A

xxtapwf1to©

K

1

owabm(o

iP

0d

Journal of Steroid Biochemistry & Molecular Biology 105 (2007) 124–130

Modulation of breast cancer cell survival by aromataseinhibiting hop (Humulus lupulus L.) flavonoids

Rosario Monteiro a,b,∗, Ana Faria a,c, Isabel Azevedo a, Conceicao Calhau a

a Department of Biochemistry (U38-FCT), Faculty of Medicine, University of Porto, Alameda Prof. Hernani Monteiro,4200-319 Porto, Portugal

b Faculty of Nutrition and Food Sciences, University of Porto, Rua Dr. Roberto Frias, 4200-465 Porto, Portugalc Chemistry Investigation Centre, Department of Chemistry, Faculty of Sciences, University of Porto, Rua do Campo Alegre 687,

4169-007, Porto, Portugal

Received 28 July 2006; accepted 27 November 2006

bstract

Hop flavonoids are being regarded as attractive molecules to prevent or treat certain forms of cancer. Studies have focused mainly onanthohumol, the most abundant prenylated chalcone existing in hops extract. However, during the production of beer, or after its ingestion,anthohumol originates different metabolites, among which isoxanthohumol and 8-prenylnaringenin. The aim of this work was to studyhe effect of the prenylflavonoids xanthohumol, isoxanthohumol and 8-prenylnaringenin on the breast cancer Sk-Br-3 cell line proliferation,poptosis and activity of the enzyme aromatase (estrogen synthase). Aromatase activity was determined by a tritiated water assay, cellroliferation was assessed by [3H]thymidine incorporation, sulforhodamine B protein measurement and Ki-67 immunostaining and apoptosisas determined by TUNEL. Our results show that all tested prenylflavonoids were able to inhibit aromatase activity and thus, estrogen

ormation. Additionally, breast cancer cell line proliferation was decreased and apoptosis induced by all three compounds. The presence of7�-estradiol in treatment medium was able to revert the effect of the prenylflavonoids on cellular proliferation. These observations strengthen

he idea that hop flavonoids may have anti-breast cancer effects and shed new light on a possible mechanism of action by which these effectsccur, namely through their ability to decrease estrogen synthesis.

2007 Elsevier Ltd. All rights reserved.

eywords: Aromatase; Beer; Breast cancer; Estrogens; Humulus lupulus L.; Isoxanthohumol; 8-Prenylnaringenin; Sk-Br-3; Xanthohumol

ci[amcr

. Introduction

Breast cancer is the most frequent form of cancer andne of the main responsible for cancer-related mortality inomen [1]. Estrogens, female reproductive hormones, playcrucial role in the development of breast cancer, having

een shown to be involved in the stages of initiation, pro-

otion and progression of carcinogenesis [2]. Aromatase

EC 1.14.13) is the enzyme responsible for the conversionf circulating androgens into estrogens in the breast. This

∗ Corresponding author at: Servico de Bioquımica da Faculdade de Medic-na da Universidade do Porto, Alameda Prof. Hernani Monteiro, 4200-319orto, Portugal. Tel.: +351 225513624; fax: +351 225513624.

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

[

vbriac

960-0760/$ – see front matter © 2007 Elsevier Ltd. All rights reserved.oi:10.1016/j.jsbmb.2006.11.026

ytochrome P450 (CYP) isoenzyme (CYP19) is expressedn several tissues where estrogens exert physiological roles3]. Breast tumors have also been demonstrated to expressbnormally high levels of the enzyme, in comparison to nor-al tissue [4]. In breast tumors, it is most usual that epithelial

ells express the lower level of aromatase, the stroma beingesponsible for the greatest amount of estrogen production5].

New data show that phytochemicals in common fruits andegetables may have potential as adjuvants in cancer therapyy their ability to (i) induce apoptosis, (ii) scavenge free

adicals (antioxidant activity), (iii) regulate gene expressionnvolved in cell proliferation, cell differentiation, oncogenesnd tumor suppressor genes, (iv) modulate enzymes impli-ated in detoxification (phase II enzymes) and/or the enzymes

emistry

ili

dbg[aiidwaadm

papw[nfTo

naaaosgaa

2

2

2McTr((DXGPS

2

CaMglb1tswEsT

2

otepfwbwfeaiwotaaagrtm

2

Stwpcw

R. Monteiro et al. / Journal of Steroid Bioch

nvolved in carcinogen activation (phase I enzymes), (v) regu-ate hormone metabolism, (vi) improve immune system activ-ty and (vii) possess antibacterial or antiviral activities [6–8].

Hop (Humulus lupulus L.) extracts are added to beeruring its production, being responsible for its flavor anditterness [9]. These extracts are obtained from the lupulinlands of hop cones and are specially rich in prenylflavonoids10]. The major prenylflavonoid in hop is xanthohumol (XN),

chalcone which corresponds to 0.1–1% of hop extractn dry weight and accounts for approximately 82–89% ofts prenylflavonoids [11]. Lately, much attention has beenrawn on this subject, since the discovery that XN bears aide array of anti-cancer effects. Currently, XN is regarded

s a ‘broad spectrum’ chemopreventive agent because it isble to inhibit initiation, promotion and progression in tumorevelopment; studies on the mechanisms of action of thisolecule are on the raising [12].In previous studies, we have described the ability of com-

ounds in hop and beer to interfere with estrogen synthesis inchoriocarcinoma cell line (JAR cells) [13]. Other importantrenylated flavonoids from hops are isoxanthohumol (IXN),hich is formed from XN during the brewing process

8] and 8-prenylnaringenin (8-PN) which may be formedon-enzymatically during drying, storage and extractionrom hops and enzymatically by demethylation of IXN [14].hese compounds share some of the anti-tumour propertiesf XN although with differing strength [6,15].

Nevertheless, the mechanisms of action of these polyphe-ols are largely unknown. Thus, in the present study weimed at testing the effect of XN, IXN and 8-PN, onromatase-expressing breast cancer cell line Sk-Br-3 (breastdenocarcinoma-derived) to investigate their putative effectsn the modulation of estrogen synthesis. To elucidate theignificance of aromatase inhibition on breast cancer cellrowth, DNA and protein synthesis, as well as proliferationnd apoptosis under the influence of prenylflavonoids werelso investigated.

. Materials and methods

.1. Materials

[1�-3H]Androst-4-ene-3,17-dione (specific activity5,3 Ci/mmol, NEN Life Science Products, Boston,A, EUA). 4-Androstene-3,17-dione, activated charcoal,

hloroform, 8-prenylnaringenin, trypsin-EDTA solution,ris–HCl (Tris–[hydroxymethyl]-aminometane hydrochlo-ide), N-2-hydroxyethylpiperazine-N′-2-ethanesulfonic acidHEPES) (Sigma, St. Louis, MO, USA). DimethylsulfoxideDMSO), Triton X-100 (Merck, Darmstadt, Germany).extran 70 (Amersham Biosciences, Uppsala, Sweeden).

anthohumol kindly supplied by Hopsteiner (Mainburg,ermany) through Instituto de Bebidas e Saude (iBeSa,ortugal). Isoxanthohumol (ALEXIS Biochemicals, Lausen,witzerland).

w(c(

& Molecular Biology 105 (2007) 124–130 125

.2. Cell culture

Sk-Br-3 cells were obtained from the American Typeulture Collection. Cells were maintained in humidifiedtmosphere of 5% CO2–95% air and were grown incCoy’s 5A culture medium (Gibco BRL, Life Technolo-

ies, Gaithersburg, MD, USA) supplemented with 2 mM-glutamine, 2.2 g/l NaHCO3, 10% heat-inactivated fetalovine serum (56 ◦C, 30 min) and 100 U/ml penicillin,00 �g/ml streptomycin and 0.25 �g/ml amphotericin. Cul-ure medium was changed every 2–3 days and the culture wasplit when cells reached confluence. For subculturing, cellsere rinsed with PBS and incubated with 0.25% trypsin-DTA solution (37 ◦C, 5 min), removed from the plateurface and cultured on 22.1 cm2 culture plates (Ø60 mm,PP, Trasadingen, Switzerland).

.3. Aromatase assay

Aromatase activity was determined as described previ-usly [16], through measuring the release of [3H]H2O duringhe aromatization of [3H]androstenedione to estrone. For thexperiments cells were split 1:5 and cultured in 12 welllates (3.66 cm2, Ø22.2 mm, TPP, Trasadingen, Switzerland)or 7 days. To study the effect of compounds, cells wereashed with 500 �l serum-free medium and the incubationegan with the addition of 300 �l of serum-free mediumith 100 nM [3H]androstenedione in the presence of dif-

erent concentrations of the test compounds or vehicle (1%thanol). Incubation occurred at 37 ◦C, in a 5% CO2–95% airtmosphere for 5 h. After incubation, plates were placed once (to stop the reaction) and 250 �l of incubation mediumere removed and added to microtubes containing 750 �lf chloroform. Samples were vortexed for 60 s and cen-rifuged at 9000 × g, 1 min. An aliquot of 200 �l of thequeous upper phase was mixed with the same volume of5% charcoal/0.5% dextran 70 suspension, vortexed for 40 snd incubated at room temperature for 10 min. After centrifu-ation for 15 min at 9000 × g, 350 �l of the supernatant wereemoved to determine the level of radioactivity after addi-ion of 4 ml of scintillation cocktail and the [3H]H2O waseasured by liquid scintillation counting.

.4. Determination of DNA synthesis

Cells were seeded onto 24-well plates (TPP, Trasadingen,witzerland) in a final volume of 500 �l culture medium con-

aining 10% FBS. After 24 h in culture, the cells were treatedith different concentrations of the prenylflavonoids in theresence or absence of estradiol (10 nM) in culture mediumontaining 5% FBS for 72 h. Control cells were incubatedith 0.1% ethanol. After treatment, the cells were incubated

ith 200 �l culture medium with methyl-[3H]thymidine

0.5 �Ci/well) for 4 h. The medium was removed and theells were fixed by incubation with 10% trichloroacetic acidTCA), 1 h at 4 ◦C. The cells were washed twice with 10%

1 emistry

Ta2ts

2

wloSowa2otctttwStrw(4tdft

2

Ka8AfwgibHs(c(tCbK

fcoC

2

st3tmuSwwBmo1f

2

d

2

Iistc

3

3

isy(b[siT

26 R. Monteiro et al. / Journal of Steroid Bioch

CA to remove unbound radioactivity. Plates were air-driednd the cells were lysed with 1 M NaOH (280 �l/well). A50 �l aliquot of the lysate was neutralized with HCl prioro the addition of scintillation fluid. The radioactivity of theamples was quantified by liquid scintillation counting.

.5. Sulforhodamine B assay

The sulforhodamine B (SRB) assay, which measureshole-culture protein content as an index of tumour cell pro-

iferation, was employed to determine the relative potenciesf the individual hop compounds in inhibiting the growth ofk-Br-3 cells as well as their effect in the presence of 10 nMf estradiol. Cell cultures were plated on 96-well plates,ith each well containing 200 �l growth medium per well

t a density of 8 × 103 cells/well. The cells were exposed to00 �l of culture medium containing different concentrationsf hop flavonoids or vehicle (0.05% ethanol). After 72 h ofreatment, 25 �l of ice-cold 50% TCA were added to theulture medium on each well to fix cells, for 1 h at 4 ◦C inhe dark. Control cells were in the proliferative stage duringreatment and semi-confluent after 4 days. The plates werehen washed five times with tap water to remove TCA. Platesere air-dried and then stained for 15 min with 0.4% (w/v)RB dissolved in 1% acetic acid. SRB was removed and cul-

ures were rinsed four times with 1% acetic acid to removeesidual dye. Plates were again air-dried and the bound dyeas then solubilized with 200 ml of 10 mM Tris base solution

pH 10.5). The absorbance of each well was determined at92 nm in a microplate reader. At day zero (day of treatment),he absorbance of the control wells in the 96-well plates wasetermined and the antiproliferative activity was calculatedrom the ratio of the absorbance readings of the treated wellso those of the control wells.

.6. Ki-67 immunocytochemistry

For immunostaining of the nuclear proliferation markeri-67, 2 × 104 cells were cultured on glass cover slips for 24 h

nd treated for another 24 h with XN (5 �M), IXN (50 �M),-PN (0.1 �M) or vehicle (0.1% ethanol) in culture medium.fter treatment, cells were fixed with ice-cold methanol

or 30 min at 4 ◦C. Cells were rinsed in PBS and blockedith 0.5% BSA in PBS followed by 1 h incubation withoat anti-human Ki-67 antibody, diluted 1:50 in 0.5% BSAn PBS. After rinsing with PBS, cells were incubated withiotinylated anti-goat IgG (1:200) for 10 min. Avidin-biotin-RP and 3,3-diaminobenzidin tetra-hydrochloride (DAB)

ubstrate chromogen were used to demonstrate the antigensVector Laboratories, California, USA). Cell nuclei wereounterstained with modified Harris’ hematoxylin solutionSigma, St. Louis, MO, USA). Negative control for immunos-

aining was carried out by omission of the primary antibody.overslips were mounted on glass slides and were observedlindly under a light microscope at 200× magnification. Fori-67 counting of stained nuclei, over 1000 cells from dif-

5TdC

& Molecular Biology 105 (2007) 124–130

erent fields (5–8) were counted and Ki-67 labelling wasalculated as percentage of positive cells over total numberf cells counted. All antibodies were purchased from Santaruz Biotechnologies (CA, USA).

.7. Apoptosis determination

For apoptosis assays, cells were cultured on glass coverlips and treated for 24 h as previously described. Afterreatment, cells were fixed with ice-cold 4% formalin for0 min at room temperature. To identify apoptotic cellshe TUNEL assay (Terminal deoxynucleotidyl transferase-

ediated deoxyuridine triphosphate nick end-labeling) wassed according to the producer’s (Roche Diagnostics, Basel,witzerland) instructions. Cell nuclei were counterstainedith DAPI (Roche Diagnostics, Basel, Switzerland). Slidesere visualized under a fluorescence microscope (Olympus,H-2, UK) with 200x magnification. Apoptosis was deter-ined as the percentage of positive cells over the total number

f cells counted (visualized with DAPI staining). A total of000 nuclei were blindly counted from different section fieldsor each treatment.

.8. Protein determination

The protein content of cell monolayers was determined asescribed [17] with bovine serum albumin as standard.

.9. Calculation and statistics

Results are expressed as arithmetic means (S.D.). TheC50s are given as geometric means with 95% confidencentervals (CI) and were calculated using non-linear regres-ion analysis in Graph Pad Prism 3.0 software. To comparereatments, Student’s t-test was applied. Differences wereonsidered to be statistically significant when p < 0.05.

. Results

.1. Aromatase activity

In previous time-course studies of aromatase activ-ty in Sk-Br-3 cells, we had found that [3H]H2Oynthesis was linear up to 7 h, and from kinetic anal-sis Km for [3H]androstenedione found to be 113 nM95% CI = 50.5–177.1 nM). Therefore, we used an incu-ation time of 5 h with 100 nM aromatase substrate3H]androstenedione for subsequent studies (results nothown). Treatment with XN, IXN and 8-PN resultedn a dose dependent inhibition of aromatase (Fig. 1).he concentration that decreased aromatase activity to

0% (IC50) was calculated for the three prenylflavonoids.he most striking effect was observed with 8-PN whichecreased aromatase activity with an IC50 of 0.08 �M (95%I = 0.02–0.27 �M; n = 6). XN and IXN did also inhibit

R. Monteiro et al. / Journal of Steroid Biochemistry & Molecular Biology 105 (2007) 124–130 127

Fig. 1. Effect of xanthohumol (XN), isoxanthohumol (IXN) and 8-pw(w

aI(

3

fbcppraldd7wIEgtc

wi7

Fig. 2. Effect of prenylflavonoids in the absence or in the presence of 10 nMof 17�-estradiol on cellular growth. A – Xanthohumol (XN). B – Isoxantho-humol (IXN). C – 8-Prenylnaringenin (8-PN). Cells were plated on 96-wellplates and treated after 24 h with the compounds or the solvent (ethanol) for

TIp

AC[

renylnaringenin (8-PN) on aromatase activity in Sk-Br-3 cells. The cellsere incubated for 5 h in the presence of the compounds or the solvent

ethanol) and 100 nM of [3H]androstenedione and the release of tritiatedater was measured. Results are means (S.D.) (n = 6).

romatase (IC50 = 3.2 �M, 95% CI = 0.9–11.8 �M, n = 6;C50 = 25.4 �M, 95% CI = 5.5–116.9 �M, n = 6, respectively)Table 1).

.2. Cellular proliferation

Proliferation of Sk-Br-3 cells after treatment with dif-erent concentrations of XN, IXN or 8-PN was determinedy SRB staining, methyl-[3H]thymidine incorporation intoellular DNA and immunocytochemical detection of Ki-67rotein expression. SRB method consists of staining cellularroteins with SRB and measuring the dye spectrophotomet-ically. This method is considered to provide informationbout cellular proliferation, since protein synthesis and cellu-ar proliferation are tightly connected. Protein synthesis wasecreased by 72 h incubation with XN, IXN or 8-PN in a doseependent manner. XN was the most potent with an IC50 of.1 �M (95% CI = 3.4–14.9 �M; n = 9), followed by 8-PNith an IC50 22.6 �M (95% CI = 7.7–66.0 �M; n = 9) and

XN with an IC50 41.0 �M (95% CI = 19.3–87.7 �M; n = 9).stradiol was able to revert the inhibition imposed on cellularrowth by XN and IXN. For 8-PN, the addition of estradiolo the treatment medium resulted in a significant increase inellular growth (Fig. 2).

The other method used to evaluate cellular proliferationas measuring the incorporation of methyl-[3H]thymidine

nto cellular DNA. Treatment of Sk-Br-3 cells for2 h with XN resulted in the most pronounced dose-

72 h. Cellular growth was measured on a microplate reader after stainingcellular protein with sulforhodamine B. Results are means (S.D.) (n = 9).

able 1C50 for the inhibition of aromatase activity, cellular growth, and [3H]thymidine incorporation by xanthohumol (XN), isoxanthohumol (IXN) and 8-renylnaringenin (8-PN) on Sk-Br-3 cells

IC50 (95% CI, �M)

XN IXN 8-PN

romatase activity 3.2 (0.9–11.8) 25.4 (5.5–116.9) 0.08 (0.02–0.27)ellular growth 7.1 (3.4–14.9) 41.0 (19.3–87.7) 22.6 (7.7–66.0)

3H]Thymidine incorporation 0.52 (0.21–1.3) 25.0 (6.7–93.7) 0.88 (0.22–3.5)

128 R. Monteiro et al. / Journal of Steroid Biochemistry & Molecular Biology 105 (2007) 124–130

Fig. 3. Effect of prenylflavonoids in the absence or in the presence of 10 nMof 17�-estradiol on [3H]thymidine incorporation. A – Xanthohumol (XN). B– Isoxanthohumol (IXN). C – 8-Prenylnaringenin (8-PN). Cells were platedon 24-well plates and treated after 24 h with the compounds or the solvent(4c

dCtnogoe(

Fig. 4. Cellular proliferation after treatment with xanthohumol (XN, 5 �M),isoxanthohumol (IXN, 50 �M), 8-prenylnaringenin (8-PN, 0.1 �M) or thesolvent (C, ethanol 0.1%). Sk-Br-3 cells were plated in; cover slips (2 x 104ctet

v8mrcan

3

cnaany of the polyphenols. 8-PN (0.1 �M) displayed the moststriking effect, increasing the percentage of apoptotic cellsto 27% (S.D. = 15.2) in comparison to 0.8% (S.D. = 0.4) incontrol cells (Fig. 5).

Fig. 5. Apoptosis in Sk-Br-3 cells after treatment with xanthohumol (XN,5 �M), isoxanthohumol (IXN, 50 �M), 8-prenylnaringenin (8-PN, 0.1 �M)

ethanol) for 72 h. Cellular proliferation was measured after incubation forh with 0.5 �Ci [3H]thymidine per well, cell lysis and liquid scintillationounting. Results are means (S.D.) (n = 9).

ependent reduction in DNA synthesis (IC50 = 0.52 �M; 95%I = 0.21–1.3 �M; n = 9). 8-PN did also inhibit DNA syn-

hesis, with an IC50 of 0.9 �M (95% CI = 0.22–3.5 �M;= 9). The effect of IXN was less pronounced with an IC50f 25 �M (95% CI = 6.7–93.7 �M; n = 9). As for cellular

rowth assessed through SRB staining, the effect of anyf the three prenylfiavonoids was attenuated by estradiol,xcept for the higher concentrations of XN, IXN and 8-PNFig. 3).

o(sR

ells) and the nuclear marker of cellular proliferation was Ki-67 revealedhrough immunocytochemistry. Cells positive for Ki-67 were counted forach 1000 cells. Results are means (S.D.) of three independent determina-ions.

Through Ki-67 immunostaining, proliferating cells wereisualized in all treatments (5 �M XN, 50 �M IXN, 0.1 �M-PN) and in control cells (30%, S.D. = 3.3, n = 3). Deter-ination of the proportion of cells stained for Ki-67

evealed once again that all three treatments signifi-antly decreased cellular proliferation (Fig. 4). In thisssay IXN showed as the most potent (21%, S.D. = 5.3,= 3).

.3. Apoptosis

The proportion of apoptotic cells was determined byounting, in several optical fields, the number of fluorescentuclei stained with TUNEL. We found that the number ofpoptotic cells was significantly higher after treatment with

r the solvent (C, ethanol 0.1%). Sk-Br-3 cells were plated in cover slips2 × 104 cells) and apoptosis was determined by TUNEL. Apoptotic cellstained with TUNEL were counted for every 1000 cells stained with DAPI.esults are means (S.D.) of three independent determinations.

emistry

4

eogtEcdarnanfdabmoih

eeeceipnGsvggwtcm

idmobtbG

flrswg

ie

opsaXmeI[t

tpstotbia1ii[

lppbtrm

cnetciitf

cwafwn

R. Monteiro et al. / Journal of Steroid Bioch

. Discussion

Growing evidence is being gathered on the anti-tumourffects of XN, a flavonoid found in the female influorescencef the plant Humulus lupulus L., commonly known as hop. Areat amount of scientific work is being carried out in ordero understand how this compound exerts its actions [9,15].ffects such as inhibition of metabolic activation of car-inogenesis, induction of carcinogen detoxifying enzymes,ecrease of inflammation and angiogenesis and antioxidantnd free radical scavenging activities have already beeneported [6,8,18]. Here, we suggest a new possible mecha-ism through which XN might block carcinogenesis initiationnd/or progression. Adding to the already described mecha-isms, recently we have found the ability of prenylflavonoidsrom hops to inhibit aromatase activity in choriocarcinomaerived JAR cells [13]. Our present results show a decrease ofromatase activity in Sk-Br-3 cells, a breast cancer cell line,y incubation with different concentrations of XN. Further-ore, we have also studied the effect of the main metabolites

f this flavonoid, IXN and 8-PN, on aromatase activity, hav-ng found that they do also decrease aromatase activity and,ence, estrogen formation.

Sk-Br-3 cells are adenocarcinoma-derived cells thatxpress aromatase, although they are usually considered asstrogen receptor (ER)-depleted [19]. As is well known,strogens may influence breast cancer cells through thelassical route, mediated by estrogen interaction with ERsxisting in the nucleus or cell membrane, or through ER-ndependent pathways that involve actions on cell signallingathways [2,20]. In this particular cell line, although ER-egative, there is a G protein-coupled orphan receptor,PR30, which displays high affinity, limited capacity and

pecific 17�-estradiol binding [19]. Evidence has been pro-ided that this receptor mediates most of estrogen actions onrowth and survival in Sk-Br-3 cell line [21]. Although estro-ens are endocrine hormones, secreted to the blood to interactith more or less distant target tissues [2], in breast cancer

he in situ production of estrogens has been described as cru-ial to tumour development, estrogens acting in a paracrineode [22].Given these facts, the inhibition of aromatase observed

n Sk-Br-3 cells is a relevant finding since estrogens pro-uced by these epithelial-like cells may act in an autocrineanner or they may as well be released to induce growth

f adipose tissue and vascular stromal cells, which wille essential to tumor nourishment and survival [5]. Onhe other hand, decreased estrogen availability is likely toe translated into lesser activation of estrogen-dependentPR30-mediated events.Explanations for the inhibitory activity of aromatase by

avonoids have been advanced by others, and include the

esemblance of flavonoid basic structure to aromatase sub-trates [23]. In good agreement with this hypothesis, 8-PN,ith its 5-hydroxyl group reminiscent of the 4-hydroxylroup of 4-hydroxyandrostenedione, a specific irreversible

fatfl

& Molecular Biology 105 (2007) 124–130 129

nhibitor of aromatase, showed the highest inhibitoryffect.

In order to confirm the impact of aromatase inhibitionn tumor cell survival, we tested the same compounds onrotein and DNA synthesis and in proliferation and apopto-is. A concentration-dependent decrease of protein synthesis,n index of cellular proliferation, was observed. 8-PN andN did also dose-dependently inhibit DNA synthesis, aseasured by [3H]thymidine incorporation, while IXN was

ffective only at some concentrations. Although for XN andXN this has been previously described in several cell lines6], ours constitutes the first report for 8-PN reduction ofumor cell proliferation.

In order to verify if the reduction of cellular prolifera-ion was related to aromatase inhibition, the effects of therenylflavonoids on protein and DNA synthesis were mea-ured in the presence of 17�-estradiol. It was observed thathis procedure was able to recover most of the inhibitionbtained with XN, IXN and 8-PN. These results revealedhat at least part of the cellular proliferation can be explainedy the lack of estrogen availability caused by aromatase-nhibitory properties of the flavonoids. However, for XNnd IXN at the higher concentrations tested, the addition of7�-estradiol was not able to fully recover DNA synthesisnhibition, what is compatible with the existence of estrogenndependent mechanisms of actions of these flavonoids on3H]thymidine incorporation.

Aiming to investigate a more specific marker of cell pro-iferation, Sk-Br-3 cells were cultivated on coverslips andrepared for immunocytochemical detection of the nuclearrotein Ki-67. This experiment showed a decreased num-er of Ki-67 stained nuclei after treatment with any ofhe flavonoids, what reinforces their action as cell cycle-egulating agents and advances aromatase inhibition as a newechanism of action for these polyphenols.As previously mentioned, estrogen actions on Sk-Br-3

ells can be mediated through GPR30 dependent mecha-isms. In these cells, activation of this membrane receptor bystrogens leads to increased cAMP formation, EGFR transac-ivation and also to c-fos expression [24]. The actions of theseellular effectors reflect on increased expression of genesnvolved in cell growth and cell cycle regulation [21]. Thisndicates that the decrease in estrogen availability obtainedhrough the action of the prenylflavonoids may well accountor the effects produced over cellular proliferation.

Concerning effects on apoptosis, some studies report theapacity of XN to stimulate this programmed cell death path-ay. On breast cancer cells, part of these effects has been

ttributed to its partial antagonistic action on ERs [25]. Datarom the present study support the existence of a second path-ay involved in apoptosis that may be influenced by XN,amely a decrease of estrogen levels. This may also be true

or the other prenylflavonoids. However, one could face theromatase-inhibiting ability of 8-PN suspiciously, knowinghat, despite being the strongest to inhibit aromatase, thisavonoid is, at the same time, a potent phytoestrogen [26].

1 emistry

Tltcc

mepfapeolaa

A

Cv(S

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

J. Clin. Endocrinol. Metab. 85 (2000) 4912–4915.

30 R. Monteiro et al. / Journal of Steroid Bioch

his fact could result in deleterious effects in terms of the pro-iferation of estrogen-dependent tumors. Still, it is possiblehat it is not quite so in classical ER-deficient breast cancerells, as we observed a concomitant significant decrease ofellular growth and DNA synthesis with 8-PN.

Although both a decrease of proliferation and an enhance-ent of apoptosis might be a consequence of decreased

strogen availability, they may as well result from other inde-endent mechanisms, some of which have been demonstratedor XN [6,12,18,26,27]. In conclusion, our work advancesn original hypothesis for the anti-breast tumor effects ofrenylflavonoids from hops, related to their ability to inhibitstrogen formation. Although this theory concerns only onef the possible explanations, given the great diversity of cellu-ar phenomena that are influenced by this class of compounds,romatase inhibition may be of importance for the preventionnd treatment of breast cancer.

cknowledgements

The authors would like to thank Eng◦ Jose Machadoruz (IBeSa—Instituto de Bebidas e Saude) for pro-iding xanthohumol. This work was supported by FCTPOCTI/POCI, Quadro Comunitario de Apoio, FEDER andFRH/BD/12622/2003).

eferences

[1] F. Levi, C. Bosetti, F. Lucchini, E. Negri, C. La Vecchia, Monitoringthe decrease in breast cancer mortality in Europe, Eur. J. Cancer Prev.14 (2005) 497–502.

[2] J.D. Yager, N.E. Davidson, Estrogen carcinogenesis in breast cancer,N. Engl. J. Med. 354 (2006) 270–282.

[3] E.R. Simpson, C. Clyne, G. Rubin, W.C. Boon, K. Robertson, K. Britt,C. Speed, M. Jones, Aromatase-a brief overview, Annu. Rev. Physiol.64 (2002) 93–127.

[4] S.J. Santner, R.J. Pauley, L. Tait, J. Kaseta, R.J. Santen, Aromataseactivity and expression in breast cancer and benign breast tissue stromalcells, J. Clin. Endocrinol. Metab. 82 (1997) 200–208.

[5] S.E. Bulun, Z. Lin, G. Imir, S. Amin, M. Demura, B. Yilmaz, R. Martin,H. Utsunomiya, S. Thung, B. Gurates, M. Tamura, D. Langoi, S. Deb,Regulation of aromatase expression in estrogen-responsive breast anduterine disease: from bench to treatment, Pharmacol. Rev. 57 (2005)359–383.

[6] C.L. Miranda, J.F. Stevens, A. Helmrich, M.C. Henderson, R.J.Rodriguez, Y.H. Yang, M.L. Deinzer, D.W. Barnes, D.R. Buhler,Antiproliferative and cytotoxic effects of prenylated flavonoids fromhops (Humulus lupulus) in human cancer cell lines, Food Chem. Toxi-col. 37 (1999) 271–285.

[7] E. Middleton Jr., C. Kandaswami, T.C. Theoharides, The effects ofplant flavonoids on mammalian cells: implications for inflammation,heart disease, and cancer, Pharmacol. Rev. 52 (2000) 673–751.

[8] C. Gerhauser, A. Alt, E. Heiss, A. Gamal-Eldeen, K. Klimo, J. Knauft,I. Neumann, H.R. Scherf, N. Frank, H. Bartsch, H. Becker, Cancer

chemopreventive activity of Xanthohumol, a natural product derivedfrom hop, Mol. Cancer Ther. 1 (2002) 959–969.

[9] J.F. Stevens, J.E. Page, Xanthohumol and related prenylflavonoidsfrom hops and beer: to your good health!, Phytochemistry 65 (2004)1317–1330.

[

& Molecular Biology 105 (2007) 124–130

10] B. Vanhoecke, L. Derycke, V. Van Marck, H. Depypere, D. DeKeukeleire, M. Bracke, Antiinvasive effect of xanthohumol, a preny-lated chalcone present in hops (Humulus lupulus L.) and beer, Int. J.Cancer 117 (2005) 889–895.

11] B.M. Dietz, Y.H. Kang, G. Liu, A.L. Eggler, P. Yao, L.R. Chad-wick, G.F. Pauli, N.R. Farnsworth, A.D. Mesecar, R.B. van Breemen,J.L. Bolton, Xanthohumol isolated from Humulus lupulus Inhibitsmenadione-induced DNA damage through induction of quinone reduc-tase, Chem. Res. Toxicol. 18 (2005) 1296–1305.

12] E.C. Colgate, C.L. Miranda, J.F. Stevens, T.M. Bray, E. Ho, Xantho-humol, a prenylflavonoid derived from hops induces apoptosis andinhibits NF-kappaB activation in prostate epithelial cells, Cancer Lett.246 (2006) 201–209.

13] R. Monteiro, H. Becker, I. Azevedo, C. Calhau, Effect of hop (Humu-lus lupulus L.) flavonoids on aromatase (estrogen synthase) activity, J.Agric. Food Chem. 54 (2006) 2938–2943.

14] D. Nikolic, Y. Li, L.R. Chadwick, G.F. Pauli, R.B. vanBreemen, Metabolism of xanthohumol and isoxanthohumol, prenylatedflavonoids from hops (Humulus lupulus L.), by human liver micro-somes, J. Mass Spectrom. 40 (2005) 289–299.

15] C. Gerhauser, Beer constituents as potential cancer chemopreventiveagents, Eur. J. Cancer 41 (2005) 1941–1954.

16] T. Nakanishi, J. Kohroki, S. Suzuki, J. Ishizaki, Y. Hiromori, S.Takasuga, N. Itoh, Y. Watanabe, N. Utoguchi, K. Tanaka, Trialkyltincompounds enhance human CG secretion and aromatase activity inhuman placental choriocarcinoma cells, J. Din. Endocrinol. Metab. 87(2002) 2830–2837.

17] M.M. Bradford, A rapid and sensitive method for the quantitation ofmicrogram quantities of protein utilizing the principle of protein-dyebinding, Anal. Biochem. 72 (1976) 248–254.

18] A. Albini, R. Dell’Eva, R. Vene, N. Ferrari, D.R. Buhler, D.M. Noo-nan, G. Fassina, Mechanisms of the antiangiogenic activity by the hopflavonoid xanthohumol: NF-kappaB and Akt as targets, FASEB J. 20(2006) 527–529.

19] P. Thomas, Y. Pang, E.J. Filardo, J. Dong, Identity of an estrogen mem-brane receptor coupled to a G protein in human breast cancer cells,Endocrinology 146 (2005) 624–632.

20] T. Simoncini, A.R. Genazzani, Non-genomic actions of sex steroidhormones, Eur. J. Endocrinol. 148 (2003) 281–292.

21] E.J. Filardo, J.A. Quinn, A.R. Frackelton Jr., K.I. Bland, Estrogenaction via the G protein-coupled receptor, GPR30: stimulation of adeny-lyl cyclase and cAMP-mediated attenuation of the epidermal growthfactor receptor-to-MAPK signaling axis, Mol. Endocrinol. 16 (2002)70–84.

22] A. Brodie, Q. Lu, J. Nakamura, Aromatase in the normal breast andbreast cancer, J. Steroid Biochem. Mol. Biol. 61 (1997) 281–286.

23] S. Chen, F. Zhang, M.A. Sherman, I. Kijima, M. Cho, Y.C. Yuan, Y.Toma, Y. Osawa, D. Zhou, E.T. Eng, Structure-function studies of aro-matase and its inhibitors: a progress report, J. Steroid Biochem. Mol.Biol. 86 (2003) 231–237.

24] M. Maggiolini, A. Vivacqua, G. Fasanella, A.G. Recchia, D. Sisci,V. Pezzi, D. Montanaro, A.M. Musti, D. Picard, S. Ando, TheG protein-coupled receptor GPR30 mediates c-fos up-regulation by17beta-estradiol and phytoestrogens in breast cancer cells, J. Biol.Chem. 279 (2004) 27008–27016.

25] N.G. Coldham, M.J. Sauer, Identification, quantitation and biologicalactivity of phytoestrogens in a dietary supplement for breast enhance-ment, Food Chem. Toxicol. 39 (2001) 1211–1224.

26] S.R. Milligan, J.C. Kalita, V. Pocock, V. Van De Kauter, J.F. Stevens,M.L. Deinzer, H. Rong, D. De Keukeleire, The endocrine activities of8-prenylnaringenin and related hop (Humulus lupulus L.) flavonoids,

27] L. Pan, H. Becker, C. Gerhauser, Xanthohumol induces apoptosis incultured 40–16 human colon cancer cells by activation of the deathreceptor- and mitochondrial pathway, Mol. Nutr. Food Res. 49 (2005)837–843.