A

iRRaoo1conbtatama©

Ka

1

ctebo

f

0d

Journal of Steroid Biochemistry & Molecular Biology 100 (2006) 117–128

Effect of ormeloxifene on ovariectomy-induced bone resorption,osteoclast differentiation and apoptosis and TGF beta-3 expression

P.S. Narayana Murthy, S. Sengupta, S. Sharma, M.M. Singh ∗

Endocrinology Division, Central Drug Research Institute, Lucknow 226001, India

Received 27 October 2005; accepted 31 March 2006

bstract

Effect of ormeloxifene, a multifunctional selective estrogen receptor modulator, on prevention of ovariectomy-induced bone resorptionn retired breeder female rats, osteoclastogenesis using bone marrow cells from adult Balb/c mice cultured in presence of M-CSF andANKL, osteoclast apoptosis using terminal deoxynucleotidyl transferase fragment end labeling and TGF beta-3 expression were investigated.aloxifene, a benzothiophene reported to mimic effects of estrogen in bone, and estradiol were used for comparison. Ormeloxifene (10−6

nd 10−8 M) significantly inhibited osteoclastogenesis (P < 0.001 versus vehicle control) as evidenced by lower number of TRAP-positivesteoclasts in bone marrow cultures and caused apoptosis of osteoclasts. The effect was almost equivalent to that observed in presencef estradiol-17 beta, except that significant number of cells undergoing apoptosis was evident even at 10−9 M concentration of estradiol-7 beta (P < 0.001). Raloxifene, though inhibited osteoclastogenesis at much lower concentrations (10−8 to 10−12 M; P < 0.001), failed toause apoptosis of osteoclasts at any of the concentrations used. While ormeloxifene, raloxifene and ethynylestradiol significantly preventedvariectomy-induced bone loss in vivo in retired breeder female rats, prevention of ovariectomy-induced decrease in BMD and trabecularetwork of proximal tibia, calcium and phosphorus levels in femur and tibia and prevention of ovariectomy-induced down-regulation of TGFeta-3 expression in lumbar vertebrae was of lower order in raloxifene- than ormeloxifene- or ethynylestradiol-supplemented females. Bothhe SERMs, however, produced considerable estrogenic effects at the uterine level as evidenced by increase in weight, total and endometrialrea and luminal epithelial cell height; the effect being generally greater in raloxifene- than ormeloxifene-treated rats. Findings demonstratehat inhibition of estrogen-deficiency osteoporosis by ormeloxifene, as in case of estradiol, was mediated via inhibition of osteoclastogenesis,poptosis of osteoclasts and up-regulation of TGF beta-3 expression. Raloxifene, though effective in inhibiting osteoclastogenesis in vitro at

uch lower concentrations, was not only less potent in preventing ovariectomy-induced bone loss in retired breeder female rats in vivo but

lso appeared to have a different mechanism of action than ormeloxifene and estradiol.2006 Elsevier Ltd. All rights reserved.

eywords: Ormeloxifene; Raloxifene; SERM; Ethynylestradiol; Antiresorptive; Bone mineral density; Bone mineral content; Osteoclastogenesis; Osteoclast

eaTte

poptosis; TGF beta-3; Retired breeder rats

. Introduction

Osteoporosis is a condition of low bone mass and microar-hitectural disruption that results in fractures with minimalrauma. In post-menopausal women, it is associated with low

strogen levels [1]. The activity of osteoclasts that resorbone, relative to bone forming osteoblasts, dictates the devel-pment of osteoporosis [2]. Several studies have shown that

∗ Corresponding author. Tel.: +91 522 2613894 (O)/2310757 (R);ax: +91 522 2623405.

E-mail address: singhmm@rediffmail.com (M.M. Singh).

rgcuagn

960-0760/$ – see front matter © 2006 Elsevier Ltd. All rights reserved.oi:10.1016/j.jsbmb.2006.03.009

strogen replacement therapy (ERT) maintains skeletal massnd reduces fracture risk in post-menopausal women [3].he protective effect of estrogens on bone tissue is believed

o be due primarily to their antiresorptive action [2]. How-ver, long term ERT/HRT has been associated with increasedisk of cancer in estrogen target tissues, including mammaryland and endometrium [4]. New synthetic, non-steroidalompounds, called the selective estrogen receptor (ER) mod-

lators or SERMs, have recently been shown to possess ERgonistic or antagonistic selectivity depending on the tar-et tissue [5]. Ormeloxifene is a member of type-1 class ofon-steroidal SERMs in clinical use [6], which possesses

1 ioche

d[caateiRidgtaiauidnimbiposae

2

2

fbduul(i

2

CiLmtwUp

18 P.S. Narayana Murthy et al. / Journal of Steroid B

esirable bone-protective [7–10] and lipid lowing activities11] of estrogen and reported to prevent advanced breast can-er in both women and men [6,12,13] without side effectsssociated with conventional HRT/ERT. The precise mech-nism by which ormeloxifene and other such SERMs exertissue-specific estrogen agonist/antagonistic effect remainsnigmatic. Molecular mechanism by which estrogen exertsts bone protective effect is also not clearly understood.ecent evidence suggests that estradiol-17 beta and ralox-

fene decrease IL-6 mRNA levels and suppress osteoclastevelopment in primary murine bone cell cultures [14]. Estro-en deficiency also affects availability of various other mul-ifunctional cytokines such as IL-1, IL-6, TGF beta, etc.,nd affecting cells in the microenvironment. TGF betas playmportant role in bone formation, induction and repair. Thesere localized at bone fracture sites during healing and can reg-late cell proliferation as well as phenotypic gene expressionn the fracture callus and initiate osteogenesis and chon-rogenesis [15]. It has been reported that ovariectomy sig-ificantly reduces mRNA expression level of TGF beta-3soform in bone, which was well maintained when supple-

ented with estrogen or raloxifene [16]. TGF beta-3 has alsoeen reported to inhibit osteoclast differentiation, suggest-ng its role in estrogen-mediated bone maintenance [17]. Theurpose of this study was to understand bone specific effectsf ormeloxifene using osteoclast differentiation and apopto-is in osteoclast cultures derived from mouse bone marrownd ovariectomy-induced bone resorption and TGF beta-3xpression in retired breeder rats.

. Materials and methods

.1. Animals

Inbred Balb/c mice (6–8 week old) and retired breederemale Sprague–Dawley rats (age: 12–14 months; parity: ±3;ody weight: 250–300 g), referred as retired after 3–6 littersue to age-related increased incidence of reproductive fail-re/litter size, of the Institute’s breeding colony maintainednder standard conditions (22 ± 1 ◦C) with alternate 12-hight:12-h dark periods and free access to regular pellet dietLipton India Ltd., Bangalore, India) and tap water were usedn this study.

.2. Chemicals

Ormeloxifene (International Nonproprietary Name forentchroman) was synthesized [18] at this institute. Ralox-

fene was a generous gift from M/s. Cadila Healthcaretd., Ahmedabad, India. All other chemicals includingacrophage colony stimulating factor (M-CSF) and recep-

or activator of nuclear factor kappa B ligand (RANKL)ere purchased from Sigma–Aldrich, St. Louis, MO,SA. Enhanced chemiluminescence (ECL-Plus) kit wasurchased from Amersham Biosciences, Amersham, UK,

mistry & Molecular Biology 100 (2006) 117–128

and TdT-FragEL kit (QIA33) and anti-TGF beta-3 anti-body (raised in mouse) (GL-16) from Oncogene, Boston,MA, USA.

2.3. Experimental design

2.3.1. Antiresorptive activity in vivoAll protocols used in this study were approved by the Insti-

tute’s Animal Ethical Committee. Retired breeder female ratsfrom the Institute’s breeding colony were randomized into 5groups of 10 each and kept in separate steel cages. Animals offour groups were bilaterally ovariectomized under light etheranesthesia and treated orally with ormeloxifene (1.25 mg/kg),raloxifene (3 mg/kg), ethynylestradiol (0.5 mg/kg) or thevehicle (gum acacia in distilled water) once daily on days1–30 post-ovariectomy (day 1: day of ovariectomy). Dose oformeloxifene used was chosen based on its contraceptive dos-ing [10] and that of raloxifene from earlier reports [19]. Doseof ethynylestradiol, though high, was aimed to delineate itsbeneficial effects on bone rather than its uterus stimulatoryeffects [20]. One group of females was sham-operated andtreated similarly with the vehicle.

2.3.2. Autopsy and collection of tissuesTwenty-four hours after the last treatment, the animals

were autopsied by excessive ether inhalation and femurand tibia bones and uterus of each rat were dissected freeof adhering tissues, fixed in 70% ethanol and stored at−20 ◦C until BMD measurements. After BMD analysis,the isolated femur and tibia bones were kept in desicca-tors containing fused CaCl2 until further analysis. Lum-bar vertebrae (nos. 2–4) were excised, thoroughly cleanedof adhering tissues and stored at −20 ◦C for TGF beta-3expression studies. The uterus of each rat was weighed andfixed in 4% formaldehyde in phosphate buffer (pH 7.4) forhistomorphometry.

2.3.3. Bone mineral density (BMD) measurementBMD measurement of isolated femur and tibia bones was

performed using DEXA Fan Beam Densitometer (Model:QDR-4500A, Hologic, USA) calibrated daily with hydrox-yapatite anthropomorphic spine phantom (Hologic #3256)using manufacturer provided high-resolution software forsmall animals. Scans of isolated bones were performed usingidentical regions of interest (femur: global and neck; tibia:global and proximal region) at a scan speed of 1 mm/s or 4lines/mm [10].

2.3.4. Drying and ashing of bonesEach dried bone kept in the desiccator was weighed and

placed in a pre-weighed clean silica crucible, soaked with

500 �l of concentrated HCl and kept in a Muffle furnace setat a constant temperature of 500 ◦C for 3–4 h, allowed tocool to room temperature and then weighed to determine theweight of bone ash.

iochem

2

si1eM5o

2p

taacsMccafiwsCAclr(pU

2v

w

tiMs[Diwad

tcetfiot(BfiaCvldgas

2

MsT

F4p

P.S. Narayana Murthy et al. / Journal of Steroid B

.3.5. Bone ash calcium and phosphorus estimationStock solution of each ash sample was prepared by dis-

olving accurately weighed (approximately 10 mg) bone ashn 200 �l of concentrated HCl and making the volume to0 ml with triple distilled water [21]. Ash calcium wasstimated using commercial kits (#1553593, Boehringer,annheim, Germany) on a pre-programmed Boehringer

010 Photometer. For ash phosphorus estimation [22], 2 mlf the stock solution was diluted to 10 ml with distilled water.

.3.6. Osteoclast culture from bone marrow derivedrogenitors

Bone marrow cells were used to isolate osteoclast progeni-ors and cultured using soluble factors of osteoclast differenti-tion [23]. Briefly, adult female Balb/c mice were decapitatednd their femur and tibia bones were aseptically isolated andleared of adherent tissues. Bone marrow was flushed withterile minimal essential medium-alpha modification (alpha-EM) and agitated using a 23-gauge needle to achieve single

ell suspension. The cell suspension was washed with fetalalf serum (FCS)-free culture medium and centrifuged twicet 250 × g (7 min each) and the pellet was resuspended inresh culture medium. Approximately 106 cells were platedn 75 cm2 flasks and cultured in alpha-MEM supplementedith M-CSF (5 ng/ml), FCS (10%), penicillin (10,000 U/ml),

treptomycin (10 mg/ml) and l-glutamine (20 mM/l) in aO2 incubator (5% CO2 in air; 95% humidity) at 37 ◦C.fter 24 h in culture, suspended cells in the medium were

entrifuged at 250 × g for 7 min and their viability was estab-ished using Trypan blue exclusion staining. The cells wereesuspended in the medium containing M-CSF and RANKL20 ng/ml each) in a flat-bottomed 96-well plate containinglastic cover slips (6 mm diameter; Thermanox, Nunc, IL,SA). The culture medium was changed every 48 h.

.3.7. Osteoclastogenesis and osteoclast apoptosis initro

The test compounds were dissolved in DMSO and dilutedith alpha-MEM to achieve the desired concentrations (10−6

dcps

ig. 1. Osteoclast cell culture from bone marrow derived progenitors of adult Balb00×. (a) Cells cultured in presence of M-CSF (20 ng/ml) alone (negative controlositive multinucleate osteoclast 7 days after culture in presence of M-CSF and RA

istry & Molecular Biology 100 (2006) 117–128 119

o 10−12 M) immediately before use. The cells after load-ng onto the cover slips were cultured for 7 days in alpha-

EM containing the test compounds or the vehicle, fixed andtained for tartrate resistant acid phosphatase (TRAP) (Fig. 1)23]. Parallel cultures with the vehicle (dimethyl sulphoxide,MSO; final concentration: 0.01%) served as correspond-

ng controls. TRAP-positive osteoclasts were counter-stainedith haematoxylin and counted using an eyepiece graticule

nd a 40× objective lens at five different sites selected ran-omly on each coverslip.

To assess the effect of test compounds on osteoclast apop-osis, the cells after loading onto plastic coverslips wereultured for 5 days, followed by culture for 48 h in pres-nce of different concentrations (10−6 to 10−12 M) of theest compounds. Parallel cultures with the vehicle (DMSO;nal concentration: 0.01%) served as control. Apoptoticsteoclasts were identified by terminal deoxynucleotidylransferase (TdT)-mediated deoxy-UTP nick end labelingTUNEL) assay, according to the manufacturer’s instructions.riefly, the cells on plastic coverslips were washed with PBS,xed in 10% formalin in PBS for 10 min, air dried, rehydratednd treated with proteinase K (20 �g/ml) at 37 ◦C for 5 min.ells were then treated with 3% H2O2 for 5 min to inacti-ate endogenous peroxidases followed by incubation in TdTabeling reaction mixture for 1.5 h at 37 ◦C, staining withiaminobenzedene (DAB) and counterstaining with methylreen for 3 min. Number of apoptotic cells was counted usingn eyepiece graticule and a 40× objective lens at five randomites on each coverslip.

.3.8. TGF beta-3 expressionProtein lysates of lumbar vertebrae were prepared using

ammalian Cell Lysis kit (MCL-1; Sigma, USA). Tissueamples were homogenized in RIPA buffer containing 50 mMris–HCl, 1 mM EDTA, 150 mM NaCl, 0.1% SDS, 0.5%

exyocholic acid and 1% Igepal along with protease inhibitorocktail (P8340; Sigma) provided with the kit. The entirerocess was conducted on ice. The tissue homogenates wereonicated at low speed and the samples were placed on ice

/c mice. Osteoclasts were identified following staining for TRAP (a and b)). Note complete absence of TRAP positive staining. (b) A typical TRAPNKL (20 ng/ml each).

1 iochem

f21Pm

2a

2TwrdfsUtwibwavoqB

2

fiwsPDwep

ws

3

Kg

4

4

iosv(bTooaTevdovcts

TEr

R

F

T

V

20 P.S. Narayana Murthy et al. / Journal of Steroid B

or another 30 min and centrifuged at 14,000 × g at 4 ◦C for0 min. The supernatants were collected and recentrifuged at4,000 × g at 4 ◦C for 20 min to obtain clear tissue lysates.rotein content of the tissue lysates was estimated by BCAethod [24].

.3.9. SDS-polyacrylamide gel electrophoresis (PAGE)nd Western blot analysis

Equal amount of protein was mixed with equal volume of× electrophoresis sample buffer containing glycerol, SDS,ris and bromophenol blue. Samples were then heated in aater bath at 100 ◦C for 2–4 min and the proteins were sepa-

ated on 9% SDS-PAGE at constant voltage (100 V) until theye front reached lowest level. The gel was retrieved care-ully and the proteins were transferred to nitrocellulose byemi-dry protein transfer method using Trans-Blot (BioRad,SA) at constant current (60 mA) for 1 h at room tempera-

ure. The blot was then incubated in 3% BSA-TBS for 1.5 h,ashed in TBS containing 0.1% Tween-20 and incubated

n primary monoclonal antibody (raised in mouse) anti-TGFeta-3 diluted (1:1500) with 0.1% BSA-TBS for 1.5 h. Afterashing, the blot was incubated in biotinylated secondary

ntibody and HRP-labeled streptavadin using mouse Extra-idin kit. The blot was then developed by chemiluminescencen X-ray film using ECL plus. Relative band intensities wereuantified using gel documentation system (Gel Doc 2000;ioRad).

.3.10. Uterine histology and histomorphometryAbout 5 mm pieces from middle segment of the uterus

rom each animal were dehydrated in ascending grades ofsopropanol, cleared in xylene and embedded in paraffinax using standard procedures. Representative transverse

ections (5 �m) were stained with haematoxylin and eosin.hotomicrographs of sections were obtained using Leica

C 300 camera and Leica IM50 Image Acquisition soft-are fitted to a Leica DMLB microscope. Histomorphom-

try measurements of total uterine area, endometrial area,ercent endometrial area and luminal epithelial cell height

o<i(

able 1ffect of ovariectomy and ormeloxifene, raloxifene or ethynylestradiol supplement

ats

egion of interest of the isolated bone Intact rats Ovariectomized

Sham + vehicle Vehicle

emurGlobal 0.2211 ± 0.0064 0.1913 ± 0.010Neck 0.1997 ± 0.0100 0.1694 ± 0.010

ibiaGlobal 0.2009 ± 0.0046 0.1914 ± 0.008Proximal regiona 0.2129 ± 0.0043 0.1761b ± 0.009

alues represent mean ± S.E.M. All other relevant comparisons were statistically na About 1/4th proximal portion of tibia.b P < 0.05 vs. corresponding sham + vehicle treatment group.c P < 0.05 vs. corresponding ovariectomized + vehicle treatment group.d P <0.01 vs. corresponding ovariectomized + vehicle treatment group.

istry & Molecular Biology 100 (2006) 117–128

ere done using Leica Qwin-Semiautomatic Image Analysisoftware.

. Statistical analysis

The data were analyzed by one-way ANOVA. Newman–eul’s test was used to identify differences between treatmentroups.

. Results

.1. Bone mineral density

Bilateral ovariectomy was accompanied with decreasen BMD of isolated femur and tibia bones 30 days post-variectomy. The decrease was 15% (statistically notignificant) in neck region of femur and 17% (P < 0.05;ersus sham + vehicle treatment group) in proximal tibiaTable 1), which are constituted primarily of the trabecularone. This was confirmed by analysis of BMD data on-score basis with decrease in BMD more than 2.5 × S.D.f sham intact control group depicting osteoporosis statef the proximal tibia, whereas BMD of femur (globalnd neck region) reached the osteopenic levels (Table 2).his was prevented by ormeloxifene, raloxifene as well asthynylestradiol supplementation. Although the mean BMDalues of femur (global and neck) were not significantlyifferent in rats supplemented with ormeloxifene, raloxifener ethynylestradiol when compared to that of ovariectomizedehicle treated group, T-score analysis of the data revealedomplete reversal of osteopenia to normal levels in all thereatment groups, except in proximal tibia in raloxifeneupplemented females where the levels reached only the

steopenic levels with BMD levels being >1.0 × S.D., but2.5 × S.D. of sham intact control value. Maximum increase

n BMD of proximal tibia was observed after ethynylestradiol26%, P < 0.01), followed by ormeloxifene (19%, P < 0.05)

ation on mineral density (BMD, g/cm2) of isolated bones of retired breeder

rats

Ormeloxifene Raloxifene Ethynyl estradiol

6 0.2154 ± 0.0064 0.2174 ± 0.0076 0.2200 ± 0.00840 0.1913 ± 0.0058 0.1871 ± 0.0101 0.1873 ± 0.0070

2 0.2029 ± 0.0072 0.2147 ± 0.0043 0.2099 ± 0.00614 0.2100c ± 0.0100 0.2007c ± 0.0115 0.2219d ± 0.0064

on-significant.

P.S. Narayana Murthy et al. / Journal of Steroid Biochemistry & Molecular Biology 100 (2006) 117–128 121

Table 2Evaluation of effect of ovariectomy and ormeloxifene, raloxifene or ethynylestradiol supplementation on extent of osteoporosis based on T-/Z-score in retiredbreeder rats

Parameter Femur Tibia

Global Neck Global Proximala region

Mean ± S.D. of sham intact controls 0.2211 ± 0.0171 0.1997 ± 0.0266 0.2009 ± 0.0130 0.2129 ± 0.01202.5 × S.D. of sham intact controls 0.0428 0.0665 0.0325 0.0300

Sham + vehicle vs. ovariectomized + vehicle treatment groupDifference in BMD −0.0298b −0.0303b −0.0095 −0.0368c

Sham + vehicle vs. ovariectomized + ormeloxifene treatment groupDifference in BMD −0.0057 −0.0084 +0.0020d −0.0029

Sham + vehicle vs. ovariectomized + raloxifene treatment groupDifference in BMD −0.0037 −0.0126 +0.0138d −0.0122b

Sham + vehicle vs. ovariectomized + ethynylestradiol treatment groupDifference in BMD −0.0011 −0.0124 +0.0090d +0.0090d

a About 1/4th proximal portion of tibia.trol valvalue, a

aitWdtsifar

4

wa

godimobPptl(

TEb

P

BA

A

A

V

b Decrease in BMD by >1.0 × S.D. but <2.5 × S.D. of sham + vehicle conc Decrease in BMD > 2.5 × S.D. of corresponding sham + vehicle controld Increase in BMD over corresponding sham + vehicle control value.

nd raloxifene (14%, P < 0.05) supplementation, with levelsn ethynylestradiol supplemented group reaching even higherhan that of the corresponding sham intact level (Table 1).

hile BMD of tibia (global) did not exhibit discernibleecrease following ovariectomy, the levels reached higherhan that of the corresponding sham intact group followingupplementation with all the three agents (Table 2). Max-mum increase in BMD of femur neck, which is the mostracture prone area in aged men and women, was observedfter ormeloxifene (13%), followed by ethynylestradiol andaloxifene (∼11% each) (Tables 1 and 2).

.2. Bone weight and ash weight

Bilateral ovariectomy produced marked decrease in femureight (19%; statistically insignificant) as well as femur

sh weight (29%, P < 0.01; versus sham + vehicle treatment

4

c

able 3ffect of ovariectomy and ormeloxifene, raloxifene or ethynylestradiol supplementaones of retired breeder rats

arameter Intact rats Ovariectomize

Sham + vehicle Vehicle

one weight (mg) 564.6 ± 20.1 456.7 ± 26.8sh weight (mg) 346.0 ± 16.7 247.1 ± 13.8a

sh calciumTotal content (mg) 66.13 ± 4.17 35.98 ± 3.68b

Concentration (mg/100 mg ash) 18.81 ± 0.91 14.93 ± 2.14b

sh phosphorusTotal content (mg) 41.73 ± 0.72 21.28 ± 0.56a

Concentration (mg/100 mg ash) 9.77 ± 0.41 7.57 ± 0.25

alues represent mean ± S.E.M. All other relevant comparisons were statistically na P < 0.01 vs. corresponding sham + vehicle treatment group.b P < 0.05 vs. corresponding sham + vehicle treatment group.c P < 0.01 vs. corresponding ovariectomized + vehicle treatment group.d P < 0.05 vs. corresponding ovariectomized + raloxifene treatment group.e P < 0.01 vs. corresponding ovariectomized + raloxifene treatment group.

ue, a condition commonly referred to as osteopenia [10].condition commonly referred to as osteoporosis.

roup) in retired breeders female rats. Supplementation withrmeloxifene, raloxifene or ethynylestradiol beginning on theay of ovariectomy prevented ovariectomy-induced decreasen femur weight as well as ash weight and the levels were

aintained at almost the sham intact level (Table 3). In casef tibia, in comparison, there was a significant decrease inone weight (21%, P < 0.05) as well as ash weight (28%,< 0.01) and its prevention by SERMs/ethynylestradiol sup-

lementation was observed 30 days after bilateral ovariec-omy, the levels remained significantly or non significantlyower in all treatment groups than sham + vehicle groupTable 4).

.3. Ash calcium and phosphorus

Bilateral ovariectomy significantly decreased total ashalcium in the femur (46%; P < 0.05) as well as tibia (48%;

tion on weight, ash weight and ash calcium and phosphorus levels in femur

d rats

Ormeloxifene Raloxifene Ethynylestradiol

475.1 ± 15.4 521.1 ± 38.1 576.9 ± 48.1278.4 ± 8.8 298.5 ± 24.5 303.7 ± 26.3

87.18 ± 4.28c 79.94 ± 6.69c 84.95 ± 6.96c

31.30 ± 1.11a,c 26.76 ± 0.49b,c 28.26 ± 1.68b,c

39.19 ± 1.91c,e 22.43 ± 3.59b 45.48 ± 7.40c,e

13.83 ± 0.67c,d 7.43 ± 0.99 15.64 ± 3.23b,c,e

on-significant.

122 P.S. Narayana Murthy et al. / Journal of Steroid Biochemistry & Molecular Biology 100 (2006) 117–128

Table 4Effect of ovariectomy and ormeloxifene, raloxifene or ethynylestradiol supplementation on weight, ash weight and ash calcium and phosphorus levels in tibiabone of retired breeder rats

Parameter Intact rats Ovariectomized rats

Sham + vehicle Vehicle Ormeloxifene Raloxifene Ethynylestradiol

Bone weight (mg) 432.2 ± 17.9 340.5 ± 28.3a 369.3 ± 14.2 413.3 ± 21.9 410.7 ± 26.6Ash weight (mg) 269.5 ± 17.3 194.8 ± 11.5b 212.6 ± 6.2a 231.4 ± 8.5a 228.5 ± 15.8

Ash calciumTotal content (mg) 62.01 ± 3.21 32.56 ± 2.67b 58.16 ± 2.78c 60.93 ± 6.16c 69.23 ± 5.82c

Concentration (mg/100 mg ash) 23.05 ± 0.86 17.23 ± 1.56a 27.36 ± 1.08c 26.10 ± 1.76c 28.24 ± 1.51c

Ash phosphorusTotal content (mg) 26.89 ± 1.12 19.06 ± 3.38 27.22 ± 3.87 20.92 ± 2.80 30.55 ± 6.76Concentration (mg/100 mg ash) 10.27 ± 0.92 9.60 ± 1.44 12.26 ± 1.21d 8.83 ± 1.31 16.23 ± 3.08d

Values represent mean ± S.E.M. All other relevant comparisons were statistically non-significant.a P < 0.05 vs. corresponding sham + vehicle treatment group.b

.

PrecaogcdtisirIasclb

4

i(ct1Fanccm(c

4

oeew(caiocing no apoptosis, in case of ormeloxifene and estradiol-17 beta, rare methyl green stained normal cells couldbe seen.

Fig. 2. Inhibition in osteoclast generation 7 days after culture of bone marrow

P < 0.01 vs. corresponding sham + vehicle treatment group.c P < 0.01 vs. corresponding ovariectomized + vehicle treatment group.d P < 0.05 vs. corresponding ovariectomized + raloxifene treatment group

< 0.01) bones in rats. Oral administration of ormeloxifene,aloxifene or ethynylestradiol prevented to almost equalxtent (P < 0.01) ovariectomy-induced decrease in ash totalalcium levels in femur and tibia bones such that the totalsh calcium levels were maintained at almost equal to (tibia)r even higher (femur) than the sham + vehicle treatmentroup. As in case of total ash calcium, total ash phosphorusontent of femur bone exhibited significant (49%, P < 0.01)ecrease following bilateral ovariectomy. However, whilehis was prevented following supplementation with ormelox-fene and ethynylestradiol, with the latter reaching levelslightly higher (statistically non significant) than the shamntact control level, there appeared no apparent effect ofaloxifene therapy on total ash phosphorus content (Table 3).n case of tibia, too, while a decrease after ovariectomynd its prevention following ormeloxifene/ethynylestradiolupplementation was evident (Table 4), there was no dis-ernible effect of raloxifene supplementation in phosphorusevels represented in absolute or per 100 mg ash weightasis.

.4. Effect on osteoclastogenesis

Ormeloxifene (10−6 and 10−8 M) significantly inhib-ted M-CSF and RANKL induced osteoclast formation519 ± 17 and 570 ± 11, respectively, versus 693 ± 27 inase of vehicle control group; P < 0.001) and the inhibi-ion was almost equivalent to that observed with estradiol-7 beta (427 ± 11 and 440 ± 22, respectively; P < 0.001;ig. 2). Raloxifene, at the equivalent concentrations (10−6

nd 10−8 M), in comparison, failed to reduce osteoclastumber (663 ± 27 and 610 ± 23; P > 0.05 versus vehicleontrol group), but significantly (P < 0.001) inhibited osteo-

last formation in a concentration-dependent manner atuch lower concentrations of 10−9 M (418 ± 22), 10−10 M

430 ± 26), 10−11 M (369 ± 12) and 10−12 M (560 ± 15)oncentrations.

d(Vm*

.5. Osteoclast apoptosis

Significant increase in nuclear morphology and presencef apoptotic bodies were observed in cells cultured in pres-nce of ormeloxifene (185 ± 16 at 10−6 M; P < 0.001) andstradiol-17 beta (218 ± 16 at 10−6 M; P < 0.001) (Fig. 3)hen compared with corresponding vehicle control group

73 ± 13), strongly suggesting that ormeloxifene alters osteo-last lifespan similar to estradiol-17 beta [25]. Very fewpoptotic bodies were, however, observed in case of ralox-fene at any of the concentration tested. While in casef vehicle and raloxifene treated cultures, most of theells appeared normal with methyl green staining suggest-

erived progenitors of adult Balb/c mice in presence of M-CSF and RANKL20 ng/ml each) by ormeloxifene (�), raloxifene (�) and estradiol-17� (�).alues represent percent of vehicle control group. Each point representsean ± S.E.M. of three independent observations. *P < 0.01 vs. vehicle;

*P < 0.001 vs. vehicle.

P.S. Narayana Murthy et al. / Journal of Steroid Biochemistry & Molecular Biology 100 (2006) 117–128 123

Fig. 3. Number of apoptotic cells was counted using eyepiece graticule and 40× objective lens at five random sites on each cover slip and represented aspercent of vehicle control group (A). Each bar represents mean ± S.E.M. of three independent observations. Most of the cells appeared normal with methylgreen staining as in case of vehicle control group (a). In case of ormeloxifene and estradiol-17�, in comparison, rare methyl green stained cells could be seen.(A) (400×) evaluation of apoptosis of osteoclasts cultured for 48 h in the presence of 1 nM ( (web version); (print version)), 10 nM ( (web version);(print version)) and 1 �M ( (web version); (print version)) concentrations of ormeloxifene (b), raloxifene (c) and estradiol-17� (d). Apoptotic cells werei green (B

4

rii

ib

dentified by FragEL staining (arrow head) and counterstained with methyl

.6. TGF beta-3 expression in lumbar vertebrae

The level of TGF beta-3 expression in lumbar vertebrae inats of ovariectomized + vehicle treated group was 25% lowern comparison to sham + vehicle treatment group. Ormelox-fene as well as ethynylestradiol prevented ovariectomy-

gctl

).

nduced down-regulation of TGF beta-3 expression; the leveleing, respectively, 8 and 12% lower than the sham intact

roup. The prevention of down-regulation by raloxifene, inomparison, was of much lower order and level was main-ained close to the ovariectomized + vehicle group, i.e., 21%ower than the sham intact control group (Fig. 4).

124 P.S. Narayana Murthy et al. / Journal of Steroid Biochemistry & Molecular Biology 100 (2006) 117–128

Table 5Effect of ovariectomy and ormeloxifene, raloxifene or ethynylestradiol supplementation on uterine weight and histomorphometry in retired breeder rats

Treatment group Uterine weight (mg) Histomorphometry

Absolute Per 100 g bodyweight

Total areaa of the uterus Endometrium Luminal epithelialcell heightc

Total areaa Percent areab

ShamVehicle 438.5 ± 21.3 169.9 ± 7.9 5.24 ± 0.007 2.85 ± 0.060 45.2 19.09 ± 0.78

Bilaterally ovariectomizedVehicle 117.1 ± 4.5d 45.4 ± 1.8d 2.21 ± 0.005d 0.55 ± 0.080d 25.0 14.82 ± 0.25Ormeloxifene 176.1 ± 10.5d,e,f 74.4 ± 5.2d,e,f 3.24 ± 0.008d,g,h 1.41 ± 0.005d,g,h 55.9 27.88 ± 1.80d,g

Raloxifene 224.5 ± 28.8d,g 86.4 ± 7.3d,g 3.92 ± 0.009d,g 1.85 ± 0.018d,g 52.5 31.97 ± 2.06d,g

Ethynylestradiol 421.2 ± 13.6g,h 162.2 ± 7.9g,h 6.96 ± 0.022d,g,h 2.35 ± 0.010d,g,h 66.1 56.13 ± 2.77d,g,h

Values represent mean ± S.E.M. All other relevant comparisons were statistically non-significant.a mm2.b Percent of total uterus.c �m.d P < 0.05 vs. corresponding sham + vehicle treatment group.e P < 0.05 vs. corresponding ovariectomized + vehicle treatment group.f P < 0.05 vs. corresponding ovariectomized + raloxifene treatment group.g P < 0.01 vs. corresponding ovariectomized + vehicle treatment group.h P < 0.01 vs. corresponding ovariectomized + raloxifene treatment group.

Fig. 4. Effect of ovariectomy and ormeloxifene (1.25 mg/kg), raloxifene(3 mg/kg) or 17�-ethynylestradiol (0.5 mg/kg) supplementation beginningon the day of ovariectomy on TGF�-3 expression in lumbar vertebrae ofretired breeder female Sprague–Dawley rats 30 days after treatment. Relativeband intensity of the Western blots in arbitrary units has been representedas percent of sham intact + vehicle group. Each bar represents mean of twoindependent observations. Note decrease in TGF�-3 expression followingbilateral ovariectomy ( (web version); (print version)) and its preventionby ormeloxifene ( (web version); (print version)) and ethynylestradiol( (web version); (print version)) therapy. Raloxifene ( (web version);

(print version)), in comparison, was much less effective.

4

utaPapmsbwutodhiltoiit(

Fig. 5. Transverse sections of the uterus of sham operated intact (a and b) retirormeloxifene (e and f), raloxifene (g and h) or 17�-ethynylestradiol (i and j) suppluterine diameter and luminal epithelial cell height following ovariectomy and theirThe effect in ethynylestradiol treated rats was maximal and was associated with maglands, endometrial folding and vacuolization of luminal epithelial cells. (a, c, e, g,

.7. Uterine weight, histology and histomorphometry

There was a marked (73%, P < 0.05, Table 5) decrease interine weight 30 days after ovariectomy. Weak, but statis-ically significant, increase in uterine weight was observedfter ormeloxifene (50%, P < 0.05) and raloxifene (92%,< 0.01); the increase being significantly greater (P < 0.05)

fter raloxifene than that observed after ormeloxifene sup-lementation. Ethynylestradiol, in comparison, produced aassive (P < 0.001) increase in uterine weight when repre-

ented in absolute (260%) or /100 g body weight (257%)asis. Histologically, uterine lumen in sham intact femalesas lined with low cuboidal epithelial cells with vac-olated cytoplasm. The stroma was compact with infil-ration of leucocytes into the stroma and epithelium. Invariectomized vehicle treated rats, there was a markedecrease in uterine diameter and luminal epithelial celleight 30 days after ovariectomy. Ormeloxifene and ralox-fene treatment increased total uterine as well as subepithe-ial endometrial stroma and luminal epithelial cell height,he increase being generally more after raloxifene thanrmeloxifene treatment. Ethynylestradiol, in comparison,nduced marked increase in all the parameters studied,n addition to increase in number of glands, endome-

rial folding and vacuolization of luminal epithelial cellsFig. 5).

ed breeder female rats and effect of bilateral ovariectomy (c and d) andementation beginning on the day of ovariectomy. Note marked decrease inmoderate increase in ormeloxifene and raloxifene supplemented females.rked increase in uterine diameter, luminal epithelial cell height, number ofi) 25×; (b, d, f, h, j) 400×.

P.S. Narayana Murthy et al. / Journal of Steroid Biochemistry & Molecular Biology 100 (2006) 117–128 125

1 iochem

5

tbipsoasfhstTtmou

mteisavcuIdogws

bTeerewiai

ocalirof

icmeplwosna(

tiepiatstanwanaobaepiooahioWnre5oadtirg

26 P.S. Narayana Murthy et al. / Journal of Steroid B

. Discussion

Under physiological conditions, bone mass is well main-ained by a strictly balanced coupling of bone formation andone resorption. Loss of this balance causes severe disordern bone tissues. Osteoclasts are multinucleated giant cells thatlay specialized role in bone resorption and no other cells canubstitute for osteoclasts. Therefore, specific suppressors ofsteoclast function as well as stimulators of bone formationre therapeutically promising in the treatment of osteoporo-is. To date, some low-molecular weight compounds, growthactors, peptide hormones including calcitonin and estrogensave served as anti-resorptive agents. All these moleculeshow strong inhibitory effect on osteoclastic bone resorp-ion, either directly or indirectly, by distinct mechanisms.his, according to Hall et al. [8] might be due to inhibi-

ion of osteoclast activity. However, the precise cellular andolecular mechanisms underlying bone protective activity

f ormeloxifene as well as estradiol remain incompletelynderstood.

Recent in vivo studies have suggested that estradiol treat-ent in ovariectomized rats increases number of TRAP posi-

ive multinucleate cells undergoing apoptosis [25]. A similarvidence of estrogenic regulation of osteoclast productionn bone marrow through apoptosis has been provided in thistudy. Using mouse bone marrow cultures, both ormeloxifenend estradiol not only reduced generation of osteoclasts initro resulting in decreased number of differentiated osteo-lasts, there was also an increase in the number of cellsndergoing apoptosis in a concentration-dependent manner.n case of raloxifene, in contrast, no such apoptosis was evi-ent at any of the concentrations tested, clearly indicating lackf detrimental effect of raloxifene on osteoclast life span, sug-esting different mechanisms of action for the two SERMs,ith ormeloxifene appears to be more beneficial at bone tis-

ue level.TGF betas have been reported to play important role in

one formation, induction and repair. Of the three isoforms,GF beta-3 has been reported to inhibit osteoclast differ-ntiation [17]. While an elevation in TGF beta-3 proteinxpression in lumbar vertebrae in bilaterally ovariectomizedats following treatment with ormeloxifene, raloxifene andthynylestradiol for 30 days beginning on day of ovariectomyas observed in the present study, prevention of ovariectomy-

nduced down-regulation of expression appeared maximumfter ormeloxifene followed by ethynylestradiol and ralox-fene supplementation.

A similar correlation was evidenced using preventionf ovariectomy-induced decrease in BMD and mineralontent of long bones, particularly in regions (femur necknd proximal tibia) known to primarily contain trabecu-ar bone, which is lost during estrogen-deficiency states

ncluding post-menopause [19]. Thus while ormeloxifene,aloxifene as well as ethynylestradiol significantly preventedvariectomy-induced bone loss in vivo in retired breederemale rats, prevention of ovariectomy-induced decrease

([ao

istry & Molecular Biology 100 (2006) 117–128

n BMD and trabecular network of proximal tibia andalcium and phosphorus levels in femur and tibia was ofuch lower order in raloxifene- than ormeloxifene- or

thynylestradiol-supplemented females. An almost similaricture was observed when ash calcium and phosphorusevels were represented on mg/100 mg ash weight basis,ith significant decrease occurring following bilateralvariectomy and its prevention by SERMs/ethynylestradiolupplementation, with levels in femur reaching sig-ificantly higher (ormeloxifene: P < 0.01; raloxifenend ethynylestradiol: P < 0.05) than sham intact levelsTables 3 and 4).

One of the major concerns of the long-term ERT/HRTherapy has been the associated increased risk of cancern estrogen target tissues, including mammary gland andndometrium [4]. In the present study, both the SERMsroduced considerable estrogenic effects as evidenced byncrease in uterine weight, total uterine and endometrialrea and uterine luminal epithelial cell height. However,he increase at their respective effective (antiosteoporo-is) doses was generally significantly greater in raloxifene-han ormeloxifene-treated rats. Pertinently, raloxifene haslso been reported to exhibit weak, but statistically sig-ificant, uterotrophic response (1.7-fold at 0.1 mg/kg dose)ith increase in uterine dry weight and height, mitotic

ctivity, vacuolization and degeneration of uterine lumi-al epithelial cells and number of glands in immaturend ovariectomized rats [26]. Its potent estrogenic effectn hypothalamo-pituitary-ovarian axis was also evidencedy decreased gonadotrophin secretion, hyperprolactinemia,dvanced vaginal opening, persistent presence of cornifiedpithelial cells in vaginal smears, anovulation, inhibition inositive feedback between estradiol and LH and infertilityn neonatal rats treated with 50–500 �g doses of raloxifenen days 1–5 of age [27]. Clinically, it increases incidencef hot flashes, deep vein thrombosis, pulmonary embolismnd leg cramps in women similar to that associated withormone replacement therapy [14]. However, at least threendependent clinical studies [28–30] report apparent lackf its proliferative effects on endometrial tissue in women.hile such information in case of ormeloxifene is so far

ot available, a study on evaluation in 60 patients of AUBeported no evidence of endometrial hyperplasia or ovariannlargement [31]. When administered at a very high dose of00 mg/rat (subcutaneous) to male and female young onesn day 5 of post-natal life, ormeloxifene does not produceny effect on ovarian and testicular histology at 60 and 90ays of age, except reduction in diameter of seminiferousubules [32,33]. Levormeloxifene, the l-isomer of ormelox-fene, which is a more potent receptor binder with higherelative binding affinity (RBA to rat uterine cytosol estro-en receptors) of 15.7 ± 3.1% of estradiol-17 beta than the d-

RBA: 2.10 ± 0.9) or the dl-ormeloxifene (RBA: 5.24 ± 1.45)6] and reported to prevent bone loss in estrogen-deficientnimal models, without stimulation of endometrial glandsr epithelium [6,7] has, however, been reported to induce

Biochem

epmpppiwh

atuetciittiawsobo

A

cMiHafDm

R

[

[

[

[

[

[

[

[

[

[

[

[

[

P.S. Narayana Murthy et al. / Journal of Steroid

ndometrial thickening [34–38] in ∼40% of the around 4000ost-menopausal women in phases II (12 months) and III (3–9onths) clinical trials, while showing its estrogen-like bone

reserving effect based on decrease in serum total alkalinehosphatase and urinary excretion of C-terminal extensioneptides [34]. The observed endometrial thickening, accord-ng to these investigators, was due only to fluid retention,ithout evidence of proliferation or hyperplasia even at veryigh doses.

Pertinently, immature rat bioassay using uterine weightnd histomorphometry as parameters has routinely been usedo evaluate estrogen agonistic profile of promising agentsnder development for human use and welfare [10]. How-ver, based on the above information, it appears that despitehe fact that both ormeloxifene and raloxifene exhibitedonsiderable estrogenic effects in rat models, with ralox-fene presenting greater estrogenic responses than ormelox-fene, studies involving post-menopausal women, in con-rast, report lack of their effect on endometrial prolifera-ion or hyperplasia. Thus while further comparative stud-es with ormeloxifene and raloxifene to precisely evalu-te their effect on uterus/endometrium in post-menopausalomen appears imminent, the present study clearly demon-

trates that raloxifene was not only less potent in preventingvariectomy-induced bone loss in retired breeder female rats,ut also appears to have a different mechanism of action thanrmeloxifene and estradiol.

cknowledgements

The authors thank Mr. P.K. Dasgupta for help in osteoclastulture, Prof. Kartar Singh, Director, Dr. V.L. Bhatia and Mr.anoj Dubey, Sanjay Gandhi PostGraduate Institute of Med-

cal Sciences, Lucknow for BMD measurements, M/s. Cadilaealthcare Ltd., Ahmedabad, for generous gift of raloxifene

nd the Ministry of Health & Family Welfare, Govt. of India,or financial support. One of us (PSN) thanks the CSIR, Newelhi, for award of Senior Research Fellowship. CDRI com-unication no. 6697.

eferences

[1] S.R. Cummings, W.S. Browner, D. Bauer, K. Stone, K. Ensurd, S.Jamal, B. Ettinger, Endogenous hormones and the risk of hip andvertebral fractures among older women, N. Engl. J. Med. 339 (1998)733–738.

[2] B.L. Riggs, S. Khosla, L.J. Melton, Sex steroids and the constructionand conservation of adult skeleton, Endocr. Rev. 23 (2002) 279–302.

[3] P.D. Delmas, Treatment of postmenopausal osteoporosis, Lancet 359(2002) 2018–2026.

[4] D. Grady, T. Grebretsadik, V. Ernestwr, D. Petitti, Hormone replace-

ment therapy and endometrial cancer risk: a meta-analysis, Obstet.Gynecol. 85 (1995) 304–313.

[5] B.L. Riggs, L.C. Hartmann, Selective estrogen receptor modulators:mechanism of action and application to clinical practice, N. Engl. J.Med. 348 (2003) 618–629.

[

istry & Molecular Biology 100 (2006) 117–128 127

[6] M.M. Singh, Centchroman: a selective estrogen receptor modulator,as a contraceptive and in the management of hormone related clinicaldisorder, Med. Res. Rev. 21 (2001) 302–347.

[7] S.D. Bain, D.L. Celino, M.C. Baily, M.J. Strachan, J.R. Piggott,V.M. Labroo, Centchroman, a nonsteroidal antiestrogen, is a bone-specific estrogen agonist in the ovariectomized rat, Calcif. TissueInt. 54 (1995) 338.

[8] T.J. Hall, H. Nyugen, M. Schaueblin, B. Fournier, The bonespecific estrogen centchroman inhibits osteoclastic bone resorp-tion in vitro, Biochem. Biophys. Res. Commun. 216 (1995) 662–668.

[9] V.M. Labroo, M. Creek, J.R. Piggott, Bothall, S.D. Bain, Methodfor inhibiting bone loss using centchroman derivatives, US Patentno. 5,464,862 (1995). ZymoGenetics Inc., Seattle, WA.

10] M. Arshad, S. Sengupta, S. Sharma, R. Ghosh, V. Sawlani, M.M.Singh, In vitro anti-resorptive activity and prevention of ovariectomy-induced osteoporosis in female Sprague–Dawley rats by ormelox-ifene, a selective estrogen receptor modulator, J. Steroid Biochem.Mol. Biol. 91 (2004) 67–78.

11] H.U. Bryant, J.A. Dodge, Methods of lowering serum cholesterol andinhibiting smooth muscle cell proliferation, restenosis, endometriosisand uterine fibrinoid disease, US Patent #5,407,955, Eli Lilly &Company, Indianapolis, IN, 1995.

12] N.C. Misra, P.K. Nigam, R. Gupta, A.K. Agarwal, V.P. Kamboj,Centchroman:, A non-steroidal anticancer agent for advanced breastcancer: phase II study, Int. J. Cancer 43 (1989) 781–783.

13] N.C. Misra, O.P. Asthana, S. Misra, A. Chaturvedi, S.N. Gupta,M.K. Mittal, V.P. Kamboj, R. Kumari, N.N. Khanna, I.D. Sharma,Centchroman—a new antiestrogenic and non-steroidal agent in themanagement of advanced breast cancer, in: Proceedings of EighthInternational Cong. Anti-Cancer Treatment, Paris, France, 3–6 Febru-ary, 1998 (abstract).

14] D. Clemett, C.M. Spencer, Raloxifene: a review of its use in post-menopausal osteoporosis, Drugs 60 (2000) 379–411.

15] M.E. Joyce, A.B. Roberts, M.B. Sporn, M.E. Bolander, Transforminggrowth factor � and the initiation of chondrogenesis and osteogenesisin the rat femur, J. Cell Biol. 110 (1990) 2195–2207.

16] R.D. Finkelman, N.H. Bell, D.D. Strong, L.M. Demers, D.J. Baylink,Ovariectomy selectively reduces the concentration of transforminggrowth factor-� in rat bone: implications for estrogen deficiency-associated bone loss, Proc. Natl. Acad. Sci. U.S.A. 89 (1992)12190–12193.

17] N.N. Yang, H.U. Bryant, S. Hardikar, M. Sato, R.J.S. Galvin, A.L.Glasebrook, J.D. Termine, Estrogen and raloxifene stimulate trans-forming growth factor-�3 gene expression in rat bone: a potentialmechanism for estrogen- or raloxifene-mediated bone maintenance,Endocrinology 137 (1996) 2075–2084.

18] A. Saeed, N. Durani, S. Durani, S. Ray, R.S. Kapil, Cis-isomerof centchroman—a selective ligand for the microsomal antiestrogenbinding site, Biochem. Biophys. Res. Commun. 125 (1984) 346–352.

19] C.H. Turner, M. Sato, H.U. Bryant, Raloxifene preserves bonestrength and bone mass in ovariectomized rats, Endocrinology 135(1994) 2001–2005.

20] A. Galleghar, T.J. Chambers, J.H. Tobias, The estrogen antago-nist ICI 182,780 reduces cancellous bone volume in female rats,Endocrinology 133 (1993) 2787–2791.

21] S.R. Cooper, D.R. Topliff, D.W. Freeman, M.A. Collier, O.K. Balch,Evaluation of bone mineral content in equine cadavers and pregnantmares, Animal Science Research Report, Oklahoma State University,1998, pp. 25–31.

22] J.B. Sumner, A method for determination of phosphorus, Science

100 (1944) 413–414.

23] M.R. Wani, K. Fuller, N.S. Kim, Y. Choi, T. Chambers, ProstaglandinE2 cooperates with TRANCE in osteoclast induction from haemopoi-etic precursors: synergistic activation of differentiation, cell spread-ing, and fusion, Endocrinology 140 (1999) 1927–1935.

1 Biochem

[

[

[

[

[

[

[

[

[

[

[

[[[[

28 P.S. Narayana Murthy et al. / Journal of Steroid

24] P.K. Smith, R.I. Krohn, G.T. Hermanson, A.K. Mallia, F.H. Gart-ner, M.D. Provenzano, E.K. Fujimoto, N.M. Goeke, B.J. Olson,D.C. Klenk, Measurement of protein using bicinchoninic acid, Anal.Biochem. 150 (1985) 76–85.

25] D.E. Hughes, A. Dai, J.C. Tiffee, H.H. Li, G.R. Mundy, B.F. Boyce,Estrogen promotes apoptosis of murine osteoclasts mediated byTGF-beta, Nat. Med. 2 (1996) 1132–1136.

26] J. Ashby, J. Odum, J.R. Foster, Activity of raloxifene in immatureand ovariectomized rat uterotrophic assays, Regul. Toxicol. Pharma-col. 25 (1997) 226–231.

27] L. Pinilla, L.C. Gonzalez, F. Geytan, M. Tena-Sempere, E.A.Guilar, Oestrogenic effect of neonatal administration of raloxifeneon hypothalamic–pituitary–gonadal axis in male and female rats,Reproduction 121 (2001) 915–924.

28] J.A. Scott, C.C. Da Camara, J.E. Early, Raloxifene: a selective estro-gen receptor modulator, Am. Fam. Physician 60 (1999) 1131–1139.

29] S.M. Boss, W.J. Huster, J.A. Neild, M.D. Glant, C.C. Eisen-hunt, M.W. Draper, Effects of raloxifene hydrochloride on theendometrium of postmenopausal women, Am. J. Obstet. Gynecol.

177 (1997) 1458–1464.

30] M.W. Draper, D.E. Flowers, W.J. Huster, J.A. Neild, K.D. Harper,C. Arnaud, A controlled trial of raloxifene (LY139481) HCl: impacton bone turnover and serum lipid profile in healthy postmenopausalwomen, J. Bone Miner. Res. 11 (1996) 835–842.

istry & Molecular Biology 100 (2006) 117–128

31] U. Subramanium, S.A. Gupte, A.S. Gupte, Centchroman in AUB,in: Proceedings of 43rd All India Cong. Obstet. Gynecol., Lucknow,India, 26–30 December, 1999, p. 28 (abstract 28M/M-1).

32] J.D. Dhar, B.S. Setty, Sexual maturation in male rats after centchro-man and naphthofuran administration during the neonatal period,Andrologia 15 (1983) 463–467.

33] J.D. Dhar, B.S. Setty, Effect of centchroman on the genital organs ofthe offsprings of rats when administered during lactation or duringthe neonatal period: a comparative study with ethynylestradiol anddiethylstilbestrol, Acta Eur. Fertil. 17 (1986) 285–290.

34] S.D. Bain, D. Grennspan, R. Kurman, M. Shalmi, B. Guldham-mer, N. Korsgaard, Levormeloxifene, a nonsteroidal, partial estrogenagonist, prevents bone loss, reduces serum cholesterol and exerts anon-proliferative action on uterine tissue in the ovariectomized rat,J. Bone Miner. Res. 12 (Suppl. 1) (1997) (abstract S347).

35] Novo Nordisk revises SERM predictions, Scrip 2347 (1998a) 19.36] Novo Nordisk denies SERM failure, Scrip 2365 (1998b) 16.37] Novo Nordisk drops levormeloxifene, Scrip 2374 (1998c) 18.38] K. Bjarnason, B.K. Skrumsager, B. Kiehr, P.C. Pederson, Lev-

ormeloxifene, a new partial estrogen receptor agonist, demonstratesantiresorptive and antiatherogenic properties in post-menopausalwomen, in: Proceedings of 19th Ann. Meet. American Soc. BoneMiner. Res., Cincinnati, September 10–14, J. Bone Miner. Res. 12(Suppl. 1) (1997) (abstract S346).