Nutrition, Metabolism & Cardiovascular Diseases (2011) xx, 1e7

+ MODEL

ava i lab le at www.sc iencedi rec t .com

journal homepage : www.e lsev ie r . com/ locate /nmcd

The endothelial plasma membrane transporterbilitranslocase mediates rat aortic vasodilationinduced by anthocyanins

L. Ziberna a, M. Lunder a, F. Tramer b, G. Dreven�sek a, S. Passamonti b,*

a Institute of Pharmacology and Experimental Toxicology, Faculty of Medicine, University of Ljubljana,Korytkova 2, 1000 Ljubljana, SloveniabDepartment of Life Sciences, University of Trieste, via L. Giorgieri 1, 34127 Trieste, Italy

Received 21 September 2010; received in revised form 17 December 2010; accepted 4 February 2011

KEYWORDSRat aorta;Vasodilation;Flavonoids;Membrane transport;Bilitranslocase

Abbreviations: Ach, acetylcholine; AUPMSF, phenylmethylsulphonyl fluoride* Corresponding author. Tel.: þ39 04E-mail addresses: lovro.ziberna@m

drevensek@mf.uni-lj.si (G. Dreven�sek

Please cite this article in press as: Zvasodilation induced by anthocyanin

0939-4753/$ - see front matter ª 201doi:10.1016/j.numecd.2011.02.005

Abstract Background and aims: Anthocyanins, a sub-class of flavonoids, induce endothelium-dependent vasorelaxation, by activating endothelial nitric oxide synthase and consequentlyincreasing production of the vasorelaxant agent nitric oxide. It is not yet clear if anthocy-anin-induced vasorelaxation starts with their interaction with plasma membrane receptorsin the extracellular compartment, or with their membrane transport toward intracellularmolecular targets. We therefore investigated the possible role of bilitranslocase (TC2.A.65.1.1), an endothelial plasma membrane carrier that transports flavonoids, in the vasodi-lation activity induced by anthocyanins.Methods and results: Vascular reactivity was assessed in thoracic aortic rings obtained frommale Wistar rats. Pre-treatment of aortic rings with anti-sequence bilitranslocase antibodiestargeting the carrier, decreased vasodilation induced by cyanidin 3-glucoside and bilberryanthocyanins.Conclusion: Here we show for the first time that bilitranslocase mediates a critical step invasodilation induced by anthocyanins. This offers new insights into the molecular mechanisminvolved in endothelium-dependent vasorelaxation by flavonoids, and the importance of theirspecific membrane carriers.ª 2011 Elsevier B.V. All rights reserved.

C, area under the curve; C3G, cyanidin 3-glucoside; L-NNA, N(u)-nitro-l-arginine; NO, nitric oxide;; SNP, sodium nitro prusside.05583680; fax: þ39 0405583691.f.uni-lj.si (L. Ziberna), mojca.lunder@mf.uni-lj.si (M. Lunder), ftramer@units.it (F. Tramer), gorazd.), spassamonti@units.it (S. Passamonti).

iberna L, et al., The endothelial plasma membrane transporter bilitranslocase mediates rat aortics, Nutrition, Metabolism & Cardiovascular Diseases (2011), doi:10.1016/j.numecd.2011.02.005

1 Elsevier B.V. All rights reserved.

2 L. Ziberna et al.

+ MODEL

Introduction

Epidemiological studies show that subjects whose diet is richin flavonoids have a lower cardiovascular disease risk [1].Since cardiovascular disease is anticipated by an impairmentof the endothelial function, the impact of flavonoids on thelatter has been examined by various approaches, rangingfrom in vivomeasurement of flow-mediated dilation [2] to invitro tests of vasodilation of isolated arterial rings [3]. Dataobtained have consistently demonstrated that endothelium-dependent vasorelaxation induced by flavonoids is mainlyNO-dependent [4], via activation of the Src kinase/phos-phoinositide 3-kinase/Akt pathway [5,6]. However, it is notfully clear if flavonoids interact with some surface receptorexpressed at the level of the endothelial plasma membraneor/and if they penetrate into the cells, where they may bindto intracellular molecular targets. In the last hypothesis,translocation of flavonoids through the cell plasmamembrane must occur via specific transport proteins [7].One among the carriers possibly involved is bilitranslocase(TC #2.A.65.1.1), a plasma membrane flavonoid transporterexpressed in various rat organs [8], including the vascularendothelium [9].

The aim of this study was thus to address the generalquestion whether vasodilation critically depends on bili-translocase-mediated membrane transport of vasoactiveflavonoids into the endothelium. The involvement ofa single carrier in a complex biological response, such asvasoreactivity, could be investigated by exploiting the dualproperties of bilitranslocase antibodies, that both bind toan extracellular domain of the carrier protein and inhibit itstransport function [9]. We found that pre-treatment ofaortic rings with these antibodies severely reduced vaso-dilation, suggesting that bilitranslocase plays a key role inflavonoid vasoactivity.

Methods

Animals

Adult, male Wistar rats weighing 240e280 g were bred andhoused at the Animal House of the Faculty of Medicine atthe University of Ljubljana under standard conditions, ina temperature-controlled environment (22 � 1 �C, 60%humidity) maintained on a 12 h light/dark cycle, and givenad libitum access to food and water. All animal proceduresand study protocols were conducted in accordance withpermission issued by the Veterinary Administration of theRepublic of Slovenia (permit SI-No. 34401-23/2009/3),which conforms with the Guide for the Care and Use ofLaboratory Animals published by the US National Institutesof Health (NIH Publication No. 85-23, revised 1996).

Reagents

N(u)-nitro-L-arginine (L-NNA), phenylmethylsulphonyl fluo-ride (PMSF) and rabbit immunoglobulins were purchasedfrom Sigma-Aldrich (Steinheim, Germany). Cyanidin3-glucoside (C3G) was purchased from Polyphenols Labo-ratories AS (Sandnes, Norway). Stock solutions were

Please cite this article in press as: Ziberna L, et al., The endothelialvasodilation induced by anthocyanins, Nutrition, Metabolism & Cardio

prepared in deionized Milli-Q water (Millipore, Bedford,USA) except PMSF, which was dissolved in dimethylsulfoxide.

Bilberry anthocyanins

Bilberry anthocyanins were isolated from the purifiedbilberry extract, andwere provided as generous gift by �SpelaMo�ze (PhD candidate, Biotechnical Faculty, University ofLjubljana). Anthocyanins composition was analyzed byHPLC-DAD method (l Z 520 nm) and was the following(presented as relative abundances in %): 14.3% of delphinidin3-galactoside, 14.0% of delphinidin 3-glucoside, 9.2% ofcyanidin 3-galactoside, 12.1% of delphinidin 3-arabinoside,10.1% of cyanidin 3-glucoside, 4.0% of petunidin 3-galacto-side, 7.7% of cyanidin 3-arabinoside, 8.8% of petunidin 3-glucoside, 1.1% of peonidin 3-galactoside, 2.6% of petunidin3-arabinoside, 3.7% of peonidin 3-glucoside, 2.5% ofmalvidin3-galactoside; 0.5% of peonidin 3-arabinoside, 7.9% of mal-vidin 3-glucoside, 1.5% of malvidin 3-arabinoside. Concen-tration of total anthocyanins in the bilberry extract wasexpressed in mg/L as equivalents of cyanidin 3-glucoside.Just prior to vascular experiments, the aliquot of purifiedmethanolic bilberry anthocyanins was evaporated to drynessunder reduced pressure at 38 �C and then redissolved in Milli-Q water.

Bilitranslocase antibodies

Polyclonal bilitranslocase antibodies were obtained fromrabbit sera immunized with a peptide (EDSQGQHLSSF)corresponding to segment 65e75 of the primary structureof bilitranslocase. Antibodies were purified by affinitychromatography from immune sera of rabbits, as describedearlier [9e11].

Vascular reactivity studies

Rats were anaesthetized by intraperitoneal application ofurethane (1.5 g/kg) and then exsanguinated by vena cavainferior transection. Thoracic rat aorta was dissected,cleaned of fat and connective tissue, and cut transversallyinto cylindrical rings (3e4 mm in length). The endotheliumwas preserved by cautious dissection of the rings. Inexperiments with endothelium-denuded rings, the endo-thelium was mechanically removed by gentle rubbing of theaortic lumen with a thin metallic wire. Aortic rings wererinsed of blood and immediately mounted in standard organbaths filled with KrebseHenseleit solution, as describedpreviously [12].

After mounting, rings were allowed to equilibrate at20 mN resting tension for at least 60 min. During this period,tensionwas periodically adjusted to the desired level and theKrebseHenseleit solution was changed every 15 min. Afterthe equilibration period, the rings were contracted with60 mM KCl until we obtained reproducible contractileresponses. Then, the ringswere pre-contractedwith 1 mmol/L phenylephrine (PE) before addition of 1 mmol/L acetyl-choline (ACh) to check the presence of a functionalendothelium; however, in experiments with the endothe-lium-denuded rings, the absence of ACh-induced

plasma membrane transporter bilitranslocase mediates rat aorticvascular Diseases (2011), doi:10.1016/j.numecd.2011.02.005

Bilitranslocase-mediated vasodilation 3

+ MODEL

vasorelaxant effects was taken as evidence that the prepa-ration was effectively stripped of endothelium. Afterwashout, rings were incubated for 30-min with 0.24 mg/mLrabbit immunoglubulins (control group), 0.24 mg/mL specificanti-sequence bilitranslocase antibodies (BTL group), 1 mMphenylmethylsulphonyl fluoride (PMSF group), or 0.1 mMN(u)-nitro-L-arginine (L-NNA group) in separate sets ofexperiments, respectively. Afterwashout and further 30-minequilibration period, rings were again constricted with1 mmol/L phenylephrine and 60 mmol/L KCl until a stableplateauwas reached. Then, concentration-relaxation curvesto cyanidin 3-glucoside (1 nmol/Le10 mmol/L) and bilberryanthocyanins (0.01e20 mg/L expressed as equivalents ofcyanidin 3-glucoside) were constructed.

To test the influence of rabbit immunoglobulins, specificanti-sequence bilitranslocase antibodies and PMSF onendothelium function, we examined vasorelaxation ofpreconstricted rings with phenylephrine (1 mmol/L) byadding cumulative concentrations of endothelium-depen-dent dilator acetylcholine (1 nmol/Le0.1 mmol/L), orendothelium-independent dilator sodium nitro prusside(0.1 nmol/Le10 mmol/L). Vasorelaxations were examinedbefore the 30-min incubation in each of the tested solu-tions, and after incubation followed by 30-min equilibrationperiods to avoid Ach tolerance.

Data analysis

All data were expressed as means � SEM for n animals andm observations, where n represents the number ofanimals and m the number of aortic rings used in eachof the studied groups. Relative relaxation responses ofarterial rings were expressed as the percentage of KCl orphenylephrine pre-contracted aortic rings. In experimentswith ACh and SNP, results were fitted and graphed usingsigmoidal concentration-response curves. From thesecurves, the potency (�log EC50) and the maximal efficacy(Emax) were calculated and compared using extra sum-of-squares F test. In experiments with C3G and bilberryanthocyanins, two-way analysis of variance (ANOVA) andBonferroni’s post-test were used to make inter-groupcomparisons. For each vasorelaxation curve the areaunder the curve (AUC) was calculated and normalized tocontrol group. One-way analysis of variance (ANOVA) andDunnet’s post-test were used to make inter-groupcomparisons for normalized AUC values. All data analysisand statistical comparisons were done using GraphPadPrism software (version 5.00, GraphPad Software, SanDiego, USA). A P value less than 0.05 was consideredsignificant.

Results

Effect of antibodies and PMSF on endothelialfunction

In the experiments described below, vasodilation inducedby vasoactive flavonoids was tested under novel experi-mental conditions; that was by pre-treating aortic ringswith either the serine-specific reagent PMSF, or surfaceantigen-binding antibodies. These reagents have already

Please cite this article in press as: Ziberna L, et al., The endothelialvasodilation induced by anthocyanins, Nutrition, Metabolism & Cardio

been used in isolated cultured cells, such as humanhepatocytes (HepG2) [13] and human vascular endothelialcells [9], to inhibit flavonoid uptake. However, the impactof these treatments per se on endothelial function inaortic rings needed to be assessed. Thus, a series ofcontrol vasodilation experiments using PMSF, specificbilitranslocase antibodies, and pure non-specific immu-noglobulins were carried out in the absence ofanthocyanins.

In control experiments with rabbit immunoglobulins(Fig. 1, panels A1-A2) and bilitranslocase antibodies (Fig. 1,panels B1-B2), we observed no alterations in endothelium-dependent (ACh-induced) and endothelium-independent(SNP-induced) vasorelaxations of aortic rings. However,maximal ACh-induced vasorelaxation was reduced after thePMSF incubation (Fig. 1, panel C1) (Emax before 65.8 � 1.3%,Emax after 59.2 � 1.0%; P < 0.05). Furthermore, althoughmaximal SNP-induced vasorelaxation (Fig. 1, panel C2) ofaortic rings before and after incubation in PMSF did notdiffer, reduced potency (�log EC50) in endothelium-inde-pendent vasorelaxation was detected (before 8.33 � 0.05,after 7.60 � 0.04; P < 0.001).

Effect of antibodies and PMSF on endothelium-dependent vasorelaxation induced by C3G

Vasorelaxation tests were carried out in aortic rings pre-contracted by the addition of 1 mM phenylephrine. For thefirst time, we show here that C3G induced vasorelaxationstarting from as low as 0.1 mM. Such effect was abolishedin experiments with either denuded endothelium or L-NNA, an NO-synthase inhibitor, which shows that C3Grelaxation was endothelium- and NO-dependent. Bothbilitranslocase antibodies and PMSF decreased C3Ginduced vasorelaxation significantly (Fig. 2, panel A). Toassess the overall activity of C3G under different condi-tions, we compared normalized AUC of vasorelaxationcurves for each studied group. Relative vasorelaxationactivity by C3G in the studied groups was found as follows:endothelium-denuded 0.04-fold, L-NNA 0.10-fold, PMSF0.38-fold, and BTL 0.52-fold (Fig. 2, panel B) of thecontrol vasorelaxation activity.

However, in aortic rings pre-contracted with KCl, weobtained vasorelaxation response of less than 5% in thetested concentration range of C3G (data not shown), sug-gesting that the mechanisms whereby C3G induces vaso-relaxation are inactivated under certain unphysiologicalexperimental conditions.

Effect of antibodies and PMSF on endothelium-dependent vasorelaxation induced by bilberryanthocyanins

To test the vasodilatation activity of other anthocyaninsapart from C3G, we used chemically characterized bilberryanthocyanins. The rationale for using them was that theyrepresent a nutritionally relevant source of anthocyanins.We obtained similar results as above, showing that bilberryanthocyanins induced endothelium- and NO-dependentvasorelaxation. Aortic rings pre-contracted with phenyl-ephrine (1 mmol/L) (Fig. 3, panel A) were more sensitive to

plasma membrane transporter bilitranslocase mediates rat aorticvascular Diseases (2011), doi:10.1016/j.numecd.2011.02.005

Figure 1 Effect of studied solutions on vasorelaxations of isolated rat aortic rings. Acetylcholine-induced endothelium-depen-dent (panels A1, B1, C1) and sodium nitro prusside (SNP)-induced endothelium-independent (panels A2, B2, C2) vasorelaxationbefore (C) and after (B) 30 min of incubation. The tested solutions were: control (IgG), specific bilitranslocase antibodies (BTL)and phenylmethylsulfonyl fluoride(PMSF). Each point on the curves represents means � SEM; n is the number of animals; m is thenumber of aortic rings.

4 L. Ziberna et al.

+ MODEL

anthocyanins, in terms of vasorelaxation activity, comparedto those pre-contracted with KCl (60 mmol/L) (Fig. 3, panelC), even though both vasoconstrictors induced similar sub-maximal constriction. Both bilitranslocase antibodies andPMSF reduced vasorelaxation induced by bilberry anthocy-anins (Fig. 3, panels A and C). Normalized AUC values ofvasorelaxation curves for each of the studied groups indi-cated that, when pre-contracted with phenylephrine,

Please cite this article in press as: Ziberna L, et al., The endothelialvasodilation induced by anthocyanins, Nutrition, Metabolism & Cardio

aortic rings in endothelium-denuded group had 0.02-fold, inL-NNA group 0.04-fold, in PMSF group 0.30-fold, and in BTLgroup 0.56-fold vasorelaxation activity of the rings in thecontrol group (Fig. 3, panel B). When pre-contracted withKCl, aortic rings in endothelium-denuded group had 0.04-fold, in L-NNA group 0.09-fold, in PMSF group 0.25-fold, andin BTL group 0.61-fold vasorelaxation activity of the rings inthe control group (Fig. 3, panel D).

plasma membrane transporter bilitranslocase mediates rat aorticvascular Diseases (2011), doi:10.1016/j.numecd.2011.02.005

Figure 2 Vasorelaxation of isolated rat aortic rings inducedby C3G. Panel A: effect of C3G after precontraction withphenylephrine (1 mmol/L). Studied groups after 30-min ofincubation under control conditions (0.24 mg/mL rabbitimmunoglobulins), specific bilitranslocase antibodies (0.24 mg/mL, BTL), phenylmethylsulfonyl fluoride (1 mM, PMSF), orN(u)-nitro-L-arginine (0.1 mM, L-NNA). Relaxations areexpressed as percentages of the precontraction to phenyl-ephrine. Panel B: normalized areas under the curve (AUC)values for each of the vasorelaxation curves. All data aremean � SEM. )P < 0.05, ))P < 0.01, and )))P < 0.001. E-,endothelium-denuded aortic rings. n is the number of animals;m is the number of aortic rings.

Figure 3 Vasorelaxation of isolated rat aortic rings induced bphenylephrine (panel A) or 60 mmol/L KCl (panel C). Correspondvasorelaxation curves are shown in panels B and D, respectively. ExP < 0.01, and )))P < 0.001. E-, endothelium-denuded aortic ring

Bilitranslocase-mediated vasodilation 5

+ MODEL

Please cite this article in press as: Ziberna L, et al., The endothelialvasodilation induced by anthocyanins, Nutrition, Metabolism & Cardio

Discussion

Thepresenteddata show for thefirst time that C3G relax pre-contracted aortic rings in a significantly low concentrationrange. This observation is consistent with the well knowneffect of berry anthocyanins, which induce endothelium-and NO-dependent vasorelaxation, [3]. In our study, wereproduced those effects by using bilberry anthocyanins;even under these conditions, the inhibitory effect of bili-translocase-specific antibodies was evident. Thus, the novelfinding of this work is that anthocyanin-induced vasoactivitystrongly depends on a bilitranslocase-mediated membranetransport mechanism. In facts, when the transporter waschallenged with specific antibodies, we observed reducedvasodilation activity. The effect is solely dependent on theinhibition of bilitranslocase transporter activity since weshowed that antibodies per se do not influence endotheliumfunction. In addition, this finding opens an important novelapproach in studying the vascular reactivity by using specificantibodies against endothelial plasma membrane proteins.

In another set of experiments with PMSF, we alsoobserved diminished vasodilation. PMSF strongly inhibitsbilitranslocase transport activity by covalently binding tothe active serine residues, and thereby forming the irre-versible bilitranslocaseePMSF complex. In fact, previousstudies have shown that PMSF inhibited electrogenic BSPuptake in rat liver plasma membrane vesicles [14], as wellas inhibited both bilirubin uptake into hepatic HepG2 cells[15] and quercetin uptake into endothelial Ea.hy 926 cells[9]. In control vascular studies, PMSF affected endotheliumfunction per se. Reduced vasodilation activity could bepartially attributed to the fact that PMSF is not selectiveand could diffuse into the cells, from the intima to themedia, and therefore cause modification of intracellularproteins involved in vasodilation. Reduced potency inendothelium-independent relaxation (Fig. 1F) could mean

y bilberry anthocyanins. Rings were pre-contracted by 1 mMing normalized areas under the curve (AUC) for each of theperimental conditions as in Fig. 2. All data are mean � SEM. ))s. n is the number of animals; m is the number of aortic rings.

plasma membrane transporter bilitranslocase mediates rat aorticvascular Diseases (2011), doi:10.1016/j.numecd.2011.02.005

6 L. Ziberna et al.

+ MODEL

decreased sensitivity to NO, which can indeed arise due tomodification of intracellular smooth muscle proteins.

Anthocyanins represent a class of glycosylated flavo-noids, with minor differences in the basic chemical struc-ture consisting of different hydroxylation or methoxylationpatterns. They are substrates of bilitranslocase, asconfirmed in previous studies by electrogenic transport inplasma membrane vesicles or rat liver [16] and uptakestudies [9]. Interestingly mono- and di- glucosyl anthocya-nins, which are abundant in various vegetables and fruits,are better ligands than the corresponding aglycones [16].Furthermore, previous findings showed that anthocyanin-enriched fractions are the most potent fractions isolatedfrom red wine polyphenolic compounds, suggesting thatthey greatly contribute to the endothelium-dependentvasodilatory effect of red wine polyphenols [17]. Similarly,another study found that, although there is a strongcorrelation between vasodilation activity and total quan-tified phenolics in red wine, the only correlation witha phenolic family was with the total anthocyanins [4].

In pharmacokinetic studies, systemic bioavailabilities ofC3G and total anthocyanins were 1.7 and 3.3%, respectively[18], thus implying that the plasma concentration is verylow, i.e. <1 mM [19]. In our studies, we observed vasodi-lation activity of C3G from 10 nM on, thereby in the range ofreported post-absorption plasma concentration [19]. Ourdata thus allow to mechanistically interpret the effect ofacute intake of red wine polyphenols by healthy subjects,which was associated with improved endothelium-depen-dent vasodilation of the brachial artery induced by reactivehyperemia [20]. In spite of their low bioavailability,anthocyanins can indeed induce pronounced in vivo vaso-activity in this plasma concentration range, becausea specific plasma membrane transporter mediates theiruptake into the endothelium. This emphasizes the impor-tance of transporter-mediated flavonoid uptake intoendothelium as a key determinant of the impact of the dieton the vascular function.

Anthocyanins are especially abundant in bilberries,comprising on the average 80e90% of total phenoliccompounds [21,22]. Moreover, anthocyanin composition ofbilberry is more complex than of most other anthocyanin-containing fruits, as a total amount of 15 different mono-glycosilated anthocyanins have been identified [21]. On thisbasis, we decided to use bilberry extract for the screeningpurposes to study the effect of multiple anthocyanins, andthereby enhance the value of obtained results with C3G.Comparison of both vascular studies shows that vasculareffects lie in similar concentration range, for example C3G(10 mmol/L) induced 33.7% vasorelaxation, while bilberryanthocyanins (5 mg/L expressed as C3G equivalents corre-sponds to 10.3 mmol/L) induced 42.5%.

The protective activity of anthocyanins in endothelialdysfunction can be also explained in terms of acute acti-vation of eNOS and heme oxygenase, which increase theproduction of endogenous antioxidants such as NO andbilirubin, respectively. The effect is evident in lowconcentrations and occurs within minutes, as was shown inthe study where C3G (as low as 10 nM) regulated phos-phorylation of eNOS and Akt, thereby increasing NOproduction in endothelial cells [23]. This is in coherencewith our data, showing that C3G has an acute effect on

Please cite this article in press as: Ziberna L, et al., The endothelialvasodilation induced by anthocyanins, Nutrition, Metabolism & Cardio

endothelial NO production, which is responsible forobserved vasodilation. In addition, the inhibition of bili-translocase activity by anthocyanins might also hinder thebilirubin efflux from the endothelial cells [9], so increasingthe intracellular antioxidant activity. Together with NO,they form an important defense against various sources ofreactive oxygen species (ROS). Moreover, this observationcan partially explain the paradox of redox-sensitive flavo-noid-induced vasorelaxation, where flavonoids prior toactivation of eNOS increase the intracellular production ofROS, particularly superoxide anions and H2O2 [24]. Thesource of ROS in response to flavonoids remains to bedetermined. It could be due to activation of oxidaseenzymes, such as NADPH-oxidase, xanthine oxidase, mito-chondria oxidases, or due to the flavonoids itself [25]. It canbe speculated that the vasodilation activity of flavonoids ismerely the side effect of the endothelium defense mech-anisms, incorporating increased NO production, against thesudden increase in cellular oxidative stress. In any case,bilitranslocase seems to play an important role in regulatingthe intracellular redox balance inside the endothelial cells.

The involvement of other plasma membrane trans-porters on the endothelium in flavonoid transport can notbe ruled out, e.g. the ubiquitous GLUT1 which areresponsible for quercetin uptake into erythrocytes [26],sodium-glucose co-transporter 1 (SGLT1) responsible forphloridzin transport [27], organic anion transporting poly-peptide (OATP) family with a broad substrate spectrumincluding many flavonoids [28,29]. Moreover, anthocyaninsemerged as substrates for efflux transporters, such asmultidrug resistance protein 1 (MDR1) and breast cancerresistance protein (BRCP) in endothelial cells [30], whichnot only shows that anthocyanins may be actively trans-ported out of endothelial cells, but also suggests the exis-tence of specific plasma membrane transporters mediatingtheir influx into the cells, as proposed by Dobson et al. [7].

In conclusion, the present study demonstrates thatvasorelaxation induced by anthocyanins is a complex seriesof events, starting at the level of the cell plasma membraneand continuing on intracellular targets. On one hand,anthocyanins acted by interfering with membrane trans-porter bilitranslocase, and on the other by triggeringintracellular redox-dependent cascade leading to activa-tion of eNOS and release of NO from vascular endothelium.These results indicate that bilitranslocase has an importantrole in endothelium-dependent vasorelaxation induced byflavonoids, thereby offering a novel mechanistic approachfor studying vasoactivity of flavonoids. In particular, itseems important to elucidate the physiological and phar-macological role of bilitranslocase in modulating theendothelial function.

Conflict of interest

None declared.

Acknowledgments

Authors are grateful for the financial support by theSlovenian Research Agency [research project J3-0024];

plasma membrane transporter bilitranslocase mediates rat aorticvascular Diseases (2011), doi:10.1016/j.numecd.2011.02.005

Bilitranslocase-mediated vasodilation 7

+ MODEL

grant for international mobility of students (Ad Futura,Slovenia); Universita di Trieste (Italy); Fondazione Cassa diRisparmio di Trieste and Ministero degli Affari Esteri(Cooperazione scientifica e tecnologica Italia-Slovenia2006e2009).

References

[1] Dohadwala MM, Vita JA. Grapes and cardiovascular disease.J Nutr 2009;139:1788Se93S.

[2] Hooper L, Kroon P, Rimm E, Cohn J, Harvey I, Le Cornu K, et al.Flavonoids, flavonoid-rich foods, and cardiovascular risk:a meta-analysis of randomized controlled trials. Am J Clin Nutr2008;88:38e50.

[3] Bell D, Gochenaur K. Direct vasoactive and vasoprotectiveproperties of anthocyanin-rich extracts. J Appl Physiol 2006;100:1164e70.

[4] Andriambeloson E, Magnier C, Haan-Archipoff G, Lobstein A,Anton R, Beretz A, et al. Natural dietary polyphenoliccompounds cause endothelium-dependent vasorelaxation inrat thoracic aorta. J Nutr 1998;128:2324e33.

[5] Ndiaye M, Chataigneau M, Lobysheva I, Chataigneau T, Schini-Kerth VB. Red wine polyphenol-induced, endothelium-depen-dent NO-mediated relaxation is due to the redox-sensitivePI3-kinase/Akt-dependent phosphorylation of endothelial NO-synthase in the isolated porcine coronary artery. FASEB J 2005;19:455e7.

[6] Madeira SVF, Auger C, Anselm E, Chataigneau M,Chataigneau T, Soares de Moura R, et al. eNOS activationinduced by a polyphenol-rich grape skin extract in porcinecoronary arteries. J Vasc Res 2009;46:406e16.

[7] Dobson PD, Kell DB. Carrier-mediated cellular uptake ofpharmaceutical drugs: an exception or the rule? Nat Rev DrugDiscov 2008;7:205e20.

[8] Passamonti S, Terdoslavich M, Franca R, Vanzo A, Tramer F,Braidot E, et al. Bioavailability of flavonoids: a review of theirmembrane transport and the function of bilitranslocase inanimal and plant organisms. Curr Drug Metab 2009;10:369e94.

[9] Maestro A, Terdoslavich M, Vanzo A, Kuku A, Tramer F,Nicolin V, et al. Expression of bilitranslocase in the vascularendothelium and its function as a flavonoid transporter. Car-diovasc Res 2010;85:175e83.

[10] Battiston L, Passamonti S, Macagno A, Sottocasa GL. Thebilirubin-binding motif of bilitranslocase and its relation toconserved motifs in ancient biliproteins. Biochem Biophys ResCommun 1998;247:687e92.

[11] Passamonti S, Cocolo A, Braidot E, Petrussa E, Peresson C,Medic N, et al. Characterization of electrogenic bromo-sulfophthalein transport in carnation petal microsomes and itsinhibition by antibodies against bilitranslocase. FEBS J 2005;272:3282e96.

[12] Ziberna L, Lunder M, Kuzner J, Drevensek G. Normothermicand hypothermic models for studying the deleterious effectsof hypoxia-reoxygenation on EDHF-mediated relaxation inisolated porcine coronary arteries. J Pharmacol ToxicolMethods 2009;59:1e6.

[13] Passamonti S, Vanzo A, Vrhovsek U, Terdoslavich M, Cocolo A,Decorti G, et al. Hepatic uptake of grape anthocyanins andthe role of bilitranslocase. Food Res Int 2005;38:953e60.

Please cite this article in press as: Ziberna L, et al., The endothelialvasodilation induced by anthocyanins, Nutrition, Metabolism & Cardio

[14] Passamonti S, Battiston L, Sottocasa GL. Arylsulfonylation ofbilitranslocase in plasma membranes from rat liver enables todiscriminate between natural and artificial substrates. Bio-chim Biophys Acta 1997;1323:130e6.

[15] Passamonti S, Terdoslavich M, Margon A, Cocolo A, Medic N,Micali F, et al. Uptake of bilirubin into HepG2 cells assayed bythermal lens spectroscopy. Function of bilitranslocase. FEBS J2005;272:5522e35.

[16] Passamonti S, Vrhovsek U, Mattivi F. The interaction ofanthocyanins with bilitranslocase. Biochem Biophys ResCommun 2002;296:631e6.

[17] Burns J, Gardner PT, O’Neil J, Crawford S, Morecroft I,McPhail DB, et al. Relationship among antioxidant activity,vasodilation capacity, and phenolic content of red wines. JAgric Food Chem 2000;48:220e30.

[18] Marczylo TH, Cooke D, Brown K, Steward WP, Gescher AJ.Pharmacokinetics and metabolism of the putative cancerchemopreventive agent cyanidin-3-glucoside in mice. CancerChemother Pharmacol 2009;64:1261e8.

[19] Galvano F, La Fauci L, Vitaglione P, Fogliano V, Vanella L,Felgines C. Bioavailability, antioxidant and biological proper-ties of the natural free-radical scavengers cyanidin andrelated glycosides. Ann Ist Super Sanita 2007;43:382e93.

[20] Hashimoto M, Kim S, Eto M, Iijima K, Ako J, Yoshizumi M,et al. Effect of acute intake of red wine on flow-mediatedvasodilatation of the brachial artery. Am J Cardiol 2001;88:1457e60. A9.

[21] Heinonen M. Antioxidant activity and antimicrobial effect ofberry phenolicsda Finnish perspective. Mol Nutr Food Res2007;51:684e91.

[22] Kahkonen MP, Hopia AI, Heinonen M. Berry phenolics and theirantioxidant activity. J Agric Food Chem 2001;49:4076e82.

[23] Xu J, Ikeda K, Yamori Y. Cyanidin-3-glucoside regulatesphosphorylation of endothelial nitric oxide synthase. FEBSLett 2004;574:176e80.

[24] Ndiaye M, Chataigneau T, Andriantsitohaina R, Stoclet J-C,Schini-Kerth VB. Red wine polyphenols cause endothelium-dependent EDHF-mediated relaxations in porcine coronaryarteries via a redox-sensitive mechanism. Biochem BiophysRes Commun 2003;310:371e7.

[25] Ullrich V, Bachschmid M. Superoxide as a messenger of endo-thelial function. Biochem Biophys Res Commun 2000;278:1e8.

[26] Cunningham P, Afzal-Ahmed I, Naftalin R. Docking studiesshow that D-glucose and quercetin slide through the trans-porter GLUT1. J Biol Chem 2006;281:5797e803.

[27] Walle T, Walle U. The beta-D-glucoside and sodium-depen-dent glucose transporter 1 (SGLT1)-inhibitor phloridzin istransported by both SGLT1 and multidrug resistance-associ-ated proteins 1/2. Drug Metab Dispos 2003;31:1288e91.

[28] Mandery K, Bujok K, Schmidt I, Keiser M, Siegmund W,Bettina B, et al. Influence of the flavonoids apigenin,kaempferol and quercetin on the function of organic aniontransporting polypeptides 1A2 and 2B1. Biochem Pharmacol;2010.

[29] Wang X, Wolkoff A, Morris M. Flavonoids as a novel class ofhuman organic anion-transporting polypeptide OATP1B1(OATP-C) modulators. Drug Metab Dispos 2005;33:1666e72.

[30] Dreiseitel A, Oosterhuis B, Vukman KV, Schreier P, Oehme A,Locher S, et al. Berry anthocyanins and anthocyanidins exhibitdistinct affinities for the efflux transporters BCRP and MDR1.Br J Pharmacol 2009;158:1942e50.

plasma membrane transporter bilitranslocase mediates rat aorticvascular Diseases (2011), doi:10.1016/j.numecd.2011.02.005