A

K

I

chgcIqtl

nb

3

0d

Basic nutritional investigation

Red wine increases adipose tissue aromatase expression and regulatesbody weight and adipocyte size

Rosário Monteiro, Ph.D.a,b,*, Raquel Soares, Ph.D.a, Susana Guerreiro, M.Sc.a,Diogo Pestana, B.Sc.a, Conceição Calhau, Ph.D.a, and Isabel Azevedo, Ph.D.a

a Department of Biochemistry (U38-FCT), Faculty of Medicine of the University of Porto, Porto, Portugalb Faculty of Nutrition and Food Sciences of the University of Porto, Porto, Portugal

Manuscript received March 11, 2008; accepted January 3, 2009.

bstract Objective: Obesity is an important component of the metabolic syndrome in constituting a riskfactor for cardiovascular disease, diabetes, and cancer. Estrogens influence lipid accumulation inadipocytes, acting indirectly or directly on adipose tissue. In this study we aimed to investigate theinfluence of red wine ingestion on the expression of aromatase (estrogen synthase) in adipose tissue.Methods: Red wine or ethanol solution, in the concentration found in red wine, was provided toWistar rats as the sole drinking fluid for 8 wk. Food and drink intakes and body weight weremonitored throughout treatment and adipocyte size and aromatase expression in the adipose tissuewere determined at the end of the experimental period.Results: Red wine and ethanol increased aromatase expression in the adipose tissue and red winedecreased adipocyte size (P � 0.05). In addition, animals treated with red wine or ethanol hadsignificantly lower weight gain than controls, despite a similar energy intake.Conclusion: Thus, the ingestion of red wine may alter the production of estrogens by adiposetissue, body weight gain, and adipocyte size. Some of these red wine effects are attributable toethanol. This relation among estrogen availability, adipocyte biology, and weight gain is mostinteresting and deserves further study because it may lead to new strategies to reduce metabolicsyndrome incidence. © 2009 Published by Elsevier Inc.

Nutrition 25 (2009) 699–705www.nutritionjrnl.com

eywords: Adipocyte size; Aromatase; Estrogen; Obesity; Red wine

mptpiSemdsimwst

ntroduction

The incidence of obesity in developed and developingountries is high and increasing. This is becoming a seriousealth problem, mainly because of the metabolic and patho-enic consequences that are associated with it, such asardiovascular disease, type 2 diabetes, and cancer [1].nadequate physical activity and energy intake are fre-uently claimed to be the main lifestyle factors that con-ribute to the installation of obesity, often starting in earlyife [1].

This work was supported by FCT (POCTI, Feder, Programa Comu-itário de Apoio and SFRH/BD/12622/2003) and iBeSa (Instituto de Be-idas e Saúde).

* Corresponding author. Tel.: �351-22-551-3624; fax: �351-22-551-624.

sE-mail address: rosariom@med.up.pt (R. Monteiro).

899-9007/09/$ – see front matter © 2009 Published by Elsevier Inc.oi:10.1016/j.nut.2009.01.001

Important scientific advances in the field of energy ho-eostasis have allowed setting aside the idea that the adi-

ocyte simply stores energy as triacylglycerols. Currently,hese cells are known to be active endocrine units capable ofroducing and secreting hormones, adipokines, and peptidesn response to changes in the surrounding environment [2].ubcutaneous adipose tissue is a significant producer ofstrogens in agreement with its high expression of aro-atase, the enzyme responsible for the conversion of an-

rogens to estrogens [2]. Although not such a significantource of estrogens in the premenopausal woman, this tissues responsible for most of the production of estrogens inen and postmenopausal women [2,3]. Estrogens haveell-documented effects on body weight regulation and

eem to be related to the amount and location of adiposeissue accumulation. Apart from exerting central nervous

ystem effects on the hypothalamus to regulate appetite,

ermtammToo

ileawa

wTtiscntomewtu

Rtaiop

M

A

St2pEpgt[(

lwiamfpwaflapcfwR(amsa

M

aaowdmufw

W

uUdwlceamBbTi1p(

700 R. Monteiro et al. / Nutrition 25 (2009) 699–705

nergy expenditure, and metabolism, estrogens also act di-ectly on the adipose tissue or on organs that regulate itsetabolism [4]. Estrogens seem to directly act on adipose

issue by decreasing lipogenesis and stimulating lipolysisnd fatty acid �-oxidation. The overall effect of estrogens inales and females seems to be a decrease in adipose tissueass, particularly of that located in the visceral depot [4].his is further supported by the development of severebesity in aromatase knockout mice [5] and by the tendencyf postmenopausal women to accumulate visceral fat [6].

It is clear that food has a major influence in obesityncidence. Recently, some particular foods have been re-ated to the regulation of body weight in epidemiologic andxperimental studies. Tea and soybean-based foods are ex-mples for this interaction and have been shown to interfereith cell differentiation and proliferation, lipid metabolism,

nd the regulation of food intake [7–9].Less attention has been drawn to the influence of red

ine (RW) or its components on adipose tissue metabolism.he exploration of this subject becomes even more attrac-

ive given the known protective effect of moderate RWngestion on metabolic syndrome components [10]. Mosttudies on this subject have been conducted by Ardevol andolleagues [11–13] focusing on the effects of wine procya-idins on adipocyte metabolism. However, mechanisms forhe reported interference are still not fully understood. We andther investigators have reported the presence of aromatase-odulating compounds in RW on different experimental mod-

ls [14,15]. Nevertheless, the effect of long-term ingestion ofine on adipose tissue aromatase expression or the impact of

he putative modulation of aromatase on body weight isnknown.

In the present study, we aimed to determine if prolongedW intake could influence aromatase expression in adipose

issue. Moreover, the significance of estrogen production todipocyte size and body weight was addressed in rats drink-ng RW or an ethanol solution with the same concentrationf ethanol found in RW as the sole fluid source for an 8-wkeriod.

aterials and methods

nimals and diets

Fifteen male Wistar rats (Harlan Iberica, Barcelona,pain) with mean body weight of 226 � 15 g were main-

ained under standard temperature and light conditions (20–2°C, 12-h light/dark cycle). Animal handling and housingrotocols followed European Union guidelines (86/609/EC) and the Portuguese Act (129/92) for the use of ex-erimental animals. The animals were divided into threeroups of five animals each that were subjected to differentreatments: 1) water (control), 2) RW (with 13% ethanolEtOH]; Douro region, Portugal), and 3) EtOH in water

v/v) solution. The beverages were supplied to the rats ad c

ibitum in dark bottles. Body weight gain was monitoredeekly. Beverages and animal pellet food were renewed and

ntake was assessed every 2–3 d. After 8 wk of treatment, thenimals were anesthetized with sodium pentobarbital (60g/kg of body weight, intraperitoneally), and blood was drawn

rom the left ventricle to heparin-containing tubes to extractlasma. Two samples of inguinal subcutaneous adipose tissueere removed and one sample was frozen in liquid nitrogen

nd stored at �80°C for protein extraction, and the other wasormalin-fixed, paraffin-embedded, and sectioned for histo-ogic analysis. Plasma samples were used to measure aspartateminotransferase, alanine aminotransferase, and alkaline phos-hatase activities as markers of hepatic function and estradioloncentrations. The hepatic activities of aspartate aminotrans-erase, alanine aminotransferase, and alkaline phosphataseere determined by kinetic methods according to the Nordicecommendation [16] in an Olympus AU-600 analyzer

Olympus, Lisbon, Portugal). Activity was expressed asctivity units of milliliters of plasma. Plasma estradiol waseasured with a chemiluminescent microplate immunoas-

ay (Architect, Abbott Diagnostics, Longford, Ireland) withsensitivity of 10 pg/mL and a precision of 5 pg/mL.

easurement of adipocyte size

Hematoxylin- and eosin-stained 4-�m-thick sections ofdipose tissue were observed under identity occultation anddipocyte measurement was performed as described previ-usly [17]. Briefly, camera lucida drawings of all adipocytesere made at a final magnification of 250�. The largestiameter of each cell and the axis perpendicular to it wereeasured on paper drawings. These two values were then

sed to calculate the mean diameter. Fifty to 80 adipocytesrom four to five randomly selected, different optical fieldsere drawn per animal.

estern blotting

Protein extraction from adipose tissue was performedsing Tripure Isolation Reagent (Roche, Indianapolis, IN,SA). Extracted proteins were dissolved in 1% sodiumodecylsulfate (w/v) and stored at �80°C until used forestern blotting. The protein solution was diluted (1:1) in

oading buffer (Bio-Rad Laboratories, Hercules, CA, USA)ontaining 2% mercaptoethanol (v/v), separated (15 �g) bylectrophoresis in a 10% sodium dodecylsulfate polyacryl-mide electrophoretic, and transferred to a nitrocelluloseembrane (Hybond C-Extra, Amersham, GE Healthcare,uckinghamshire, United Kingdom). The membrane waslocked in phosphate buffered saline with 0.05% (v/v) ofween-20 containing 5% (w/v) fat-free powdered milk,

ncubated with goat anti-aromatase polyclonal antibody (1:000), washed, and then incubated with bovine anti-goatolyclonal antibody conjugated to horseradish peroxidase1:2000). Detection was performed with an enhanced

hemiluminescence reagent (Amersham, GE Healthcare).

�sC

C

dSbbwGUds

R

A

otnt1mfTsastktckabwtk

Esnpct

A

w

twsrysa2

A

eci2a3

FE(a0

701R. Monteiro et al. / Nutrition 25 (2009) 699–705

-Actin primary antibody hybridization was done by theame procedure. All antibodies were purchased from Santaruz Biotechnologies (Santa Cruz, CA, USA).

alculations and statistical analysis

Results are expressed as mean � standard deviation. Theifferences between groups were evaluated by unpairedtudent’s t test or Mann-Whitney test whenever variancesetween measurements differed. To calculate correlationsetween outcome variables, data from the three groupsere pooled and Spearman’s correlations were plotted onraphPad Prism 3.0, GraphPad Software, San Diego, CA,SA. Correlation coefficients are shown with 95% confi-ence intervals. Differences were considered statisticallyignificant at P � 0.05.

esults

nimals and diets

Body weight was similar among groups at the beginningf treatment and increased gradually throughout the 8 wk ofreatment (Fig. 1A). However, the rate of weight gain wasot the same for all groups. Body weight gain was lower forhe animals treated with RW or EtOH (114.8 � 22.4 and08.0 � 22.7 g, respectively) compared with control ani-als (160.4 � 10.5 g; Fig. 1B). The daily ingestion of pellet

ood was also different among groups of animals (Table 1).he RW- and EtOH-treated groups ingested a significantlymaller amount of pellet food than controls. As to daily fluidnd ethanol intake, the RW- and EtOH-treated rats ingestedimilar amounts. Energy intake was calculated by addinghe energy provided by the amount of chow ingested (15.38J/g) to the energy derived from ethanol, taking into accounthe amount of ethanol ingested, ethanol density (1 mLorresponds to 0.79 g), and that 1 g of ethanol supplies 29.3J. Comparison of energy intake revealed no differencesmong groups (Table 1). The feed efficiency or the ratioetween grams of body weight gain per energy consumedas also calculated. Despite having the same energy intake,

he RW- and EtOH-treated animals gained less weight perilojoule of energy consumed (Table 1).

No signs of hepatotoxicity were detected in the RW- andtOH- treated rats in comparison with controls, as shown byimilar plasma concentrations of the enzymes aspartate ami-otransferase, alanine aminotransferase, and alkaline phos-hatase (data not shown). In addition, plasma estradiol con-entrations were similar in controls and after RW and EtOHreatments (53.8 � 2.6, 49.4 � 3.3, and 52.5 � 2.6 pg/mL).

dipocyte size

Measurement of adipocyte size revealed that rats treated

ith RW for 8 wk had smaller adipocyte mean diameters E

han controls (Fig. 2A–C). This reduction in adipocyte sizeas approximately 85% of controls. The same tendency was

een in EtOH-treated rats, although this difference did noteach statistical significance (Fig. 2A,B,D). Histologic anal-sis also revealed that smaller adipocytes in adipose tissueections from RW- and EtOH-treated animals surroundedreas of fibroblast-like cells, probably preadipocytes (Fig.C,D).

romatase expression in adipose tissue

The results obtained showed that the ratio between thexpression of aromatase and �-actin in the adipose tissue ofontrol animals was 0.4 � 0.28. Treatment with EtOHnduced a marked increase in aromatase expression (3.9 �.2; Fig. 3A,B). RW also significantly increased the relativeromatase expression in the adipose tissue to 2.6 � 1.7 (Fig.A,B) when compared with controls. Although the effect of

ig. 1. Evolution of body weight of controls or animals treated with RW ortOH throughout 8 wk of treatment. Results are presented as mean � SD.

A) Body weight curves. *RW significantly different from control. **RWnd EtOH significantly different from control. (B) Body weight gain. *P �.05 versus control. EtOH, 13% ethanol solution; RW, red wine.

tOH seems more pronounced than the effect of RW, there

wt

Ca

piwwwabb

tw

D

aistsdAt

TS

CRE

c

FAdosa

702 R. Monteiro et al. / Nutrition 25 (2009) 699–705

as no statistically significant difference between these tworeatments concerning aromatase expression.

orrelations of adipocyte size, body weight, andromatase expression

To plot correlations, data from the animals of each ex-erimental group were pooled. We found a significant pos-tive correlation between adipocyte mean diameter and bodyeight, i.e., the larger the adipocytes, the greater the bodyeight (r � 0.77; Fig. 4A). A significant inverse correlationas verified between adipose tissue aromatase expression

nd body weight (r � �0.75; Fig. 4B). This associationecame even stronger when body weight gain instead ofody weight was considered (r � �0.83), i.e., the greater

able 1ummary of food, fluid, ethanol, and energy intakes*

Food (g/d per rat) Fluid (mL/d per rat) Ethanol(mL · kg

ontrol 20.62 � 0.70 27.05 � 4.61 —W 14.77 � 1.72† 16.40 � 1.87† 7.35 � 0tOH 15.63 � 1.41† 19.99 � 3.27† 8.97 � 1

BW, body weight; EtOH, 13% ethanol solution; RW, red wine* Rats were treated for 8 wk with water (control), RW, or EtOH and th

alculated taking into account body weight gain and energy intake during† P � 0.05 versus control.

ig. 2. Adipocyte size in animals from different groups. (A) Adipocytedipocyte diameter was measured after drawing in camera lucida. The largrawings. These two values were then used to calculate the mean diametef adipose tissue sections from control (B), RW-treated, and (C) EtOH-treatained with hematoxylin and eosin. Scale bar � 50 �m. Note the existe

dipose tissue sections. EtOH, 13% ethanol solution; RW, red wine.

he aromatase expression in the adipose tissue, the lesser theeight gain (Fig. 4C).

iscussion

Excessive adipose tissue accumulation has serious met-bolic consequences. Moreover, obesity incidence is grow-ng worldwide in adults and children [18]. The obese stateeems to be associated with low-grade systemic inflamma-ion, because inflammatory markers are increased in obeseubjects, establishing a link among obesity, cardiovascularisease, type 2 diabetes, and other associated pathologies.dipose tissue is currently regarded as a metabolically ac-

ive secretory organ sending out and integrating signals to

· d�1)Energy(kJ · kg�1 BW · d�1)

Feed efficiency (g BW gain/kJ)

994.49 � 35.73 0.0100 � 0.0004981.24 � 60.32 0.0082 � 0.0008†

1050.49 � 77.22 0.0072 � 0.0011†

od and fluid ingestions were monitored every 2–3 d. Feed efficiency wasxperimental period. Results are presented as mean � SD.

diameter in water-treated (control), RW-treated, and EtOH-treated rats.meter of each cell and the axis perpendicular to it were measured on paperults are presented as mean � SD. *P � 0.05 versus control. Micrographs) rats. Tissue sections were prepared from paraffin-embedded samples andsmaller adipocytes surrounding stromal areas in RW- and EtOH-treated

�1 BW

.65

.50

eir fothe e

meanest diar. Rested (Dnce of

iei

aitpliifmmd

eEfmsite

Tar

Frem

Ffw�

E

703R. Monteiro et al. / Nutrition 25 (2009) 699–705

nfluence energy expenditure, appetite, insulin signaling,ndocrine and reproductive systems, bone metabolism, andmmunity [18].

Adipose tissue is the main producer of estrogens in mennd in postmenopausal women and these hormones arenvolved in the determination of the amount of adiposeissue in the body [3,4]. They are usually regarded as op-osing fat accumulation because they are known to increaseipolysis, inhibit lipogenesis, and interfere with satiety andnvoluntary energy expenditure [4]. Confirmation of thenfluence of estrogens on body weight homeostasis is givenrom models of estrogen insufficiency. The increased accu-ulation of fat, particularly visceral fat, in women afterenopause is an example [6] as is the severe obese state

eveloped by aromatase knockout mice [5].A key finding in this study was the increase in aromatase

xpression in adipose tissue of rats induced by RW andtOH treatments. In these groups, body weight gain and

eed efficiency were lower than in control rats. In the treat-ent model used, rats had RW or EtOH as the only fluid

ources, but no signs of toxicity were present, thus discard-ng the possibility of decreased weight gain due to ethanoloxicity. After 8 wk of drinking RW or EtOH, aromatase

ig. 3. Relative aromatase expression in adipose tissue of animals treatedor 8 wk with water (control), RW, or EtOH. The expression of aromataseas determined by western blotting and normalized for the expression of-actin. Results are presented as mean � SD. *P � 0.05 versus control.tOH, 13% ethanol solution; RW, red wine.

xpression in adipose tissue increased more than 2.5-fold. fi

his suggests an increase in estrogen production by thedipose tissue and raises the possibility of body weightegulation through this mechanism. The association be-

ig. 4. Correlation between adipocyte mean diameter and body weight (A),elative aromatase expression and body weight (B), and relative aromatasexpression and body weight gain (C). Data from animals of the three experi-ental groups were pooled to plot Spearman’s correlations. Correlation coef-

cients are shown with 95% confidence intervals. *Statistically significant.

tledEplqcctaOtTattpa

iicwsfcmrprbosoopstbci

aeedrm

aAfrd

spatasc(pard

lsHboRgrbaaeAsccfAaaw

C

ieatur

A

aam

704 R. Monteiro et al. / Nutrition 25 (2009) 699–705

ween alcohol consumption and the increase in circulatingevels of estrogens has been demonstrated [19,20]. How-ver, in the present study, we were not able to detectifferences in plasma estradiol concentrations after RW ortOH treatment, although the pronounced increase in adi-ose tissue aromatase expression suggests an increase inocal estradiol concentrations. A similar observation is fre-uent in other estrogen-producing tissues, where estrogenoncentration may be 10- to 20-fold higher than plasmaoncentration [21,22]. Indeed, the lower body weight inhese groups is in good agreement with the increase indipose tissue aromatase and local estrogen availability.ur results indicate that part of the RW effects on adipose

issue aromatase expression may be attributable to ethanol.he lower level of aromatase expression in RW-treatednimals, in comparison with EtOH, is most probably relatedo the wine polyphenolic compounds. As a matter of fact,he modulation of aromatase activity by polyphenolic com-ounds from wine such as quercetin, myricetin, resveratrol,nd procyanidins has been reported [14,15,23,24].

Most interesting was the observation that the presumablencrease in estrogen production by the adipose tissue, asndicated by a higher local aromatase expression, was ac-ompanied by a lower body weight gain in the animals. Thisas observed in RW- and EtOH-treated rats, despite a

imilar energy intake in comparison with controls, i.e., ratsrom the RW and EtOH groups had lower feed efficienciesompared with controls. Bargallo et al. [12] found thatoderate RW ingestion could partly prevent weight gain in

ats fed with a hyperlipidic diet. Our results are thus com-atible with treatments interfering with energy metabolismather than satiety. Alcohol seems to induce some effects onody weight gain. Several epidemiologic studies have dem-nstrated an inverse association of moderate alcohol con-umption with body mass index [10,25]. This also reflectsn obesity complications, namely decreasing the prevalencef type 2 diabetes, insulin resistance, and the vascular riskrofile [26]. In experimental settings, ethanol has beenhown to interfere with insulin action by decreasing theranslocation of glucose transporter-4 to the plasma mem-rane, inhibiting glucose uptake by cells and its posterioronversion to lipids, inhibiting basal and adrenaline-nduced lipolysis, and increasing leptin secretion [27].

The effect of RW on weight gain was similar to that ofn equivalent ethanol solution in water, despite a lesserffect on aromatase expression. This may indicate that theffect of wine on body weight is most likely not exclusivelyue to its ability to alter aromatase expression and may alsoesult from other effects of phenolic compounds on energyetabolism.The RW-treated animals had smaller adipocytes in their

dipose tissue, which may significantly add to their health.s we have demonstrated, the larger the adipocyte, the more

ragile it becomes, being more prone to rupture [28]. Theeported accumulation of macrophages in the vicinity of

ead adipocytes, suggesting that adipocyte death may be the a

ource of obesity related-inflammation [29], strongly sup-orts our interpretation. This is of particular significance fordipocytes that are enclosed within the visceral cavity andhus subject to sudden pressure variation during physicalctivities, cough, or obstructive sleep apnea [30]. In theame view, adipocyte size has been shown to be positivelyorrelated with circulating levels of inflammatory cytokinestumour necrosis factor-�, interleukin-6, and C-reactiverotein) and negatively correlated with anti-inflammatorydipokine levels [31]. Consistent with these results is theeported increase in adiponectin circulating levels in menrinking RW for 4 wk [32].

The reduction of adipocyte size may be due to increasedipolysis and decreased lipogenesis, as demonstrated intudies about estrogen effects on energy metabolism [4].owever, a further and more interesting mechanism coulde the recruitment of preadipocytes. Indeed, the observationf larger areas of stroma in adipose tissue sections fromW- and EtOH-treated rats may indicate that there is areater availability of preadipocytes. Conversely, a putativeeduction of oxidative stress by RW polyphenols woulde expected to favor preadipocyte differentiation [33]. Inny case, adipose tissue expandability is now recognizeds an important buffering intervenient to deal with nutri-nt excess and obesity-induced metabolic complications.romatase data are in good agreement with this hypothe-

is because aromatase expression is higher in preadipo-ytes compared with adipocytes [34]. It also becomeslear that a great part of these effects probably resultsrom or is facilitated by the ethanol present in RW.lthough we cannot be sure that those effects are caus-

lly related to aromatase induction, the observed alter-tions in body weight and adipocyte size are compatibleith an increase in estrogen production.

onclusions

The beneficial potential of RW ingestion to human healths emphasized, as is the large contribution of ethanol to itsffects. Although the effects of wine ingestion on health arelways intimately connected with the dose, RW was showno be capable of decreasing adipocyte size, which may be oftmost significance for the improvement of metabolic pa-ameters in obesity.

cknowledgments

The excellent technical support of Mrs. Luísa Vasquesnd Mr. Abílio Ferreira is gratefully acknowledged. Theuthors thank Prof. João Tiago Guimarães from the Depart-ent of Clinical Pathology, S. João Hospital, for the valu-

ble help with plasma measurements.

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

705R. Monteiro et al. / Nutrition 25 (2009) 699–705

eferences

[1] Stein CJ, Colditz GA. The epidemic of obesity. J Clin EndocrinolMetab 2004;89:2522–5.

[2] Kershaw EE, Flier JS. Adipose tissue as an endocrine organ. J ClinEndocrinol Metab 2004;89:2548–56.

[3] Simpson ER. Role of aromatase in sex steroid action. J Mol Endo-crinol 2000;25:149–56.

[4] Cooke PS, Naaz A. Role of estrogens in adipocyte development andfunction. Exp Biol Med (Maywood) 2004;229:1127–35.

[5] Jones ME, Thorburn AW, Britt KL, Hewitt KN, Misso ML, WrefordNG, et al. Aromatase-deficient (ArKO) mice accumulate excess ad-ipose tissue. J Steroid Biochem Mol Biol 2001;79:3–9.

[6] Kotani K, Tokunaga K, Fujioka S, Kobatake T, Keno Y, Yoshida S,et al. Sexual dimorphism of age-related changes in whole-body fatdistribution in the obese. Int J Obes Relat Metab Disord 1994;18:207–2.

[7] Bhathena SJ, Velasquez MT. Beneficial role of dietary phytoestro-gens in obesity and diabetes. Am J Clin Nutr 2002;76:1191–201.

[8] Lin J, Della-Fera MA, Baile CA. Green tea polyphenol epigallocat-echin gallate inhibits adipogenesis and induces apoptosis in 3T3-L1adipocytes. Obes Res 2005;13:982–90.

[9] Wu CH, Lu FH, Chang CS, Chang TC, Wang RH, Chang CJ.Relationship among habitual tea consumption, percent body fat, andbody fat distribution. Obes Res 2003;11:1088–95.

10] Djousse L, Arnett DK, Eckfeldt JH, Province MA, Singer MR,Ellison RC. Alcohol consumption and metabolic syndrome: does thetype of beverage matter? Obes Res 2004;12:1375–85.

11] Ardevol A, Blade C, Salvado MJ, Arola L. Changes in lipolysis andhormone-sensitive lipase expression caused by procyanidins in3T3-L1 adipocytes. Int J Obes Relat Metab Disord 2000;24:319–24.

12] Bargallo MV, Grau AA, Fernandez-Larrea JdeD, Anguiano GP, Se-garra MC, Rovira MJ, et al. Moderate red-wine consumption partiallyprevents body weight gain in rats fed a hyperlipidic diet. J NutrBiochem 2006;17:139–42.

13] Pinent M, Blay M, Blade MC, Salvado MJ, Arola L, Ardevol A.Grape seed–derived procyanidins have an antihyperglycemic effectin streptozotocin-induced diabetic rats and insulinomimetic activity ininsulin-sensitive cell lines. Endocrinology 2004;145:4985–90.

14] Eng ET, Williams D, Mandava U, Kirma N, Tekmal RR, Chen S.Suppression of aromatase (estrogen synthetase) by red wine phyto-chemicals. Breast Cancer Res Treat 2001;67:133–46.

15] Monteiro R, Azevedo I, Calhau C. Modulation of aromatase ac-tivity by diet polyphenolic compounds. J Agric Food Chem 2006;54:3535– 40.

16] SSCC. The committee on enzymes of the Scandinavian Society forClinical Chemistry and Clinical Physiology: recommended methodsfor the determination of four enzymes in blood. Scand J Clin LabInvest 1974;33:291–306.

17] Sarmento A, Soares-da-Silva P, Teixeira AA, Azevedo I. Effects of

denervation induced by 6-hydroxydopamine on cell nucleus activity

of arterial and cardiac cells of the dog. J Auton Pharmacol1987;7:119–26.

18] Fantuzzi G. Adipose tissue, adipokines, and inflammation. J AllergyClin Immunol 2005;115:911–9.

19] Ginsburg ES, Mello NK, Mendelson JH, Barbieri RL, Teoh SK,Rothman M, et al. Effects of alcohol ingestion on estrogens inpostmenopausal women. JAMA 1996;276:1747–51.

20] Purohit V. Moderate alcohol consumption and estrogen levels inpostmenopausal women: a review. Alcohol Clin Exp Res 1998;22:994–7.

21] van Landeghem AA, Poortman J, Nabuurs M, Thijssen JH. Endoge-nous concentration and subcellular distribution of estrogens in normaland malignant human breast tissue. Cancer Res 1985;45:2900–6.

22] Smith IE, Dowsett M. Aromatase inhibitors in breast cancer. N EnglJ Med 2003;348:2431–42.

23] Eng ET, Ye J, Williams D, Phung S, Moore RE, Young MK, et al.Suppression of estrogen biosynthesis by procyanidin dimers in redwine and grape seeds. Cancer Res 2003;63:8516–22.

24] Wang Y, Lee KW, Chan FL, Chen S, Leung LK. The red winepolyphenol resveratrol displays bilevel inhibition on aromatase inbreast cancer cells. Toxicol Sci 2006;92:71–7.

25] Wannamethee SG, Field AE, Colditz GA, Rimm EB. Alcohol intakeand 8-year weight gain in women: a prospective study. Obes Res2004;12:1386–96.

26] Dixon JB, Dixon ME, O’Brien PE. Alcohol consumption in theseverely obese: relationship with the metabolic syndrome. Obes Res2002;10:245–52.

27] Szkudelski T, Bialik I, Szkudelska K. Adipocyte lipolysis, hormonaland metabolic changes in ethanol-drinking rats. J Anim Physiol AnimNutr (Berl) 2004;88:251–8.

28] Monteiro R, de Castro PM, Calhau C, Azevedo I. Adipocyte size andliability to cell death. Obes Surg 2006;16:804–6.

29] Cinti S, Mitchell G, Barbatelli G, Murano I, Ceresi E, Faloia E, et al.Adipocyte death defines macrophage localization and function inadipose tissue of obese mice and humans. J Lipid Res 2005;46:2347–55.

30] Monteiro R, Calhau C, Azevedo I. Obstructive sleep apnoea andadipocyte death. Eur J Heart Fail 2007;9:103–4.

31] Bahceci M, Gokalp D, Bahceci S, Tuzcu A, Atmaca S, Arikan S. Thecorrelation between adiposity and adiponectin, tumor necrosis factoralpha, interleukin-6 and high sensitivity C-reactive protein levels. Isadipocyte size associated with inflammation in adults? J EndocrinolInvest 2007;30:210–4.

32] Beulens JW, van Beers RM, Stolk RP, Schaafsma G, Hendriks HF.The effect of moderate alcohol consumption on fat distribution andadipocytokines. Obesity (Silver Spring) 2006;14:60–6.

33] Galinier A, Carriere A, Fernandez Y, Carpene C, Andre M, Caspar-Bauguil S, et al. Adipose tissue proadipogenic redox changes inobesity. J Biol Chem 2006;281:12682–7.

34] Rubin GL, Zhao Y, Kalus AM, Simpson ER. Peroxisome proliferator-activated receptor gamma ligands inhibit estrogen biosynthesis in humanbreast adipose tissue: possible implications for breast cancer therapy.

Cancer Res 2000;60:1604–8.