Maturitas 56 (2007) 173–183

The effects of soy extract on the uterus of castrated adult rats

Rejane Mosquette a, Manuel de Jesus Simoes b, Ismael Dale Cotrim Guerreiro da Silva f,Celina Tizuko Fujiyama Oshima d, Ricardo Martins Oliveira-Filho c,

Mauro Abi Haidar a, Ricardo Santos Simoes a,Edmund Chada Baracat a,e, Jose Maria Soares Junior a,∗

a Department of Gynecology, Federal University of Sao Paulo, Rua Sena Madureira 1245, 04021051 Sao Paulo, Brazilb Department of Morphology, Federal University of Sao Paulo, Rua Sena Madureira 1245, 04021051 Sao Paulo, Brazil

c Department of Pharmacology, University of Sao Paulo, Rua Sena Madureira 1245, 04021051 Sao Paulo, Brazild Department of Pathology, Federal University of Sao Paulo, Rua Sena Madureira 1245, 04021051 Sao Paulo, Brazil

e Department of Obstetrics and Gynecology, University of Sao Paulo, Rua Sena Madureira 1245, 04021051 Sao Paulo, Brazil

f Department of Oncology, Federal University of Sao Paulo, Rua Sena Madureira 1245, 04021051 Sao Paulo, Brazil

Received 12 January 2006; received in revised form 26 July 2006; accepted 26 July 2006

Abstract

Objective: The aim of this study was to evaluate the effects of different doses of a standardized soy extract on the uterus ofcastrated rats.Methods: Fifty-six adult castrated female Wistar rats were randomly divided into seven groups (eight animals in each)that received: GI—drug vehicle (propylene glycol); GII—soy extract 10 mg/kg per day; GIII—soy extract 50 mg/kg perday; GIV—soy extract 100 mg/kg per day; GV—soy extract 300 mg/kg per day; GVI—soy extract 600 mg/kg per day;GVII—conjugated equine estrogens (CEE) 200 �g/kg per day. After 21 days of treatment, all animals were sacrificed andfragments of the uterine horns were immediately removed, fixed in 10% formaldehyde and submitted to routine histologicaltechniques for morphometric study. The endometrial cell proliferation index was determined with the PCNA antibody PC-10and expressed as the percentuals of the PCNA-positive nuclei relative to the total countings. Other fragments were immediatelyfrozen in liquid nitrogen for RNA extraction and VEGF analysis using RT-PCR technique.Results: The minimal dose of soy extract that produced a significant increase of the morphometric parameters was 100 mg/kg(GIV). The maximum effects on endometrial and myometrial morphometry were detected in the groups treated with 300 and600 mg/kg of soy extract (groups V and VI) and CEE (GVII). The expression of PCNA in the endometrial epithelium and stromawas increased by treatment with 100–600 mg/kg per day of soy extract (groups IV–VI) or with CCE (group VII). Doses equalto or higher than 50 mg/kg of soy extract (groups III–VI) and CEE stimulated the expression of VEGF.

Conclusion: The treatment of adult castrated rats during 21 days with doses of 100 mg/kg per day or higher of soy extract maydetermine significant proliferation in the endometrium and myometrium.© 2006 Published by Elsevier Ireland Ltd.

Keywords: Soy extract; Isoflavones; Castrated rat; Uterus; VEGF; Morphology; Endometrium

∗ Corresponding author. Tel.: +55 11 50813685; fax: +55 11 50813685.E-mail address: jsoares415@aol.com (J.M. Soares Junior).

0378-5122/$ – see front matter © 2006 Published by Elsevier Ireland Ltd.doi:10.1016/j.maturitas.2006.07.011

1 aturita

1

tmewbbbtmet

JcA[dewbhr

a[iteitia[

stgetptp

am

1oitv(ttcptgmp

tpsg4t(iSafacf

atptlniiwt

osT

74 R. Mosquette et al. / M

. Introduction

Hormone therapy (HT) and unopposed estrogenherapy (ET) have been the most common therapies for

enopausal and genitourinary symptoms [1–3]. How-ver, these protocols have been respectively associatedith increased risk for breast and endometrial cancers,esides presenting a number of side effects, such asreast tenderness, thromboembolic events and uterineleeding [4–6]. These facts may be accounted for byhe low compliance seen with these hormonal treat-

ents. Accordingly, alternative strategies are in duevaluation, aiming the relieve of menopausal symp-oms, mainly the hot flashes [7,8].

Epidemiological data indicate that less than 25%apanese and 18% Chinese climacteric women [9,10]omplain of hot flushes, compared to 85% Northmerican women [11] and 75% European women

12]. Furthermore, Asian women have a lower inci-ence of estrogen-related cancers, cardiovascular dis-ase and osteoporosis when compared to Westernomen [13,14]. Among the many theories that haveeen proposed to explain these differences, alimentaryabits including isoflavones-rich diets have not beenuled out.

Isoflavones belong to a class of phytoestrogens thatre found in soybeans, lentils, chickpeas and red clover15]. Soybeans are a rich source of isoflavones whichnclude genistein, daidzein, and glycitein. The struc-ure of these molecules closely resembles that of manystrogenic and antiestrogenic compounds [16], includ-ng the endogenous estrogen 17�-estradiol and the syn-hetic, selective estrogen receptor modulator tamox-fen. Therefore, isoflavones may have estrogenic orntiestrogenic activities, depending on the target tissue17,18].

Isoflavones have a common phenolic structure thateems to be a prerequisite for binding to estrogen recep-ors (ERs). These so-called phytoestrogens, mainlyenistein, seem to have a higher binding affinity for thestrogen receptor � than for the � subtype; thereforeissue-specific effects can be displayed by these com-ounds, depending on the differential tissue distribu-ion of those ER subtypes [19]. Also, isoflavones have

otentially important non-hormonal effects [16,19].

Isoflavones may be antagonists of the ERs in uterinend breast tissues [20,21]. Alternatively, isoflavonesay combine with ERs, albeit with lower affinity than

e[

o

s 56 (2007) 173–183

7�-estradiol [22], thus having an estrogenic effectn bone and blood vessels [23]. In fact, some exper-mental studies in female rats showed that pure genis-ein (50–100 mg/kg) stimulated the uterine horn andaginal cell growth [21,24,25]. However, low doses<2.5 mg/kg per day) of this hormone did not haverophic activity on the rat uterus [25]. Also, genis-ein may act as an agonist on the MCF-7 mammaryancer cell line [26] and mammary tissue [27]. Inostmenopausal women, some studies suggested thathe treatment with isoflavones (15–20 mg/kg with aenistein content of about 70–75%) may decrease theenopausal symptoms without significant endometrial

roliferation for a short time [8,28].Morton et al. [29] determined the serum concen-

rations of isoflavones in a large group of Japaneseeople over 40 years of age as compared to serumamples from a United Kingdom population. The meanenistein concentration in the Japanese samples ranged93–502 nmol/L, as compared to 28–33 nmol/L forhe UK samples. The mean concentrations of daidzeinanother main isoflavone in soy) were 247–283 nmol/Ln Japanese as compared to 12–12.5 nmol/L in UK.imilar values were reported by Yamamoto et al. [30] ingroup of Japanese people with estimated intakes (food

requency questionnaire) for genistein of 31 mg/daynd for daidzein of 18 mg/day, resulting in serum con-entrations of 475 nmol/L for genistein and 120 nmol/Lor daidzein.

In Wistar rats, McClain et al. [31] showed thatn acute treatment with high doses of pure genis-ein (500 mg/kg per day) resulted in free genisteinlasma levels of about 550 nmol/L. This blood concen-ration is similar to that found in humans who consumearge amounts of soy isoflavones [29]. However, it isot known whether a short-term, high-dose isoflavonentake could induce endometrial growth. Being so, annvestigation on the effect of a short period of treatmentith isoflavones on the uterine tissue, mainly focusing

he effects of high doses, is worthwhile.The estrogen receptor is involved with the activation

f vascular endothelial growth factor (VEGF) expres-ion that is directly related with angiogenesis [32].herefore this factor may be used as a marker of prolif-

ration and of the estrogen effects on the endometrium33].

Although some studies confirm the health benefitsf soy isoflavones for menopause-related conditions

aturita

[atfasoc2

2

Umf

2

t(si6fe

((ba

2

(labwmtGaepo

3doeareatwdwiosi4e

2

gb(−(cUwca0cl

2

wHtde

R. Mosquette et al. / M

8,34,35], many conflicting results have been reportedbout the effects on postmenopausal symptoms andhe risks of this therapy. One explanation for theseacts is the administration of low amounts of genisteinnd daidzein soy derivatives. Therefore, the aim of thistudy was to compare the effects of increasing dosesf a standardized soy extract with a 42.6% isoflavonesontent on the uterus of castrated rats upon an acute,1-day treatment protocol.

. Material and methods

The Laboratory Animal Care Committee of Federalniversity of Sao Paulo approved this study; rats wereaintained in accordance with the Guiding Principles

or The Care and Use of Animals.

.1. Drugs

Soy isoflavones: A fermented soybean extract rich inhe major isoflavone aglycones, genistein and daidzeinZhonshan Road, Dalian, China), was used as theource of these isoflavones. This soy extract with 42.6%soflavones contained approximately 36% of genistein,2% of daidzein and 2% of glycitein (including the iso-orms of isoflavones). The proteins present in the soyxtract averaged 57.4%.

The preparation of conjugated equine estrogensCEE, Wyeth, Pennsylvania, USA) included estrone50%), equiline (25%) and dihydroequiline (15%),esides minor amounts of 17�-estradiol, equinelinend 17�-dihydroequilenine.

.2. Animals

Fifty-six adult (3-month old), female Wistar rats180–210 g body weight) were kept under a 14-hight:10-h dark cycle (lights on at 08:00 a.m.) at 23 ◦C,nd maintained on a standard diet free from any soy-ean compounds and tap water ad libitum. All animalsere subjected to bilateral ovariectomy (OVx). Oneonth later, the rats were randomly divided into seven

reatment groups: GI—vehicle (propylene glycol);II—soy extract 10 mg/kg per day (corresponding to

n isoflavones dosing of 4.3 mg/kg per day); GIII—soyxtract 50 mg/kg per day (21.3 mg of isoflavones/kger day); GIV—soy extract 100 mg/kg per day (42.6f isoflavones mg/kg per day); GV—soy extract

sI

t

s 56 (2007) 173–183 175

00 mg/kg per day (127.8 mg of isoflavones/kg peray); GVI—soy extract 600 mg/kg per day (255.6 mgf isoflavones/kg per day), and GVII—conjugatedquine estrogens (CEE), 200 �g/kg per day. Allnimals were given the drugs once a day by oraloute (gavage) during 21 consecutive days. At thend of the treatment period the rats were weighednd sacrificed by decapitation under deep anesthesia;runk blood was collected, and the resulting serumas separated and stored at −20 ◦C until hormoneeterminations. Fragments of the right uterine hornere carefully dissected out and immediately frozen

n liquid nitrogen for further RNA extraction. Samplesf the left horn were fixed in 10% formaldehydeolution for 24 h. Each fixed tissue was dehydratedn graded alcohol, embedded in paraffin, sectioned at�m-thick sections and stained with hematoxylin andosin (H.E.) for light microscopy analysis.

.3. Hormone analysis

The serum levels of 17�-estradiol (E2) and pro-esterone (P4) were determined. Upon collection, thelood samples were immediately centrifuged at 4 ◦C1500 × g) for 10 min and the supernatants frozen at20 ◦C for subsequent analysis by radioimmunoassay

RIA) using the double antibody technique in pre-oated tubes (ICN Biomedical Inc., Costa Mera, CA,SA) according the manufacturer’s directions. Assaysere done in duplicate. The samples were all pro-

essed on the same day. The detection limit of the E2nd P4 assays averaged approximately 0.1 pg/ml and.1 ng/ml, respectively. The maximum percentual ofross-reactivity with other steroids for E2 and P4 wasower than 0.01%.

.4. Morphometric analysis

Formaldehyde-fixed and paraffin-embedded tissuesere submitted to routine 4 �m-thick sectioning and.E. staining. Morphometric measurements [36] were

hen performed using digitalized images obtainedirectly from the light microscope via a videocam-ra and stored in magnetic medium. Afterwards, mea-

urements were automatically performed using themagelab-Softium® software (Sao Paulo, Brazil).

Ten separate measurements of total uterine, myome-rial and endometrial areas of each animal were made

1 aturita

bwtbbubattecAtemsto

2

ecgPclp

2

paaRipbwtv1mTAt

11CG

N1mCtrsT(r1sbtaB

2

mmaTs

3

3a

wefaG

76 R. Mosquette et al. / M

y tracing around the on-screen image of the sectionsith a cursor. The outer uterine perimeter (P1) and

he inner luminal surface (P2) were traced: the areaounded by the latter was subtracted from that boundedy the former, and the result was taken as the totalterine cross-sectional area. Similarly, the interfaceetween the myometrial and endometrial layers waslso traced; the inner luminal surface area was sub-racted from it, and this value was taken as the endome-rial area in order to determine the thickness of ratndometria; the endometrial thickness index (ETI) wasalculated using the formula: ETI = 2A/(P1 + P2) [37].

similar method was used to calculate the myome-rial area and thickness index (MTI). The average ofndometrial glands, vessels and eosinophils was deter-ined by counting all structures of every uterine cross-

ection. Two independent investigators measured theotal uterine and endometrial areas on two separateccasions.

.5. PCNA analysis

For determination of the endometrial cell prolif-ration index, the sections were immunohistochemi-ally stained with anti-proliferating cell nuclear anti-en (PCNA) antibody PC-10 (Dako, Denmark). TheCNA-positive nuclei (PCNA-labeling index) in 1000ells of the endometrium (superficial and gland epithe-ium as well as stroma) were counted and expressed asercentages of the total countings.

.6. VEGF analysis (RT-PCR)

For the semiquantitative reverse transcriptase-olymerase chain reaction (RT-PCR) analysis, equalmounts of total RNA from every group were poolednd reverse transcribed using the SuperscriptTM IINAse H reverse transcriptase (Invitrogen®) accord-

ng to the manufacturer’s instructions. The PCR VEGFrimers were as described (GeneBank®, access num-er NM 031836). Cytoplasmic �-actin (ACTB) geneas used to normalize the reactions. For PCR, 2 �l of

he reverse transcribed cDNA were amplified in a finalolume of 100 �l by PCR under standard conditions:.5 mM MgCl , 125 �M dNTP and 2.5 U Taq poly-

2erase using specific primers at 10 �M concentration.he expected product size, termocycling steps (Genemp PCR System 9700, PE Applied Biosystems, Fos-

er City, CA, USA) and primers were respectively:

e(gg

s 56 (2007) 173–183

Cytoplasmic beta actin: GeneBank V01217,89 bp; 40 cycles (94 ◦C, 3 min/94 ◦C, 1 min/55 ◦C,:50 min/72 ◦C, 1:00 min/72 ◦C 7:00 min) S: 5′-TTCTG GGT AAG TTG TAG TC-3′; AS: 5′-AGC ACTTG TTG GCA TAG AG-3′.Vascular endothelial growth factor: GeneBank

M 031836, 568 bp; 32 cycles (94 ◦C, 5 min/94 ◦C,min/58 ◦C, 1:00 min/72 ◦C, 1:00 min/72 ◦C, 7:00in) S: 5′-tgc acc cac gac aga agg aga-3′; AS: 5′-TCACG CCT TGG CTT GTC ACA T-3′. Negative con-

rols were also included in order to avoid false-positiveesults. Following amplification, 10 �l of the cDNA inolution were mixed with BlueJuice (Invitrogen Lifeechnologies, California) and applied to a 2% agaroseInvitrogen) gel stained with ethidium bromide (Invit-ogen). The used molecular weight standard was the00 bp DNA ladder (Invitrogen). After electrophore-is, the intensity of RT-PCR products was visualizedy an ultraviolet Tran® illuminator and semiquantita-ively analyzed by using an image analysis computer-ssisted program (ImageLab Softium®, Sao Paulo,razil).

.7. Statistical analysis

Data from morphometric analyses and serum hor-ones (estrogen and progesterone) are given asean ± standard deviation (S.D.); the differences were

nalyzed using one-way ANOVA followed by theurkey’s post hoc test, with the level of significanceet at P < 0.05.

. Results

.1. Weight (body and uterus) and hormonenalysis

Table 1 shows the values of body and uterineeights as well as the hormone analyses (blood lev-

ls of estradiol and progesterone). No significant dif-erences regarding body weights among the groupsfter the treatments. However, the uterine weights ofIV–VII rats were higher than those of GI–III. As

xpected, the circulating levels of estradiol in GVII32.1 ± 6.3 pg/ml) were higher than those of the otherroups. Conversely, the levels of progesterone in allroups were similar.

R. Mosquette et al. / Maturitas 56 (2007) 173–183 177

Table 1Ponderal data (body and uterus weight) and hormone blood levels (mean ± S.D.) upon completion of the treatment period

Treatment groups

GI GII GIII GIV GV GVI GVII

Body weight (g) 238.5 ± 24.4 227.3 ± 33.9 209.7 ± 20.2 214.2 ± 22.7 202.0 ± 13.8 215.0 ± 18.2 212.7 ± 3.4Uterine weight (g) 2.7 ± 0.5a 3.2 ± 0.8b 4.2 ± 0.8c 6.2 ± 0.7c 7.7 ± 0.9 6.2 ± 0.2c 7.5 ± 0.1Estradiol (pg/ml) 2.5 ± 0.9 3.1 ± 1.1 2.8 ± 1.3 3.5 ± 0.8 3.4 ± 1.5 3.2 ± 1.6 32.1 ± 6.3d

Progesterone (ng/ml) 1.3 ± 0.3 1.7 ± 0.9 1.9 ± 0.8 1.9 ± 0.6 1.7 ± 0.7 1.5 ± 0.6 2.1 ± 1.1

Groups of ooforectomized rats (n = 8 throughout) were treated during 21 days as follows: GI, drug vehicle (propylene glycol); GII, soy extract10 mg/kg per day; GIII, soy extract 50 mg/kg per day; GIV, soy extract 100 mg/kg per day; GV, soy extract 300 mg/kg per day; GVI, soy extract600 mg/kg per day; GVII, conjugated equine estrogens (CEE) 200 �g/kg per day.

a P < 0.01 compared to GIV, GV, GVII.b P < 0.01 compared to GV, and GVII.

ey test)

3

mawtwmt(luLwtafedaGGpem

bwctw

Gtgwl

3

aaaptg

3a

ite(

3g

c P < 0.01 compared to GV, and GVII.d P < 0.001 compared to other groups (ANOVA and post hoc Turk

.2. Histological analysis

The endometrium of ovariectomized control ani-als (GI) was dense and thin. Glands were small

nd closed (Fig. 1A, GI). The luminal epitheliumas unorganized, without columnar morphology and

he nuclei randomly positioned in the cells. Thereas little, if any, interstitial space between the stro-al cells. Treatment with estrogens (GVII) increased

he size of endometrial epithelial and stromal cellsFig. 1A, GVII). The endometrial glands had largeumina, many of which contained secretory prod-cts. Interstitial space appeared between stromal cells.uminal epithelial cells presented a columnar aspect,ith basally located nuclei. The low-dosing isoflavone

reatment (GII and GIII, see Fig. 1A, GII, GIII) did notffect the uterine morphology, being the histologicaleatures very similar to those of GI. In addition, thendometrium in GIV–VI (Fig. 1A GIV–VI) was bettereveloped than in GII and GIII. In GIV it was observed

slightly proliferative endometrium, and GV andVI showed a markedly proliferative endometrium.IV–VI showed many endometrial glands, with sim-le cylindrical epithelium bearing a large amount ofosinophils. The endometrial enlargement of GV wasore similar to GVII than to GVI or other groups.The myometrium of GI and GII showed a low num-

er of collagen fibers among the uterine myocytes,

hich had elliptic, heterochromatic and small nuclei,

ompared to other groups (Fig. 1B, GI, GII). In addi-ion, GIII, GIV and GVI presented a large cytoplasmith euchromatic and large nuclei (Fig. 1B, GIII, GIV,

iTb

.

VI). GV and GVII showed the greatest increase inhe number of myocytes and a large amount of colla-en fibers in myometrium (Fig. 1B, GV, GVII), whichas noticeably thicker and infiltrated with numerous

eukocytes, mainly eosinophils.

.3. Morphometric analysis

The morphometric analysis is shown in Table 2. GVnd GVII presented the largest values of endometrialrea and thickness index, number of endometrial glandsnd eosinophils, as well as of myometrial area. Thesearameters in GVI and GIV were even greater thanhose of GI, GII and GIII. The number of vessels ofroups I–III was fewer than that of groups IV–VII.

.4. Proliferating cell nuclear antigen (PCNA)nalysis

The expression of PCNA in the rat endometrium wasncreased in the groups IV–VII (Fig. 2) compared tohat found in the groups I–III regarding the endometrialpithelium (superficial and glandular) and the stromaTable 3).

.5. RT-PCR analysis (Vascular endothelialrowth factor—VEGF)

The expression of VEGF in the rat uterus wasncreased in GIII, GIV, GV, GVI and GVII (Fig. 3).here was no differences regarding the cDNA of VEGFetween the groups GI and GII.

178 R. Mosquette et al. / Maturitas 56 (2007) 173–183

Fig. 1. (A) Histological analysis of the endometrium of castrated rats (HE, 180×); (B) histological analysis of the myometrium of castratedrats (HE, 120×). GI—vehicle (propylene glycol); GII—soy extract 10 mg/kg per day; GIII—soy extract 50 mg/kg per day; GIV—soy extract100 mg/kg per day; GV—soy extract 300 mg/kg per day; GVI—soy extract 600 mg/kg per day; GVII—conjugated equine estrogens (CEE)200 �g/kg per day. GE: glandular epithelium; LE: luminal epithelium; ST: stroma.

R. Mosquette et al. / Maturitas 56 (2007) 173–183 179

Table 2Morphometric analyses of endometrium and myometrium in histological samples from control, soy extract- or estrogen-treated groups ofooforectomized rats at the end of the treatment period

Treatment groups

GI GII GIII GIV GV GVI GVII

EndometriumArea × 103 (�m2) 7.4 ± 1.6 7.7 ± 1.6 9.5 ± 1.4 14.9 ± 6.2a 31.9 ± 7.1b 20.4 ± 8.8a 32.0 ± 13.8b

Endometrial thickness index × 102 12.7 ± 0.3 13.1 ± 0.2 17.9 ± 0.2 36.3 ± 1.8c 75.7 ± 1.2d 42.6 ± 1.7c 75.8 ± 2.4d

Epithelial thickness index × 102 2.2 ± 0.2 2.4 ± 0.7 2.6 ± 0.3 6.6 ± 2.1e 16.1 ± 4.9f 10.3 ± 4.3e 16.8 ± 2.2f

Endometrial glands/area (n/�m2) 12.7 ± 8.8 13.1 ± 8.1 17.9 ± 9.5 36.3 ± 27.8g 75.7 ± 23.6h 52.6 ± 20.8g 75.8 ± 21.9h

Endometrial gland area (�m2) 0.5 ± 0.1 0.5 ± 0.2 0.6 ± 0.1 1.1 ± 0.4i 2.5 ± 0.8 j 1.2 ± 0.5i 2.5 ± 0.5j

MyometriumArea × 103 (�m2) 5.5 ± 0.5 5.4 ± 1.4 5.8 ± 0.6 11.3 ± 0.4k 27.6 ± 1.0L 18.5 ± 2.0k 27.5 ± 7.9L

Number of vessels/area (n/�m2) 6.5 ± 0.6 3.9 ± 0.6 5.7 ± 0.9 8.7 ± 1.5m 10.2 ± 3.6m 9.5 ± 1.5m 12.9 ± 3.9m

Data are given as mean ± S.D. Groups of ooforectomized rats (n = 8 throughout) were treated during 21 days as described in Table 1.a,c,e,g,i,k,mP < 0.05 compared to GI, GII, GIII, GV and GVII; b,d,f,h,j,LP < 0.01 compared to GI, GII and GIII and P < 0.05 compared to GIVand GVI; mP < 0.05 compared to GI, GII and GIII (ANOVA and post hoc Turkey test).

Table 3Proliferating cell nuclear antigen (PCNA) analysis in the endometrial cells after the treatment

Treatment groups

GI GII GIII GIV GV GVI GVII

Luminal epithelium 4.2 ± 2.8 5.7 ± 3.3 4.0 ± 2.1 59.9 ± 39.7a 74.9 ± 23.7a 65.8 ± 30.5a 79.3 ± 26.2a

Glandular epithelium 7.1 ± 6.6 8.1 ± 6.6 5.3 ± 4.9 74.6 ± 23.4b 95.7 ± 24.9b 82.3 ± 28.6b 947.1 ± 29.3b

S 15

T y 1000r .01 com

4

tPctomka

rtidwuiw1

tt2os9

gaaouobds

troma 0.2 ± 0.3 1.1 ± 0.7 1.9 ± 2.9

he values are given as percentages of PCNA-positive nuclei for everats (n = 8 throughout) were treated as described in Table 1. a,b,cP < 0

. Discussion

The potential adverse or beneficial impact of phy-oestrogens on human health is under much debate.hytoestrogens, particularly the isoflavones, are dis-ussed as a potential treatment for menopausal symp-oms [8,28,38]. However, the real effect of isoflavonesn the endometrium is uncertain. Although experi-ental data obtained in animals may contribute to the

nowledge on this issue, the effects produced in womenre largely unknown.

In our conditions, low doses of soy extract (cor-esponding to low doses of the isoflavones containedherein) did not show any effect on ovariectomy-nduced uterine atrophy. Accordingly, Diel et al. [39]emonstrated that the oral administration of genisteineakly stimulated uterine weight in doses from 25

p to 100 mg/kg per day and produced a significantncrease in the uterine and the vaginal epithelial heighthen OVx animals were treated during 3 days with00 mg/kg genistein per day. Taking into account that

ietr

.1 ± 7.1c 20.4 ± 6.8c 26.7 ± 8.9c 36.2 ± 9.7c

endometrial cells counted (mean ± S.D.). Groups of ooforectomizedpared to GI, GII, GIII and GIV (ANOVA and post hoc Turkey test).

he soy extract used herein contained ca. 36% genis-ein out of the total isoflavones content (see Section), these data are in accordance with our own, sinceur GVI rats (treated with 600 mg/kg per day of theoy extract) were actually been treated with just about2 mg/kg per day of genistein.

Overall, even though these results suggest thatenistein can behave as a weak estrogen receptorgonist in the rat uterus, literature data reported thatnother phytoestrogen, daidzein, has important actionsn rat uterus [40]. In addition, equol, which is nat-rally obtained by the gut microbial transformationf daidzein in rats and humans [41,42], is known toe at least 10–100- and 10-fold more estrogenic thanaidzein and genistein, respectively, in the Siberianturgeon [43].

Vascular endothelial growth factor (VEGF) is an

mportant angiogenic molecule that is regulated bystrogens and is considered to be a prognostic fac-or for many tumors, including those of endocrine-esponsive tissues such as the uterus. VEGF is also

180 R. Mosquette et al. / Maturitas 56 (2007) 173–183

F metrium1 xtract 1e E) 200S

taopcptpic

n[

udb

ig. 2. Proliferating cell nuclear antigen (PCNA) analysis of the endo0 mg/kg per day; GIII—soy extract 50 mg/kg per day; GIV—soy extract 600 mg/kg per day; GVII—conjugated equine estrogens (CET: stroma.

he most potent known regulator of capillary perme-bility and growth [44]. Thus, the biological activityf this growth factor in vivo could play a role in theroliferation of uterine cells by causing changes inapillary permeability to water, small molecules, androteins, that have been reported previously in response

o estrogens. Increased angiogenesis and changes inermeability enhance tissue perfusion and lead to anncrease in nutrient supply and in the local tissue con-entration of plasma proteins. This facts are recog-

(rie

of castrated rats. GI—vehicle (propylene glycol); GII—soy extract00 mg/kg per day; GV—soy extract 300 mg/kg per day; GVI—soy�g/kg per day. GE: glandular epithelium; LE: luminal epithelium;

izedly essential for tumor expansion and metastasis45,46].

In rats, it has been shown that estrogens increaseterine VEGF expression after ovariectomy [36]. Ourata showed that the effects of soy extract resem-le those of estrogens on the rat uterus proliferation

morphology, VEGF and PCNA). Two subtypes of theat, mouse and human estrogen receptor (ER) weredentified, namely ER � and ER �, that have differ-nt tissue distributions and transcriptional regulatory

R. Mosquette et al. / Maturita

Fig. 3. Result of agarose gel electrophoresis of RT-PCR products.(A) Detection of vascular endothelial growth factor (VEGF) RNAmexpression in the uterus of the several studied groups. VEGF expres-sions in uterine tissue (RT-PCR assay). M: marker; GI—vehicle(propylene glycol); GII—soy extract 10 mg/kg per day; GIII—soyextract 50 mg/kg per day; GIV—soy extract 100 mg/kg per day;GV—soy extract 300 mg/kg per day; GVI—soy extract 600 mg/kgper day; GVII—conjugated equine estrogens (CEE) 200 �g/kg perday. The marker has eleven lanes from 80 to 1031 bp. The 600, 500,450 and 370 bp products of VEGF. The internal control of �-actinproduced a 189 bp product. (B) Graphic representation of the amountof DNAc of the VEGF gene after subtraction of beta actin DNAcafter 32 RT-PCR cycles in the studied groups. *p < 0.01 comparedtc

eetwa

sbtsrbeH

eiIdd[e

hoiam0ivpcsotdoefcwa

5

em

A

C

R

o other groups, **p < 0.05 compared to GIII, GIV, GVI and p < 0.01ompared to GI and GII.

ffects on a wide number of target genes [47]; ER � isxpressed most abundantly in the female reproductiveract (uterus, vagina, ovaries) and mammary glands,hereas ER � is expressed mostly in prostate gland

nd ovaries.Kuiper et al. [16] showed that some phytoestrogens

uch as genistein competed more with estradiol forinding to ER � than to ER � in vitro. Also, genis-ein stimulated the transcriptional activity of both ERubtypes at 1–10 nM concentrations. There is a distinct

anking of the estrogenic potency of phytoestrogens foroth ER subtypes in the transactivation assay; that is,stradiol > genistein > daidzein for ER � and ER � [15].owever, even though many isoflavone effects may be

s 56 (2007) 173–183 181

strogenic in nature, there is evidence of the partic-pation of other mechanisms, such as topoisomeraseI activation [48], platelet-activating factor- and epi-ermal growth factor-induced expression of c-fos [49],iacylglycerol synthesis and tyrosine kinase activation50]. These effects may explain, at least in part, thendometrial proliferation obtained in our study.

Short-term clinical trials have shown that soy extractave a preventive effect on hot flushes [8,28]. On thether hand, a double-blind, placebo-controlled trialnvolving 376 women and lasting 5 years showed that

low-dose daily intake of phytoestrogens (approxi-ately 1.06 mg/kg genistein; 1.06 mg/kg diadzein and

.38 mg/kg glycitein per day) was associated with anncreased incidence of endometrial hyperplasia (3.8%ersus 0%) [51]. In practice, the risk–benefit ratio ofhytoestrogens has not been adequately assessed, espe-ially considering that an acute high-dose intake ofoy extract can result in peak levels of genistein andther phytoestrogens much higher than those reportedo have only moderate effects on the female repro-uctive tract. In addition, our study showed that theral consumption of soy extract may proliferate thendometrium, even in a short-course regimen. There-ore, before more solid clinical data are obtained, theonsumption of high doses of soy phytoestrogens byomen might be viewed as potentially harmful as far

s the endometrium proliferation is concerned.

. Conclusion

Our data are indicative that orally administered soyxtract, most presumably due to its isoflavones content,ay proliferate the uterus of castrated rats.

cknowledgements

Financial support was provided by CNPq, 041/2003APES (Brasılia-Brazil), and FAPESP 2003/13407-0.

eferences

[1] Judd HL. Estrogen replacement therapy: indications and com-plications. Ann Intern Med 1983;98:195–205.

[2] Erlite Y. Association of waking episodes with menopausal hotflushes. JAMA 1981;245:1741–4.

1 aturita

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

82 R. Mosquette et al. / M

[3] Semmens JP, Wagner G. Estrogen reproduction and vagi-nal function in postmenopausal women. JAMA 1982;248:445–8.

[4] Woodruf JD, Pickard JH, The Menopause Study Group. Inci-dence of endometrial hyperplasia in postmenopausal womentaking conjugated estrogens (Premarin) with medroxyproes-terone acetate or conjugated estrogens alone. Am J ObstetGynecol 1994;170:1213.

[5] Collaborative Group on Hormonal Factors in breast cancer.Breast cancer and hormone replacement therapy: collaborativereanalysis of data from 51 epidemiological studies of 52 705women with breast cancer and 108 411 women without breastcancer. Lancet 1997;350:1047.

[6] WHI-Writing Group for the WHI (Women’s Health Initia-tive) Investigators. Risk and benefits of estrogen plus progestinin healthy postmenopausal women: principal results from theWomen’s Health Iniciative randomized controlled trial. JAMA2002;288:321–33.

[7] Nachtigall LE. Isoflavones in the management of menopause. JBr Menopause Soc 2001;S1:8–11.

[8] Han KK, Soares Jr JM, Haidar MA, de Lima GR, Baracat EC.Benefits of soy isoflavone therapeutic regimen on menopausalsymptoms. Obstet Gynecol 2002;99:389–94.

[9] Boulet M, Oddens B, Lehert P, Vemer H, Visser A. Climactericand menopause in seven south-east Asian countries. Maturitas1994;19:157–76.

10] Lock M. Encounters with aging: mythologies of menopause inJapan and North America. Berkeley, CA: University of Califor-nia Press; 1993.

11] Notelovitz M. Estrogen replacement therapy indications, con-traindications and agent selection. Am J Obstet Gynecol1989;167:8–17.

12] Rekers H. Mastering the menopause. In: Burger H, Boulet M,editors. A portrait of the menopause. Park Bridge, New Jersey:The Parthenon Publishing Group; 1991. p. 23–43.

13] Adlercreutz H. Phytoestrogens: epidemiology and a pos-sible role in cancer protection. Environ Health Perspect1995;103:103–12.

14] Grady D, Rubin S, Petitti D. Hormone therapy to prevent diseaseand prolong life in postmenopausal women. Ann Intern Med1992;117:1016–37.

15] Vincent A, Fitzpatrick LA. Soy isoflavones: are they useful inmenopause? Mayo Clin Proc 2000;75:1174–84.

16] Kuiper GG, Lemmen JG, Carlsson B, et al. Interaction of estro-genic chemicals and phytoestrogens with estrogen receptorbeta. Endocrinology 1998;130:4252–63.

17] Cline JM, Paschold JC, Anthony MS, Obasanjo IO, Adams MR.Effects of hormonal therapies and dietary soy phytoestrogenson vaginal cytology in surgically postmenopausal macaques.Fertil Steril 1996;65(5):1031–5.

18] Erlandsson MC, Islander U, Moverare S, Ohlsson C, CarlstenH. Estrogenic agonism and antagonism of the soy isoflavone

genistein in uterus, bone and lymphopoiesis in mice. APMIS2005;113(5):317–23.

19] Kuiper GG, Carlsson B, Grandien K, et al. Comparison of the ligand binding specificity and transcript tissue distribution of estro-gen receptors alpha and beta. Endocrinology 1997;138:863–70.

[

s 56 (2007) 173–183

20] Makela S, Davis VL, Tally WC, et al. Dietary estrogens actthrough estrogen receptor-mediated processes and show noantiestrogenicity in cultured breast cancer cells. Environ HealthPerspect 1994;102:572–8.

21] Santell RC, Chang YC, Nair MG, Helferich WG. Dietarygenistein exerts estrogenic effects upon the uterus, mam-mary gland and the hypothalamic/pituitary axis in rats. J Nutr1997;127:263–9.

22] Mikisicek R. 1994 Interaction of naturally occurring nons-teroidal estrogens with expressed recombinant human estrogenreceptor. J Steroid Biochem Mol Biol 1997;49:153–60.

23] Schonherr E, Kinsella M, Wight T. Genistein selectively inhibitsplatelet derived growth factor-stimulated versican biosynthesisin monkey arterial smooth muscle cells. Arch Biochem Biophys1997;339:353–61.

24] Diel P, Smolnikar K, Schulz T, Laudenbach-Leschowski U,Michna H, Vollmer G. Phytoestrogens and carcinogenesis—differential effects of genistein in experimental modelsof normal and malignant rat endometrium. Hum Reprod2001;16:997–1006.

25] Makela SI, Pylkkanen LH, Santti RS, Adlercreutz H. Dietarysoybean may be antiestrogenic in male mice. J Nutr1995;125:437–45.

26] Hsieh CY, Santell RC, Haslam SZ, Helferich WG. Estrogeniceffects of genistein on the growth of estrogen receptor-positivehuman breast cancer (MCF-7) cells in vitro and in vivo. CancerRes 1998;58:3833–8.

27] Allred CD, Allred KF, Ju YH, et al. Dietary genistein resultsin larger MNU-induced, estrogen-dependent mammary tumourfollowing ovariectomy of Sprague–Dawley rats. Carcinogene-sis 2004;25:211–28.

28] Kaari C, Haidar MA, Soares Jr JM, et al. Randomizedclinical trial comparing conjugated equine estrogens andisoflavones in postmenopausal women: a pilot study. Maturi-tas 2006;53(1):49–58.

29] Morton MS, Arisaka O, Miyake N, Morgan LD, EvansBA. Phytoestrogen concentrations in serum from Japanesemen and women over forty years of age. J Nutr 2002;132:3168–71.

30] Yamamoto S, Sobue T, Sasaki S, et al. Validity and repro-ducibility of a self-administered food frequency questionnaireto assess isoflavone intake in a Japanese population in compar-ison with dietary records in blood and urine isoflavones. J Nutr2001;131:2741–7.

31] McClain RM, Wolzb E, Davidovichc A, Pfannkuchd F,Edwardsb JA, Bauschb J. Acute, subchronic and chronicsafety studies with genistein in rats. Food Chem Toxicol2006;44(1):56–80.

32] Buteau-Lozano H, Ancelin M, Lardeux B. Transcriptionalregulation of vascular endothelial growth factor by estra-diol and tamoxifen in breast cancer cells: a complex inter-play between estrogen receptors alpha and beta. Cancer Res

2002;62:4977–84.

33] Koos RD, Kazi AA, Roberson MS, Jones JM. New insight intothe transcriptional regulation of vascular endothelial growth fac-tor expression in the endometrium by estrogen and relaxin. AnnN Y Acad Sci 2005;1041:233–47.

aturita

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

R. Mosquette et al. / M

34] Brzezinski A, Adlercreutz H, Shaoul R, Rosler A, Shmueli A,Tanus Schenker JG. Short-term effects of phytoestrogen-richdiet on postmenopausal women. Menopause 1997;4:89–94.

35] Faure ED, Chantre P, Mares P. Effects of a standardized soyextract on hot flushes: a multicenter, double-blind, randomized,placebo-controlled study. Menopause 2002;9:324–9.

36] Andrade PM, Silva ID, Borra RC, Lima GR, Baracat EC. Estro-gen and selective estrogen receptor modulator regulation ofinsulin-like growth factor binding protein 5 in the rat uterus.Gynecol Endocrinol 2002;16:265–70.

37] Rossi AG, Soares Jr JM, Motta EL, et al. Metoclopramide-induced hyperprolactinemia affects mouse endometrial mor-phology. Gynecol Obstet Invest 2002;54:185–90.

38] Petri Nahas E, Nahas Neto J, De Luca L, Traiman P, PontesA, Dalben I. Benefits of soy germ isoflavones in post-menopausal women with contraindication for conventional hor-mone replacement therapy. Maturitas 2004;48:372–80.

39] Diel P, Smolnikar K, Schulz T, Laudenbach-Leschowski U,Michna H, Vollmer G. Phytoestrogens and carcinogenesis-differential effects of genistein in experimental models of nor-mal and malignant rat endometrium. Hum Reprod 2001;16:997–1006.

40] Caldarelli A, Diel P, Vollmer G. Effect of phytoestrogens ongene expression of carbonic anhydrase II in rat uterus and liver.J Steroid Biochem Mol Biol 2005;97:251–6.

41] Lephart ED, Setchell KD, Handa RJ, Lund TD. Behavioraleffects of endocrine-disrupting substances: phytoestrogens.ILAR J 2004;45:443–54.

42] Axelson M, Sjovall J, Gustafsson BE, Setchell KD. Soya—adietary source of the non-steroidal oestrogen equol in man andanimals. J Endocrinol 1984;102:49–56.

43] Pelissero C, Bennetau B, Babin P, Le Menn F, DunoguesJ. The estrogenic activity of certain phytoestrogens in the

[

s 56 (2007) 173–183 183

Siberian sturgeon Acipenser baeri. J Steroid Biochem Mol Biol1991;38:293–9.

44] Hyder SM, Nawaz Z, Chiappetta C, Stangel GM.Endocrinologh identification of functional estrogen responseelements in the gene coding for the potent angiogenic factorvascular endothelial growth factor. Cancer Res 2000;60:3183–90.

45] Dvorak HF, Brown LF, Detmar M, Dvorak AM. Vascu-lar permeability factor/vascular endothelial growth factor,microvascular hyperpermeability, and angiogenesis. Am JPathol 1995;146:1029–39.

46] Hyder SM, Stancel GM. Regulation of angiogenic growth fac-tors in the female reproductive tract by estrogens and progestins.Mol Endocrinol 1999;13:806–11.

47] Hanstein B, Djahansouzi S, Dall P, Beckmann MW, BenderHG. Insights into the molecular biology of the estrogen recep-tor define novel therapeutic targets for breast cancer. Eur JEndocrinol 2004;150:243–55.

48] Okura A, Arakawa H, Oka H, Yoshinari T, Monden Y. Effectof genistein on topoisomerase activity and on the growth of[VAL12] Ha-ras-transformed NIH 3T3 cells. Biochem BiophysRes Commun 1988;157:183–9.

49] Tripathi YB, Lim RW, Fernandez-Gallardo S, Kandala JC, Gun-taka RV, Shukla SD. Involvement of tyrosine kinase and proteinkinase C in platelet-activating-factor-induced c-fos gene expres-sion in A-431 cells. Biochem J 1992;286:527–33.

50] Akiyama T, Ishida J, Nakagawa S, et al. Genistein, a spe-cific inhibitor of tyrosine-specific protein kinases. J Biol Chem

1987;262:5592–5.

51] Unfer V, Casini ML, Costabile L, Mignosa M, Gerli S, Di RenzoGC. Endometrial effects of long-term treatment with phytoe-strogens: a randomized, double-blind, placebo-controlled study.Fertil Steril 2004;82:145–8.