B

AMJRa

b

c

d

e

a

ARRAA

KGLAAATH

1

iesFbSooa

0d

Journal of Ethnopharmacology 122 (2009) 384–393

Contents lists available at ScienceDirect

Journal of Ethnopharmacology

journa l homepage: www.e lsev ier .com/ locate / je thpharm

ioactivity studies and chemical profile of the antidiabetic plant Genista tenera

mélia P. Rautera,∗, Alice Martinsa, Rui Lopesa, Joana Ferreiraa, Luísa M. Serralheiroa,aria-Eduarda Araújoa, Carlos Borgesa, Jorge Justinob, Filipa V. Silvab, Margarida Goulartb,

ane Thomas-Oatesc, João A. Rodriguesc, Emma Edwardsc, João P. Noronhad,ui Pintoe, Hélder Mota-Filipee

Departamento de Química e Bioquímica/Centro de Química e Bioquímica, Faculdade de Ciências da Universidade de Lisboa, Ed. C8, 5◦ piso, 1749-016 Lisboa, PortugalEscola Superior Agrária de Santarém, Instituto Politécnico de Santarém, Apartado 279, 2001-904 Santarém, PortugalDepartment of Chemistry, University of York, Heslington, York, Y10 5DD, UKDepartamento de Química/REQUIMTE-CQFB, Faculdade de Ciências e Tecnologia da Universidade Nova de Lisboa, 2829-516 Caparica, PortugalLaboratório de Farmacologia, Faculdade de Farmácia da Universidade de Lisboa, Av. das Forcas Armadas 1649-019 Lisboa, Portugal

r t i c l e i n f o

rticle history:eceived 3 January 2008eceived in revised form 3 October 2008ccepted 6 October 2008vailable online 1 November 2008

eywords:enista teneraeguminosaentihyperglycaemic activityntioxidant activitycetylcholinesterase inhibition

a b s t r a c t

Aim of the study: Genista tenera is a plant endemic to the island of Madeira and is used in folk medicine tocontrol diabetes. In the present work we evaluate the antihyperglycaemic activity of its n-butanol extractand determine its chromatographic profile. In addition, this extract, the ethyl acetate and diethyl etherplant extracts were studied in order to assess the plant antioxidant and acetylcholinesterase inhibitoryactivities, as well as its cyto- and genotoxicities.Materials and methods: HPLC-DAD-ESI-MS was used to analyze the flavonoid profile of the n-butanolextract. The antihyperglycaemic activity of this extract was performed over streptozotocin induceddiabetic Wistar rats (200 mg/kg, bw/day), for 15 days. Antioxidant activity (DPPH assay) and acetyl-cholinesterase inhibitory effect (Ellman method) were also performed. Acute cytotoxicity and genotoxicitywere assessed by proliferative index quantification and the short-term chromosomal aberration tech-nique, after exposure of lymphocytes to the extracts.

oxicityPLC–DAD–ESI-MS

Results and conclusions: The n-butanol extract, where 21 monoglycosyl and 12 diglycosyl flavonoids weredetected, significantly lowered blood glucose levels, bringing them to normal values after 15 days oftreatment. The best radical scavenging activity was observed for the ethyl acetate extract (48.7% at139.1 �g/mL), which was also the most effective one at the minimal concentration tested. The highestacetylcholinesterase inhibitory activity (77.0% at 70.0 �g/mL) was also obtained with the ethyl acetateextract. In vitro toxicity studies showed no evidence for acute cytotoxicity or genotoxicity. This is the first

vity o

etfl3waab

report on antidiabetic acti

. Introduction

Genista tenera (Jacq. ex Murr) O. Kuntze is a bush endemic to thesland of Madeira, Portugal. It has characteristic small yellow flow-rs and grows on rocky soils between sea level and 1500 m aboveea level. Flowering occurs between March and July (Jardim andrancisco, 2000). An infusion of the aerial parts of the plant is usedy the local population as an adjuvant for the treatment of diabetes.

tudy of the plant’s chemical composition showed the presencef alkaloids and flavonoids, which are chemotaxonomic markersf the genus Genista (Harborne, 1994). The alkaloid content of theerial parts of the plant has been previously reported (Martins

∗ Corresponding author. Tel.: +351 21 7500952; fax: +351 21 7500088.E-mail address: aprauter@fc.ul.pt (A.P. Rauter).

aritflpwh

378-8741/$ – see front matter © 2008 Elsevier Ireland Ltd. All rights reserved.oi:10.1016/j.jep.2008.10.011

f genus Genista.© 2008 Elsevier Ireland Ltd. All rights reserved.

t al., 2005). In addition FAB-MS/MS has been used to identifyhe flavonoids in a diethyl ether extract of Genista tenera. Theavones apigenin and chrysoeriol, and the isoflavones genistein,′-O-methylorobol, 5-O-methylgenistein and alpinumisoflavoneere identified (Borges et al., 2001; Martins et al., 2002). The ethyl

cetate extract has been previously analysed by LC–DAD–ESI-MSnd NMR, genistein-8-C-glucoside and genistein-7-O-glucosideeing identified as major constituents, luteolin-7-O-glucosides a minor compound, while luteolin-7,3′-di-O-glucoside andutin were found in trace amounts (Rauter et al., 2005). Cap-llary electrophoresis–mass spectrometric characterisation of

he n-butanol extract indicated the presence of 26 differentavonoid components (Edwards et al., 2006). These extracts areotentially useful sources of bioactive compounds, due to theell-documented activity of flavonoids (Havsteen, 2002). Manyerbal medicines have been recommended for the treatment of

ophar

dptAitOcrtcflcpccO

lapAMaApGpogo

dcsost

aLdbat

2

2

(vJ

2

eraten

va

2

iaLl(0

(T9e

M(wSw

u(tt

u5amwtsdt

dol2(iaoraa(Bat211a

(

A.P. Rauter et al. / Journal of Ethn

iabetes (Jung et al., 2006) and the antidiabetic activity of severallant extracts has been correlated with their antioxidant proper-ies and flavonoid content (Sezik et al., 2005; Aslan et al., 2007).s antioxidants, these compounds may prevent the progressive

mpairment of pancreatic beta-cell function due to oxidative stress,hus reducing the occurrence of type 2 diabetes (Song et al., 2005).xidative stress is also associated with Diabetes mellitus secondaryomplications (Vitor et al., 2004) such as vascular diseases, andetinopathies. Several epidemiological studies have reported thathe regular intake of flavonoids is effective against cancer andardiovascular diseases (Mariappan et al., 2006). Numerous dietaryavonoids have shown a beneficial impact over parameters asso-iated with arteriosclerosis, including lipoprotein oxidation, bloodlatelet aggregation and vascular reactivity (Peluso, 2006). Patentslaiming the health benefits of plant extracts and/or nutraceuti-als containing flavonoids have also emerged (Kim et al., 2006;moigui, 2006; Nishizono et al., 2007; Raederstorff et al., 2007).

Reactive oxygen species are responsible for various cellu-ar anomalies such as protein damage, deactivation of enzymes,lteration of DNA and lipid peroxidation which lead to severalathological conditions, including neurological disorders such aslzheimer’s and Parkinson’s diseases (Kaur and Geetha, 2006).ultiple pathogenic factors including oxidative stress and reduced

cetylcholine levels have been implicated in Alzheimer’s disease.mong the potential agents, flavonoids are promising lead com-ounds to combat this neurological disorder (Ji and Zhang, 2006).enistein, one of the components of the plant diethyl ether extract,revents glucose autoxidation-mediated atherogenic modificationf low-density lipoprotein (LDL) (Exner et al., 2001), and neurode-enerative diseases (Sonee et al., 2004; McCarty, 2006), by virtuef its antioxidant properties.

One of the most promising approaches for treating Alzheimer’sisease is to enhance acetylcholine levels in the brain using acetyl-holinesterase inhibitors (Wilkinson et al., 2004). However someynthetic drugs used to treat the cognitive dysfunctions and mem-ry loss associated with Alzheimer’s disease have only moderateymptomatic benefits, encouraging research into new inhibitors ofhis enzyme.

The aim of this work was to evaluate the antihyperglycaemicctivity of Genista tenera n-butanol extract, and to perform itsC–DAD–ESI-MS analysis. In addition, evaluation of the antioxi-ant activity and of the inhibition of acetylcholinesterase promotedy its diethyl ether, ethyl acetate and n-butanol extracts was alsoccomplished, as well as in vitro toxicity studies, in order to accesshe biological potential of the plant.

. Materials and methods

.1. Plant material

The plant was identified and collected on the island of MadeiraPortugal) by Dra. Susana Fontinha and Padre Manuel Nóbrega. Aoucher specimen (MADJ 2508) is deposited in the Herbarium ofardim Botânico da Madeira, Funchal.

.2. Extraction

The powdered aerial parts (162.71 g) were exhaustivelyxtracted in a Soxhlet apparatus with ethanol (3 × 1000 mL, Pan-

eac). The crude extract was concentrated to dryness under vacuumt 40 ◦C and the residue (40.49 g) was dissolved in water at roomemperature, filtered and extracted by partition, first with diethylther (2 × 500 mL) and ethyl acetate (2 × 500 mL), and then with-butanol (2 × 500 mL) (Merck). The solvent was evaporated under

ico01

macology 122 (2009) 384–393 385

acuum to give the diethyl ether (2.87 g), the ethyl acetate (2.41 g)nd the n-butanol (7.41 g) extracts.

.3. LC–DAD–ESI-MS analyses

Separations were carried out using a liquid chromatographnterfaced to an ion trap mass spectrometer. The LC system was

Surveyor (Thermo Electron, San Jose, CA, USA), consisting of anC pump and autosampler, equipped with an injector with a 10 �Loop. Chromatography was performed on a Zorbax Eclipse column3.0 mm × 150 mm, 5 �m) using a linear gradient at a flow rate of.5 mL/min.

The initial composition of the mobile phase was 90% A and 10% BA = H2O + 0.5% formic acid and B = acetonitrile + 0.1% formic acid).he amount of B was increased to 50% over 20 min and then to0% over 10 min. Finally, the %B was lowered and the column re-quilibrated.

Electrospray ionisation (ES) tandem mass spectrometry (ES-S/MS) was carried out on an ion trap mass spectrometer (ITMS)

LCQ Deca XP Plus, Thermo Electron, San Jose, CA, USA) equippedith a Thermo Electron orthogonal electrospray interface. Xcalibur

oftware (version 1.3, Thermo Electron) was used to control thehole LC–MS system.

Ionisation and mass spectrometric conditions were optimizedsing the TunePlus optimization function of the Xcalibur Softwareversion 1.3), using the instrument’s syringe pump (20 �L/min) toee the plant extract into the LC eluent (0.5 mL/min, 50:50 A:B) anduning on m/z 285 (a dihydroxy methoxy flavone).

The positive ion mode was used for MS and MS/MS experiments,sing the following conditions: spray voltage 4 kV, tube lens offset0 V, transfer capillary temperature 330 ◦C, 60 units of nebulisingnd 20 units of auxiliary gas. The scan range was m/z 150–1500, theaximum injection time was set to 100 ms and two microscansere used. While performing the MS/MS analyses, two scan func-

ions were used; the first scan event was to obtain a full-scan masspectrum and the second was an MS/MS experiment, using the dataependent acquisition mode. MS/MS data were acquired on thehree most intense ions, using a dynamic exclusion list.

An alternative HPLC–DAD–ESI-MS analysis was performed inifferent conditions in order to tentatively assign the structuref the major flavonoids detected. These conditions were as fol-ows: the n-butanol extract was dissolved in methanol (HPLC grade,5 �g/mL). Samples (30 �L) were analysed using a HPLC systemWaters) coupled with a pump and controller (Waters 600), ann-line degasser (Waters), an autosampler (Waters 717 plus) andphotodiode array detector (DAD) (Waters 996). The system wasperated with MassLynkTM 3.0 software (Micromass, UK Ltd.). Aeversed-phase �Bondapak C-18 column (300 mm × 3.9 mm i.d.nd particle size 10 �m) was used. The Waters pump was oper-ted at 1 mL/min using the following eluents: water–formic acid99.5:0.5) (eluent A) and acetonitrile–formic acid (99.9:0.1) (eluent). The elution program used was based on the following percent-ge changes of the eluent B: from 6% until 16% in 14 min; from 16%o 17% until 16 min; from 17% to 18% until 18 min; isocratic until0 min; from 18% until 19% until 22 min; isocratic until 24 min; from9% to 20% until 27 min; from 20% to 100% until 45 min and from00% to 6% until 46 min. DAD detector was operated between 200nd 600 nm.

LC–DAD–ESI-MS were carried in a quadrupole VG PlatformMicromass, UK Ltd.) spectrometer equipped with an electrospray

onisation source operating in positive and negative modes. Thehromatographic separation was performed with the same typef equipment and conditions described above, with a flow of.1 mL/min, to the MS. The ionisation conditions were adjusted at20 ◦C and 3.5 kV for capillary temperature and voltage, respec-

3 ophar

t1m

2

2

(Lnaa1

2

imsFplbia

•

•

•

8wtdfid

c

2

gear2l

2

vsdtP

2

s

wo1dT0ac

2

oDbiop(cbfbsCc5paa

2

eiebaH4scca3caaptcRm

3

3

86 A.P. Rauter et al. / Journal of Ethn

ively. Nitrogen was used as drying and nebulising gas at 300 and0 mL/min, respectively. The full-scan mass covered the range from/z 100 up to 700.

.4. Antihyperglycaemic activity

.4.1. AnimalsMale Wistar rats weighing 178–270 g were used in this study

handled throughout at Faculdade de Farmácia da Universidade deisboa). Animal experiments were carried out following the inter-ationally accepted ethical guidelines for the care of laboratorynimals. They were fed with astandard pellets and water ad-libitumnd maintained at 22 ◦C temperature, 85% relative humidity, and a2 h day and night cycle.

.4.2. Streptozotocin-induced diabetic ratsDiabetes was induced in rats by a single intraperitoneal (i.p.)

njection of streptozotocin (50 mg/kg b.w., Sigma) dissolved in nor-al saline. One group of six identical rats was kept without

treptozotocin administration as the normal control (group I).orty-eight hours after streptozotocin administration, blood sam-les were obtained from the tips of the rats’ tails and glucose

evels determined to confirm diabetes. The diabetic rats exhibitinglood glucose levels above 180 mg/dL were selected for the stud-

es. These rats were sub-divided randomly into three groups (sixnimals/group) and treated orally as follows:

Group II (untreated diabetic rats) received 5% Tween 80 (0.5 mL,Bio Rad).Group III (diabetic rats) received Genista tenera n-butanol extract(200 mg/kg b.w.) in 0.5 mL 5% Tween 80.Group IV (diabetic rats) received 5% Tween 80 (0.5 mL) containingglibenclamide (0.5 mg/kg b.w., Sigma).

The normal control group of rats (group I) received 5% Tween0 (0.5 mL). Treatments were continued daily for 15 days. Bloodas collected from the tips of the rats’ tails for glucose estima-

ion just before drug administration on the 1st day and 1 h afterrug administration on days 4, 7 and 10. The animals were sacri-ced after blood collection, under Eutasil anesthesia on the 15thay.

Blood glucose level was determined using a glucometer Accu-hek sensor (Roche, USA).

.4.3. Glucose tolerance testA glucose tolerance test was performed for all experimental

roups of rats to screen for possible occurrence of hypoglycaemicffects. The rats were loaded with a glucose solution 20% (oraldministration). Blood samples were collected from the tips of theats’ tails before glucose administration, and at 15, 30, 60, 120 and40 min after glucose loading. Serum was separated and glucoseevels were measured immediately.

.4.4. Statistical evaluationStatistical evaluation was carried out using one-way analyses of

ariance (ANOVA) followed by Bonferroni’s test using the GraphPadoftware, GraphPad Prism version 3.00 for Windows, except for theata concerning the effect of the extract on body weight and in vitrooxicity studies, which were evaluated using the Student’s t-test. Avalue <0.05 was considered significant.

.5. Antioxidant activity (DPPH assay)

The extracts were dissolved in dimethylsulfoxide to preparetock solutions over the concentration range 0.01–1.30 mg/mL,

bb2tt

macology 122 (2009) 384–393

hich were then used for the radical scavenging assay. The capacityf the prepared extracts to scavenge the free radical 2,2-diphenyl--picrylhydrazyl (DPPH) was monitored according to the methodescribed by Ferreira et al. (2006) and references cited therein.he extract (50 �L) was added to a methanolic solution of DPPH.002% in methanol (5 mL), and the mixture left at room temper-ture for 30 min. Absorption was measured at 517 nm against theorresponding blank.

.6. Acetylcholinesterase inhibitory activity

The assay for measuring AChE activity used was basedn the method reported by Ingkaninan et al. (2003). 5,5′-ithiobis(2-nitrobenzoic acid) (DTNB, Sigma) was dissolved inuffer containing 0.1 M NaCl and 0.02 M MgCl2. Acetylthiocholine

odide (ATCI, Sigma) was dissolved in deionised water. Solutionsf DTNB (500 �L, 3 mM), ATCI (100 �L, 15 mM), buffer (Tris–HClH 8.0, Sigma, 275 �L), and solutions of sample extracts in DMSOMerck) (100 �L) were prepared, leading to a final extract con-entration ranging from 1 to 130 �g/mL. The reaction was startedy adding a solution of acetylcholinesterase (AChE type VI-S,rom electric eel 349 U/mg solid, 411 U/mg enzyme, Sigma) inuffer (25 �L, 0.28 U/mL). The resulting solutions were placed in apectrophotometer (M350 Double Beam UV–Visible Spectrometeramspec). The thiocholine formed during hydrolysis of acetyl-holine reacts rapidly with DTNB and releases a yellow compound,-thio-2-nitrobenzoic acid. The production of this coloured com-ound was recorded over 5 min on a spectrophotometer at 405 nm,gainst a reagent blank, to control for non-enzymatic hydrolysis ofcetylcholine.

.7. Toxicity studies

Direct cytotoxicity and genotoxicity of the three extracts (diethylther, ethyl acetate and n-butanol) were determined by mitoticndex evaluation and chromosomal aberration assay in periph-ral human lymphocytes from healthy donors. Venous peripherallood was collected aseptically from healthy donors using heparins anti-coagulant. 500 �L of whole blood were added to 4.5 mLam’s F-10 medium supplemented with 15% fetal calf serum,�g/mL 2% (v/v) phytohemagglutinnin, 1% l-glutamine, 100 �g/mL

treptomycin and 100 IU/mL penicillin. After 24 h culture the testompounds were added to the culture medium, and incubationontinued for a further 24 h. 2% Hydrogen peroxide was used aspositive control for acute cytotoxicity and genotoxicity. In the lasth of incubation, colcemid was added to the medium to a final con-entration of 0.5 �g/mL. Metaphase spreads were then obtaineds described by Rueff et al. (1993), scored in a Leitz microscopet 500× magnification for mitotic index (number of metaphaseser 1000 cells) and 1.250× magnification for chromosomal aberra-ions, classified as chromatid and chromosome gaps, chromatid andhromosome breaks, chromosome interchanges and intrachanges.esults are presented as total percentage of chromosome abnor-alities, excluding gaps.

. Results and discussion

.1. Chemical analysis

We have used an LC–UV–MS method for the analysis of an n-

utanol extract of the aerial parts of Genista tenera. The resultingase peak chromatogram and the UV chromatograms obtained at60 and 350 nm are shown in Fig. 1. Tentative identifications ofhe components have been made by examination of their UV spec-ra and mass spectrometric data (Table 1). The PDA profiles were

A.P. Rauter et al. / Journal of Ethnopharmacology 122 (2009) 384–393 387

Fig. 1. LC–UV–MS base peak chromatogram and PDA profiles of Genista tenera n-butanol extract.

Table 1UV absorption, mass spectral data and putative identification of Genista tenera n-butanol extract components.

tr m/z Wavelength (nm) Diagnostic ions Putative identity

260 350

4.27 595√

577 541 499 475 433 415 367 313 C-Dihexosyl-trihydroxyflavone4.90 447

√ √429 411 393 381 351 327 285 C- or -O-Hexosyl-dihydroxy-methoxyflavone

4.90 449√

430 412 382 287 C- or -O-Hexosyl tetrahydroxyflavone5.01 433

√415 367 271 O-Hexosyl trihydroxyflavone

5.01 595√

577 475 433 415 271 C-Dihexosyl trihydroxyflavone5.31 463

√ √301 O-Hexosyl-trihydroxy-methoxyflavone

5.52 595√ √

577 541 499 475 433 415 367 313 C-Dihexosyl-trihydroxyflavone5.59 449

√431 413 287 C- or -O-Hexosyl tetrahydroxyflavone

5.86 595√ √

577 559 529 511 457 433 409 379 313 C-Dihexosyl-trihydroxyflavone6.05 449

√431 413 383 329 287 C- or -O-Hexosyl-tetrahydroxyflavone

6.34 611√

449 287 O-Dihexosyl-tetrahydroxyflavone6.36 595

√577 433 301 271 O-Dihexosyl-trihydroxyflavone

6.50 449√

287 O-Hexosyl-tetrahydroxyflavone6.53 447

√285 O-Hexosyl-dihydroxy-methoxyflavone

6.68 565√

433 415 367 337 313 C-Diglycosyl-trihydroxyflavone (Hex1Pent1)6.88 417

√255 O-Hexosyl-dihydroxyflavone

6.96 447√

429 411 381 327 285 C- or -O-hexosyl-dihydroxy-methoxyflavone7.02 477

√315 O-Hexosyl-dihydroxy-dimethoxyflavone

7.08 595√

577 559 540 511 456 433 427 379 313 C-Dihexosyl-trihydroxyflavone7.46 449

√ √* 431 287 O-hexosiyl-tetrahydroxyflavone7.50 565

√ √* 433 415 396 376 313 271 C-Diglycosyl-trihydroxyflavone (Hex1Pent1)7.69 433

√415 397 379 367 367 313 C-Hexosyl-trihydroxyflavone

7.84 447√

285 O-Hexosyl-dihydroxy-methoxyflavone7.94 463

√423 411 393 381 327 285 C- or -O-Hexosyl-trihydroxy-methoxyflavone

8.27 611√

593 591 535 465 449 Dihexosyl-tetrahydroxyflavone8.33 433

√ √415 397 379 367 337 313 295 283 C-Hexosyl-trihydroxyflavone

8.85 449√

287 O-Hexosyl-tetrahydroxyflavone8.91 595

√577 558 484 433 415 313 271 C- or -O-Dihexosyl-trihydroxyflavone

8.99 463√

445 427 409 397 367 343 C-Hexosyl-trihydroxy-methoxyflavone9.08 433

√271 O-Hexosyl-trihydroxyflavone

9.33 595√

577 535 433 287 C-Dihexosyl-trihydroxyflavone10.27 433

√271 O-Hexosyl-trihydroxyflavone

10.32 433√

271 O-Hexosyl-trihydroxyflavone11.96 285

√Dihydroxy-methoxyflavone

388 A.P. Rauter et al. / Journal of Ethnophar

Fg

uFaI(

ii

t(lwifTrF4ue4

ibCfCeaptiwtt

m

ig. 2. Typical LC–UV–MS selected ion chromatograms of flavonoids and flavonoidlycosides in n-butanol extract of Genista tenera.

sed to distinguish between isoflavones and flavones and flavonols.lavones and flavonols typically exhibit an intense band II (350 nm)bsorption with a shoulder or low intensity peak representing band

(260 nm), while isoflavones lack the intense band II (350 nm)

Łuczkiewicz et al., 2004).Thorough examination of the UV, mass spectral and product

on data allowed recognition of 34 components, which are listedn Table 1. It is noteworthy that the majority of components in

ct(mO

Fig. 3. Structure of the flavonoids tent

macology 122 (2009) 384–393

he n-butanol extract have UV absorbances typical of isoflavonesTable 1). Many of the ions have m/z values corresponding to estab-ished flavonoid structures. One peak for a component giving ions

ith typical m/z values for an aglycone, together with peaks indicat-ng the presence of 21 variously glycosylated and 12 diglycosylatedorms of the basic established aglycone structures were detected.ypical selected ion chromatograms (SICs) chosen to highlight theange of different types of compounds detected are plotted inig. 2: aglycone (m/z 285), flavone hexosides (m/z 433, 447 and49) and flavone-dihexosides (m/z 565 and 595). Some m/z val-es were observed more than once in the chromatograms. Forxample, there are at least six separate peaks for an ion of m/z49.

Previous studies of glycosylated flavonoids have shown that its possible to distinguish between O- and C-glycosylated flavonoidsased on their fragmentation behavior (Cuyckens & Claeys, 2004).ollision induced dissociation (CID) gives rise to Y0 and B0 ions

or O-glycosides but results in cross ring cleavages (X ions) for-glycosides. In an attempt to identify the various isomers andnable structural elucidation of the analytes detected during LC–MSnalysis of the crude plant extract, LC–MS/MS experiments wereerformed. A data dependent acquisition method was utilized andhe product ion data obtained allowed proposals regarding putativedentities to be made. All these results are summarized in Table 1

hich presents a detailed list of all 34 components in elution order,ogether with spectroscopic, product ion data and proposed iden-ity for each component.

In order to tentatively assign structures for some of the above-entioned flavonoids, further HPLC–DAD–ESI-MS studies were

onducted under different experimental conditions. Based onhe fact that genistein-8-C-glucoside (1), apigenin-7-O-glucoside2), luteolin-7-O-glucoside (3), genistein-7-O-glucoside (4), 5-O-

ethylgenistein (5), apigenin-7-O-(2-O-glucosylglucoside) (6), 5--methylgenistein-7-O-glucoside (7), hesperidin (8) and luteolin-

atively identified by LC–UV–MS.

A.P. Rauter et al. / Journal of Ethnopharmacology 122 (2009) 384–393 389

F nisatiO (RT = 2( oside

7cMwpamUlfio2g(m76ac(p4p3wtsoacgIlYpa5i8wog

3

(cTgt((tt

bl

egsvcsfec

3

dtts(t4d2

ig. 4. HPLC–DAD of n-butanol extract of Genista tenera (with ESI-MS in positive io-(2-O-glucosylglucoside) (RT = 27.10 min), 7 – 5-O-methylgenistein-7-O-glucoside

RT = 34.33 min), 3 – luteolin-7-O-glucoside (RT = 37.84 min), 4 – genistein-7-O-gluc

,3′-di-O-glucoside (9) (Fig. 3) are already known to be majorompounds produced by Genista species (Borges et al., 2001;artins et al., 2002; Luczkiewicz et al., 2004; Rauter et al., 2005),e propose the following tentative assignments of some of theeaks. Compounds are tentatively assigned based on consider-tion of elution order together with the positive and negativeode electrospray MS data (Figs. 4 and 5, respectively) and theV spectra acquired. The UV profiles clearly indicated flavonoid-

ike character of the compounds. Isoflavones may be distinguishedrom flavones and flavonols by their UV spectra which typ-cally exhibit an intense band II absorption with a shoulderr low intensity peak representing band I (Łuczkiewicz et al.,004). The major components are proposed to correspond toenistein-8-C-glucoside (1), apigenin-7-O-(2-O-glucosylglucoside)6) and 5-O-methylgenistein-7-O-glucoside (7); [M+H]+ ions at/z 433 (compound 1), 595 (compound 6) and 447 (compound

), and [M−H]− at m/z 431 (compound 1), 593 (compound) and 445 (compound 7) are consistent with these proposedssignments. The minor constituents are tentatively proposed toorrespond to apigenin-7-O-glucoside (2), luteolin-7-O-glucoside3), genistein-7-O-glucoside (4), 5-O-methylgenistein (5), hes-eridin (8) and luteolin-7-3′-di-O-glucoside (9); [M+H]+ at m/z33 (compounds 2 and 4), 449 (compound 3) and 285 (com-ound 5), and [M−H]− at m/z 431 (compound 2), 447 (compound), 431 (compound 4), 609 (compounds 8 and 9) are consistentith these proposals. For the species proposed to correspond

o genistein-7-O-glucoside (4), an in source fragment ion corre-ponding to the loss of the monosaccharide residue gives thebserved protonated and deprotonated aglycones, Y0

+ at m/z 271,nd Y0

− at m/z 269 (Fig. 6). The same was observed for theomponent proposed to correspond to 5-O-methylgenistein-7-O-lucoside (7): ions Y0

+ at m/z 285 and Y0− at m/z 283 were observed.

n the spectrum of the component proposed to correspond touteolin-7-O-glucoside (3) a peak at m/z 287 was observed for0

−. The loss of water molecules and hydroxyl radical in com-ounds apigenin-7-O-(2-O-glucosylglucoside) (6), hesperidin (8)nd luteolin-7-3′-di-O-glucoside (9) gave the peaks detected at m/z77 and 578 (compound 1), respectively, for ESI-MS in positive

onisation mode, and the peaks at m/z 591 and 592 (compounds

and 9), for ESI-MS in negative ionisation mode. Hesperidin (8)as detected by ESI-MS in negative ionisation mode, being alsobserved a peak at m/z 594 due to the loss of a methyl radical,iving the ion fragment [M−•CH3]•−.

3

c

on mode) with proposed assignments of major Genista flavonoids; 6 – apigenin-7-9.17 min), 1 – genistein-8-C-glucoside (RT = 31.64 min), 2 – apigenin-7-O-glucoside(RT = 38.47 min), 5 – 5-O-methylgenistein (RT = 50.00 min).

.2. Antihyperglycaemic activity

The effects of the Genista tenera n-butanol extract200 mg/kg b.w. per day) on body weight and on blood glu-ose levels of normal and STZ–diabetic rats are shown in Table 2.his extract induced a significant reduction (53.4–62.5%) in serumlucose levels in diabetic rats, from the 7th day of administra-ion. After 15 days treatment, glucose levels of diabetic animals61.20 ± 8.80 mg/dL) were quite similar to those of normal controls64.50 ± 11.80 mg/dL), the extract appearing to be more effectivehan the reference drug, glibenclamide, that lowered the glycaemiao 88.00 ± 16.00 mg/dL.

The evolution of body weight of all groups was also followed,ut no significant changes were observed between the first and the

ast day treatment (Table 2).On day 15 a glucose tolerance test was performed on all

xperimental groups (Fig. 7). It was observed that, 240 min afterlucose administration, diabetic rats treated with the extracthowed a glucose level of 65.40 ± 13.97 mg/dL, this value beingery close to basal glucose levels (64.50 ± 11.83 mg/dL) of theontrols (group I). The extract caused a significant decrease inerum glucose levels (42.1%) when compared to those obtainedor untreated diabetic rats. In addition, no acute hypoglycaemicffects were observed with the extract under these experimentalonditions.

.3. Antioxidant activity (DPPH assay)

Evaluation of antioxidant activity is carried out by different stan-ard methods, the DPPH assay being one of the most common. Inhis work the radical scavenging activity of the plant extracts wasested over the concentration range of 1.1–139.1 �g/mL. All extractshowed similar scavenging activities between 10.7 and 107.0 �g/mLca. 20%). At the highest assayed dose, the percent bleaching forhe diethyl ether, ethyl acetate and n-butanol extracts were 39.1%,8.7% and 24.0%, respectively. Under the same experimental con-itions, the inhibition of quercetin, a reference flavonol, was 96% at5 �g/mL (Pourmorad et al., 2006).

.4. Acetylcholinesterase inhibitory activity

The diethyl ether, ethyl acetate and n-butanol extracts at con-entrations ranging from 1.0 to 130 �g/mL were screened for

390 A.P. Rauter et al. / Journal of Ethnopharmacology 122 (2009) 384–393

Fig. 5. HPLC–DAD of n-butanol extract of Genista tenera (with ESI-MS in negative ionisation mode) with proposed assignments of major Genista flavonoids; 8 – hes-p ), 7 –( glucosg

AtetaciTmA11

tCflii

3

srtpmamaoe

eridin (RT = 27.01 min), 6 – apigenin-7-O-(2-O-glucosylglucoside) (RT = 27.77 minRT = 32.71 min), 2 – apigenin-7-O-glucoside (RT = 35.91 min) 9 – luteolin-7,3′-di-O-lucoside (RT = 40.44 min).

ChE inhibitory activity using Ellman’s colorimetric method. Athe highest assayed dose, the percent inhibitory effect for thethyl acetate and n-butanol extracts were 77.0% and 23.5%, respec-ively. However the ethyl acetate extract displayed a nearly constantcetylcholinesterase inhibitory activity of 77% with extract con-entrations equal to or greater than 70 �g/mL, while 19% enzymenhibition was achieved with an extract concentration of 1 �g/mL.he diethyl ether extract was inactive. Under the same experi-ental conditions, rivastigmine, a drug used for medication oflzheimer’s disease patients, showed 95% inhibitory activity at00 �g/mL but did not inhibit the enzyme at the concentration of0 �g/mL (Rauter et al., 2007).

The results obtained are promising, in particular those for

he ethyl acetate extract, since it contains mainly O- and-glycosylflavones and -isoflavones (Rauter et al., 2005) andavonoids have previously been reported to be multipotent agents

n combatting Alzheimer’s disease (Ji and Zhang, 2006) by enhanc-ng acetylcholine levels.

tpmiw

7-O-glucoside-5-O-methylgenistein (RT = 30.14 min), 1 – genistein-8-C-glucosideide (RT = 37.97 min), 3 – luteolin-7-O-glucoside (RT = 40.11 min), 4 – genistein-7-O-

.5. Toxicity studies

In vitro toxicity studies have become a valid alternative to clas-ical animal toxicity evaluations, and are now accepted or evenequired in safety assessment by food, cosmetic, and pharmaceu-ical regulatory agencies (Bernauer et al., 2005). We conducted areliminary in vitro toxicity evaluation including a direct assess-ent of acute toxicity and genotoxicity. Cell morphology was

nalysed visually at 1.250× magnification (for three parameters:embrane integrity, presence or absence of vacuoles in cytoplasm

nd approximate cell size) and the mitotic index of lymphocytesn metaphase spreads was obtained (Table 3). By both criteria,xposure to the Genista tenera extracts did not significantly affect

he morphology or viability of cells. Furthermore, exposed lym-hocytes seemed to have an increased stability of the cytoplasmicembrane when exposed to hypotonic shock, a preliminary find-

ng that requires further investigation. Potential genotoxic activityas evaluated by scoring chromosomal aberrations on metaphase

A.P. Rauter et al. / Journal of Ethnopharmacology 122 (2009) 384–393 391

Fig. 6. ESI-MS spectra in positive (ESI+) or negative (ESI−) modes, at the retention time indicated in the chromatogram of the n-butanol extract (see Figs. 4 and 5): (a)RT = 27.10, ESI+; (b) RT = 29.17, ESI+; (c) RT = 31.64, ESI+; (d) RT = 37.84, ESI+; (e) RT = 38.47, ESI+; (f) RT = 50.00, ESI+; (g) RT = 27.01, ESI−; (h) RT = 27.77, ESI−; (i) RT = 30.14, ESI−;(j) RT = 32.71, ESI−; (k) RT = 35.91, ESI−; (l) RT = 37.97, ESI−; (m) RT = 40.11, ESI−; (n) RT = 40.44, ESI−.

Table 2Effect of Genista tenera n-butanol extract (200 mg/kg b.w./day, orally) in body weight and in blood glucose levels on non-diabetic and streptozotocin-induced diabetic rats.

Group Mean weight ± S.E.M. (g) Mean blood glucose concentration ±S.E.M. (mg/dL)

Initial (1st day) Final (15th day) Treatment day

1st 4th 7th 10th 15th

I 192.5 ± 6.7 198.8 ± 9.2 51.7 ± 7.7 53.3 ± 6.8 58.2 ± 9.5 53.8 ± 6.7 64.5 ± 11.8II 189.2 ± 8.2 160.8 ± 18.3 304.6 ± 24.3*** 245.8 ± 13.3** 206.0 ± 18.6*** 183.5 ± 8.0*** 143.5 ± 14.5**

III 246.6 ± 20.4 225.2 ± 29.3 215.2 ± 19.7* 133.4 ± 40.6 96.0 ± 21.6### 68.8 ± 6.6### 61.2 ± 8.8##

IV 187.8 ± 6.3 145.2 ± 9.2 279.4 ± 25.2*** 229.2 ± 28.0** 151.2 ± 6.2** 146.8 ± 16.3*** 88.0 ± 16.0

Animal groups. I: control (untreated); II: diabetic rats; III: diabetic rats + 200 mg/kg n-butanol extract; IV: diabetic rats + glibenclamide (0.5 mg/kg). S.E.M.: standard errormean.

* P < 0.05 (significant from Group I).** P < 0.01 (significant from Group I).

*** P < 0.001 (significant from Group I).# P < 0.05 (significant from Group II).

## P < 0.01 (significant from Group II).### P < 0.001 (significant from Group II).

392 A.P. Rauter et al. / Journal of Ethnophar

Fig. 7. Glucose tolerance curves for Group I (non-diabetic rats given 0.5 mL 5% Tween80), Group II (untreated diabetic rats given 0.5 mL 5% Tween 80), Group III [dia-betic rats given n-butanol extract (200 mg/kg b.w.) in 0.5 mL 5% Tween 80], GroupIV [diabetic rats given glibenclamide (0.5 mg/kg b.w.) in 0.5 mL 5% Tween 80].

Table 3In vitro toxicity study of extracts of Genista tenera (2 mg/mL) in human lymphocytes.

Mitotic index Chromosomal aberrations

Blank 23.25 ± 7.85 1.75 ± 0.96H2O2 5% 3.00 ± 8.21 7.25 ± 2.22DEn

se

4

iadrbgeHaiPnflatt

A

CpftaE1a

R

A

B

B

C

E

E

F

H

H

I

J

J

J

K

K

Ł

M

M

M

M

N

O

P

P

R

iethyl ether 22.50 ± 9.00 1.50 ± 1.29thyl acetate 19.75 ± 5.19 1.75 ± 0.50-Butanol 20.75 ± 3.90 1.25 ± 0.96

preads; no evidence for genotoxic risk was found for any of thextracts (P > 0.05 for all three extracts in comparison to the blank).

. Conclusions

In this work we have evaluated the antihyperglycaemic activ-ty of the n-butanol extract of the Genista tenera, which contains

complex mixture of O- and C-glycosylflavonoids, including 12iglycosylated species. Treatment with the extract (200 mg/kg b.w.)estored the serum glucose levels of streptozotocin-induced dia-etic rats to almost normal levels, being more effective thanlibenclamide, a standard antidiabetic oral drug. This extract alsoxhibited antioxidant and acetylcholinesterase inhibitory effects.owever the ethyl acetate extract, the major components of whichre mono-O-/C-glycosyl flavonoids (Rauter et al., 2005), exhib-ted the highest antioxidant and antiacetylcholinesterase activities.reliminary toxicity studies suggest that the extracts tested haveo cytotoxic or genotoxic effects. Our results indicate that theavonoid- and flavonoid glycoside-rich extracts of Genista tenerare promising agents for combating neurological disorders, oxida-ive stress injuries and diabetes and give clear evidence to supporthe medicinal use of this plant.

cknowledgements

The authors thank The British Council and the Ministério daiência e do Ensino Superior for financial support and Fundacãoara a Ciência e Tecnologia for a grant. Jane Thomas-Oates grate-ully acknowledges Thermo Finnigan’s support and funding from

he Analytical Chemistry Trust Fund, the RSC Analytical Division,nd EPSRC. Studentship support for Joao A. Rodrigues and Emmadwards from the University of York and the EU (ICA4-CT-2001-0056), and the BBSRC and Merck, Sharp & Dohme, respectively,re gratefully acknowledged.

R

macology 122 (2009) 384–393

eferences

slan, M., Orhan, D., Orhan, N., Sezik, E., Yesilada, E., 2007. In vivo antidiabeticand antioxidant potential of Helichrysum plicatum ssp. plicatum capitulumsin streptozotocin-induced-diabetic rats. Journal of Ethnopharmacology 109,54–59.

ernauer, U., Oberemm, A., Madle, S., Gundert-Remy, U., 2005. The use of in vitro datain risk assessment. Basic and Clinical Pharmacology and Toxicology 96, 176–181.

orges, C., Martinho, P., Martins, A., Rauter, A.P., Almoster-Ferreira, M.A., 2001. Struc-tural characterisation of flavonoids and flavonoid-O-glycosides extracted fromGenista tenera by fast-atom bombardment tandem mass spectrometry. RapidCommunications in Mass Spectrometry 15, 1760–1767.

uyckens, F., Claeys, M., 2004. Mass spectrometry in the structural analysis offlavonoids. Journal of Mass Spectrometry 39, 1–15.

dwards, E., Rodrigues, J., Ferreira, J., Goodall, D., Rauter, A., Justino, J., Thomas-Oates, J., 2006. Capillary electrophoresis–mass spectrometry characterisationof secondary metabolites from the anti-hyperglycaemic plant Genista tenera.Electrophoresis 27, 2164–2170.

xner, M., Hermann, M., Hofbauer, R., Kapiotis, S., Quehenberger, P., Speiser, W., Held,I., Gmeiner, B.M.K., 2001. Genistein prevents the glucose autoxidation mediatedatherogenic modification of low density lipoprotein. Free Radical Research 34,101–112.

erreira, A., Proenca, C., Serralheiro, M.L.M., Araújo, M.E.M., 2006. The in vitro screen-ing for acetylcholinesterase inhibition and antioxidant activity of medicinalplants from Portugal. Journal of Ethnopharmacology 108, 31–37.

arborne, J.B., 1994. In: Harborne, J.B. (Ed.), Phytochemical Dictionary of Legumi-nosae. Chapman & Hall, London, pp. 313–330.

avsteen, B.H., 2002. The biochemistry and medical significance of the flavonoids.Pharmacology & Therapeutics 96, 67–202.

ngkaninan, K., Temkitthawon, P., Chuenchom, K., Yuyaem, T., Thongnoi, W., 2003.Screening for acetylcholinesterase inhibitory activity in plants used in Thai tra-ditional rejuvenating and neurotonic remedies. Journal of Ethnopharmacology89, 261–264.

ardim, R., Francisco, D., 2000. Flora Endémica da Madeira. Múchia Publicacões,Funchal, pp. 164–166.

i, H.F., Zhang, H.Y., 2006. Theoretical evaluation of flavonoids as multipotent agentsto combat Alzheimer’s disease. Journal of Molecular Structure: Theochemistry767, 3–9.

ung, M., Park, M., Lee, H.C., Kang, Y.H., Kang, E.S., Kim, S.K., 2006. Antidiabetic agentsfrom medicinal plants. Current Medicinal Chemistry 13, 1203–1218.

aur, I.P., Geetha, T., 2006. Screening methods for antioxidants. A review. Mini-Reviews in Medicinal Chemistry 6, 305–312.

im, J.I., Kang, M.J., Kim, J.C., Ha, Y.R., Lee, J.H., Kwon, T.W., Bae, S.Y., Joo, H.J., Kim,Y.S. Composition comprising chongkukjang extract or apigenin isolated from it,having blood glucose-lowering activities. Patent nr. KR2005071101-A (2006).

uczkiewicz, M., Głód, D., Baczek, T., Bucinski, A., 2004. LC–DAD UV and LC–MS forthe analysis of isoflavones and flavones from in vitro and in vivo biomass ofGenista tinctoria L. Chromatographia 60, 179–185.

cCarty, M.F., 2006. Toward prevention of Alzheimers disease–potential nutraceu-tical strategies for suppressing the production of amyloid beta peptides. MedicalHypotheses 67, 682–697.

ariappan, D., Winkler, J., Parthiban, V., Doss, M.X., Hescheler, J., Sachinidis, A., 2006.Dietary small molecules and large- scale gene expression studies: an experimen-tal approach for understanding their beneficial effects on the development ofmalignant and non-malignant proliferative diseases. Current Medicinal Chem-istry 13, 1481–1489.

artins, A., Almoster-Ferreira, M.A., Borges, C., Rauter, A.P., Brum-Bousquet, M., Tille-quin, F., González, A.G., Bermejo, J., 2002. In: Rauter, A.P., Palma, F.B., Justino,J., Araújo, M.E., Santos, S.P. (Eds.), Proceedings of the Phytochemical Society ofEurope, vol. 47. Kluwer Academic Publishers, Dordrecht, p. 111.

artins, A., Wink, M., Tei, A., Brum-Bousquet, M., Tillequin, F., Rauter, A.P., 2005.Phytochemical study of quinolizidine alkaloids from Genista tenera by gas–liquidchromatography/mass spectrometry. Phytochemical Analyses 16, 264.

ishizono, S., Tanaka, K., Okada, T., Murai, H. Diabetes prevention composition for usein pharmaceutical or food/beverage products, contains rutin as active ingredient.Patent nr. JP2007070263-A (2007).

moigui, O.S. Prevention and treatment of aging and age-related disorders e.g.peripheral vascular disease, coronary artery disease, arthritis, Azheimer diseaseinvolves administration of components inhibiting specific interleukin mediatedinflammation. Patent nrs. US2006078533-A1 and WO2006121558-A2 (2006).

eluso, M.R., 2006. Flavonoids attenuate cardiovascular disease, inhibit phos-phodiesterase and modulate lipid homeostasis in adipose tissue and liver.Experimental Biology and Medicine 231, 1287–1299.

ourmorad, F., Hosseinimehr, S.J., Shahabimajd, N., 2006. Antioxidant activity, phe-nol and flavonoid contents of some selected Iranian medicinal plants. AfricanJournal of Biotechnology 5, 1142–1145.

aederstorff, D., Wang-Schmidt, Y., Wolfram, S. Use of resveratrol in combinationwith (−)-epigallocatechin gallate, coenzyme Q-10. genistein, lycopene, hydrox-ytyrosol and/or polyunsaturtated fatty acids for delaying aging and/or for the

treatment or prevention of age-related diseases, Patent nr. WO2007042272-A2(2007).

auter, A.P., Martins, A., Borges, C., Ferreira, J., Justino, J., Bronze, M.-R., Coelho,A.V., Choi, Y.H., Verpoorte, R., 2005. Liquid chromatography-diode arraydetection-electrospray ionisation mass spectrometry/nuclear magnetic reso-nance analyses of the anti-hyperglycemic flavonoid extract of Genista tenera.

ophar

R

R

S

S

S

V

lial protection against oxidative injury. Journal of Ethnopharmacology 93, 363–370.

A.P. Rauter et al. / Journal of Ethn

Structure elucidation of a flavonoid-C-glycoside. Journal of Chromatography A1089, 59–64.

auter, A.P., Branco, I., Lopes, R.G., Justino, J., Silva, F.V.M., Noronha, J.P., Cabrita,E.J., Brouard, I., Bermejo, B., 2007. A new lupene triterpenetriol and anti-cholinesterase activity of Salvia sclareoides. Fitoterapia 78, 474–481.

ueff, J., Brás, A., Cristovão, L., Mexia, J., Sá da Costa, M., Pires, V., 1993. DNA strandbreaks and chromosomal aberrations induced by H2O2 and 60Co �-radiation.Mutation Research 289, 197–204.

ezik, E., Aslan, M., Yesilada, E., Ito, S., 2005. Hypoglycaemic activity of Gentianaolivieri and isolation of the active constituent through bioassay-directed frac-tionation techniques. Life Sciences 76, 1223–1238.

onee, M., Sum, T., Wang, C.Y., Mukherjee, S.K., 2004. The soy isoflavone, genistein,protects human cortical neuronal cells from oxidative stress. Neurotoxicology25, 885–891.

W

macology 122 (2009) 384–393 393

ong, Y.Q., Manson, J.E., Buring, J.E., Sesso, H.D., Liu, S.M., 2005. Associations of dietaryflavonoids with risk of type 2 diabetes, and markers of insulin resistance and sys-temic inflammation in women: a prospective study and cross-sectional analysis.Journal of the American College of Nutrition 24, 376–384.

itor, R.F., Mota-Filipe, H., Teixeira, G., Borges, C., Rodrigues, A.I., Teixeira, A., Paulo,A., 2004. Flavonoids of an extract of Pterospartum tridentatum showing endothe-

ilkinson, D.G., Francis, P.T., Schwam, E., Payne-Parrish, J., 2004. Cholinesteraseinhibitors used in the treatment of Alzheimer’s disease the relationshipbetween pharmacological effects and clinical efficacy. Drugs & Aging 21, 453–478.