J =11.9,4.6) 4.52(1H, dd,J =11.9,4.6) 1’ 4.37(1H, d,J =7.8) 103.6 4.35(1H, d,J= 7.3) 103.5 4.37(1H, d,J =7.8) 103.5 4.35(1H, d,J= 7.3) 103.5 2’ 3.23(1H, m) 75.2 3.20(1H, m) 75.4 3.23(1H, m) 75.1 3.20(1H, m) 75.6 3’ 3.34(1H, m) 78.0 3.26(1H, m) 78.4 3.34(1H, m) 78.0 3.26(1H, m) 78.2 4’ 3.32(1H, m) 71.6 3.27(1H, m) 72.0 3.32(1H, m) 71.7 3.27(1H, m) 71.9 5’ 3.39(1H, m) 77.5 3.35(1H, m) 78.4 3.39(1H, m) 77.0 3.35(1H, m) 78.4 3.73(1H, dd,J =11.4,5.8) 69.4 3.66(1H,d d,J= 12.2,4.8) 63.0 3.73(1H, dd,J =11.4,5.8) 69.5 3.66(1H,d d,J= 12.2,4.8) 62.9 6’ 4.11(1H, dd,J =11.4,2.2) 3.86(1H,d d,J= 12.2,1.9) 4.11(1H, dd,J =11.4,2.2) 3.86(1H,d d,J= 12.2,1.9) 1’’ 4.82(1H, d,J =2.0) 107.6 4.34(1H, d,J =6.8) 105.2 2’’ 3.65(1H, dd,J =2.0,6.8) 74.6 3.61(1H, dd,J =8.8,6.8) 72.4 3’’ 3.52(1H, m) 71.1 3.52(1H, m) 74.2 4’’ 3.89(1H, m) 83.1 3.79(1H, m) 69.5 3.44(1H, m) 63.2 3.80(1H, m) 67.2 5’’ 3.52(1H, m) 3.51(1H, m) 6’’

1P 13 P P H and C spectra recorded at 500 MHz and 125 MHz, respectively, in DMSO-d6B B using TMS as internal standard, δ values given in ppm, J values in Hz

Table 5-10 Table 5-11 1 13 1 13

P P P P P H and P C NMR spectral data of compound RR7 P H and P C NMR spectral data of compound RR12

RR7 RR12 1 13 1 13 No P HP P CP No P HP P CP 118.2 1 4.33(1H, m) 65.9 1 4.37(1H, m) 2 112.6 2 5.57(1H, t,J =6.6) 122.7 3 6.46(1H, t,J =6.6) 145.0 3 142.7 4 4.44(2H, m) 68.4 4 3.97(2H, t,J =6.6) 77.9 5 1.98(3H, s) 20.2 5 2.24(2H, d,J =6.8) 34.6 1’ 4.33(1H, d,J= 7.8) 104.0 6 5.11(1H, t,J =7.0) 121.4 2’ 3.21(1H, t, J= 8.3) 75.0 7 133.9 3’ 3.27(1H, m) 78.0 8 1.69(3H, s) 25.8 4’ 3.35(1H, m) 71.5 9 1.62(3H, s) 17.8 5’ 3.37(1H, m) 78.0 10 1.67(3H, s) 11.8 6’ 3.85(2H, d,J= 11.7) 62.6 1’ 4.29(2H, d,J =7.8) 102.7 2’ 3.18(1H, m) 74.9 3’ 3.25(1H, m) 77.8 4’ 3.28(1H, m) 71.5 5’ 3.34(1H, m) 77.8 6’ 3.66(2H,d d,J =11.9,5.6) 62.6

1P 13 P P H and C spectra recorded at 500 MHz and 125 MHz, respectively, in DMSO-d6B B using TMS as internal standard, δ values given in ppm, J values in Hz

5. EXPERIMENTAL PART

5.8.3. Compounds isolated from the MeOH extract of Sedum dasyphyllum L .

Compound SD1 Structure OH

3 2' HO 1' 3' 1 2 O

4' HO Chemicalname (E)2Propenoicacid,3(3,4dihydroxyphenyl) Trivialname Caffeicacid(LuandFoo,1997)

Formula C9B HB 8B OB 4B B ExactMass 180.0417Da Molecularweight 180.1606Da Aspect Whiteamorphoussolid SH SH

λB B P P UV max 220,242 P ,296 P ,324nm 1 P HNMRdataP SeeTable512 13 P CNMRdataP SeeTable512 + HRESIMS m/z 181.0547 (CP 9B HB 9B OB 4B [M+H]B P requires181.0455)

173 5. EXPERIMENTAL PART

Compound SD2 Structure CO H 6'' 2 OH O HO 3' OH HO 1'' O 4' OH HO 8 O 2 1' 7

6 3 5 OH

OH O Chemicalname 4H1Benzopyran4one,8[( βglucopyranosiduronicacid)oxy] 3,5,7trihydroxy2(3,4dihydroxyphenyl) Trivialname Hibifolinorgossypetin8Oβglucuronide(Lai etal. ,2007)

Formula C21B HB 18B OB 14B B ExactMass 494.0691Da Molecularweight 494.3656Da Aspect Yellowamorphoussolid SH

λB P UV max 258,272 P ,378 nmB 1 P HNMRdataP SeeTable516 13 P CNMRdataP SeeTable516 +P HRESIMS m/z 495.0756(C21B BH19B BO14B [M+H] P requires B495.0775)

174 5. EXPERIMENTAL PART

Compound SD3 Structure OH

6'' O

HO 8 O O 1'' HO 2 OH 7

3 6 1' 5

OH O 4' 3' OH

OCH3 Chemicalname 4H1Benzopyran4one,7(βglucopyranosyloxy)5hydroxy3(4 hydroxy3methoxyphenyl) Trivialname 3’Omethylorobol7OβDglucoside(AdinarayandRajasekh, 1974;Viscardi etal. ,1984)

Formula C22B HB 22B OB 11B B ExactMass 462.1162Da Molecularweight 462.4106Da Aspect Yellow,amorphoussolid

UV λmaxB B 262nm 1 P HNMRdataP SeeTable518 13 P CNMRdataP SeeTable518

P HRESIMS m/z 507.1118(C23B HB 23B OB 13B [M+HCOO] P requires507.1139)

175 5. EXPERIMENTAL PART

Compound SD4 Structure

6''CO2H O HO OH HO 1'' O 2' OH

8 HO O 1' 2 7

3 6

5 OH

OH O Chemicalname 4H1Benzopyran4one,8[( βglucopyranosiduronicacid)oxy] 3,5,7trihydroxy2(4hydroxyphenyl) Trivialname Melocorin(Wolbis,1989)

Formula C21B HB 18B OB 13B B ExactMass 478.0747Da Molecularweight 478.3598Da Aspect Yellowneedles SH SH

λB B P P UV max 224,248 P ,270,320 P ,372nm 1 P HNMRdataP SeeTable516 13 P CNMRdataP SeeTable516

P HRESIMS m/z 477.0638(C21B HB 17B O13B [MH] P requires477.0669)

176 5. EXPERIMENTAL PART

Compound SD5 Structure OH 4'

HO 8 O 1' 2 7

6 3 5 O

1'' OH O OH OH O OH CH 6'' 3 Chemicalname 4H1Benzopyran4one,3[(6deoxyαmannopyranosyl)oxy]5,7 dihydroxy2(4hydroxyphenyl) Trivialname Afzeloside(Sethi etal. ,1983)

Formula C21B HB 20B OB 10B B ExactMass 432.1051Da Molecularweight 432.3843Da Aspect Yellowamorphoussolid

UV λmaxB B 264,346nm 1 P HNMRdataP SeeTable516 13 P CNMRdataP SeeTable516 HRESIMS m/z 431.0949(C21B HB 19B OB 10B [MH]B P requires431.0978)P

177 5. EXPERIMENTAL PART

Compound SD6 Structure OH

6'' O

HO 8 1'' O O HO 2 OH 7

6 3

H3CO 5 1'

OH O 4'

3' OCH3

OCH3 Chemicalname 4H1Benzopyran4one,3(3,4dimethoxyphenyl)7(β glucopyranosyloxy)5hydroxy6methoxy Trivialname Dalspinosin7Oβglucoside(Sethi etal. ,1983)

Formula C24B HB 26B OB 12B B ExactMass 506.1419Da Molecularweight 506.4639Da Aspect Greenpaleamorphoussolid SH

λB P UV max 264,346 P nmB 1 P HNMRdataP SeeTable518 13 P CNMRdataP SeeTable518

P HRESIMS m/z 551.1369(C25B HB 27B OB 14B [M+CHOO]B P requires551.1401)

178 5. EXPERIMENTAL PART

Compound SD7 Structure 8 HO O 2 7

6 3 1' H3CO 5

OH O 4' 3' OCH3

OH Chemicalname 4H1Benzopyran4one,5,7dihydroxy3(3hydroxy4 methoxyphenyl)6methoxy Trivialname IristectorigeninB(Hanawa etal. ,1991;Shawl etal. ,1984)

Formula C17B HB 14B OB 7B B ExactMass 330.0734Da Molecularweight 330.2944Da Aspect Yellowamorphoussolid

UV λ maxB B 200,266nm 1 P HNMRdataP SeeTable518 13 P CNMRdataP SeeTable518

P HRESIMS m/z 329.2336(C17B HB 13B OB 7B [MH]B P requires329.2345)

179 5. EXPERIMENTAL PART

Compound SD8 Structure OH

3' OH 4'

HO 8 O 2 1' 7

6 3 1'' HO 5 O OH OH OH O O 6''

OH Chemicalname 4H1Benzopyran4one,2(3,4dihydroxyphenyl)3(β glucopyranosyloxy)5,7dihydroxy Trivialname Hirsutrinorquercetin3Oβglucoside(MarkhamandChari,1982)

Formula C21B HB 20B OB 12B B ExactMass 464.0950Da Molecularweight 464.3830Da Aspect Yellowamorphoussolid SH

λB B P UV max 258,296 P ,354nm 1 P HNMRdataP SeeTable516 13 P CNMRdataP SeeTable516 +

P HRESIMS m/z 465.1150(C21B HB 21B OB 12B [M+H] P requires 465.1175)B

180 5. EXPERIMENTAL PART

Compound SD9 Structure OH OH 6''' O 4' HO 8 1''' O O HO 2 1' O 7 O HO 1'''' 6 3

6'''' 5 O OH

OH O 1'' OH HO HO OH O OH CH3 6'' Chemicalname 4H1Benzopyran4one,3[(6deoxyαmannopyranosyl)oxy]7 [(2Oβglucopyranosylglucopyranosyl)oxy]5hydroxy2(4 hydroxyphenyl) Trivialname Kaempferol3 Oαrhamnoside7Oβsophoroside

Formula C33B HB 40B OB 20B B ExactMass 756.2107Da Molecularweight 756.6693Da Aspect Greenyellowishamorphoussolid 20°C [α]DB PB P 114.7°(MeOH, c1.0)

UV λmaxB (logB εB )B 266(4.27),322(4.05),366(4.28)nm 1 P HNMRdataP SeeTable519 13 P CNMRdataP SeeTable520 HRESIMS m/z 755.2037(C33B HB 39B OB 20B [MH]B P requires755.2035)

181 5. EXPERIMENTAL PART

Compound SD10 Structure H3C 6''' OH O HO

HO 1''' OH

OH 4'

O 8 O 2 1' 7

HO 6 3 1'' OH 5 O OH O OH O 6''

OH Chemicalname 4H1Benzopyran4one,3(βglucopyranosyloxy)7[(6deoxyα mannopyranosyl)oxy]5,8dihydroxy2(4hydroxyphenyl) Trivialname Herbacetin3Oβglucoside7Oαrhamnoside(Yoshikawa,2006)

Formula C27B HB 30B OB 16B B ExactMass 610.1528Da Molecularweight 610.5261Da Aspect Yellowamorphouspowder

UV λmaxB B 202,276,328nm 1 P HNMRdataP SeeTable517 13 P CNMRdataP SeeTable517 HRESIMS m/z 609.1428(C27B HB 29B OB 16B [MH] P requires 609.1456)B

182 5. EXPERIMENTAL PART

Compound SD11 Structure HO 6'

O HO 5' HO O OH 4' 6' 9' 1' OH 1' 2 3' 8' 7 OMe 8 3 MeO 2' 7' 1

4 9 6 HO OH 5 Chemicalname βDglucopyranoside,4[(2R,3R)4hydroxy3[(4hydroxy3 methoxyphenyl)methyl]2(hydroxymethyl)butyl]2methoxyphenyl Trivialname ()Secoisolariciresinol4Oβglucopyranoside(Yuan etal. ,2002)

Formula C26B HB 36B OB 11B B ExactMass 524.22521Da Molecularweight 524.5664Da Aspect Yellowamorphoussolid

UV λmaxB B 230,270nm 1 P HNMRdataP SeeTable513 13 P CNMRdataP SeeTable513

P HRESIMS m/z 523.2134(C26B HB 35B OB 11B [MH]B P requires523.2179)

183 5. EXPERIMENTAL PART

Compound SD12 Structure OH

6' N O 1 HO O 2a HO 1' OH 2

3 4 Chemicalname Butanenitrile,2(βglucopyranosyloxy)2methyl,(2R) Trivialname Lotaustralin(Akgul etal. ,2004)

Formula C11B HB 19B NOB 6B B ExactMass 261.1206Da Molecularweight 261.2755Da Aspect Whiteamorphoussolid

UV λmaxB B 216nm 1 P HNMRdataP SeeTable514 13 P CNMRdataP SeeTable514

P HRESIMS m/z 306.1200(C12B HB 20B NO8B [M+CHOO]B P requires306.1189)

184 5. EXPERIMENTAL PART

Compound SD13 Structure 6''''CO2H O HO OH OH HO 1'''' O OH 3' OH 6''' O 4' HO O 8 1''' O HO 2 1' OH 7 HO 6 3 1'' OH 5 O OH O OH O 6''

OH Chemicalname 4H1Benzopyran4one,3(βglucopyranosyloxy)7(β glucopyranosyloxy)8[( βglucopyranosiduronicacid)oxy]5 hydroxy2(3,4dihydroxyphenyl) Trivialname Gossypetin3,7diOβglucoside8Oβglucuronide

Formula C33B HB 38B OB 24B B ExactMass 818.1747Da Molecularweight 818.6506Da Aspect Yellowamorphoussolid 20°C

P [α]DB PB +29.3°(MeOH, c1.0) SH

λB B B B P UV max (log ε) 261(4.19),270 P (4.17),366(3.97)nm 1 P HNMRdataP SeeTable519 13 P CNMRdataP SeeTable520 HRESIMS m/z 817.1631(C33B HB 37B OB 24B [MH] P requiresP 817.1675)B

185 5. EXPERIMENTAL PART

Compound SD14 Structure 6''''CO2H O HO OH HO 1'''' O OH OH 6''' O 4' HO O 8 1''' O HO 2 1' OH 7

OCH3 HO 6 3 1'' OH 5 O OH O OH O 6''

OH Chemicalname 4H1Benzopyran4one,3(βglucopyranosyloxy)7(β glucopyranosyloxy)8[( βglucopyranosiduronicacid)oxy]5 hydroxy2(4hydroxyphenyl) Trivialname Herbacetin3,7diOβglucoside8Oβglucuronide

Formula C33B HB 38B OB 23B B ExactMass 802.1798Da Molecularweight 802.6513Da Aspect Yellowamorphoussolid 20°C

P [α]DB PB +13.4°(MeOH, c1.0) SH

λB B B B P UV max (log ε) 271(4.51),328 P (4.36),358(4.31)nm 1 P HNMRdataP SeeTable519 13 P CNMRdataP SeeTable520 HRESIMS m/z 801.1667(C33B HB 37B OB 23B [MH]B P requires801.1726)P

186 5. EXPERIMENTAL PART

Compound SD15 Structure OH

6'' O HO O O OH HO 1'' 4' OH 1

1' 2 3'

O 2' 3

Chemicalname 2Propenoicacid,3[4(β −glucopyranosyloxy)3methoxyphenyl] Trivialname Ferulicacid βglucoside(BaderschneiderandWinterhalter,2001)

Formula C16B HB 20B OB 9B B ExactMass 356.1101Da Molecularweight 356.3298Da Aspect Brownishamorphoussolid

UV λmaxB B 198,218,294nm 1 P HNMRdataP SeeTable515 13 P CNMRdataP SeeTable515

P HRESIMS m/z 355.1015(C16B HB 19B OB 9B [MH] P requires 355.1029)B

187 5. EXPERIMENTAL PART

Compound SD16 Structure CO H 6''' 2 OH O HO 3' OH 1''' O HO 4' OH HO 8 O 2 1' 5' 7 OH

HO 6 3 1'' OH 5 O OH O OH O 6''

OH Chemicalname 4H1Benzopyran4one,3(βglucopyranosyloxy)8[( β glucopyranosiduronicacid)oxy]5,7dihydroxy2(3,4,5 trihydroxyphenyl) Trivialname Hibiscetin3Oβglucoside8Oβglucuronide

Formula C27B HB 28B OB 20B B ExactMass 672.1168Da Molecularweight 672.5074Da Aspect Yellowamorphoussolid 20°C

P [α]DB PB +57.5°(MeOH, c1.0)

UV λmaxB (logB εB )B 270(3.99),366(3.83)nm 1 P HNMRdataP SeeTable517 13 P CNMRdataP SeeTable517 HRESIMS m/z 671.1068(C27B HB 27B OB 20B [MH]B P requires671.1096)

188 5. EXPERIMENTAL PART

Compound SD17 Structure

6''''CO2H O HO OH HO 1'''' O OH OH 6''' O 4' HO 8 1''' O O HO 2 1' OH 7 OH 6 3 1'' OCOCH 5 O 3 OH OH O O 6''

OH Chemicalname 4H1Benzopyran4one,3(3acetylβglucopyranosyloxy)7(β glucopyranosyloxy)8[( βglucopyranosiduronicacid)oxy]5 hydroxy2(4hydroxyphenyl) Trivialname Herbacetin3Oβ(3’’acetylglucoside)7Oβglucoside8Oβ glucuronide

Formula C35B HB 40B OB 24B B ExactMass 844.1904Da Molecularweight 844.6886Da Aspect Yellowamorphoussolid 20°C

P [α]DB PB –18.8°(MeOH,c1.0)

UV λmaxB (logB εB )B 270(4.10),356(3.91)nm 1 P HNMRdataP SeeTable519 13 P CNMRdataP SeeTable520

P HRESIMS m/z 843.1755(C35B HB 39B O24B [MH]B P requires843.1831)

189 5. EXPERIMENTAL PART

Compound SD18 Structure

6'''CO2H O HO OH 1''' HO O 4' OH HO 8 O 2 1' 7

HO 6 3 1'' OH 5 O OH O OH O 6''

OH Chemicalname 4H1Benzopyran4one,3(βglucopyranosyloxy)8[( β glucopyranosiduronicacid)oxy]5,7dihydroxy2(4hydroxyphenyl) Trivialname Herbacetin3Oβglucoside8Oβglucuronide(Sikorska etal. , 2004)

Formula C27B HB 28B OB 18B B ExactMass 640.1270Da Molecularweight 640.5088Da Aspect Palebrownishamorphoussolid

UV λmaxB B 198,270,352nm 1 P HNMRdataP SeeTable517 13 P CNMRdataP SeeTable517

P HRESIMS m/z 639.1183 (C27B HB 27B OB 18B [MH]B P requires639.1197)

190 5. EXPERIMENTAL PART

Compound SD19 Structure

6'''CO2H O HO OH 1''' O HO 4' OH HO 8 O 2 1' 7

OH 6 3 1'' OCOCH 5 O 3 OH O OH O 6''

OH Chemicalname 4H1Benzopyran4one,3(3acetylβglucopyranosyloxy)8[( β glucopyranosiduronicacid)oxy]5,7dihydroxy2(4hydroxyphenyl) Trivialname Herbacetin3Oβ(3’’acetylglucoside)8Oβglucuronide

Formula C27B HB 28B OB 18B B ExactMass 682.1375Da Molecularweight 682.5461Da Aspect Brownishamorphoussolid 20°C [α]DB PB P –139.6°(MeOH, c1.0)

UV λmaxB (logB εB )B 271(4.06),355(3.88)nm 1 P HNMRdataP SeeTable517 13 P CNMRdataP SeeTable517

P HRESIMS m/z 681.1305(C27B HB 27B OB 18B [MH]B P requires681.1303)

191 5. EXPERIMENTAL PART

Table 5-12 1 13

P P PP P H and P C NMR spectral data of compound SD1

SD1 No P1HP NMR P13 CP NMR 1 170.4 2 6.25(1H ,d , J=15.6) 115.7 3 7.58(1H, d, J=16.0) 146.7 1’ 127.8 2’ 7.07(1H, d, J=2.0) 115.1 3’ 146.1 4’ 149.4 5’ 6.82(1H, d, J=8.3) 116.5 6’ 6.21(1H ,dd , J=2.9,8.3) 122.8 1P 13 P H and C spectra recorded at 500 MHz and 125 MHz, respectively, in CD 3B BOD using TMS as internal standard, δ values given in ppm, J values in Hz

Table 5-13 1 13 P H and C NMR spectral data of compound SD11

SD11 No P1HP NMR P13 CP NMR 1 135.2 2 6.69(1H ,d , J=2.0) 113.2 3 148.6 3OMe 3.69(3H, s) 55.5 4 144.3 5 6.95(1H ,d , J=7.8) 115.1 6 6.62(1H ,dd , J=8.3,2.0) 120.9 7 2.51(2H ,m ) 33.8 8 3.28(1H ,m ) 60.2 9 1.85(1H ,d , J=5.4) 42.5 1’ 132.1 2’ 6.64(1H ,d , J=2.4) 112.9 3’OMe 3.69(3H, s) 55.5 3’ 147.2 4’ 144.6 5’ 6.65(1H ,d , J=7.3) 115.0 6’ 6.50(1H ,dd , J=8.0,1.7) 121.1 7’ 2.51(2H ,m ) 33.8 8’ 3.28(1H ,m ) 60.2 9’ 1.85(1H ,d , J=5.4) 42.5 1’’ 4.82(1H ,d , J=7.3) 100.4 2’’ 3.24(1H, m) 73.2 3’’ 3.25(1H, m) 76.9 4’’ 3.18(1H ,d , J=8.8) 69.8 5’’ 3.25(1H, m) 76.9

6’’CH 2 3.45(1H, m) 60.7 3.66(1H, m,) 1P 13 P P H and C spectra recorded at 500 MHz and 125 MHz, respectively, in DMSO-d6B B using TMS as internal standard, δ values given in ppm, J values in Hz

192 5. EXPERIMENTAL PART

Table 5-14 1 13

P P H and P C NMR spectral data of compound SD12

SD12 No P1HP NMR P13 CP NMR 1 21.5 2 75.4 2a 1.53(3H, s) 23.9 3 1.80(1H ,m ) 32.9 1.88(1H ,m ) 4 0.98(3H, t, J=7.3) 8.1 1’ 4.45(1H, d, J=7.3) 99.3 2’ 2.95(1H, m) 73.1 3’ 3.13(1H, m) 76.6 4’ 3.05(1H, m) 69.8 5’ 3.13(1H,m) 76.8

6’CH 2B B 3.43(1H, m) 60.9 3.64(1H, m) 1 13 PPP P H and C spectra recorded at 500 MHz and 125 MHz, respectively, in DMSO-d using TMS as internal standard, δ values given in ppm, J values in Hz

Table 5-15 1H and 13 C NMR spectral data of compound SD15

SD15 No P1HP NMR P13 CP NMR 1 168.4 2 6.82(1H ,d , J=12.7) 119.5 3 5.85(1H, d, J=12.7) 141.5 1’ 129.2 2’ 7.59(1H, d, J=2.0) 115.0 3’ 148.0 4’ 148.8 5’ 7.08(1H, d, J=8.7) 115.1 6’ 7.21(1H ,dd , J=2.0,8.3) 124.5

OCH 3B B 3.27(3H, s) 56.2 1’’ 4.98(1H, d, J=7.1) 100.4 2’’ 3.27(1H, m) 73.9 3’’ 3.28(1H, m) 77.6 4’’ 3.16(1H, m) 70.3 5’’ 3.33(1H, m) 77.8

6’’CH 2 3.46(1H, m) 61.3 3.66(1H, m) 1 13

P P P H and P C spectra recorded at 500 MHz and 125 MHz, respectively, in DMSO-d6B B using TMS as internal standard, δ values given in ppm, J values in Hz

193

Table 5-16 1 13 P HP and P CP NMR spectral data of flavonols SD2, SD4, SD5 and SD8

SD2 aP P SD4 aP P SD5 aP P SD8 bP P No P1HP P13 CP P1HP P13 CP P1HP P13 CP P1HP P13 CP 1 2 145.8 148.7 159.3 156.1 3 137.2 137.3 136.2 133.3 4 177.3 177.4 179.6 177.4 5 158.4 158.1 163.3 161.2 5OH 12.55(1H ,brs ) 6 6.26(1H ,s ) 99.4 6.25(1H ,s ) 99.59 6.21(1H ,d , J=1.4) 99.8 6.21(1H ,d , J=2.0) 98.6 7 157.9 158.7 165.8 164.1 8 126.5 126.7 6.47(1H ,d , J=2.0) 94.7 6.40(1H ,d , J=2.0) 93.4 9 150.1 150.2 158.5 156.3 10 104.5 104.6 105.9 103.9 1’ 124.0 123.8 122.6 121.5 2’ 7.98(1H, s) 116.6 8.32(2H, d, J=8.8) 131.5 7.78(2H, d, J=8.3) 131.8 7.59(1H, m) 116.2 3’ 148.6 6.93(2H, d, J=8.8) 116.5 6.95(2H, d, J=8.3) 116.5 144.7 4’ 148.8 160.6 161.6 148.4 5’ 6.93(1H, d, J=8.3) 116.5 6.93(2H, d, J=8.8) 116.5 6.95(2H, d, J=8.3) 116.5 6.85(1H, m) 115.1 6’ 7.86(1H, d, J=7.8) 122.5 8.32(2H, d, J=8.8) 131.5 7.78(2H, d, J=8.3) 131.8 7.59(1H, m) 121.1 1’’ 4.79(1H, d, J=7.8) 108.3 4.79(1H, d, J=7.1) 108.3 4.24(1H, m) 103.5 5.45(1H, d, J=7.3) 100.8 2’’ 3.70(1H, m) 75.4 3.66(1H, m) 75.4 3.86(1H, m) 71.9 3.23(1H, m) 74.0 3’’ 3.53(1H, m) 77.3 3.50(1H, m) 77.3 3.73(1H, m) 72.1 3.08(1H, m) 77.5 4’’ 3.66(1H, m) 73.3 3.73(1H, m) 73.2 3.35(1H, m) 73.2 3.10(1H,d, J=3.4) 69.9 5’’ 3.86(1H, m) 77.3 3.90(1H, m) 77.3 3.86(1H, m) 72.0 3.20(1H, m) 76.5 6’’ 171.0 171.0 0.94(3H, d, J=5.4) 17.6 3.31(1H, m) 60.9 3.57(1H, d,J =11.7)

1P 13 a) P b) P P H and C spectra recorded at 500 MHz and 125 MHz, respectively, in P CD 3B ODB or P DMSO-d6B B using TMS as internal standard, δ values given in ppm, J values in Hz

Table 5-17 1 13

P P P H and P C NMR spectral data of flavonols diglycosides SD10, SD 16, SD18 and SD19 SD10 SD16 SD18 SD19 No P1HP P13 CP P1HP P13 CP P1HP P13 CP P1HP P13 CP 2 156.6 156.6 156.1 156.9 3 133.2 133.3 133.0 132.9 4 177.9 177.2 177.3 177.3 5 152.0 156.2 157.0 156.3 5OH 12.55(1H ,brs ) 12.50(1H ,brs ) 12.55(1H ,brs ) 6 6.61(1H ,s ) 98.8 6.28(1H ,s ) 98.7 6.24(1H ,s ) 99.2 6.29(1H ,s ) 99.0 7 156.6 156.2 157.0 156.3 8 127.0 124.3 125.1 124.9 9 144.4 148.4 148.4 148.4 10 105.4 103.7 103.7 103.6 1’ 121.0 119.9 120.7 120.5 2’ 8.13(1H, d, J=8.3) 131.0 7.38(2H, s) 109.1 8.25(1H, d, J=9.1) 131.4 8.21(1H, d, J=9.1) 131.3 3’ 6.91(1H, d, J=8.3) 115.0 145.1 6.85(1H, d, J=9.1) 115.2 6.88(1H, d, J=9.1) 115.3 4’ 160.0 136.7 160.0 160.0 5’ 6.91(1H, d, J=8.3) 115.0 145.1 6.85(1H, d, J=9.1) 115.2 6.88(1H, d, J=9.1) 115.3 6’ 8.13(1H, d, J=8.3) 131.0 7.38(2H, s) 109.1 8.25(1H, d, J=9.1) 131.4 8.21(1H, d, J=9.1) 131.3 1’’ 5.47(1H, d, J=7.3) 100.8 5.50(1H, d, J=7.5) 100.8 5.49(1H, d, J=7.5) 100.1 5.60(1H, d, J=7.5) 100.6 2’’ 3.20(1H, m) 74.2 3.34(1H, m) 73.8 3.22(1H, m) 74.2 3.38(1H, m) 72.0 3’’ 3.21(1H, m) 76.4 3.231H, d,J =8.3) 76.5 3.10(1H, m) 77.5 4.86(1H, m) 77.5

3’’OCOCH 3B B 2.05(3H, s) 169.6 21.1 4’’ 3.08(1H, m) 69.8 3.171H, d,J =3.1) 69.9 3.12(1H, m) 69.9 3.35(1H, m) 67.7 5’’ 3.07(1H, m) 77.5 3.13(1H, m) 77.7 3.24(1H, m) 76.5 3.25(1H, m) 77.2

6’’CH 2B B 3.31(1H, m) 60.8 3.30(1H, m) 61.1 3.35(1H, m) 60.8 3.421H, d,J =7.9) 60.4 3.55(1H, m) 3.67(1H, d,J =11.9) 3.60(1H, m) 3.601H, d,J =10.1) 1’’’ 5.50(1H, d, J=7.5) 99.3 4.82(1H, d, J=7.5) 105.0 4.76(1H, d, J=7.5) 106.2 4.76(1H, d, J=7.5) 106.1 2’’’ 3.95(1H, m) 69.8 3.45(1H, d,J =8.5 ) 71.3 3.43(1H, m) 73.5 3.441H, d,J =7.9) 73.7 3’’’ 3.82(1H, m) 70.0 3.45(1H, d,J =8.5 ) 73.8 3.31(1H, m) 75.5 3.33(1H, m) 75.3 4’’’ 3.30(1H, m) 71.7 3.31(1H, m) 75.5 3.48(1H, m) 71.6 3.50(1H, t,J =9.1) 71.4 5’’’ 3.53(1H, m) 69.8 3.691H, d,J =9.5) 77.5 3.75(1H, m) 76.5 3.78(1H, d,J =9.5) 76.0 6’’’ 1.12(3H, d, J=6.3) 17.8 170.9 170.4 169.9

1P 13 P P H and P C spectra recorded at 500 MHz and 125 MHz, respectively, in DMSO-d6B B using TMS as internal standard, δ values given in ppm, J values in Hz

Table 5-18 1 13

P P P H and P C NMR spectral data of isoflavonoids SD3, SD6 and SD7

SD3 SD6 SD7

No P1HP P13 CP P1HP P13 CP P1HP P13 CP 1 2 8.18(1H, s) 155.5 155.7 8.22(1H, s) 155.1 3 125.0 124.6 125.0 4 181.0 182.5 182.2 5 165.1 151.3 151.0 6 6.71(1H ,d , J=2.0) 95.8 130.7 129.3

6OCH 3B B 3.92(3H, s) 62.3 3.90(3H, s) 61.9 7 164.0 157.7 158.8 8 6.53(1H ,d , J=2.0) 101.1 6.69(1H ,s ) 100.2 6.31(1H, s) 100.3 9 159.2 158.7 157.2 10 108.0 107.7 106.1 1’ 123.6 124.9 124.4 2’ 7.19(1H, d, J=1.5) 113.9 7.23(1H, s) 114.2 7.21(1H, d, J=1.5) 114.2 3’ 148.8 150.2 150.2

3’OCH 3B B 3.91(3H, s) 56.47 3.89(3H, s) 56.5 4’ 148.0 150.8 150.7

4’OCH 3B B 3.98(3H, s) 56.5 3.88(3H, s) 56.4 5’ 6.86(1H, d, J=8.3) 116.2 7.13(1H, d, J=8.8) 112.8 7.03(1H, d, J=8.3) 112.8 6’ 7.0(1H, dd , J=1.5,8.3) 122.9 7.03(1H, d, J=8.3) 122.8 7.1(1H, dd , J=1.7,8.1) 122.8 1’’ 5.05(1H, d, J=6.8) 101.6 5.08(1H, d, J=7.8) 102.0 2’’ 3.50(1H, m) 74.7 3.55(1H, m) 74.8 3’’ 3.51(1H, m) 77.0 3.49(1H,m) 78.4 4’’ 3.43(1H, m) 71.2 3.45(1H, m) 71.1 5’’ 3.52(1H, m) 77.4 3.51(1H, m) 78.0 6’’ 3.74(1H, m) 62.4 3.74(1H, d, J=5.4) 62.3 3.92(1H, m) 3.75(1H, d, J=5.4)

1P 13 P H and C spectra recorded at 500 MHz and 125 MHz, respectively, in CD 3B BOD using TMS as internal standard, δ values given in ppm, J values in Hz

Table 5-19 1 P HP NMR spectral data of flavonols triglycosides SD9, SD13, SD14 and SD17

P1HP SD9 SD13 SD14 SD17 5OH 12.55(1H ,brs ) 12.55(1H ,brs ) 12.55(1H ,brs) 12.55(1H ,brs ) 6 6.52(1H ,d , J=2.0) 6.72(1H ,s ) 6.73(1H ,s ) 6.74(1H ,s ) 7 8 6.79(1H ,d , J=2.0) 9 10 1’ 2’ 6.95(1H, d, J=8.3) 7.84(1H, d, J=2.2) 8.20(1H, d, J=8.8) 8.20(1H, d, J=9.1) 3’ 7.84(1H, d, J=8.3) 6.88(1H, d, J=8.7) 6.88(1H, d, J=8.7) 4’ 5’ 7.84(1H, d, J=8.3) 6.85(1H, d, J=8.5) 6.88(1H, d, J=8.7) 6.88(1H, d, J=8.7) 6’ 6.95(1H, d, J=8.3) 7.66(1H, dd , J=2.2,8.5) 8.20(1H, d, J=8.8) 8.20(1H, d, J=9.1) 1’’ 5.44(1H, d, J=1.5) 5.55(1H, d, J=7.3) 5.55(1H, d, J=7.3) 5.64(1H, d, J=7.3) 2’’ 4.24(1H, m) 3.38(1H, m) 3.38(1H, m) 3.38(1H, d, J=7.8) 3’’ 3.72(1H, d,J =7.6) 3.62(1H, d,J =9.5) 3.61(1H, d, J=11.7) 4.87(1H, m) 4’’ 3.34(1H, m) 3.34(1H, m) 3.34(1H, m) 3.38(1H, d, J=7.8) 5’’ 3.35(1H, d,J =2.4) 3.27(1H, m) 3.27(1H, m) 3.27(1H, m) 6’’ 0.94(3H, d, J=5.4) 3.52(1H, m) 3.52(1H, m) 3.61(1H, m) 3.75(1H, d,J =11.7) 3.75(1H, d, J=10.7) 3.44(1H, d, J=5.4)

6’’OCOCH 3B B 2.06(3H, s) 1’’’ 5.29(1H, d, J=7.3) 4.98(1H, d, J=7.3) 4.99(1H, d, J=7.3) 5.00(1H, d, J=7.3) 2’’’ 3.78(1H, d,J =8.8) 3.40(1H, m) 3.40(1H, m) 3.40(1H, m) 3’’’ 3.70(1H, d,J =8.8) 3.31(1H, m) 3.31(1H, m) 3.31(1H, m) 4’’’ 3.39(1H, m) 3.15(1H, dd,J =11.7,1.8) 3.15(1H, m) 3.18(1H, d, J=4.9) 5’’’ 3.29(1H, m) 3.26(1H, m) 3.27(1H, m) 3.44(1H, d, J=5.4)

6’’’CH 2B B 3.73(1H, m) 3.75(1H, d,J =11.7) 3.75(1H, d, J=10.7) 3.74(1H, d, J=9.5) 3.93(1H, m) 3.50(1H, d,J =9.1) 3.51(1H, m) 3.50(1H, d, J=9.6) 1’’’’ 4.67(1H, d, J=7.8) 4.82(1H, d, J=7.8) 4.81(1H, d, J=7.8) 4.82(1H, d, J=7.8) 2’’’’ 3.24(1H, m) 3.40(1H, m) 3.40(1H, m) 3.38(1H, m) 3’’’’ 3.41(1H, m) 3.65(1H, d,J =11.9) 3.65(1H, m) 3.65(1H, dd , J=9.9,2.8) 4’’’’ 3.45(1H, d,J =2.9) 3.50(1H, d,J =9.1) 3.50(1H, d, J=9.8) 3.50(1H, d, J=9.8) 5’’’’ 3.55(1H, m) 3.32(1H, m) 3.32(1H, d, J=9.8) 3.32(1H, m)

6’’’’CH 2B B 3.59(1H, m), 3.65(1H, d,J =4.4) 1P P H spectra recorded at 500 MHz, in DMSO-d6B B using TMS as internal standard, δ values given in ppm, J values in Hz

5. EXPERIMENTAL PART

Table 5-20 13 P CP NMR spectral data of flavonols triglycosides SD9, SD13, SD14 and SD17

13 P CP SD9 SD13 SD14 SD17 2 159.8 160.9 157.3 158.7 3 136.4 135.8 132.9 133.7 4 179.8 177.6 177.6 177.5 5 162.8 156.5 156.7 157.4 6 100.9 98.9 98.9 99.04 7 164.4 155.3 155.4 156.5 8 95.9 125.6 125.6 126.5 9 158.1 148.6 148.4 148.4 10 107.7 105.5 105.4 105.4 1’ 122.4 125.6 120.6 120.6 2’ 116.6 120.9 131.7 131.7 3’ 132.1 144.5 115.1 115.1 4’ 161.8 148.4 160.1 160.2 5’ 132.1 117.4 115.1 115.1 6’ 116.6 122.1 131.7 131.7 1’’ 103.5 100.7 100.6 100.4 2’’ 71.9 74.0 74.1 72.0 3’’ 77.1 75.6 75.7 77.4 4’’ 73.2 69.9 69.6 67.6 5’’ 72.1 76.5 76.3 77.2

6’’ 17.1 61.1 60.6 60.4

6’’OCOCH 3B B 169.7 21.1 1’’’ 100.2 101.4 101.3 101.4 2’’’ 83.6 73.2 73.1 73.2 3’’’ 77.5 75.6 75.7 74.0 4’’’ 71.2 69.7 69.8 70.0 5’’’ 78.0 76.5 76.3 77.3

6’’’ 62.4 60.7 60.8 60.7 1’’’’ 105.5 105.5 104.9 104.8 2’’’’ 76.0 74.0 73.9 74.0 3’’’’ 77.7 77.7 77.6 75.7 4’’’’ 70.9 71.3 71.3 71.7 5’’’’ 78.1 77.3 77.2 75.4 6’’’’ 62.1 170.1 169.9 169.9 13

P P C spectra recorded at 125 MHz, in DMSO-d6B B using TMS as internal standard, δ values given in ppm

198

6. REFERENCES

6. REFERENCES

A

Abidov M., Crendal F., Grachev S., Seifulla R. and Ziegenfuss T. (2003). Effect of extracts from Rhodiola rosea and Rhodiola crenulata (Crassulaceae) roots on ATP content in mitochondria of skeletal muscles. Bulletin of Experimental Biology and Medicine 136 : 585-587.

Abidov M., Grachev S., Seifulla R.D. and Ziegenfuss T.N. (2004). Extract of Rhodiola rosea radix reduces the level of C-reactive protein and creatinine kinase in the blood. Bulletin of Experimental Biology and Medicine 138 : 63-64.

Adinaray D. and Rajasekh J. (1974). Synthesis and study of 3'-O-methylorobol. Indian Journal of Chemistry 12 : 911-913.

Ahmad R., Sethi M.S., Qureshi M.T. and Ahmad A. (1994). Studies on indigenous cyanogenetic plants. Part-II. Science International 6: 353-354.

Akgul Y., Ferreira D., Abourashed E.A. and Khan I.A. (2004). Lotaustralin from Rhodiola rosea roots. Fitoterapia 75 : 612-614.

Alaluf S., Heinrich U., Stahl W., Tronnier H. and Wiseman S. (2002). Dietary carotenoids contribute to normal human skin color and UV photosensitivity. Journal of Nutrition 132 : 399-403.

Ali Z., Fronczek F.R. and Khan I.A. (2008). Phenylalkanoids and monoterpene analogues from the roots of Rhodiola rosea . Planta Medica 74 : 178-181.

Alm T. (1996). Bruk av rosenrot (Rhodiola rosea ) mot skjorbuk. Polarflokken 20 : 29-32.

Altavilla D., Polito F., Bitto A., Minutoli L., Miraldi E., Fiumara T., Biagi M., Marini H., Giachetti D., Vaccaro M. and others (2008). Anti-inflammatory effects of the methanol extract of Sedum telephium ssp. maximum in lipopolysaccharide-stimulated rat peritoneal macrophages. Pharmacology 82 : 250-256.

Ampong-Nyarko K., Brown J., De Mlder J., Chaudhary N., Zhang Z. and Jiao A. 2006. Rhodiola rosea commercialization in Alberta. Alberta: Crop diversification Centre North.

Amrein R., Hetzel W., Stabl M. and Schmid-Burgk W. (1993). RIMA: a new concept in the treatment of depression with moclobemide. International Clinical Psychopharmacology 7: 123-132.

Anilakumar P.K.R., Khanum F. and Bawa A.S. (2006). Phytoconstituents and antioxidant potency of Rhodiola rosea - A versatile adaptogen. Journal of Food Biochemistry 30 : 203-214.

Arora R., Chawla R., Sagar R., Prasad J., Singh S., Kumar R., Sharma A., Singh S. and Sharma R.K. (2005). Evaluation of radioprotective activities of Rhodiola imbricata Edgew - A high altitude plant. Molecular and Cellular Biochemistry 273 : 209-223.

201 6. REFERENCES

B

Baderschneider B. and Winterhalter P. (2001). Isolation and characterization of novel benzoates, cinnamates, flavonoids, and lignans from riesling wine and screening for antioxidant activity. Journal of Agricultural and Food Chemistry 49 : 2788-2798.

Barnham K.J., Masters C.L. and Bush A.I. (2004). Neurodegenerative diseases and oxidative stress. Nature Reviews Drug Discovery 3: 205-214.

Batchelor J.G., Cushley R.J. and Prestegard J.H. (1974). Carbon-13 Fourier transform nuclear magnetic resonance. VIII. Role of steric and electric field effects in fatty acid spectra. Journal of Organic Chemistry 39 : 1698-1705.

Bate-Smith E.C. and Swain T. (1963). Comparative biochemistry: A comprehensive treatise Mason HS and Florkin AM (eds). New York: Academic Press.

Berger A. (1930). Crassulaceae. pp. 352-483. In: Engler A and Prantl K (eds). Die natürlichen Pflanzenfamilien . 2 ed. Leipzig: Verlag Wilhelm Engelmann.

Birks J. and Flicker L. (2003). Selegiline for Alzheimer's disease. Cochrane Database of Systematic Reviews Issue 1 : Art. No.: CD000442.

Bjarnholt N., Rook F., Motawia M.S., Cornett C., Jorgensen C., Olsen C.E., Jaroszewski J.W., Bak S. and Moller B.L. (2008). Diversification of an ancient theme: Hydroxynitrile glucosides. Phytochemistry 69 : 1507-1516.

Blois M.S. (1958). Antioxidant determinations by the use of a stable free radical. Nature 181 : 1199- 1200.

Bocharova O.A., Matveev B.P., Baryshnikov A, Figurin K.M., Serebriakova R.V. and Bodrova N.B. (1995). The effect of a Rhodiola rosea extract on the incidence of recurrences of a superficial bladder cancer (experimental clinical research). Urologiia I Nefrologiia 2: 46-47.

Bonina F., Puglia C., Tomaino A., Saija A., Mulinacci N., Romani A. and Vincieri F.F. (2000). In- vitro antioxidant and in-vivo photoprotective effect of three lyophilized extracts of Sedum telephium L. leaves. Journal of Pharmacy and Pharmacology 52 : 1279-1285.

Borissova A.G. (1939). Crassulaceae. In: Komarov VL (eds). Flora of USSR 9 . Liningrad: Academia Scientiarum.

Bortolato M., Chen K. and Shih J.C. (2008). Monoamine oxidase inactivation: From pathophysiology to therapeutics. Advanced Drug Delivery Reviews 60 : 1527-1533.

Brekhman L. and Dardymov I.V. (1969). New substances of plant origin which increase nonspecific resistance. Annual Review of Pharmacology 9: 419-430.

Briegel M. and Beguin C. (1926). Isolation of rutoside, asparagines and a new glycoside, hydrolysable by emulsion, salidroside from Salix triandra L. Comptes Rendus Hebdomadaires des Séances de l'Académie des Sciences 183 : 321-323.

202 6. REFERENCES

Brown R.P., Gerbarg P.L. and Ramazanov Z. (2002). Rhodiola rosea . A phytomedicinal overview. HerbalGram 56 : 40-52.

Bruneton J. (2005). Pharmacognosie, phytochimie, plantes médicinales . 3rd ed. Tec & Doc and EMinter (eds). Paris: Lavoisier.

Byung S.M., Ki H.B., Young H.K., Hirotsugu M., Masao H. and Kunitada S. (1999). Screening of Korean plants against human immunodeficiency virus type 1 protease. Phytotherapy Research 13 : 680-682.

C

Chen J.H. and Ho C.-T. (1997). Antioxidant activities of caffeic acid and its related hydroxycinnamic acid compounds. Journal of Agricultural and Food Chemistry 45 : 2374-2378.

Chen T.S., Liou S.Y. and Chang Y.L. (2008). Antioxidant evaluation of three adaptogen extracts. American Journal of Chinese Medicine 36 : 1209-1217.

Clifford M.N. and Brown J.E. (2006). Flavonoids. Chemistry, biochemistry and applications . Markham KR and Andersen ØM (eds). Boca Raton: CRC Press Taylor & Francis.

Committee on herbal medicinal products. 2008. Reflection paper on the adaptogenic concept. London: EMEA.

Cronquist A. (1988). The evolution and classification of flowering plants . 2nd ed. New York: Columbia University Press.

Cuendet M., Hostettmann K., Potterat O. and Dyatmiko W. (1997). Iridoid glucosides with free radical scavenging properties from Fagraea blumei . Helvetica Chimica Acta 80 : 1144-1152.

Cyr M. and Farrar M.K. (2000). Treatment for posttraumatic stress disorder. Annals of Pharmacotherapy 34 : 366-376.

D

Dahlgreen R. (1983). General aspects of angiosperm evolution and macrosystematics. Nordic Journal of Botany 3: 119-150.

Darbinyan V., Aslanyan G., Amroyan E., Gabrielyan E., Malmstrom C. and Panossian A. (2007). Clinical trial of Rhodiola rosea L. extract SHR-5 in the treatment of mild to moderate depression. Nordic Journal of Psychiatry 61 : 343-348.

Darbinyan V., Kteyan A., Panossian A., Gabrielian E., Wikman G. and Wagner H. (2000). Rhodiola rosea in stress induced fatigue - A double blind cross-over study of a standardized extract SHR-5 with a repeated low-dose regimen on the mental performance of healthy physicians during night duty. Phytomedicine 7: 365-371.

Davies K. and Schwimm K. (2006). Molecular biology and biotechnology of flavonoids biosynthesis. In: Markham KR and Andersen ØM (eds). Flavonoids. Chemistry, biochemistry and applications . Boca Raton: CRC Press Taylor & Francis.

203 6. REFERENCES

De Bock K., Eijnde B.O., Ramaekers M. and Hespel P. (2004). Acute Rhodiola rosea intake can improve endurance exercise performance. International Journal of Sport Nutrition and Exercise Metabolism 14 : 298-307.

De Candolle A.P. (1828). Prodormus naturalis regni vegetabilis . Paris: Treuttel & Wurtz.

De Melo G.O., Malvar D.d.C., Vanderlinde F.A., Pires P.A., Cortes W.S., Germano Filho P., Muzitano M.F., Kaiser C.R. and Costa S.S. (2005). Phytochemical and pharmacological study of Sedum dendroideum leaf juice. Journal of Ethnopharmacology 102 : 217-220.

De Paiva L.S., Hayashi E.A., De Melo G.O., Costa S.S., Koatz V.L.G. and Nobrega A. (2008). Inhibition of B cell development by kalanchosine dimalate. International Immunopharmacology 8: 828-835.

De Rijk M.C., Breteler M.M.B., denBreeijen J.H., Launer L.J., Grobbee D.E., vanderMeche F.G.A. and Hofman A. (1997). Dietary antioxidants and Parkinson disease - The Rotterdam study. Archives of Neurology 54 : 762-765.

De Sanctis R., De Bellis R., Scesa C., Mancini U., Cucchiarini L. and Dacha M. (2004). In vitro protective effect of Rhodiola rosea extract against hypochlorous acid-induced oxidative damage in human erythrocytes. BioFactors 20 : 147-159.

Dementieva L.A. and Laremenko K.V. (1987). Effect of a Rhodiola extract on the tumor process in an experiment. Voprosy Onkologii 33 : 57-60.

E

Ellman G.L., Courtney K.D., Andres jr V. and Featherstone R.M. (1961). A new and rapid colorimetric determination of acetylcholinesterase activity. Biochemical Pharmacology 7: 88- 90.

Emerit J., Edeas A. and Bricaire F. (2004). Neurodegenerative diseases and oxidative stress. Biomedicine & Pharmacotherapy 58 : 39-46.

F

Fan W., Tezuka Y., Ni K.M. and Kadota S. (2001). Prolyl endopeptidase inhibitors from the underground part of Rhodiola sachalinensis . Chemical & Pharmaceutical Bulletin 49 : 396- 401.

Fang S.-D., Yan X.-Q., Li C.-F., Fan Z.-Y., Xu X.-Y. and Xu J.-S. (1979). Separation and structure of the active principles of Sedum sarmentosum Bunge. Kexue Tongbao 24 : 431-432.

Fang S., Yan X., Li J., Fan Z., Xu X. and Xu R. (1982). Studies on the chemical constituents of Sedum sarmentosum Bunge. IV. The structures of sarmentosine and isosarmentosine. Huaxue Xuebao 40 : 273-280.

Finkel T. and Holbrook N.J. (2000). Oxidants, oxidative stress and the biology of ageing. Nature 408 : 239-247.

204 6. REFERENCES

Fitch W.H. (1924). Illustrations of the British Flora . http://caliban.mpiz- koeln.mpg.de/~stueber/fitch/high/IMG_8651.html. Retrieved March, 30th 2009.

Fossen T., Larsen Å., Kiremire B.T. and Andersen Ø.M. (1999). Flavonoids from blue flowers of Nymphaea caerulea . Phytochemistry 51 : 1133-1137.

Franck B. (1958). Alkaloide in Sedum acre und verwandten Sedumarten. Chemische Berichte 91 : 2803-2818.

Frankel E.N., Waterhouse A.L. and Teissedre P.L. (1995). Principal phenolic in selected California wines and their antioxidant activity in inhibiting oxidation of human low- density lipoproteins. Journal of Agricultural and Food Chemistry 43 : 890-894.

Fröderström H. (1930). The genus Sedum . 1. Acta Horti Gothoburgensis 5: 1-75.

Fröderström H. (1931). The genus Sedum . 2. Acta Horti Gothoburgensis 6: 1-111.

Fu K.J. and Ohba H. (2001). Rhodiola (Crassulaceae). pp. 253-263. In: Wu ZY and Raven P (eds). Flora of China . Beijing: Science Press.

Fung S.Y., Schripsema J. and Verpoorte R. (1990). alpha, beta -unsaturated-gamma -lactones from Sedum telephium roots. Phytochemistry 29 : 517-519.

Furmanowa M., Hartwich M., Alfermann A.W., Kozminski W. and Olejnik M. (1999). Rosavin as a product of glycosylation by Rhodiola rosea (roseroot) cell cultures. Plant Cell, Tissue and Organ Culture 56 : 105-110.

Furmanowa M., Oledzka H., Michalska M., Sokolnicka I. and Radomska D. (1995). Rhodiola rosea L. (Roseroot): In vitro regeneration and the biological acivity of roots. In: Bajaj YP (eds). Biotechnology in Agriculture and Forestry, Medicinal and Aromatic Plants VIII . Heidelberg: Springer.

Furmanowa M., Skopinska-Rozewska E., Rogala E. and Hartwich M. (1998). Rhodiola rosea in vitro culture - phytochemical analysis and antioxidant action. Acta Societatis Botanicorum Poloniae 67 : 69-73.

G

Galambosi B. (2006). Demand and availibility of Rhodiola rosea L. raw material. pp. 223-226. In: Bogers RJCraker LE and Lange D (eds). Medicinal and Aromatic Plants . Netherlands: Springer.

Galambosi B., Galambosi Z. and Slacanin I. (2007). Comparison of natural and cultivated roseroot (Rhodiola rosea L.) roots in Finland. Zeitschrift Fur Arznei- & Gewurzpflanzen 12 : 141-147.

Gambelunghe C., Aroni K., Rossi R., Moretti L. and Bacci M. (2008). Identification of N,N- dimethyltryptamine and beta -carbolines in psychotropic ayahuasca beverage. Biomedical Chromatography 22 : 1056-1059.

GBIF Data Portal (2009). www.gbif.net . Retrieved March, 30 th , 2009.

205 6. REFERENCES

Glover V., Liebowitz J., Armando I. and Sandler M. (1982). beta -Carbolines as selective monoamine oxidase inhibitors: in vivo implications. Journal of Neural Transmission 54 : 209-218.

Gnerre C. (2000) Novel monoamine oxidase inhibitors of natural and synthetic origin: mechanism of action and 3D-QSAR. Lausanne: Université de Lausanne. 171 p.

Goad L.J. (1991). Phytosterols. pp. 369-434. In: Dey PM and Harborne JB (eds). Methods in Plant Biochemistry . London: Academic Press.

Godin P. (1954). A new spray reagent for paper chromatography of polyols and cetones. Nature 154 : 134.

Grasso S., Siracusa L., Spatafora C., Renis M. and Tringali C. (2007). Hydroxytyrosol lipophilic analogues: Enzymatic synthesis, radical scavenging activity and DNA oxidative damage protection. Bioorganic Chemistry 35 : 137-152.

Gupta A., Kumar R., Upadhyay N.K., Pal K., Kumar R. and Sawhney R.C. (2007). Effects of Rhodiola imbricata on dermal wound healing. Planta Medica 73 : 774-777.

Gupta R.N. and Spenser I.D. (1967). The biosynthesis of sedamine. Canadian Journal of Chemistry 45 : 1275-1285.

Gupta R.N. and Spenser I.D. (1969). Biosynthesis of N-methylpelletierine. Phytochemistry 8: 1937- 1944.

Gupta V., Saggu S., Tulsawani R.K., Sawhney R.C. and Kumar R. (2008). A dose dependent adaptogenic and safety evaluation of Rhodiola imbricata Edgew, a high altitude rhizome. Food and Chemical Toxicology 46 : 1645-1652.

Gyoergy Z., Tolonen A., Pakonen M., Neubauer P. and Hohtola A. (2004). Enhancing the production of cinnamyl glycosides in compact callus aggregate cultures of Rhodiola rosea by biotransformation of cinnamyl alcohol. Plant Science 166 : 229-236.

György Z. (2006) Glycoside production by in vitro Rhodiola rosea cultures. Oulu: University of Oulu.

H

Halliwell B. and Gutteridge J.M.C. (2007). Free radicals in biology and medicine . 4th ed. New York: Oxford University Inc.

Han X.H., Hong S.S., Hwang J.S., Lee M.K., Hwang B.Y. and Ro J.S. (2007a). Monoamine oxidase inhibitory components from Cayratia japonica . Archives of Pharmacal Research 30 : 13-17.

Han X.H., Hong S.S., Lee D., Lee J.J., Lee M.S., Moon D.-C., Han K., Oh K.-W., Lee M.K., Ro J.S. and others (2007b). Quinolone alkaloids from Evodiae Fructus and their inhibitory effects on monoamine oxidase. Archives of Pharmacal Research 30 : 397-401.

Hanawa F., Tahara S. and Mizutani J. (1991). Isoflavonoids produced by Iris pseudacorus leaves treated with cupric chloride. Phytochemistry 30 : 157-163.

206 6. REFERENCES

Hansen L.I. and Schmidt T. (1985). Major Peter Schnitlers grenseeksaminasjonsprotokoller . Oslo: Norsk Historisk Kjeldeskriftinstitutt.

Hare M.L.C. (1928). Taramine oxidase: New enzyme in liver. Biochemistry Journal 22 : 968-979.

Havsteen B.H. (2002). The biochemistry and medical significance of the flavonoids. Pharmacology & Therapeutics 96 : 67-202.

He A., Wang M., Hao H., Zhang D. and Lee K.-H. (1998). Hepatoprotective triterpenes from Sedum sarmentosum . Phytochemistry 49 : 2607-2610.

Hellenbrand W., Boeing H., Robra B.P., Seidler A., Vieregge P., Nischan P., Joerg J., Oertel W.H., Schneider E. and Ulm G. (1996). Diet and Parkinson's disease. 2. A possible role for the past intake of specific nutrients - Results from a self-administered food-frequency questionnaire in a case-control study. Neurology 47 : 644-650.

Heo B.-G., Park Y.-S., Chon S.-U., Lee S.-Y., Cho J.-Y. and Gorinstein S. (2007). Antioxidant activity and cytotoxicity of methanol extracts from aerial parts of Korean salad plants. BioFactors 30 : 79-89.

Hillhouse B., Ming D.S., French C. and Towers N.G.H. (2004). Acetylcholine esterase inhibitors in Rhodiola rosea . Pharmaceutical Biology 42 : 68-72.

Hostettmann K., Marston A. and Hostettmann M. (1998). Preparative chromatography techniques. Applications in natural product isolation . 2nd ed. Berlin: Springer-Verlag.

Huber L. and George S.A. (1993). Diode array detection in HPLC . 2nd ed. New York: Marcel Dekker.

Hwang J.-S., Lee S.A., Hong S.S., Lee K.S., Lee M.K., Hwang B.Y. and Ro J.S. (2005). Monoamine oxidase inhibitory components from the roots of Sophora flavescens . Archives of Pharmacal Research 28 : 190-194.

J

Jung H.-J., Choi J., Nam J.-H. and Park H.-J. (2007). Anti-ulcerogenic effects of the flavonoid-rich fraction from the extract of Orostachys japonicus in mice. Journal of Medicinal Food 10 : 702- 706.

Jung H.-J., Kang H.-J., Song Yun S., Park E.-H., Kim Y.-M. and Lim C.-J. (2008a). Anti- inflammatory, anti-angiogenic and anti-nociceptive activities of Sedum sarmentosum extract. Journal of Ethnopharmacology 116 : 138-143.

Jung W.-K., Lee D.-Y., Kim J.-H., Choi I., Park S.-G., Seo S.-K., Lee S.-W., Lee C.-M., Park Y.-M., Jeon Y.-J. and others (2008b). Anti-inflammatory activity of caffeic acid phenethyl ester (CAPE) extracted from Rhodiola sacra against lipopolysaccharide-induced inflammatory responses in mice. Process Biochemistry 43 : 783-787.

207 6. REFERENCES

K

Kanupriya, Prasad D., Ram M.S., Kumar R., Sawhney R.C., Sharma S.K., Ilavazhagan G., Kumar D. and Banerjee P.K. (2005). Cytoprotective and antioxidant activity of Rhodiola imbricata against tert-butyl hydroperoxide induced oxidative injury in U-937 human macrophages. Molecular and Cellular Biochemistry 275 : 1-6.

Keli S.O., Hertog M.G.L., Feskens E.J.M. and Kromhout D. (1996). Dietary flavonoids, antioxidant vitamins, and incidence of stroke - The zutphen study. Archives of Internal Medicine 156 : 637-642.

Khaidaev Z. and Menshikkova T.A. (1978). Medicinal plants in Mongolian medicine . Ulan-Bator: Mongolia.

Khanum F., Bawa A.S. and Singh B. (2005). Rhodiola rosea : a versatile adaptogen. Comprehensive Reviews in Food Science and Food Safety 4: 55-62.

Kim Dong W., Son Kun H., Chang Hyeun W., Bae K., Kang Sam S. and Kim Hyun P. (2004). Anti- inflammatory activity of Sedum kamtschaticum . Journal of ethnopharmacology 90 : 409-14.

Kishida M. and Akita H. (2005). Synthesis of rosavin and its analogues based on a Mizoroki-Heck type reaction. Tetrahedron-Asymmetry 16 : 2625-2630.

Kissling J., Ioset J.R., Marston A. and Hostettmann K. (2005). Bio-guided isolation of cholinesterase inhibitors from the bulbs of Crinum x powellii . Phytotherapy Research 19 : 984-987.

Knoll J. (2000). (−) Deprenyl (selegiline): past, present and future. Neurobiology 8: 179-199.

Kong L.D., Cheng C.H.K. and Tan R.X. (2001). Monoamine oxidase inhibitors from rhizoma of Coptis chinensis . Planta Medica 67 : 74-76.

Kooy J.H. (1976). Alkaloids of Sedum Acre . Planta Medica 30 : 295-296.

Kovaleva N.P., Tikhomirov A.A. and Dolgushev V.A. (2003). Specific characteristics of Rhodiola rosea growth and development under photoculture conditions. Russian Journal of Plant Physiology 50 : 527-531.

Krasnov E.A., Duvidzon L.M., Khnykina L.A. and Evstigneeva R.P. (1966). Photochemical study of Rhodiola rosea . Stimulyatory Tsent Nervis Sistem : 72-76.

Kucinskaite A., Poblocka-Olech L., Krauze-Baranowska M., Sznitowska M., Savickas A. and Briedis V. (2007). Evaluation of biologically active compounds in roots and rhizomes of Rhodiold rosea L. cultivated in Lithuania. Medicina-Lithuania 43 : 487-494.

Kurkin V.A. and Zapesochnaya G.G. (1986). Chemical composition and pharmacological studies of Rhodiola plants. Khimiko-Farmatsevticheskii Zhurnal 20 : 1231-1244.

Kurkin V.A., Zapesochnaya G.G., Gorbunov Y.N., Nukhimovskii E.L., Shreter A.I. and Shchavlinskii A.N. (1986). Chemical investigation of some species of the genera Rhodiola L. and Sedum L. and problems of their chemotaxonomy. Rastitel'nye Resursy 22 : 310-19.

208 6. REFERENCES

Kurkin V.A., Zapesochnaya G.G., Nukhimovskii E.L. and Klimakhin G.I. (1988). Chemical composition of rhizomes of a Mongolian Rhodiola rosea L. from districts near Moscow. Khimiko-Farmatsevticheskii Zhurnal 22 : 324-326.

Kurkin V.A., Zapesochnaya G.G. and Shchavlinskii A.N. (1984a). Flavonoids from Rhodiola rosea rhizomes. III. Khimiya Prirodnykh Soedinenii : 390.

Kurkin V.A., Zapesochnaya G.G. and Shchavlinskii A.N. (1984b). Flavonoids from the aerial parts of Rhodiola rosea L. Khimiya Prirodnykh Soedinenii : 657-658.

Kurkin V.A., Zapesochnaya G.G. and Shchavlinskii A.N. (1985). Terpenoids of Rhodiola rosea rhizomes. Khimiya Prirodnykh Soedinenii : 632-636.

L

Lagerberg T., Holmboe J. and Nordhagen R. (1955). Rosenroot. pp. 231-237. In: Lagerberg THolmboe J and Nordhagen R (eds). Vare ville planter . Oslo: Tanum.

Lai X.Y., Zhao Y.Y. and Liang H. (2007). Flavonoid glucuronide from Abelmoschus manihot (L.) Medik. Biochemical Systematics and Ecology 35 : 891-893.

Lange D. (1998). Europe's medicinal and armoatics plants: their use, trade and conservation. Traffic .

Lauber K. and Wagner G. (2007). Flora Helvetica . 3rd ed. Berne: Haupt.

Lee M.-H., Lin R.-D., Shen L.-Y., Yang L.-L., Yen K.-Y. and Hou W.-C. (2001). Monoamine oxidase B and free radical scavenging activities of natural flavonoids in Melastoma candidum D. Don. Journal of Agricultural and Food Chemistry 49 : 5551-5555.

Lee S.M., Na M.K., An R.B., Min B.S. and Lee H.K. (2003). Antioxidant activity of two phloroglucinol derivatives from Dryopteris crassirhizoma . Biological & Pharmaceutical Bulletin 26 : 1354-1356.

Li S. and Zuo C. (1991). Chemical constituents of linear stonecrop ( Sedum lineare ). Zhongcaoyao 22 : 438-440.

Li T., Xu G., Wu L. and Sun C. (2007). Pharmacological studies on the sedative and hypnotic effect of salidroside from the Chinese medicinal plant Rhodiola sachalinensis . Phytomedicine 14 : 601- 604.

Li X., Nakamura S., Matsuda H. and Yoshikawa M. (2008). Bioactive constituents from chinese natural medicines. XXIX. Monoterpene and monoterpene glycosides from the roots of Rhodiola sachalinensis . Chemical & Pharmaceutical Bulletin 56 : 612-615.

Lin L.C. and Chen K.T. (2004). New phenylpropanoid glycoside from Boschniakia rossica . Chinese Pharmaceutical Journal 56 : 77-85.

Linnaei C. (1749). Materia Medica . Stockholm: Laurentii Salvii.

Linnaei C. (1753). Species Plantarum . Stockholm: Laurentii Salvii.

209 6. REFERENCES

Lishmanov I., Maslova L.V., Maslov L.N. and Danshina E.N. (1993). The anti-arrhythmia effect of Rhodiola rosea and its possible mechanism. Biulleten Eksperimentalnoi Biologii i Meditsiny 116 : 175-176.

Lishmanov I.B., Naumova A.V., Afanasev S.A. and Maslov L.N. (1997). Contribution of the opioid system to realization of inotropic effects of Rhodiola rosea extracts in ischemic and reperfusion heart damage in vitro . Eksperimentalnaia I Klinicheskaia Farmakologiia 60 : 34- 36.

Lu Y.R. and Foo L.Y. (1997). Identification and quantification of major polyphenols in apple pomace. Food Chemistry 59 : 187-194.

M

Ma G., Li W., Dou D., Chang X., Bai H., Satou T., Li J., Sun D., Kang T., Nikaido T. and others (2006). Rhodiolosides A-E, monoterpene glycosides from Rhodiola rosea . Chemical & Pharmaceutical Bulletin 54 : 1229-1233.

Mackova Z., Koblovska R. and Lapcik O. (2006). Distribution of isoflavonoids in non-leguminous taxa - An update. Phytochemistry 67 : 849-855.

Magyar K. and Szende B. (2004). (-)-Deprenyl, a selective MAO-B inhibitor, with apoptotic and non- apoptotic properties. Neurotoxicology 25 : 233-242.

Mann J. (1987). Secondary metabolism . 2nd ed. Oxford: Clarendon Press.

Mariani E., Polidori M.C., Cherubini A. and Mecocci P. (2005). Oxidative stress in brain aging, neurodegenerative and vascular diseases: An overview. Journal of Chromatography B- Analytical Technologies in the Biomedical and Life Sciences 827 : 65-75.

Markham K.R. (1982). Techniques of flavonoids identification . Treherne JE and P.H. R (eds). London: Academic Press INC.

Markham K.R. and Chari V.M. (1982). Carbon-13 NMR spectroscopy of flavonoids. pp. 19-134. In: Harborne JB and J. MT (eds). The Flavonoids: Advances in Research . New York: Chapman and Hall.

Marston A., Kissling J. and Hostettmann K. (2002). A rapid TLC bioautographic method for the detection of acetylcholinesterase and butyrylcholinesterase inhibitors in plants. Phytochemical Analysis 13 : 51-54.

Maslova L.V., Kondratev B., Maslov L.N. and Lishmanov I. (1994). The cardioprotective and antiadrenergic activity of an extract of Rhodiola rosea in stress. Eksperimentalnaia I klinicheskaia farmakologiia 57 : 61-63.

Mayuzumi S. and Ohba H. (2004). The phylogenetic position of eastern Asian Sedoideae (Crassulaceae) inferred from chloroplast and nuclear DNA sequences. Systematic Botany 29 : 587-598.

210 6. REFERENCES

Mell C.D. (1938). Dyes, tannins, perfumes, and medicines from Rhodiola rosea . Textile Colorist 60 : 483-484.

Men Y.D. (1986). Studies on the constituents of Sedum alfredii . Taiwan Yaoxue Zazhi 38 : 52-66.

Mishra K.P., Padwad Y.S., Jain M., Karan D., Ganju L. and Sawhney R.C. (2006). Aqueous extract of Rhodiola imbricata rhizome stimulates proinflammatory mediators via phosphorylated lkappaB and transcription factor nuclear factor-kappaB. Immunopharmacology and immunotoxicology 28 : 201-212.

Mitchell J.J., Paiva M. and Heaton M.B. (1999). Vitamin E and beta-carotene protect against ethanol combined with ischemia in an embryonic rat hippocampal culture model of fetal alcohol syndrome. Neuroscience Letters 263 : 189-192.

Mitsui N., Noro T., Kuroyanagi M., Miyase T., Umehara K. and Ueno A. (1989). Studies of Enzyme- Inhibitors. 6. Monoamine-Oxidase Inhibitors from Cinchonae Cortex . Chemical & Pharmaceutical Bulletin 37 : 363-366.

Molokovskij D.S., Davydov V.V. and Khegay M.D. (2002). Comparative estimation of antidiabetic activity of different adaptogenic vegetative preparations and extractions from plant material of some official medicinal plants. Rastitel´nye Resursy 38 : 15-28.

Mort M.E., Soltis D.E., Soltis P.S., Francisco-Ortega J. and Santos-Guerra A. (2001). Phylogenetic relationships and evolution of Crassulaceae inferred from matK sequence data. American Journal of Botany 88 : 76-91.

N

Nakai T. (1938). On the japanese species of the genus Rhodiola L. Journal of Japanese Botany 14 : 491-508.

Nakamura S., Li X., Matsuda H., Ninomiya K., Morikawa T., Yamaguti K. and Yoshikawa M. (2007). Bioactive constituents from Chinese natural medicines. XXVI. Chemical structures and hepatoprotective effects of constituents from roots of Rhodiola sachalinensis . Chemical & Pharmaceutical Bulletin 55 : 1505-1511.

Narhrstedt A., Walther A. and Wray V. (1982). Sarmentosin epoxide, a new cyanogenic compound from Sedum cepaea . Phytochemistry 21 : 107-110.

Narimanian M., Badalyan M., Panosyan V., Gabrielyan E., Panossian A., Wikman G. and Wagner H. (2005). Impact of Chisan (ADAPT-232) on the quality-of-life and its efficacy as an adjuvant in the treatment of acute non-specific pneumonia. Phytomedicine 12 : 723-729.

Nguelefack T.B., Nana P., Atsamo A.D., Dimo T., Watcho P., Dongmo A.B., Tapondjou L.A., Njamen D., Wansi S.L. and Kamanyi A. (2006). Analgesic and anticonvulsant effects of extracts from the leaves of Kalanchoe crenata (Andrews) Haworth (Crassulaceae). Journal of Ethnopharmacology 106 : 70-75.

211 6. REFERENCES

Nguyen D.T.T., Guillarme D., Rudaz S. and Veuthey J.-L. (2006). Chromatographic behaviour and comparison of column packed with sub-2 µm stationary phases in liquid chromatography. Journal of Chromatography A 1128 : 105-113.

Nielsen Allan H., Olsen Carl E. and Moller Birger L. (2005). Flavonoids in flowers of 16 Kalanchoe blossfeldiana varieties. Phytochemistry 66 : 2829-35.

Nierenberg A.A., Alpert J.E., Pava J., Rosenbaum J.F. and Fava M. (1998). Course and treatment of atypical depression. Journal of Clinical Psychiatry 59 : 5-9.

Niessen W.M.A. (1999). Liquid chromatography-mass spectrometry . 2nd ed. New York: Marcel Dekker.

Ninomiya K., Morikawa T., Zhang Y., Nakamura S., Matsuda H., Muraoka O. and Yoshikawa M. (2007). Bioactive constituents from Chinese natural medicines. XXIII. Absolute structures of new megastigmane glycosides, sedumosides A4, A5, A6, H, and I, and hepatoprotective megastigmanes from Sedum sarmentosum . Chemical & Pharmaceutical Bulletin 55 : 1185- 1191.

Novaroli L., Reist M., Favre E., Carotti A., Catto M. and Carrupt P.A. (2005). Human recombinant monoamine oxidase B as reliable and efficient enzyme source for inhibitor screening. Bioorganic & Medicinal Chemistry 13 : 6212-6217.

O

Ohba H. (1978). Generic and infrageneric classification of the Old World Sedoideae. Journal of the Faculty Sciences, Tokyo University Sec. III 12 : 139-198.

Ohba H. (2005). Rhodiola . pp. 210-227. In: Eggli U (eds). Illustrated Handbook of Succulent Plants: Crassulaceae New York: Springer.

Ohsugi M., Fan W., Hase K., Xiong Q., Tezuka Y., Komatsu K., Namba T., Saitoh T., Tazawa K. and Kadota S. (1999). Active-oxygen scavenging activity of traditional nourishing-tonic herbal medicines and active constituents of Rhodiola sacra . Journal of Ethnopharmacology 67 : 111- 9.

Olsson E.M.G., von Scheele B. and Panossian A.G. (2009). A randomised, double-blind, placebo- controlled, parallel-group study of the standardised extract SHR-5 of the roots of Rhodiola rosea in the treatment of subjects with stress-related fatigue. Planta Medica 75 : 105-112.

P

Panossian A., Hambardzumyan M., Hovhanissyan A. and Wikman G. (2007). The adaptogens Rhodiola and Schizandra modify the response to immobilization stress in rabbits by suppressing the increase of phosphorylated stress-activated protein kinase, nitric oxide and cortisol. Drug Target Insights 1: 39-54.

Panossian A., Hovhannisyan A., Abrahamyan H., Gabrielyan E. and Wikman G. (2009). Pharmacokinetic and pharmacodynamic study of interaction of Rhodiola rosea SHR-5 extract with warfarin and theophylline in rats. Phytotherapy Research 23 : 351-357.

212 6. REFERENCES

Panossian A., Nikoyan N., Ohanyan N., Hovhannisyan A., Abrahamyan H., Gabrielyan E. and Wikman G. (2008a). Comparative study of Rhodiola preparations on behavioral despair of rats. Phytomedicine 15 : 84-91.

Panossian A. and Wagner H. (2005). Stimulating effect of adaptogens: an overview with particular reference to their efficacy following single dose administration. Phytotherapy Research 19 : 819-838.

Panossian A., Wikman G., Andreeva L., Boykova A., Nikiforova D. and Timonina N. (2008b). Adaptogens exert a stress protective effect by modulation of expression of molecular chaperones. Planta Medica 74 : 1018-1018.

Perfumi M. and Mattioli L. (2007). Adaptogenic and central nervous system effects of single doses of 3% rosavin and 1% salidroside Rhodiola rosea L. extract in mice. Phytotherapy Research 21 : 37-43.

Petsalo A., Jalonen J. and Tolonen A. (2006). Identification of flavonoids of Rhodiola rosea by liquid chromatography-tandem mass spectrometry. Journal of Chromatography A 1112 : 224-231.

Pogorelyi V.E. and Makarova L.M. (2002). Rhodiola rosea extract for prophylaxis of the ischemic cerebral circulation disorders. Eksperimentalnaia I Klinicheskaia Farmakologiia 65 : 19-22.

Priest R.G., Gimbrett R., Roberts M. and Steinert J. (1995). Reversible and selective inhibitors of monoamine oxidase A in mental and other disorders. Acta Psychiatrica Scandinavica. Supplementum 386 : 40-43.

Provalova N.V., Skurikhin E.G., Pershina O.V., Suslov N.I., Minakova M.Y., Dygai A.M. and Goldberg E.D. (2002). Mechanisms underling the effects of adaptogens on erythropoiesis during paradoxical sleep deprivation. Bulletin of Experimental Biology and Medicine 133 : 428-432.

R

Revina T.A., Krasnov E.A., Sviridova T.P., Stepanyuk G.Y. and Surov Y.P. (1976). Biological characteristics and chemical composition of Rhodiola rosea L. grown in Tomsk. Rastitel'nye Resursy 12 : 355-360.

Rohloff J. (2002). Volatiles from rhizomes of Rhodiola rosea L. Phytochemistry 59 : 655-661.

Ryu S.Y., Han Y.N. and Han B.H. (1988). Monoamine oxidase A inhibitors from medicinal plants. Archives of Pharmacal Research 11 : 230-239.

S

Saratikov S.A. and Krasnov E.A. (1987). Rhodiola rosea is a valuable medicinal plant (Golden root) . Tomsk: Monograph Tomsk State University Press.

Saura J., Luque J.M., Cesura A.M., Daprada M., Chanpalay V., Huber G., Loffler J. and Richards J.G. (1994). Increased monoamine oxidase B activity in plaque-associated astrocytes of Alzheimer bnrains revealed by quantitative enzyme autoradiography. Neuroscience 62 : 15-30.

213 6. REFERENCES

Scarpini E., Scheltens P. and Feldman H. (2003). Treatment of Alzheimer’s disease: current status and new perspectives. Lancet Neurology 2: 539-547.

Schaufelberger D. and Hostettmann K. (1988). Chemistry and pharmacology of Gentiana lactea . Planta Medica : 219-221.

Schönland S. (1891). Crassulaceae. pp. 23-38. In: Engler A and Prantl K (eds). Die natürlichen Pflanzenfamilien Leipzig: Verlag Wilhelm Engelmann.

SciFinder Scholar. 2007. American Chemical Society. Retrieved April, 23rd, 2009.

Scopoli G.A. (1777). Introductio ad historicum naturalem . Praha: Wolfgangum Gerle.

Sethi M.L., Taneja S.C., Dhar K.L. and Atal C.K. (1983). Three isoflavone glycosides from Juniperus macropoda . Phytochemistry 22 : 289-292.

Shawl A.S., Vishwapaul, Zaman A. and Kalla A.K. (1984). Isoflavones of Iris spuria . Phytochemistry 23 : 2405-2406.

Shevtsov V.A., Zholus B.I., Shervarly V.I., Vol'skij V.B., Korovin Y.P., Khristich M.P., Roslyakova N.A. and Wikman G. (2003). A randomized trial of two different doses of a SHR-5 Rhodiola rosea extract versus placebo and control of capacity for mental work. Phytomedicine 10 : 95- 105.

Shih C.-D., Kuo D.-H., Huang C.-W., Gu Y.-H. and Chen F.-A. (2008). Autonomic nervous system mediates the cardiovascular effects of Rhodiola sacra radix in rats. Journal of Ethnopharmacology 119 : 284-90.

Shih J.C. and Thompson R.F. (1999). Monoamine oxidase in neuropsychiatry and behavior. American Journal of Human Genetics 65 : 593-598.

Sies H. (1985). Oxidative stress . Sies H (eds). London: Academic Press.

Sikorska M., Matlawska I. and Franski R. (2004). 8-hydroxyflavonoid glucuronides of Malope trifida . Acta Physiologiae Plantarum 26 : 291-297.

Small E. and Catling P.M. (1999). Canadian Medicinal Crops . Ottawa: NRC Research Press.

Son J.K., Rosazza J.P.N. and Duffel M.W. (1990). Vinblastine and vincristine are inhibitors of monoamine oxidase B. Journal of Medicinal Chemistry 33 : 1845-1848.

Spasov A.A., Wikman G.K., Mandrikov V.B., Mironova I.A. and Neumoin V.V. (2000). A double- blind, placebo-controlled pilot study of the stimulating and adaptogenic effect of Rhodiola rosea SHR-5 extract on the fatigue of students caused by stress during an examination period with a repeated low-dose regimen. Phytomedicine 7: 85-89.

Stancheva S.L. and Mosharrof A. (1987). Effect of the extract of Rhodiola rosea L. on the content of the brain biogenic monoamines. Dokladi Na Bolgarskata Akademiya Na Naukite 40 : 85-87.

Stephenson R. (1994). Sedum cultivated stonecrops . Portland: Timber Press, Inc.

214 6. REFERENCES

Stevens J.F. (1995). Chemotaxonomy of the Eurasian Sedoideae and Sempervivoideae. In: 't Hart H and Eggli U (eds). Evolution and Systematics of the Crassulaceae . Leiden: Backhuys Publishers.

Stevens J.F., Hart H., Elema E.T. and Bolck A. (1996). Flavonoid variation in eurasian Sedum and Sempervivum . Phytochemistry 41 : 503-512.

Stevens J.F., Hart H., Van Ham R., Elema E.T., Van Den Ent M., Wildeboer M. and Zwaving J. (1995). Distribution of alkaloids and tannins in the Crassulaceae. Biochemical Systematics and Ecology 23 : 157-165.

Su X.-F., Zhang H., Shao J.-X. and Wu H.-Y. (2007). Theoretical study on the structure and properties of crenulatin molecule in herb Rhodiola crenulata . THEOCHEM 847 : 59-67.

Su Z.-h. and Hua H.-m. (2005). Chemical constituents and pharmacology of Sedum sarmentosum . Zhongyao Yanjiu Yu Xinxi 7: 19-20.

Suzuki O., Katsumata Y., Oya M., Chari V.M., Vermes B., Wagner H. and Hostettmann K. (1981). Inhibition of type A and B monoamine oxidase by naturally occuring xanthones. Planta Medica 39 : 19-23.

Szabo M.R., Iditoiu C., Chambre D. and Lupea A.X. (2007). Improved DPPH determination for antioxidant activity spectrophotometric assay. Chemical Papers 61 : 214-216.

T

Tadeg H., Mohammed E., Asres K. and Gebre-Mariam T. (2005). Antimicrobial activities of some selected traditional Ethiopian medicinal plants used in the treatment of skin disorders. Journal of Ethnopharmacology 100 : 168-175.

Tagami M., Yamagata K., Ikeda K., Nara Y., Fujino H., Kubota A., Numano F. and Yamori Y. (1998). Vitamin E prevents apoptosis in cortical neurons during hypoxia and oxygen reperfusion. Laboratory Investigation 78 : 1415-1429.

Takaya Y., Furukawa T., Miura S., Akutagawa T., Hotta Y., Ishikawa N. and Niwa M. (2007). Antioxidant constituents in distillation residue of Awamori spirits. Journal of Agricultural Food and Chemistry 55 : 75-79.

Takhtajan A. (1997). Diversity and Classification of Flowering Plants . New York: Columbia University Press.

Tao Li H.Z. (2008). Application of microscopy in authentication of traditional Tibetan medicinal plants of five Rhodiola (Crassulaceae) alpine species by comparative anatomy and micromorphology. Microscopy Research and Technique 71 : 448-458.

Taubert D., Breitenbach T., Lazar A., Censarek P., Harlfinger S., Berkels R., Klaus W. and Roesen R. (2003). Reaction rate constants of superoxide scavenging by plant antioxidants. Free Radical Biology and Medicine 35 : 1599-1607.

215 6. REFERENCES

Tebbe B. (2001). Relevance of oral supplementation with antioxidants for prevention and treatment of skin disorders. Skin Pharmacology and Applied Skin Physiology 14 : 296.

't Hart H. (1982). The white-flowered Sedum species, 1. Principles of a phylogenetic classification of the Sedoideae (Crassulaceae) and the position of the white-flowered Sedum species. Proceedings of the Koninklijke Nederlandse Academie van Wetenschappen. Series C 85 : 663- 675.

't Hart H. (1995). Infrafamilial and generic classification of the Crassulaceae. pp. 159-172. In: 't Hart H and Eggli U (eds). Evolution and Systematics of the Crassulaceae . Leiden: Backhuys.

't Hart H. (1997). Crassulaceae. pp. 20-22. In: Oldfield S (eds). Cactus and succulent plants - status survey and conservation action plan. Gland: IUCN/SSC Cactus and Succulent Specialist Group.

't Hart H. (2003). of Europe, stonecrops and wallpeppers . Eggli U (eds). Lisse: A.A. Balkema Publishers.

't Hart H. and Bleij B. (2005). Sedum . pp. 234-260. In: Eggli U (eds). Illustrated Handbook of Succulent Plants: Crassulaceae . New York: Springer.

't Hart H. and Eggli U. (1988). Cytotaxonomic studies on Rosularia (Crassulaceae). Botanica Helvetica 98 : 223-234.

't Hart H. and Eggli U. (1995). Evolution and Systematics of the Crassulaceae . 't Hart H and Eggli U (eds). Leiden: Backhuys.

't Hart H., van Ham R.D.H.J., Stevens J.F., Elema E.T., van der Klis H. and Gadella T.W.J. (1999). Biosystematic, molecular and phytochemical evidence for the multiple origin of sympetaly in Eurasian Sedoideae (Crassulaceae). Biochemical Systematics and Ecology 27 : 407-426.

The Angiosperm Phylogeny Group (1998). An ordinal classification for the families of flowering plants. Annals of the Missouri Botanical Garden : 88 : 531-553.

Thiede J. and Eggli U. (2007). Crassulaceae. pp. 83-119. In: Kubitzki K (eds). Flowering plants. . New York: Springer.

Thieme H., Walewska E. and Winkler H.J. (1969). Isolation of salidroside from leaves of Rhododendron ponticum x catawbiense . Pharmazie 24 : 118-119.

Thorne R.F. (1992). Classification and geography of flowering plants. Botanical Review 58 : 225-348.

Thull U. (1995) Monoamine oxidase inhibitors of natural and synthetic origin: biological assay and 3D-QSAR. Lausanne: Université de Lausanne. 178 p.

Thuong P.T., Ho J.K., MinKyun N., Wenyi J., Ui J.Y., Yeon H.S., Kyung-Sik S., Byung-Sun M. and Bae K. (2007). Anti-oxidant constituents from Sedum takesimense . Phytochemistry 68 : 2432- 2438.

216 6. REFERENCES

Tolonen A., Gyoergy Z., Jalonen J., Neubauer P. and Hohtola A. (2004). LC/MS/MS identification of glycosides produced by biotransformation of cinnamyl alcohol in Rhodiola rosea compact callus aggregates. Biomedical Chromatography 18 : 550-558.

Tolonen A., Hohtola A. and Jalonen J. (2003a). Comparison of electrospray ionization and atmospheric pressure chemical ionization techniques in the analysis of the main constituents from Rhodiola rosea extracts by liquid chromatography/mass spectrometry. Journal of Mass Spectrometry 38 : 845-853.

Tolonen A., Pakonen M., Hohtola A. and Jalonen J. (2003b). Phenylpropanoid glycosides from Rhodiola rosea . Chemical & Pharmaceutical Bulletin 51 : 467-470.

Troshchenko A.T. and Kutikova G.A. (1967). Rhodioloside from Rhodiola rosea and R. quadrifida L. Khimiya Prirodnykh Soedinenii 3: 244-9.

Tu Y., Roberts L., Shetty K. and Schneider S.S. (2008). Rhodiola crenulata induces death and inhibits growth of breast cancer cell lines. Journal of Medicinal Food 11 : 413-423.

U

Udintsev S.N. and Shakhov V.P. (1991). The role of humoral factors of regenerating liver in the development of experimental tumors and the effect of Rhodiola rosea extract on this process. Neoplasma 38 : 323-331.

Uhl C.H. (1992). Polyploidy, dysploidy, and chromosome pairing in Echeveria (Crassulaceae) and its hybrids. American Journal of Botany 79 : 556-56.

Urbain A., Marston A., Grilo L.S., Bravo J., Purev O., Purevsuren B., Batsuren D., Reist M., Carrupt P.A. and Hostettmann K. (2008). Xanthones from Gentianella amarella ssp acuta with acetylcholinesterase and monoamine oxidase inhibitory activities. Journal of Natural Products 71 : 895-897.

V

Van Acker S.A.B.E., Van Den Berg D.-j., Tromp M.N.J.L., Griffioen D.H., Van Bennekom W.P., Van Der Vijgh W.J.F. and Bast A. (1996). Structural aspects of antioxidant activity of flavonoids. Free Radical Biology and Medicine 20 : 331-342.

Van Ham R. (1995). Phylogenetic relationships in the Crassulaceae inferred from chloroplast DNA variation. In: 't Hart H and Eggli U (eds). Evolution and systematics of the Crassulaceae . Leiden: Backhuys Publishers.

Van Ham R. and 't Hart H. (1998). Phylogenetic relationships in the Crassulaceae inferred from chloroplast DNA restriction-site variation. American Journal of Botany 85 : 123-134.

Van Sumere C.F. (1989). Phenols and phenolic acids, pp. 29-74, in Dey P.M. and Harborne J.B. (eds). Methods in plant biochemistry. London: Academic Press.

Van Wyk B.-E. and Winter P.J.D. (1995). The homology of red flower color in Crassula , Cotyledon and Tylecodon (Crassulaceae). Biochemical Systematics and Ecology 23 : 291-293.

217 6. REFERENCES

Vanderwal R., Kooy J.H. and Vaneijk J.L. (1981). Phytochemical investigation of Sedum acre L. Planta Medica 43 : 97-99.

Varma R.K., Garg B.D. and Ahmad A. (1986). Phytochemical studies on Kalanchoe integra . Fitoterapia 57 : 433-434.

Vetter J. (2000). Plant cyanogenic glycosides. Toxicon 38 : 11-36.

Viscardi P., Reynaud J. and Raynaud J. (1984). A new isoflavone glucoside from the flowers of Cytisus scoparius Link (Leguminosae). Pharmazie 39 : 781-782.

Vollhardt K. and Schore N. (1995). Traité de chimie organique . 2nd ed. Bruxelles: De Boek-Wesmael S.A.

W

Wagner H. and Bladt S. (1996). Plant drug analysis. A thin layer chromatography atlas . 2nd ed. New York: Springer-Verlag.

Walker T.B. and Robergs R.A. (2006). Does Rhodiola rosea possess ergogenic properties? International Journal of Sport Nutrition and Exercise Metabolism 16 : 305-315.

Wang H., Zhou G., Gao X., Wang Y. and Yao W. (2007). Acetylcholinesterase inhibitory-active components of Rhodiola rosea L. Food Chemistry 105 : 24-27.

Wang K., Chen L., Li N. and Yu X. (2006). Antioxidant and radical-scavenging activity of flavonoids from Solidago canadensis . Zhongguo Yaoxue Zazhi 41 : 493-497.

Wang S. and Wang F.P. (1992). Chemical components of Rhodiola crenulata . Yaoxue Xuebao 27 : 117-120.

Westlund K.N., Denney R.M., Rose R.M. and Abell C.W. (1988). Localization of distinct monoamine oxidase A and monoamine oxidase B cell populations in human brainstem. Neuroscience 25 : 439-456.

Wiedenfeld H., Dumaa M., Malinowski M., Furmanowa M. and Narantuya S. (2007). Phytochemical and analytical studies of extracts from Rhodiola rosea and Rhodiola quadrifida . Pharmazie 62 : 308-311.

Wolbis M. (1987). Polyphenolic compounds in certain species of Sedum . I. Coumarins and phenolic acids in Sedum acre , Sedum album , and Sedum maximum . Acta Poloniae Pharmaceutica 44 : 563-566.

Wolbis M. (1989). Flavonol glycosides from Sedum album . Phytochemistry 28 : 2187-2189.

218 6. REFERENCES

Y

Yang H., Mei S., Peng L., Lin Z. and Sun H. (2002). New glucoside from Rhodiola fastigiata (Crassulaceae). Acta Botanica Sinica 44 : 224-226.

Yaremii I.N. and Grigoreva N.F. (2002). Hepatoprotector properties of a liquid extract from Rhodiola rosea . Eksperimentalnaia I Klinicheskaia Farmakologiia 65 : 57-59.

Yoshikawa M.; Nihon Yakuyo Shokuhin Kenkyusho K. K., Japan; Harima Kanpo Seiyaku Co., Ltd., assignee. 2006 20050302. Sinocrassula -derived components and usage thereof. Japan patent 2005-58051; 2006241054.

Yoshikawa M., Shimada H., Horikawa S., Murakami T., Shimoda H., Yamahara J. and Matsuda H. (1997). Bioactive constituents of Chinese natural medicines. IV. Rhodiolae Radix (2): on the release inhibitors from the underground part of Rhodiola sacra (Prain ex Hamet) S. H. Fu (Crassulaceae): chemical structures of rhodiocyanoside D and sacranosides A and B. Chemical & Pharmaceutical Bulletin 45 : 1498-1503.

Yoshikawa M., Shimada H., Shimoda H., Matsuda H., Yamahara J. and Murakami N. (1995). Rhodiocyanosides A and B, new antiallergic cyanoglycosides from Chinese natural medicine "si lie hong jing tian", the underground part of Rhodiola quadrifida (Pall.) Fisch. et Mey. Chemical & Pharmaceutical Bulletin 43 : 1245-1247.

Yoshikawa M., Shimada H., Shimoda H., Murakami N., Yamahara J. and Matsuda H. (1996). Bioactive constituents of Chinese natural medicines. II. Rhodiolae Radix. (1). Chemical structures and antiallergic activity of rhodiocyanosides A and B from the underground part of Rhodiola quadrifida (Pall.) Fisch. et Mey. (Crassulaceae). Chemical & Pharmaceutical Bulletin 44 : 2086-2091.

Yousef G.G., Grace M.H., Cheng D.M., Belolipov I.V., Raskin I. and Lila M.A. (2006). Comparative phytochemical characterization of three Rhodiola species. Phytochemistry 67 : 2380-2391.

Yuan Z., Tezuka Y., Fan W.Z., Kadota S. and Li X. (2002). Constituents of the underground parts of Glehnia littoralis . Chemical & Pharmaceutical Bulletin 50 : 73-77.

Z

Zapesochnaya G.G. and Kurkin V. (1983a). Glycosides of cinnamyl alcohol from the rhizomes of Rhodiola rosea . Chemistry of Natural Compounds 18 : 685-688.

Zapesochnaya G.G. and Kurkin V.A. (1982). Cinnamic glycosides of Rhodiola rosea rhizomes. Khimiya Prirodnykh Soedinenii : 723-727.

Zapesochnaya G.G. and Kurkin V.A. (1983b). Flavonoids of Rhodiola rosea rhizomes. II. and herbacetin glycosides. Khimiya Prirodnykh Soedinenii : 23-32.

Zapesochnaya G.G., Kurkin V.A. and Shchavlinskii A.N. (1985). Flavonoids of the above-ground part of Rhodiola rosea . II. Structure of novel glycosides of herbacetin and gossypetin. Khimiya Prirodnykh Soedinenii : 496-507.

219 6. REFERENCES

Zapesochnaya G.G., Kurkin V.A. and Shchavlinskii A.N. (1987). The chemical study of Rhodiola rosea L. Chemistry and Biotechnology of Biological Active Natural Product 4: 404-408.

Zeller E.A. (1938). Ueber den enzymatischen Abbau von Hisatmin and Diamine. Helvetica Chimia Acta 21 : 880-890.

Zhang L., Yu H., Sun Y., Lin X., Chen B., Tan C., Cao G. and Wang Z. (2007a). Protective effects of salidroside on hydrogen peroxide-induced apoptosis in SH-SY5Y human neuroblastoma cells. European Journal of Pharmacology 564 : 18-25.

Zhang Y., Gao Y., Fu R., Mikiro T. and Zhu B. (2004). Study on the antioxidant activity of the Chinese herb Rhodiola Sachalinensis A. Bor extract. Shipin Kexue 25 : 315-318.

Zhang Y., Morikawa T., Nakamura S., Ninomiya K., Matsuda H., Muraoka O. and Yoshikawa M. (2007b). Bioactive constituents from chinese natural medicines. XXV. New flavonol bisdesmosides, sarmenosides I, II, III, and IV, with hepatoprotective activity from Sedum sarmentosum (Crassulaceae). Heterocycles 71 : 1565-1576.

220