Full text of DOI:10.1359/jbmr.2001.16.1.89

JOURNAL OF BONE AND MINERAL RESEARCH
Volume 16, Number 1, 2001
© 2001 American Society for Bone and Mineral Research
Effects of a New Selective Estrogen Receptor Modulator
(MDL 103,323) on Cancellous and Cortical Bone in
Ovariectomized Ewes: A Biochemical, Histomorphometric,
and Densitometric Study
PASCALE CHAVASSIEUX,1 PATRICK GARNERO,1 FRANCOIS DUBOEUF,1 PHILIPPE VERGNAUD,1
FRANCOISE BRUNNER-FERBER,2 PIERRE D. DELMAS,1 and PIERRE J. MEUNIER1
ABSTRACT
The aims of this study performed in ewes were: (1) to confirm in this animal model the effects on bone of
ovariectomy (OVX) alone or associated with Lentaron (L), a potent peripheral aromatase inhibitor, used to amplify
the effects of OVX and (2) to evaluate the effects of a new selective estrogen receptor modulator (SERM; MDL
103,323) on bone remodeling. Thirty-nine old ewes were divided into five groups: sham (n ؍ 7); OVX (n ؍ 8); OVX
؉ L (n ؍ 8); OVX ؉ L ؉ MDL; 0.1 mg/kg per day (n ؍ 8); and OVX ؉ L ؉ MDL 1 mg/kg per day (n ؍ 8). The
animals were treated for 6 months. Biochemical markers of bone turnover (urinary excretion of type 1 collagen
C-telopeptide [CTX], serum osteocalcin [OC], and bone alkaline phosphatase [BAP]) were measured each month.
Bone biopsy specimens were taken at the beginning and after death at the end of the experiment. Bone mineral
density (BMD) was measured by dual-energy X-ray absorptiometry (DXA) on the lumbar spine and femur. OVX
induced a significant increase in biochemical markers. This effect was the highest after 3 months for CTX (؉156%
vs. sham) and after 4 months for OC and BAP (؉74% and ؉53% vs. sham, respectively). L tended to amplify the
effect of OVX on OC and BAP. OVX induced significant increases in the porosity, eroded, and osteoid surfaces in
cortical bone but no effect was observed in cancellous bone. MDL treatment reduced the bone turnover as assessed
by bone markers, which returned to sham levels as well as histomorphometry both in cortical and in cancellous
bone. Cancellous osteoid thickness decreased by 27% (p < 0.05), mineralizing perimeter by 81% (p < 0.05), and
activation frequency by 84% (p < 0.02) versus OVX ؉ L. Femoral and spinal BMD were increased by MDL and
tended to return to the sham values. The effects of OVX on bone turnover were different on cortical and cancellous
bone. These effects on cortical bone were reflected by changes in biochemical markers. MDL markedly reduces
bone turnover and increases BMD suggesting that this new agent may prevent postmenopausal bone loss. (J Bone
Miner Res 2001;16:89–96)
Key words: ewe, ovariectomy, selective estrogen receptor modulator, bone marker, bone histomorphometry
humans and the bone effects of fluoride and glucocorticoids
have been found similar to those observed in humans.(1–5)
However, the effects of ovariectomy (OVX) on ewe bone
INTRODUCTION
HE EWE HAS been used in preclinical experiments as a
T
model for therapeutic research in osteoporosis because are not clearly established. Previous studies have shown an
this large animal has a bone remodeling similar to that of increase in biochemical markers of bone turnover,(6–9) but
1Institut National de la Sante´ et de la Recherche Me´dicale Unite´ 403, Faculte´ R.T.H. Lae¨nnec and Hoˆpital Edouard Herriot, Lyon, France.
2Hoechst Marion Roussel, Romainville, France.
89

90
CHAVASSIEUX ET AL.
the decrease in bone mineral density (BMD) varies in spine,
ilium, calcaneus, and distal radius.(8,9) Seasonal variations,
calcium intake, and the number of lambs suckled influence
the bone remodeling in ewe, and, more importantly, a large
part of estradiol is not of ovarian origin but comes from the
adrenal secretion and peripheral aromatization.
Selective estrogen receptor modulators (SERMs) are a
new category of therapeutic agents that bind with high
affinity to estrogen receptors and mimic the effect of estro-
gen in some tissues but act as estrogen antagonists in
others.(10,11) Modulator (MDL) 103,323 is a new SERM that
acts as an estrogen agonist in the prevention of bone loss in
rats(12) but as an antagonist on mammary estrogen receptors
with an antitumoral effect. Furthermore, it decreased the
serum cholesterol levels in rats.(12)
The purpose of this study performed in ewes was to
evaluate the consequence of OVX, the influence of the
inhibition by Lentaron (L), a specific inhibitor of the pe-
ripheral aromatase, and the effects on bone of MDL
103,323. These effects were assessed by biochemical mark-
ers of bone remodeling, bone histomorphometry on iliac
crest biopsy specimens, and BMD measured by dual-energy
X-ray absorptiometry (DXA).
Biochemistry
Blood and urine samples were drawn before the OVX and
monthly after the beginning of the treatment. The following
biochemical parameters were measured:
(1) Serum osteocalcin (OC) level was determined by radio-
immunoassay with a specific antibody against ovine
OC.(13)
(2) Specific bone alkaline phosphatase (BAP) was deter-
mined in the serum by immunoradiometric assay.(14)
(3) Urinary excretion of type I collagen C-telopeptide
(CTX) was measured by ELISA (CrossLaps, Cis Bio
International, Gif/Yvette, France).(15) This assay is
based on an immobilized synthetic peptide with an
amino acid sequence specific for a part of the
C-telopeptide of the ␣1-chain of type I collagen (Glu-
Lys-Ala-His-Asp-Gly-Gly-Arg).
Bone biopsy specimens
Before the beginning of the experiment, a double tetra-
cycline labeling was performed. The animals received an
intramuscular injection of 20 mg/kg per day oxytetracycline
(Terramycin 100; Pfizer, Orsay, France) according to the
following schedule: 2 days on, 10 days off, and 4 days on.
Five days later a transfixing bone biopsy was performed on
the right iliac crest, after local anesthesia. Before slaughter-
ing, a second double labeling was performed with calcein
with the same schedule. A vertical slice of the left iliac crest
was cut corresponding to the biopsy site (i.e., 2 cm behind
the anterosuperior iliac spine and 2 cm below the summit of
the crest) and the histomorphometric analysis was per-
formed on the site of biopsy.
MATERIALS AND METHODS
Animals
Forty old ewes (mean age, 8 Ϯ 1 years; mean weight,
66 Ϯ 7 kg; retired breeder) were selected. They were
divided into five groups of 8 animals matched for their
breed, age, and number of lambs suckled. The animals were
penned individually in an enclosed sheep barn at the Institut
National de Recherche Agronomique (INRA) Research
Center (Saint-Genest-Champannelle, France) and kept un-
der a constant photoperiod (16 h of light per day) for 3
months before the beginning of the experiment and for the
whole duration of the experiment. At the end of the exper-
iment, the animals were slaughtered. The animals were
housed, treated, and then slaughtered by persons authorized
to handle animals in buildings approved by the European
Community.
Bone samples were prepared according to previously
described histological procedures.(16) Bone samples were
embedded in methylmethacrylate. Seven-micrometer-thick
sections were obtained with a Polycut microtome (Reichert-
Jung) and stained either with Solochrome Cyanin R or
Goldner’s trichrome. Twenty-micrometer-thick unstained
sections were prepared for observations under UV light.
Bone histomorphometry(16)
: Bone histomorphometry
measurements were performed in cancellous and cortical
tissue by either a semiautomatic (Ibas I; Leica, Heerburgg,
Switzerland) or an automatic color image analyzer (Visio-
lab; Biocom, Les Ulis, France). This system is equipped
with a specific software for the measurement of bone struc-
ture and microarchitecture parameters.(17) The histomorpho-
metric abbreviations used are derived from the American
Society for Bone and Mineral Research (ASBMR) nomen-
clature.(18) The histomorphometric data are reported in two
dimensions.
Treatment
The first group of 7 animals was considered as a control
group. The 32 animals from the other groups were OVX. To
amplify the effects of OVX, three groups were injected
twice a month with 250 mg of L (OVX ϩ L), a potent
peripheral aromatase inhibitor. Ewes from two OVX ϩ L
groups received 0.1 mg/kg per day (MDL 0.1, n ϭ 8) or 1
mg/kg per day (MDL 1, n ϭ 8) of a new SERM, MDL
103,323 (Hoechst Marion Roussel, Romainville, France).
MDL was administered fasting through a gastric fistula for
6 months. Treatments began immediately after OVX.
All animals received a normal calcium intake (4.5 g/day)
during the experimental period.
Cancellous bone histomorphometry: The following static
measurements were performed:
● The bone area (Cn-B.Ar/T.Ar): The whole spongy
bone area of three sections stained with Solochrome
Cyanin R was measured with the Biocom automatic
image analyzer.
● The eroded perimeter (Cn-E.Pm/B.Pm): Representing
the percentage of total cancellous bone surfaces cre-

EFFECTS OF SERM ON BONE IN OVARIECTOMIZED EWES
91
nated by Howship’s lacunae with or without oste-
oclasts. E.Pm/B.Pm was measured on the whole
spongy bone area of three sections stained with Gold-
ner’s trichrome, with the semiautomatic method at
ϫ80.
● The formation period (Cn-FP) represents the mean time
required to rebuild a new bone structure unit and was
calculated as W.Wi over Aj.AR. The active formation
period [Cn-FP(aϩ)] is given as the ratio W.Wi over
MAR.
● The number of osteoclasts (Cn-N.Oc/B.Pm): Counted
and expressed per unit of cancellous bone perimeter
(per mm2).
● The osteoid area (Cn-O.Ar/B.Ar): The fraction of can-
cellous bone tissue not calcified.
● The osteoid perimeter (Cn-O.Pm/B.Pm): The percent-
age of cancellous perimeter covered with osteoid seam.
● The width of osteoid seams (Cn-O.Wi): Expressed in
micrometers and measured according to the method
described by Kragstrup et al.(19) Osteoid parameters
were measured on the whole spongy bone area on three
sections stained with Solochrome Cyanin R with the
semiautomatic image analyzer at ϫ200.
● The activation frequency (Cn-Ac.f) is the probability
that a new remodeling cycle will be initiated at any
point of the bone surface. It is calculated as BFR/W.Wi
and expressed per year.
The following parameters reflecting the microarchitecture
of spongy bone were measured with the automatic image
analyzer:
● Trabecular thickness (Tb.Th; ␮m), trabecular separa-
tion (Tb.Sp; ␮m), and trabecular number (Tb.N; per
mm) were calculated from the cancellous bone area and
volume and the length of the bone marrow interface
according to Parfitt’s formula.(18)
● The osteoblast perimeter (Cn-Ob.Pm/B.Pm): Repre-
sents the percentage of the cancellous bone covered
with osteoblasts. These parameters were measured on
the whole cancellous surface of sections stained with
Goldner’s trichrome with the semiautomatic method at
ϫ200.
● The wall width (Cn-W.Wi): Represents the amount of
bone formed in any forming center at the intermediary
level of organization of bone and is the width of a
complete cancellous bone packet. It was measured un-
der polarized light to differentiate the newly formed
bone, on Solochrome Cyanin stained sections with the
semiautomatic image analyzer at ϫ80.
● The number of nodes (N.Nd); number termini (N.Tm);
the length of struts between nodes (Nd.Nd), between
termini (Tm.Tm), or between nodes and termini
(Nd.Tm); and the total strut length (TSL) were ob-
tained after skeletonization of the trabecular network
by an automatic method. The ratio of nodes to termini
(N.Nd/N.Tm), which is an index of spatial connectiv-
ity, was calculated.(20)
Cortical bone histomorphometry: The number of active
endhaversian sites with either resorption or formation was
expressed per unit of cortical bone area (Ct-N.Ca(aϩ)/B.Ar,
per mm2). The endhaversian perimeters occupied by erosion
cavities (Ct-E.Pm/Ca.Pm, %), osteoid seams (Ct-O.Pm/
Ca.Pm, %), and single (Ct-sL.Pm/Ca.Pm) and double (Ct-
dL.Pm/Ca.Pm) labels were measured and expressed in per-
cent of the total endhaversian perimeters.
The following dynamic measurements were performed in
cancellous bone on unstained sections with the semiauto-
matic image analyzer:
The cortical porosity (Ct-Po/B.Ar, %) represents the per-
centage of cortical bone area occupied by Haversian canals.
The cortical width (Ct-Wi; ␮m) represents the mean
width of the external and internal cortices.
● The mineral apposition rate (Cn-MAR) expressed in
micrometers per day, which is the rate of progression
of the mineralization front. All double-labeled foci
observed were measured according to Kragstrup’s
method,(19) at ϫ320.
● The single (Cn-sL.Pm/B.Pm)- and double (Cn-dL.Pm/
B.Pm)-labeled perimeters are expressed as the percent-
age of total cancellous perimeters covered by a single
or a double fluorochrome label. They were measured at
ϫ80. The mineralizing perimeter (M.Pm/B.Pm) was
calculated as dL.Pm ϩ 1/2sL.Pm.
BMD
After death, femurs and lumbar spine were taken for the
evaluation of BMD by dual X-ray analysis using the per-
formance speed mode analysis (1000ϩ; Hologic, Waltham,
MA, USA) and 16-cm rice bed as soft tissue equivalent.
● The bone formation rate (Cn-BFR/B.Pm) giving the
amount of mineralized bone formed per unit of bone
surface area of trabecular bone per day. It was calcu-
lated as MAR ϫ M.Pm and is expressed in square
micrometers per micrometer per day.
Statistical methods
All results were expressed as the mean Ϯ SEM. For each
biochemical parameter, a two-way analysis of variance
● The adjusted apposition rate (Cn-Aj.AR) representing (ANOVA) was performed to test the effect of (1) time, that is,
the amount of mineralized bone made per unit of whether there was a difference between pre-OVX and 1, 2, 3,
osteoid-covered perimeter per day. It was calculated as 4, 5, and 6 months after OVX; and (2) treatment, that is,
BFR/B.Pm divided by O.Pm/B.Pm and is expressed in whether there was a difference between sham, OVX, OVX ϩ
micrometers per day.
L, and MDL-treated groups. Mann-Whitney’s U test was used
● The mineralization lag time (Cn-Mlt) in cancellous for the comparison of two means. A Wilcoxon signed rank test
bone expresses the period of time between deposition was performed for the comparison between paired values.
and subsequent mineralization of osteoid. It was cal-
Spearman correlation was performed between biochemi-
culated as O.Wi over Aj.AR and is expressed in days. cal and static parameters.

92
CHAVASSIEUX ET AL.
FIG. 1. Effects of ovariectomy and aromatase inhibition
by L on urinary CTX (f, sham; ‚, OVX; Œ, OVX ϩ L).
*p Ͻ 0.05; **p Ͻ 0.01; Mann-Whitney U test versus sham
at the same time.
FIG. 2. Effects of OVX and aromatase inhibition by L on
serum OC (f, sham; ‚, OVX; Œ, OVX ϩ L).*p Ͻ 0.05;
**p Ͻ 0.01; Mann-Whitney U test versus sham at the same
time.
RESULTS
At the beginning of the experiment, biochemical and
histomorphometric parameters were similar in all groups. In
the sham group, biochemical markers of bone remodeling
tended to decrease between the beginning and the end of the
experiment, reflecting the seasonal variation of bone remod-
eling in ewes. OC and BAP were significantly correlated
with, respectively, cancellous and cortical parameters of
bone formation. A significant correlation was found be-
tween CTX and Ct-E.Pm/B.Pm (p Ͻ 0.03).
Effects of OVX
Biochemical parameters: In contrast to the decrease ob-
served in the sham group, biochemical markers of bone
resorption and formation increased after OVX. These vari-
ations were significant for the time and treatment for CTX
(p Ͻ 0.05 and p Ͻ 0.0001, respectively) and OC (p Ͻ 0.01
and p Ͻ 0.0001, respectively), only for the treatment for
BAP (p Ͻ 0.0001). When compared with the sham group,
CTX was significantly higher in OVX and OVX ϩ L groups
at each time after OVX (Fig. 1). The highest increase in
CTX was noted after 3 months (ϩ156% of sham value).
The markers of bone formation tended to increase after
OVX (Figs. 2 and 3). The highest increases in OC and BAP
FIG. 3. Effects of OVX and aromatase inhibition by L on
serum BAP (f, sham; ‚, OVX; Œ, OVX ϩ L).*p Ͻ 0.05;
Mann-Whitney U test versus sham at the same time.
were noted after the fourth month (ϩ74% and ϩ53% when perimeters were much higher in OVX groups than in sham
compared with the sham values for OC and BAP, respec- but because of the large individual variations, the difference
tively). However, because of the wide range of BAP values was not significant.
in each group, only OC was significantly higher in OVX
In contrast, no significant effect of OVX was noted on
groups than in sham, between 3 and 6 months after OVX cancellous bone (Tables 2–4) except a trend in an increase
(p ϭ 0.03). L tended to slightly amplify the effects of OVX in N.Oc/B.Pm. Between the beginning and the end of the
on bone formation markers.
treatment, similar variations of bone structure and static
Bone histomorphometry: Six months after OVX, the remodeling parameters were noted in each group (Cn-BV/
number of active endhaversian sites, the cortical eroded and TV, Ϫ14% and Ϫ18%; Cn-O.Pm/B.Pm, ϩ6% and Ϫ1% in
osteoid perimeters, and the cortical porosity were signifi- sham and OVX ϩ L groups, respectively). However, when
cantly increased (p Ͻ 0.02, p Ͻ 0.003, p Ͻ 0.02, and p Ͻ compared with the pretreatment values, Cn-M.Pm/B.Pm
0.02, respectively). L did not significantly amplify the ef- varied by Ϫ10% and ϩ8%, Cn-Ac.f by ϩ15% and ϩ68%
fects of OVX (Table 1). The mean values of mineralizing in sham and OVX ϩ L groups, respectively.

EFFECTS OF SERM ON BONE IN OVARIECTOMIZED EWES
93
TABLE 1. EFFECTS ON CORTICAL BONE OF OVX, AROMATASE INHIBITION BY L AND MDL 103,323
AFTER 6 MONTHS (MEAN Ϯ SEM)
OVX ϩ L ϩ
MDL 0.1 (8)
OVX ϩ L ϩ
MDL 1 (8)
Group (n)
Ct-Wi (␮m)
Sham (7)
OVX (8)
OVX ϩ L (8)
487.84 Ϯ 40.34
444.73 Ϯ 40.21
418.47 Ϯ 32.27
455.46 Ϯ 19.72
427.68 Ϯ 20.02
Ct-N.Ca(aϩ)/B.Ar
(per mm2)
0.0004 Ϯ 0.0001
1.30 Ϯ 0.30
2.70 Ϯ 1.17
2.06 Ϯ 1.05
0.87 Ϯ 0.51
1.90 Ϯ 0.88
0.97 Ϯ 0.32
0.0017 Ϯ 0.0010
9.75 Ϯ 2.10†
8.74 Ϯ 1.70*
5.27 Ϯ 1.53
8.13 Ϯ 3.24*
10.77 Ϯ 3.79
2.77 Ϯ 0.78†
0.0010 Ϯ 0.0003*
7.72 Ϯ 2.10†
16.42 Ϯ 4.94†
4.28 Ϯ 1.26
10.57 Ϯ 5.08
12.71 Ϯ 5.59
3.63 Ϯ 0.67†
0.0004 Ϯ 0.0001‡
3.60 Ϯ 0.65
8.01 Ϯ 2.08
4.77 Ϯ 1.80
3.30 Ϯ 2.38
5.69 Ϯ 2.76
1.34 Ϯ 0.36§
0.0004 Ϯ 0.0001‡
6.17 Ϯ 0.88
Ct-E.Pm/Ca.Pm (%)
Ct-O.Pm/Ca.Pm (%)
Ct-sL.Pm/Ca.Pm (%)
Ct-dL.Pm/Ca.Pm (%)
Ct-M.Pm/Ca.Pm (%)
Ct-Po/B.Ar (%)
5.42 Ϯ 0.78‡
2.04 Ϯ 0.91
0.97 Ϯ 0.58‡
2.00 Ϯ 0.88
1.36 Ϯ 0.27§
* p Ͻ 0.05; † p Ͻ 0.01; Mann-Whitney U test versus sham.
‡ p Ͻ 0.05; § p Ͻ 0.01; Mann-Whitney U test versus OVX ϩ L.
TABLE 2. EFFECTS ON CANCELLOUS BONE MINERALIZATION OF OVX, AROMATASE INHIBITION BY L AND MDL 103,323
AFTER 6 MONTHS (MEAN Ϯ SEM)
OVX ϩ L ϩ
OVX ϩ L ϩ
Group (n)
Sham (7)
OVX (8)
OVX ϩ L (8)
MDL 0.1 (8)
MDL 1 (8)
Cn-O.Wi (␮m)
7.7 Ϯ 0.3
0.7 Ϯ 0.3
1.32 Ϯ 0.02
14.3 Ϯ 2.0
7.7 Ϯ 0.4
0.8 Ϯ 0.2
1.21 Ϯ 0.06
16.7 Ϯ 2.6
8.2 Ϯ 0.6
1.0 Ϯ 0.3
1.20 Ϯ 0.04
18.0 Ϯ 1.2
6.3 Ϯ 0.4*
0.3 Ϯ 0.07*
1.16 Ϯ 0.07
17.5 Ϯ 3.4
6.0 Ϯ 0.4*
0.2 Ϯ 0.05*
1.15 Ϯ 0.14
24.9 Ϯ 5.3
Cn-O.Ar/B.Ar (%)
Cn-MAR (␮m/day)
Cn-Mlt (days)
* p Ͻ 0.05 Mann-Whitney U test versus OVX ϩ L.
TABLE 3. EFFECTS ON CANCELLOUS BONE REMODELING OF OVX, AROMATASE INHIBITION BY L AND MDL 103,323
AFTER 6 MONTHS (MEAN Ϯ SEM)
OVX ϩ L ϩ
OVX ϩ L ϩ
Group (n)
Sham (8)
OVX (8)
OVX ϩ L (8)
MDL 0.1 (8)
MDL 1 (8)
Cn-O.Pm/B.Pm (%)
Cn-Ob.Pm/B.Pm (%)
Cn-N.Oc/B.Pm
7.2 Ϯ 2.2
2.3 Ϯ 0.8
7.8 Ϯ 1.7
4.0 Ϯ 1.2
7.8 Ϯ 2.6
4.3 Ϯ 1.7
3.7 Ϯ 0.7
1.5 Ϯ 0.3
2.8 Ϯ 0.9*
0.5 Ϯ 0.2*
(per mm)
0.028 Ϯ 0.007
6.2 Ϯ 1.72
3.5 Ϯ 1.2
0.046 Ϯ 0.016
0.607 Ϯ 0.076
0.44 Ϯ 0.14
68.8 Ϯ 9.9
0.035 Ϯ 0.015
5.9 Ϯ 1.3
3.1 Ϯ 0.8
0.043 Ϯ 0.01
0.529 Ϯ 0.067
0.43 Ϯ 0.10
75.5 Ϯ 8.9
0.036 Ϯ 0.017
4.3 Ϯ 0.9
3.6 Ϯ 1.2
0.069 Ϯ 0.012
1.080 Ϯ 0.353
0.67 Ϯ 0.12
73.1 Ϯ 15.8
32.8 Ϯ 2.1
0.006 Ϯ 0.003
4.1 Ϯ 0.8
1.5 Ϯ 0.5*
0.022 Ϯ 0.005*
0.525 Ϯ 0.119
0.23 Ϯ 0.05*
96.3 Ϯ 20.0
32.7 Ϯ 2.6
0.017 Ϯ 0.009
7.5 Ϯ 2.7
Cn-E.Pm/B.Pm (%)
Cn-M.Pm/B.Pm (%)
Cn-BFR (␮m3/␮m2/d)
Cn-Aj.AR (␮m/day)
Cn-Ac.f (per year)
Cn-FP (days)
0.7 Ϯ 0.4*
0.012 Ϯ 0.005†
0.346 Ϯ 0.064†
0.11 Ϯ 0.05†
145.8 Ϯ 30.0
37.6 Ϯ 5.6
Cn-FP (aϩ) (days)
27.9 Ϯ 0.9
30.6 Ϯ 1.9
* p Ͻ 0.05; † p Ͻ 0.02; Mann-Whitney U test versus OVX ϩ L.
BMD: After OVX, the BMD of the femur was signifi- per day (Figs. 5 and 6). At the dose of 1 mg/kg per day,
cantly decreased (p Ͻ 0.01). The administration of L in-
duced a significant decrease in the lumbar BMD (p Ͻ 0.015;
Fig. 4)
MDL administration significantly decreased CTX, BAP,
and OC after the first month of treatment (Figs. 5–7). At the
end of treatment, all biochemical markers returned to the
sham levels.
Bone histomorphometry: Cancellous bone tissue: (1) Ef-
fect on mineralization (Table 2): At the end of treatment,
MDL induced significant decreases in osteoid thickness and
volume when compared with OVX ϩ L group. No change
Effects of MDL treatment
Biochemical parameters: When compared with OVX ϩ
L group, MDL treatment induced a significant decrease of
CTX and OC, after the third month at the dose of 0.1 mg/kg in MAR was observed. (2) Effects on bone turnover (Table

94
CHAVASSIEUX ET AL.
TABLE 4. EFFECTS ON CANCELLOUS BONE STRUCTURE AND MICROARCHITECTURE OF OVX, AROMATASE INHIBITION BY L
AND MDL 103,323 AFTER 6 MONTHS (MEAN Ϯ SEM)
OVX ϩ L ϩ
OVX ϩ L ϩ
Group (n)
Sham (7)
OVX (8)
OVX ϩ L (8)
MDL 0.1 (8)
MDL 1 (8)
Cn-W.Wi (␮m)
Cn-B.Ar/T.Ar (%)
Tb.Th (␮m)
36.7 Ϯ 0.9
13.6 Ϯ 0.8
92.7 Ϯ 0.1
626.9 Ϯ 33.3
1.40 Ϯ 0.06
1.49 Ϯ 0.9
1.1 Ϯ 0.1
1.8 Ϯ 0.2
40.8 Ϯ 4.9
9.3 Ϯ 3.7
36.9 Ϯ 1.2
10.8 Ϯ 0.9
90.0 Ϯ 1.2
759.8 Ϯ 50.6
1.20 Ϯ 0.07
1.35 Ϯ 0.11
0.6 Ϯ 0.1
39.2 Ϯ 1.2
13.9 Ϯ 1.2
105.2 Ϯ 4.8
655.4 Ϯ 39.9
1.31 Ϯ 0.09
1.33 Ϯ 0.10
0.8 Ϯ 0.1
36.9 Ϯ 0.9
11.3 Ϯ 0.8
96.6 Ϯ 4.7
742.6 Ϯ 51.8
1.23 Ϯ 0.07
1.31 Ϯ 0.12
0.8 Ϯ 0.2
38.4 Ϯ 1.2
12.4 Ϯ 0.8
98.9 Ϯ 5.7
709.2 Ϯ 22.6
1.25 Ϯ 0.03
1.27 Ϯ 0.04
0.6 Ϯ 0.1
Tb.Sp (␮m)
Tb.N (per mm)
TSL/T.Ar (mm/mm2)
N.Nd/T.Ar (per mm2)
N.Tm/T.Ar (per mm2)
Nd.Nd/TSL (%)
Tm.Tm/TSL (%)
Nd.Tm/TSL (%)
N.Nd/N.Tm
1.4 Ϯ 0.1
1.6 Ϯ 0.2
1.7 Ϯ 0.1
1.7 Ϯ 0.2
33.9 Ϯ 6.0
15.3 Ϯ 4.0
26.5 Ϯ 2.9
0.57 Ϯ 0.16
33.2 Ϯ 4.7
13.1 Ϯ 5.2
35.5 Ϯ 3.1
0.52 Ϯ 0.07
34.3 Ϯ 7.7
15.3 Ϯ 4.0
33.8 Ϯ 3.1
0.49 Ϯ 0.11
22.0 Ϯ 4.1
13.8 Ϯ 3.8
38.9 Ϯ 5.0
0.39 Ϯ 0.06
35.9 Ϯ 3.7
0.63 Ϯ 0.08
FIG. 4. Effects of OVX, aromatase inhibition by L, and
MDL 103,323 on the BMD of the lumbar vertebrae (`) and
femur (o) after 6 months. *p Ͻ 0.05, **p Ͻ 0.01; Mann-
Whitney U test versus sham.
FIG. 5. Effects of MDL 103,323 on urinary CTX (Œ,
OVX ϩ L; F, OVX ϩ L ϩ MDL 0.1; E, OVX ϩ L ϩ
MDL 1).*p Ͻ 0.05; **p Ͻ 0.01; Mann-Whitney U test
versus OVX ϩ L at the same time.
3): The osteoclast number decreased after MDL treatment
but because of the interanimal variation, the difference was
not significant. Osteoid and osteoblastic perimeters were
significantly decreased with the highest dose. The mineral-
izing perimeter, BFR, and activation frequency significantly
DISCUSSION
OVX induced an increase in biochemical markers in
decreased whatever the dose of MDL. The total formation ewes, which depended on the time after OVX. CTX was
period tended to increase but without any change in the significantly increased since the first month after OVX. A
active formation period. (3) Effects on bone structure (Table significant increase in OC was noted only after the third
4): Bone volume, wall thickness, and parameters of micro- month. The mean values of BAP showed the same trend but
architecture were unchanged after 6 months of treatment.
because of the large range of individual values, the varia-
Bone histomorphometry: Cortical bone tissue (Table 1): tions were not significant. These changes of biochemical
As on cancellous bone, MDL 103,323 induced a dose- markers in OVX ewes confirmed those observed in previous
dependent decrease in cortical parameters, which tended to studies.(6,9) These changes mainly reflected the increase in
return to the sham levels. At the dose of 1 mg/kg per day, cortical bone histomorphometric parameters because no sig-
osteoid and mineralizing perimeters decreased, respectively, nificant effect of OVX was observed on cancellous bone.
by 68% and 84% when compared with OVX ϩ L group.
However, the higher increase in Cn-Ac.f between the be-
BMD: After 6 months, MDL treatment tended to dose- ginning and the end of the treatment in OVX groups than in
dependently increase the femoral and spinal BMD but the the sham group suggested a slight effect of OVX on can-
difference was not significant (Fig. 4).
cellous bone. However, because of the short duration of the

EFFECTS OF SERM ON BONE IN OVARIECTOMIZED EWES
95
The effects of MDL 103,323, a new selective estrogen
MDL, were characterized on cancellous bone by a decrease
in osteoid volume and thickness without any change in
mineralization. This accounted for by a decrease in the rate
of bone turnover. This is confirmed by the decrease in
mineralizing perimeter by 58% and 80% and activation
frequency by 66% and 84% at 0.1 mg/kg per day and 1
mg/kg per day, respectively. Consequently, the BFR at the
tissue level was reduced. Similar effects of MDL 103,323
also were observed on cortical bone. All cortical histomor-
phometric parameters tended to return to the sham levels.
This is confirmed by biochemical changes that also tended
to decrease to the sham levels. However, no change in bone
volume and bone microarchitecture could be observed. The
absence of significant effects on bone structure parameters
probably was caused by the too short duration of the treat-
ment and by the small number of animals in each group.
These results confirm earlier studies reporting that OVX and
estrogen treatment did not induce early detectable change in
bone volume and microarchitecture in ewes.(8,9) However,
BMD of the lumbar spine tended to increase and return to
sham values after MDL treatment. This confirms that MDL
103,323 mimics the effects of estrogen on bone tissue. In
OVX ewes, estradiol significantly increased the BMD of the
lumbar spine but did not change bone histomorphometric
parameters on iliac crest.(8,9) Other SERMs have been
shown to act as an estrogen agonist in bone and an antag-
onist in uterus. Raloxifene decreased bone turnover and
increased the BMD in animals and postmenopausal wom-
en.(22,23) Droloxifene(24–26) and idoxifene(27) have been re-
ported to prevent bone loss in OVX rats.
FIG. 6. Effects of MDL 103,323 on serum OC (Œ, OVX
ϩ L; F, OVX ϩ L ϩ MDL 0.1; E, OVX ϩ L ϩ MDL
1).*p Ͻ 0.05; **p Ͻ 0.01; ***p Ͻ 0.005; Mann-Whitney U
test versus OVX ϩ L at the same time.
In conclusion, OVX induced an increase in cortical bone
turnover in ewes, confirmed by an increase in bone markers
(CTX and OC), but no effect of OVX has been detected on
cancellous bone. MDL 103,323, a new SERM, prevented
the effects of OVX by inhibiting both cortical and cancel-
lous bone turnover as shown by the biochemical markers,
the histomorphometric parameters on iliac crest biopsy
specimens, and the BMD of the lumbar spine. Our results
suggest that MDL 103,323 may be a potentially useful agent
in the prevention of postmenopausal bone loss.
FIG. 7. Effects of MDL 103,323 on serum BAP (Œ, OVX
ϩ L; F, OVX ϩ L ϩ MDL 0.1; E, OVX ϩ L ϩ MDL
1).*p Ͻ 0.05; **p Ͻ 0.01; ***p Ͻ 0.005; Mann-Whitney U
test versus OVX ϩ L at the same time.
REFERENCES
1. Chavassieux P, Pastoureau P, Boivin G, Delmas PD, Chapuy
MC, Meunier PJ 1990 Effects on ewe bone remodeling of a
low dose of disodium monofluorophosphate. Influence of a
calcium supplement. In: Christiansen C, Overgaard K (eds.)
Osteoporosis 1990. Osteopress ApS, Copenhagen, Denmark,
pp. 721–723.
2. Chavassieux P, Pastoureau P, Boivin G, Chapuy MC, Delmas
PD, Milhaud G, Meunier PJ 1991 Fluoride-induced bone
changes in lambs during and after exposure to sodium fluoride.
Osteoporos Int 2:26–33.
3. Chavassieux P, Pastoureau P, Boivin G, Chapuy MC, Delmas
PD, Meunier PJ 1991 Dose effects on ewe bone remodeling of
short-term sodium fluoride administration—a histomorpho-
metric and biochemical study. Bone 12:421–427.
4. Chavassieux P, Pastoureau P, Chapuy MC, Delmas PD,
Meunier PJ 1993 Glucocorticoid-induced inhibition of osteo-
blastic bone formation in ewes: A biochemical and histomor-
phometric study. Osteoporos Int 3:97–102.
treatment and the interindividual variation, no significant
difference could be noted. The apparent discrepancy be-
tween biochemical markers and cancellous bone parameters
confirmed that biochemical markers reflect the entire skel-
eton, including cortical and trabecular bone. The loss of
estrogen secretion after OVX induced an increase in bone
turnover as shown by the increase in bone markers. This
effect predominates on cortical bone. A site-specific action
of estrogen on the skeleton with a difference between cor-
tical and cancellous bone already have been suggested.(21)
Furthermore, in ewes, a large part of estrogen secretion
results from adrenal synthesis and peripheral aromatization.
The administration of L, a specific inhibitor of the periph-
eral aromatase, tended to amplify slightly the effects of
OVX on bone markers.

96
5. Chavassieux P, Buffet A, Vergnaud P, Garnero P, Meunier PJ
1997 Short-term effects of corticosteroids on trabecular bone
remodeling in old ewes. Bone 20:451–455.
6. Pastoureau P, Arlot ME, Caulin F, Barlet JP, Meunier PJ,
Delmas PD 1989 Effects of oophorectomy on biochemical and
CHAVASSIEUX ET AL.
pleted remodeling sites in iliac trabecular bone. Metab Bone
Dis Rel Res 4:113–119.
20. Compston JE, Mellish R, Groucher P, Newcombe R 1989
Structural mechanisms of trabecular bone loss in man. Bone
Miner 6:339–350.
histological indices of bone turnover in ewes. J Bone Miner 21. Turner RT, Riggs BL, Spelberg TC 1994 Skeletal effects of
Res 4;S1:S237. estrogen. Endocr Rev 15:275–300.
7. Hornby SB, Ford SL, Mase CA, Evans GP 1995 Skeletal 22. Black LJ, Masahiko S, Rowley ER, Magee DE, Bekele A,
changes in the ovariectomized ewe and subsequent response to
treatment with 17␤ oestradiol. Bone 17:389S–394S.
8. Turner AS, Mallinckrodt CH, Alvis MR, Bryant HU 1995
Dose-response effects of oestradiol implants on bone mineral
density in ovariectomized ewes. Bone 17:421S–427S.
Williams DC, Cullinan GJ, Bendele R, Kauffman RF, Bensch
WR, Frolick CA, Termine JD, Bryant HU 1994 Raloxifene
(LY 139481 HCl) prevents bone loss and reduces serum cho-
lesterol without causing uterine hypertrophy in ovariectomized
rats. J Clin Invest 93:63–69.
9. Turner AS, Alvis MR, Mallinckrodt CH, Myers W, Stevens 23. Delmas PD, Bjarnason NH, Mitlak BH, Ravoux AC, Shah AS,
ML, Lundy MW 1995 Changes in bone mineral density and
bone-specific alkaline phosphatase in ovariectomized ewes.
Bone 17:395S–402S.
Huster WJ, Draper M, Christiansen C 1997 Effects of ralox-
ifene on bone mineral density, serum cholesterol concentra-
tions, and uterine endometrium in postmenopausal women.
N Engl J Med 337:1641–1647.
10. Goldstein SR 1998 Selective estrogen receptor modulators: A
new category of therapeutic agents for extending the health of 24. Chen HK, Ke HZ, Jee WSS, Ma YF, Pirie CM, Simmons HA,
postmenopausal women. Am J Obstetr Gynecol 179:1479–
1484.
11. Gustafsson JA 1998 Therapeutic potential of selective estro-
gen receptor modulators. Curr Opin Chem Biol 2:508–511.
Thompson DD 1995 Droloxifene prevents ovariectomy-
induced bone loss in tibiae and femora of aged female rats: A
dual-energy X-ray absorptiometric and histomorphometric
study. J Bone Miner Res 10:1256–1262.
12. Ammann P, Bourrin S, Bonjour JP, Brunner F, Meyer JM, 25. Ke HZ, Chen HK, Qi H, Pirie CM, Simmons HA, Ma YF, Jee
Rizzoli R 1999 The new selective estrogen receptor modulator
MDL 103,323 increases bone mineral density and bone
strength in adult ovariectomized rats. Osteoporos Int 10:369–
376.
WSS, Thompson DD 1995 Effects of droloxifene on preven-
tion of cancellous bone loss and bone turnover in the axial
skeleton of aged, ovariectomized rats. Bone 17:491–496.
26. Ke HZ, Simmons HA, Pirie CM, Crawford DT, Thompson DD
1995 Droloxifene, a new estrogen antagonist/agonist, prevents
bone loss in ovariectomized rats. Endocrinology 136:2435–
2441.
13. Pastoureau P, Merle B, Delmas PD 1988 Specific radioimmu-
noassay for ovine bone-Gla protein. Acta Endocrinol (Copenh)
113:152–160.
14. Garnero P, Delmas PD 1993 Assessment of the serum levels of 27. Nuttall ME, Bradbeer JN, Stroup GB, Nadeau DP, Hoffman
bone alkaline phosphatase with a new immunoradiometric
assay in patients with metabolic bone disease. J Clin Endocri-
nol Metab 77:1046–1053.
SJ, Zhao H, Rehm S, Gowen M 1998 Idoxifene: A novel
selective estrogen receptor modulator prevents bone loss and
lowers cholesterol levels in ovariectomized rats and decreases
uterine weight in intact rats. Endocrinology 139:5224–5234.
15. Bonde M, Quist P, Fledelius C, Riis BJ, Christiansen C 1994
Immunoassay for quantifying type I collagen degradation
products in urine evaluated. Clin Chem 40:2022–2025.
16. Chavassieux P, Arlot M, Meunier PJ 1996 Clinical use of bone
biopsy. In: Marcus R, Feldman D, Kelsey J (eds.) Osteoporo-
sis. Academic Press, San Diego, CA, USA, pp. 1113–1121.
17. Roux JP, Arlot ME, Robert JP, Meunier PJ 1996 A new
automatic method for the analysis of 2-D trabecular bone
connectivity. Bone 19(Suppl):160S.
18. Parfitt AM, Drezner MK, Glorieux FH, Kanis JA, Malluche H,
Meunier PJ, Ott S, Recker RR 1987 Bone histomorphometry:
Standardization of nomenclature, symbols and units. J Bone
Miner Res 2:595–610.
Address reprint requests to:
Pascale M. Chavassieux, M.D.
INSERM Unite´ 403
Faculte´ de Me´decine R.T.H. Lae¨nnec
Rue G. Paradin
69372 Lyon Cedex 08, France
19. Kragstrup J, Gundersen HJG, Melsen F, Mosekilde L 1982 Received in original form October 4, 1999; in revised form March
Estimation of the three-dimensional wall thickness of com- 3, 2000; accepted May 25, 2000.