Journal of Steroid Biochemistry & Molecular Biology xxx (2005) xxx–xxx

DT56a stimulates gender-specific human cultured bone cells in vitro

Dalia Somjena,∗, Sara Katzburga, M. Lieberherrd, David Hendelb, Israel Yolesc

a Institute of Endocrinology, Metabolism and Hypertension, Tel-Aviv Sourasky Medical Center and the Sackler Faculty of Medicine,Tel-Aviv University, 6 Weizman Street, Tel-Aviv 64239, Israel

b Department of Orthopaedic Surgery, Sharei-Zedek Medical Center, Jerusalem, Israelc Department of Obstetrics and Gynecology, Sheba Medical Center, Tel-Hashomer, Ramat-Gan, 52620 Israel

d INRA, Jouy-en-Josas, France

Received 15 March 2005; accepted 5 August 2005

Abstract

DT56a found to have SERM-like properties is used for the treatment of menopausal symptoms and osteoporosis. In vivo experimentsdemonstrated that DT56a displayed selective estrogenic activity; it stimulated creatine kinase (CK) specific activity in the skeletal tissues butnot on the uterus of ovariectomized rats. DT56a, when applied together with estradiol-17� (E2), completely inhibited the E2-stimulated CK,as demonstrated by other SERMs. DT56a stimulated bone formation in a rat model as measured by histological and histomorphometricalp nd relief ofm osteoblasts( ies as wella ale Obw al womenw timulatedc itsn -d its uniquen ti-resorptivea©

K

1

poTproh3

withs ofgen

com-s to

cific

uecep-,ute tomors,to

0d

7

arameters. In a clinical study, administration of DT56a (Femarelle) resulted in a considerable elevation of bone mineral density aenopausal symptoms. The aim of the present study was to analyze the effects of DT56a in vitro on human-derived bone cultured

Ob), by measuring its effects, at different concentrations, on DNA synthesis, CK and alkaline phosphatase (ALP) specific activits changes in intracellular [Ca2+]i concentrations. DT56a stimulated CK and DNA synthesis in both pre- and post-menopausal femith maximal effect at 100 ng/ml for both age groups. In addition, DT56a stimulated ALP in Ob from both pre- and post-menopausith maximal effect at lower dose of 50 ng/ml, with higher response of pre-menopausal cells. Raloxifene (Ral) inhibited all DT56a-shanges in Ob from both age groups. DT56a, when given together with E2, completely antagonized E2-stimulated effects demonstratingature as a phyto-SERM. DT56a also, dose dependency, stimulated the intracellular levels of [Ca2+]i with maximal effect at 10 ng/ml. Maleerived Ob did not respond to DT56a in any parameter. In summary, DT56a stimulated sex-specifically female-derived Ob, indicatingature compared to the compounds currently used for postmenopausal osteoporosis by being bone-forming and not only an angent.2005 Elsevier Ltd. All rights reserved.

eywords: DT56a; Estradiol; Raloxifene; Creatine kinase; Human bone cells

. Introduction

Estrogen is well known for its beneficial effect on osteo-orosis[1,2]. Of the 10 million Americans estimated to havesteoporosis, 8 million are women and 2 million are men.hirty-four million Americans have low bone mass, whichuts them at increased risk of developing osteoporosis andelated fractures. One in two women and one in four menver age 50 will have an osteoporosis-related fracture iner/his remaining lifetime. Osteoporosis is the cause of some00,000 hip fractures each year (www.nof.org). Osteoporosis

∗ Corresponding author. Tel.: +972 3 6973306; fax: +972 3 6974473.E-mail address: dalias@tasmc.health.gov.il (D. Somjen).

is characterized by reduction in bone mineral density,the result of fracture after minimal trauma. The effectestrogen on a tissue is initiated by its binding to estroreceptor (ER) in the responsive cell. The estrogen-ERplex is then translocated into the nucleus, where it bindthe DNA and modulates the rate of transcription of spegenes. Two ERs have been identified, ER� and ER�, whichdiffer in their structure and tissue distribution[3] and theirbiological effect[4]. The two key factors that control tissselectivity for the estrogen are the structure of its retor(s) and its interaction with co-regulators[5–6]. Howeverestrogen can also have non-desired effects that contribthe development and growth of estrogen-dependent tusuch as breast cancer[7]. These ominous side effects led

960-0760/$ – see front matter © 2005 Elsevier Ltd. All rights reserved.oi:10.1016/j.jsbmb.2005.08.002

SBMB-2513; No. of Pages

2 D. Somjen et al. / Journal of Steroid Biochemistry & Molecular Biology xxx (2005) xxx–xxx

extensive research aimed at finding compounds with benefi-cial estrogenic effects on selected sites, such as the bones[8],and the cardiovascular system[9] without the harmful sideeffects.

Selective estrogen receptor modulators (SERMs) exhibita pharmacological profile characterized by estrogen agonistactivity in some tissues with estrogen antagonist activity inother tissues[10,11]. Raloxifene functions as an estrogenagonist in skeletal tissues and in the cardiovascular systembut not in the uterus. It functions also as an antagonist in thepresence of estrogen in skeletal and vascular tissues and inthe uterus as well as breast cancer cells in vitro[12,13].

We used the induction of brain type isozyme of creatinekinase specific activity (CK) as a response marker of estro-genic activity[14], which is stimulated by estrogens both invivo and in vitro in tissues and in cells containing active ER(s)[15–17].

DT56a (Femarelle, Se-cure Pharmaceuticals, Dalton,Israel) is a unique enzymatic isolate of soybeans. Femarellehas been shown to increase bone mineral density in post-menopausal women[18] and to relieve vasomotor symptomswith no effect on sex hormone levels or endometrial thick-ness[19]. Its properties as a SERM were shown in previousstudies: Rats were fed for 4 days with estradiol-17� (E2) orDT56a. Both compounds increased CK activity in the dif-ferent skeletal[20] and vascular tissues[21]. On the otherhs es d,p ingt

uslyw lart riec-t he-nS lox-i rats

T56ai dif-f ausaw ctiv-i eptorr nd invt -t h isa ee cep-t ings ntra-t sw te[ an-n lum

[25].The genomic and non-genomic actions of DT56a werecompared to the known membrane and genomic effects of E2.

2. Materials and methods

2.1. Reagents

All reagents were of analytical grade. Chemicals werepurchased from Sigma (St. Louis, MO). Raloxifene wasextracted from Evista® tablets. DT56a was provided by Se-cure Pharmaceuticals, as a powder and was dissolved in 1%ethanol in saline.

2.2. Cell cultures

Human bone cells from pre- and post-menopausal womenor men, with no known health problems and no medicaltreatments, were prepared from bone explants, by a non-enzymatic method as described previously[16]. Samples ofthe trabecular surface of the iliac crest or long bones werecut into 1 mm3 pieces and repeatedly washed with phosphatebuffered saline to remove blood component: The explantswere incubated in DMEM medium without calcium (to avoidfibroblastic growth) containing 10% fetal calf serum (FCS)and antibiotics. First passage cells were seeded at a densityo freeD d at3 las-t

2o

pec-i wingt ibed[ uga-tE wasa iouslyd lued tan-d

2h

col-l timesi rifu-g ina lp1 eac-t ere

and, feeding with DT56a, unlike E2 did not result in CKtimulation in the uterus[20]. Raloxifene (Ral) blocked thtimulation of CK by either DT56a or E2 in all tissues testeointing towards a common mechanism of action, utiliz

he same type of receptor(s)[20,21].We also found that DT56a when given simultaneo

ith E2 inhibited E2’s activity in rat skeletal tissues, vascuissues and in the uterus, in both immature and ovaomized female rats (submitted for publication). This pomenon of mutual annihilation of action between E2 and aERM such as tamoxifen, tamoxifen methiodide or ra

fene, was previously described in prepubertal femalekeletal tissues and cells in culture[22,23].

In the present study, we assessed the effects of Dn-vitro on cultured human osteoblasts derived from threeerent groups: males, pre-menopausal and post-menopomen. The stimulation of creatine kinase specific a

ty was measured since it is a convenient estrogen recesponse marker and is induced by estrogens in vivo aitro [15–17]. We also measured cell proliferation by [3H]hymidine incorporation (DNA)[15] and also the stimulaion of alkaline phosphatase specific activity (ALP), whicmarker of osteoblast activity[24]. We also investigated tharly effects of DT56a, which recognized the estrogen re

or, on the interaction with non-genomic ER putative bindites measuring its effect on cytosolic free calcium conceion ([Ca2+]i ) in human female osteoblasts[25]. These actionere compared to E2 which was shown by us to stimula

Ca2+]i within 5 s by both influx through voltage-gated chels as well as mobilization from the endoplasmic reticu

l

f 3× 105 cells/35 mm tissue culture dish, in phenol red-MEM with 10% charcoal stripped FCS, and incubate7◦C in 5% CO2. The cells were characterized for osteob

ic features as described before[16].

.3. Creatine kinase extraction and assay in humansteoblasts

Cells were treated for 24 h with the various agents as sfied, scraped off and homogenized by freezing and thahree times in an extraction buffer, as previously descr15–17]. Supernatant extracts were obtained by centrifion of homogenates at 14,000× g for 5 min at 4◦C in anppendorf micro-centrifuge. Creatine kinase activityssayed by a coupled spectrophotometric assay, as prevescribed[15–17]. Protein was determined by Coomasie bye binding, using bovine serum albumin (BSA) as the sard.

.4. Alkaline phosphatase extraction and assay inuman osteoblasts

After treatment with the different agents, cells wereected and homogenized by freezing and thawing threen cold PBS and the soluble ALP was obtained by centation at 14,000× g [24]. Enzyme activity was assayedn ELISA reader, measuring the hydrolysis ofp-nitrophenyhosphate at 37◦C in a buffer containing 2 mM MgCl2 and00 mM 2-amino 2-methyl isopropanol at pH 10.3. The r

ion was stopped with 1 M NaOH and the extracts w

D. Somjen et al. / Journal of Steroid Biochemistry & Molecular Biology xxx (2005) xxx–xxx 3

analyzed at 410 nm. Enzyme specific activity was determinedas OD× 4.3/mg protein[24].

2.5. DNA synthesis in human osteoblasts

Cells were grown until sub confluence and then treatedwith various hormones or agents as indicated. Twenty-twohours following the exposure to these agents, [3H] thymidinewas added for 2 h. Cells were then treated with 10% ice-coldtrichloroacetic acid (TCA) for 5 min and washed twice with5% TCA cold with ethanol. The cellular layer was dissolvedin 0.3 ml of 0.3N NaOH, samples were aspirated and [3H]thymidine incorporation into DNA was determined[15–17].

2.6. Intracellular calcium concentration

Cells grown on cover slips were washed with Hanks’Hepes, pH 7.4 buffer, and loaded with 1�M Fura-2/AM for40 min in the same buffer at room temperature. The glasscover slips carrying the cells were inserted into a cuvettecontaining Hanks’ Hepes, pH 7.4 buffer. The cuvette wasplaced in a thermo stated (37◦C) Hitachi F-2000 spectroflu-orimeter. DT56a, E2 or vehicle alone were added directlyto the cuvette under continuous stirring. The Fura-2/AMfluorescence response to intracellular calcium concentration([Ca2+]i ) was calibrated from the ratio of 340–380 nm flu-o vels.T re-m hT

2

ntala nce( nifi-c

3

3h

ed toE Eo ,s andpc post-m

3f

aredt ith

Fig. 1. Stimulation by E2 (0.082–82 ng/ml which is equivalent to0.3–300 nM) or by DT56a (10–200 ng/ml) of creatine kinase (CK) spe-cific activity in human female bone cells. Results are mean± S.E.M.of triplicate cultures from 4 pre-menopausal and 4 post-menopausalwomen expressed as the ratio between CK specific activity inhormone-treated and control cells (* p < 0.05, ** p < 0.01). Basal activ-ity of creatine kinase: 0.020± 0.005�mol/min/mg (pre-menopausal),0.016± 0.004�mol/min/mg protein (post-menopausal).

0.3–300 nM E2, or 1–250 ng/ml DT56a. A similar increasein thymidine incorporation was found in both DT56a and E2groups, in both pre- and post-menopausal cells (Fig. 2).

3.3. The effects of DT56a on ALP specific activity inhuman female bone cells

To determine their responsiveness to DT56a compared E2,human female bone cell were incubated with 0.3–300 nME2, or 1–200 ng/ml DT56a. A similar increase in thymidine

F to0 -ro ausalw reateda -r l(

rescence values after subtraction of the background lehe values forRmax andRmin were calculated from measuents using 25�M digitonin and 4 mM EGTA and enougris base to raise the pH to 8.3 or higher[25].

.7. Statistical analysis

Differences between the mean values of experimend control group were evaluated by analysis of variaANOVA). p values less than 0.05 were considered sigant.

. Results

.1. The effects of DT56a on CK specific activity inuman female bone cells

To determine their responsiveness to DT56a compar2 human bone cells were incubated with 0.3–300 nM2,r 1–250 ng/ml DT56a. DT56s, similar to E2 treated cellshowed an increase in CK specific activity, in both pre-ost-menopausal cells (Fig. 1). The increase in the E2 treatedells was greater in the pre-menopausal compared toenopausal women.

.2. The effects of DT56a on DNA synthesis in humanemale bone cells

To determine their responsiveness to DT56a compo E2, human female bone cells were incubated w

ig. 2. Stimulation by E2 (0.082–82 ng/ml which is equivalent.30–300 nM) or by DT56a (10–200 ng/ml) of [3H] thymidine incorpoation into DNA in human femal bone cells. Results are mean± S.E.M.f triplicate cultures from 4 pre-menopausal and 4 post-menopomen expressed as the ratio between DNA synthesis in hormone-tnd control cells (* p < 0.05, ** p < 0.01). Basal [3H] thymidine incorpoation: 2550± 100�mol/min/mg (pre-menopausal), 1850± 75 dpm/welpost-menopausal).

4 D. Somjen et al. / Journal of Steroid Biochemistry & Molecular Biology xxx (2005) xxx–xxx

Fig. 3. Stimulation by E2 (0.082–82 ng/ml which is equivalent to0.3–300 nM) or by DT56a (10–200 ng/ml) of alkaline phosphatase (ALP)specific activity in human female bone cells. Results are mean± S.E.M.of triplicate culture from 4 pre-menopausal and 4 post-menopausal womenexpressed as the ratio between ALP specific activity in hormone-treated andcontrol cells (* p < 0.05,** p < 0.01). Basal activity of ALP: 0.20± 0.05 u/mg(pre-menopausal), 0.26± 0.04 u/mg protein (post-menopausal).

incorporation was found in both DT56a and E2 groups(Fig. 3). In both treatments, the maximal increase was higherin the pre-menopausal cells compared to post-menopausalcells.

3.4. The effects of DT56a on CK specific activity, DNAsynthesis and ALP specific activity in human male bonecells

To determine their responsiveness to DT56a comparedto E2 and to dihydrotestosterone (DHT), human male bonecells, were incubated with 03–300 nM E2, with 1–200 ng/mlDT56a or with 30–3000 nM DHT. An increase in all param-eters, was found only with DHT at optimum of 300 nMfor CK and of 3000 nM for ALP and DNA synthesis(Fig. 4).

3.5. The effects of raloxifene and DT56a on CK specificactivity, DNA synthesis and on ALP specific activitystimulated by E2 in human female bone cells

When human female bone cells were incubated with bothDT56a and raloxifene or E2 (Fig. 5a–c), DT56a activity inboth age groups was blocked by raloxifene. E2’s activity,was also blocked by DT56a (Fig. 5a–c) as measured by CKs inb

3c

ed toE igin,

Fig. 4. The effects of DT56a (0–200 ng/ml), E2 (0.082–82 ng/ml which isequivalent to 0.3–300 nM) or DHT (8.2–8200 ng/ml which is equivalentto 3–3000 nM) on CK, DNA and ALP in human male bone cells. Resultsare mean± S.E.M. (n = 4) expressed as the ratio between the activities inhormone-treated and control tissues (* p < 0.05, ** p < 0.01). Basal activityof creatine kinase: 0.050± 0.015�mol/min/mg, in DNA synthesis, basal3H thymidine incorporation: 3550± 250�mol/min/mg and basal activity ofALP 0.36± 0.08 u/mg protein.

were incubated with 0.01–0.10 nM E2, or with 0.01–10 ng/mlDT56a (Figs. 6a, b and 7). An increase in intracellular [Ca2+]iconcentration was found with both agents (Figs. 6 and 7) withhigher increase with DT56a and with higher sensitivity tothis treatment as determined by the E2 increased intracellular[Ca2+]i concentration at 0.01 nM by about 120% and DT56aat 10 ng/ml by about 240% (Figs. 6 and 7).

F o3o -t LP)i stb

pecific activity (5a), DNA synthesis (5b) and ALP (5c)oth pre- and post-menopausal age groups.

.6. The effects of DT56a on intracellular [Ca2+]i

oncentration in human female bone cells

To determine their responsiveness to DT56a compar2, human female bone cells from post-menopausal or

ig. 5. The effects of DT56a (200 ng/ml) or E2 (8.2 ng/ml equivalent t0 nM), or raloxifene (3000 nM), or raloxifene and DT56a or E2 and DT56an (a) creatine kinase specific activity (CK), on (b) [3H] thymidine incorpora

ion into DNA (DNA) and on (c) alkaline phosphatase specific activity (An human female bone cells. Results are mean± S.E.M. (n = 4) expressed ahe ratio between hormone-treated and controls (* p < 0.05,** p < 0.01). Theasal activities were similar to that shown inFigs. 1–3.

D. Somjen et al. / Journal of Steroid Biochemistry & Molecular Biology xxx (2005) xxx–xxx 5

Fig. 6. Stimulation by (a) E2 (2.73 ng/ml equivalent to 10 nM) or (b) DT56a (1 ng/ml) of intracellular [Ca2+]i concentration in post-menopausal human bonecells. Results are tracing of the scans described in Section2.

Fig. 7. Dose dependent stimulation by E2 (0.0273–27.3 ng/ml equivalent to0.001–0.100 nM) or by DT56a (0.01–10.00 ng/ml)) of intracellular [Ca2+]i

concentration in post-menopausal human bone cells. Results are quantita-tions of the tracing of the scans described in Section2, and are mean± S.E.M.(n = 4) (* p < 0.05,** p < 0.01).

4. Discussion

Our results demonstrate that DT56a activates osteoblastsin vitro, in all the tested parameters in a sex specific manner,namely it stimulated only female and not male-derived bonecells in both pre and postmenopausal age groups.

DT56a was shown in previous clinical studies to be aneffective agent in preventing postmenopausal osteoporosis[18], with no effect on the uterus. Bone loss induced byovariectomy in female rats, has been widely used as a modelfor studying postmenopausal osteoporosis[26–32]. Treat-ment with E2 is known to reverse the osteoporotic changescaused by ovariectomy[30–32].

In the present study, we analyzed the effects of DT56aand E2 on human cultured osteoblasts. DT56a, similarlyto E2, stimulated CK specific activity, which is a markerof estrogenic response, and DNA synthesis, which is ameasure of cell proliferation, in both pre-menopausal andpost-menopausal derived cells. In contrary, ALP activityshowed age dependence decline in the response to both treat-ments. Since ALP is a marker for osteoblastic activity, thisdecline might reflect the natural decrease in this activity withage.

DT56a like E2 did not stimulated male derived bone cells,which respond only to DHT, demonstrating that like E2,DT56a is sex-specific in its effect on bone cells in culture.DT56a’s activity was blocked by the SERM raloxifene, iden-tical to Ral’s effect on E2, and also displayed a SERM-likeactivity in blocking E2’s effect.

In the present study, we also showed non-genomic effectsof DT56a via putative membranal ER, by increasing intracel-lular cytosolic free calcium concentration [Ca2+]i . We assumet tedb d

hat the increase in [Ca2+]i in both age groups was mediay two mechanisms: (1) [Ca2+]i influx through voltage-gate

6 D. Somjen et al. / Journal of Steroid Biochemistry & Molecular Biology xxx (2005) xxx–xxx

[Ca2+]i channels and (2) [Ca2+]i mobilization from the endo-plasmic reticulum, which involved a Pertussis toxin-sensitiveG-protein[25]. There was a 10 times difference in the con-centrations needed for the membranal effects as comparedto the concentrations needed for the nuclear effect. In con-clusion, these results imply that DT56a has an estrogen-likeactivity at both the nuclear and the membrane levels in femalehuman osteoblasts. DT56a must therefore interact with mem-brane binding sites, penetrate the cells, and reach the nuclearreceptors by an as yet uncharacterized mechanism, and workat both levels.

The current anti-resorptive therapies mainly slows downosteoclastic activation[33,34], thus the balance betweenosteoclast/osteoblast activity is shifted toward bone for-mation. In the present study DT56a stimulated culturedosteoblasts, suggesting DT56a to be a bone forming agent.

Acknowledgement

We thank Professor A.M. Kaye from the Departmentof Molecular Genetics, the Weizmann Institute of Science,Rehovot, Israel, for the fruitful discussion on osteoblasticactivation.

R

ment1985)

one,

eptor

d, S.peci-lpha

torility,

ar,ogen-04)

breast23–

N.D.ne-

omen:iner.

akis,trolledomb.

[ G.the

selective estrogen receptor modulator LY117018 on growth hormonesecretion: in vitro studies, Metabolism 53 (2004) 563–570.

[11] I.R. Reid, R. Eastell, I. Fogelman, J.D. Adachi, A. Rosen, C. Nete-lenbos, N.B. Watts, E. Seeman, A.V. Ciaccia, M.W. Draper, Acomparison of the effects of raloxifene and conjugated equine estro-gen on bone and lipids in healthy postmenopausal women, Arch.Int. Med. 164 (2004) 871–879.

[12] M. Heringa, Review on raloxifene: profile of a selective estrogenreceptor modulator, Int. J. Clin. Pharmacol. Ther. 41 (2003) 331–345.

[13] F. Cosman, Selective estrogen-receptor modulators, Clin. Geriatr.Med. 19 (2003) 371–379.

[14] S.D. Malnick, A. Shaer, H. Soreq, A.M. Kaye, Estrogen-inducedcreatine kinase in the reproductive system of the immature femalerat, Endocrinology 113 (1983) 1907–1909.

[15] D. Somjen, Y. Weisman, A. Harell, E. Berger, A.M. Kaye, Directand sex specific stimulation by sex steroids of creatine kinase activityand DNA synthesis in rat bone, Proc. Natl. Acad. Sci. U.S.A. 86(1989) 3361–3365.

[16] S. Katzburg, A. Ornoy, D. Hendel, M. Lieberherr, A.M. Kaye, D.Somjen, Age and gender specific stimulation of creatine kinase spe-cific activity by gonadal steroids in human bone-derived cells inculture, J. Endocrinol. Invest. 24 (2001) 166–172.

[17] D. Somjen, E. Knoll, F. Kohen, N. Stern, Effects of phytoestrogenson DNA synthesis and creatine kinase activity in vascular cells, Am.J. Hypertens. 14 (2001) 256–1262.

[18] I. Yoles, Y. Yogev, Y. Frenkel, R. Nahum, M. Hirsch, B. Kaplan,Tofupill/Femarelle (DT56a)—a new phyto-selective estrogen recep-tor modulator-like substance for the treatment of postmenopausalbone loss, Menopause 10 (2003) 522–525.

[19] I. Yoles, Y. Yogev, Y. Frenkel, M. Hirsch, R. Nahum, B. Kaplan, Effi-cacy and safety of standard versus low-dose Femarelle (DT56a) for

ecol.

[ u-s but–98.

[ ctiv-003)

[ a, D.estro-roid.

[ ofnase9

[ jen,likenadal

[ nti-male

[ bone

[ imal991)

[ s and

[ manCal-

[ pro-ssue

eferences

[1] B. Ettinger, H.K. Genant, C.E. Cann, Long-term estroge replacetherapy prevents bone loss and fractures, Ann. Int. Med. 102 (319–324.

[2] S.C. Manolagas, S. Kousteni, R.L. Jilka, Sex steroids and bRecent Prog. Horm. Res. 57 (2002) 385–409.

[3] D.P. McDonnell, The molecular determinants of estrogen recpharmacology, Maturitas 48S (2004) 7–12.

[4] G.G. Kuiper, B. Carlsson, K. Grandien, E. Enmark, J. HaggblaNilsson, J.A. Gustafsson, Comparison of the ligand binding sficity and transcript tissue distribution of estrogen receptors aand beta, Endocrinology 138 (1997) 863–870.

[5] W. Shao, E.K. Keeton, D.P. McDonnell, M. Brown, CoactivaAIB1 links estrogen receptor transcriptional activity and stabProc. Natl. Acad. Sci. U.S.A. 101 (2004) 11599–11604.

[6] A.H. Talukder, A. Gururaj, S.K. Mishra, R.K. Vadlamudi, R. KumMetastasis-associated protein 1 interacts with NRIF3, an estrinducible nuclear receptor coregulator, Mol. Cell Biol. 424 (206581–6591.

[7] J.R. Pasqualini, The selective estrogen enzyme modulators incancer: a review, Biochim. Biophys. Acta. 1654 (2004) 1143.

[8] N. Morabito, A. Crisafulli, C. Vergara, A. Gaudio, A. Lasco,Frisina, R. D’Anna, F. Corrado, M.A. Pizzoleo, M. Cincotta,Altavilla, R. lentile, F. Squadrito, Effects of genistein and hormoreplacement therapy on bone loss in early postmenopausal wa randomized double-blind placebo-controlled study, J. Bone MRes. 17 (2002) 1904–1912.

[9] H.J. Teede, B.P. McGrath, L. DeSilva, M. Cehun, A. FassoulP.J. Nestel, Isoflavones reduce arterial stiffness: a placebo-constudy in men and postmenopausal women, Arterioscler. ThrVasc. Biol. 23 (2003) 1066–1071.

10] G. Tulipano, C. Bonfanti, C. Poiesi, A. Burattin, S. Turazzi,Barone, R. Cozzi, A. Bollati, D. Valle, A. Giustina, Effects of

the treatment of menopausal symptoms, Clin. Exp. Obstet. Gyn31 (2004) 123–126.

20] D. Somjen, I. Yoles, DT56a (Tofupill/Femarelle) selectively stimlates creatine kinase specific activity in skeletal tissues of ratnot in the uterus, J. Steroid. Biochem. Mol. Biol. 86 (2003) 93

21] D. Somjen, I. Yoles, DT56a stimulates creatine kinase specific aity in vascular tissues of rats, J. Endocrinol. Invest. 26 (2966–971.

22] A.M. Kaye, M. Spatz, A. Waisman, S. Sasson, S. Tamir, J. VaySomjen, Paradoxical interactions among estrogen receptors,gens and SERMS: mutual annihilation and synergy, J. SteBiochem. Mol. Biol. 76 (2001) 85–93.

23] D. Somjen, A. Waisman, A.M. Kaye, Tissue selective actiontamoxifen methiodide, raloxifene and tamoxifen on creatine kiB activity in vitro and in vivo, J. Steroid. Biochem. Mol. Biol. 5(1996) 389–396.

24] E. Berger, I. Bleiberg, Y. Weisman, A. Harel, A.M. Kaye, D. SomDifferentiation of cultured mice bone marrow into osteoblast-cells results in acquisition of sex-specific responsiveness to gosteroids, J. Endocr. Invest. 27 (2004) 622–629.

25] D. Somjen, F. Kohen, M. Lieberherr, Nongenomic effects of aidiotypic antibody as an estrogen mimetic in human and rat feosteoblasts, J. Cell. Biochem. 65 (1997) 53–66.

26] D.N. Kalu, The ovariectomized rat model of postmenopausalloss, Bone Miner. 15 (1991) 175–191.

27] T.J. Wronski, C.F. Yen, The ovariectomized rat model as an anmodel for postmenopausal bone loss, Cell Mat. 1 (Suppl.) (169.

28] H.M. Frost, W.S. Jee, On the rat model of human osteopeniaosteoporosis, Bone Miner. 18 (1992) 227–236.

29] A. Ornoy, S. Katzburg, Osteoporosis: animal models for the hudisease, in: A. Ornoy (Ed.), Animal Models of Human Relatedcium Metabolic Disorders, CRC Press, Florida, USA, 1995.

30] I.U. Schmidt, G.K. Wakley, R.T. Turner, Effects of estrogen andgesterone on tibia histomorphometry in growing rats, Calciff TiInt. 67 (2000) 47–52.

D. Somjen et al. / Journal of Steroid Biochemistry & Molecular Biology xxx (2005) xxx–xxx 7

[31] T. Takamo-Yamamoto, G.A. Rodan, Direct effect of 17�-estradiolon trabecular bone in ovariectomized rats, Proc. Natl. Acad. Sci.U.S.A. 87 (1990) 2172–2176.

[32] R. Turner, J.J. Vandersteenhoven, N.H. Bell, The effects of ovariec-tomy and 17�-estradiol on cortical bone histomorphometry in grow-ing rats, J. Miner. Res. 2 (1987) 115–122.

[33] M.J. Rogers, New insights into the molecular mechanism of actionof bisphosphonates, Curr. Pharm. Des. 9 (2003) 2643–2658.

[34] S.L. Teitelbaum, M.M. Tondravi, F.P. Ross, Osteoclasts,macrophages, and the molecular mechanisms of bone resorp-tion, J. Leukoc. Biol. 61 (1997) 381–388.