Full text of DOI:10.1359/jbmr.2001.16.2.386

JOURNAL OF BONE AND MINERAL RESEARCH
Volume 16, Number 2, 2001
©
2001 American Society for Bone and Mineral Research
Postnatal and Pubertal Skeletal Changes Contribute
Predominantly to the Differences in Peak Bone Density
Between C3H/HeJ and C57BL/6J Mice*
1
1
1
1
2
2
C. RICHMAN, S. KUTILEK, N. MIYAKOSHI, A.K. SRIVASTAVA, W.G. BEAMER, L.R. DONAHUE,
3
1
1
1
C.J. ROSEN, J.E. WERGEDAL, D.J. BAYLINK, and S. MOHAN
ABSTRACT
Previous studies have shown that 60–70% of variance in peak bone density is determined genetically. The
higher the peak bone density, the less likely an individual is to eventually develop osteoporosis. Therefore, the
amount of bone accrued during postnatal and pubertal growth is an important determining factor in the
development of osteoporosis. We evaluated the contribution of skeletal changes before, during, and after
puberty to the development of peak bone density in C3H/HeJ (C3H) and C57BL/6J (B6) mice. Volumetric
bone density and geometric parameters at the middiaphysis of femora were measured by peripheral quan-
titative computed tomography (pQCT) from days 7 to 56. Additionally, biochemical markers of bone
remodeling in serum and bone extracts were quantified. Both B6 and C3H mice showed similar body and
femoral weights. B6 mice had greater middiaphyseal total bone area and thinner cortices than did C3H mice.
Within strains, males had thicker cortices than did females. C3H mice accumulated more minerals throughout
the study, with the most rapid accumulation occurring postnatally (days 7–23) and during pubertal matura-
tion (days 23–31). C3H mice had higher volumetric bone density as early as day 7, compared with B6 mice.
Higher serum insulin-like growth factor I (IGF-I) was present in C3H mice postnatally at day 7 and day 14.
Until day 31, B6 male and female mice had significantly higher serum osteocalcin than C3H male and female
mice, respectively. Alkaline phosphatase (ALP) was found to be significantly higher in the bone extract of C3H
mice compared with B6 mice at day 14. These data are consistent with and support the hypothesis that the
greater amount of bone accrued during postnatal and pubertal growth in C3H mice compared with B6 mice
may be caused by increased cortical thickness, increased endosteal bone formation, and decreased endosteal
bone resorption. (J Bone Miner Res 2001;16:386–397)
Key words: bone density, puberty, insulin-like growth factor I, osteocalcin, C3H/HeJ and C57BL/6J mice
INTRODUCTION
during the active growth phases early in life. This accumu-
lation of bone mass determines the peak bone density at-
(1,2)
HE RISK of developing senile osteoporosis in men and tained during the third decade of life.
About 60–70% of
(3)
T
postmenopausal osteoporosis in women is in large mea- variance in peak bone density is determined genetically,
sure determined by the amount of bone mass accumulated and 40–50% of peak bone density is accumulated during
(4)
puberty. Elucidation of the mechanisms regulating this
dramatic increase in skeletal mass during puberty is impor-
*
The view, opinions, and/or findings contained in this report are
those of the author(s) and should not be construed as a position, tant in the effort to identify potential preventive or inter-
policy, decision, or endorsement of the Federal Government or the ventive measures that would lower the risk of developing
National Medical Technology Testbed, Inc.
osteoporosis.
1
J.L. Pettis Veterans Administration Medical Center and Loma Linda University, Loma Linda, California, USA.
The Jackson Laboratories, Bar Harbor, Maine, USA.
St. Joseph Hospital, Bangor, Maine, USA.
2
3
3
86

SKELETAL CHANGES IN MICE DURING PUBERTY
387
In humans, peak bone density is attained between 20 and
MATERIALS AND METHODS
(1,2)
3
0 years of age.
The most rapid accumulation in bone
mass occurs in postnatal growth, from birth roughly to 3
years of age, and during puberty, from 12 to 18 years of
Recombinant human (rh) IGF-I (rhIGF-I) was a gift from
Ciba-Geigy (Basel, Switzerland); rhIGF-II was purchased
from BACHEM Chemicals (Torrance, CA, USA). Rabbit
polyclonal antiserum against IGF-I (kindly provided by L.E.
Underwood and J.J Van Wyk) was obtained form the Na-
tional Hormone and Pituitary Program (Baltimore, MD,
USA). Osteocalcin synthetic peptide was obtained from
SynPep Corp. (Dublin, CA, USA). Paranitrophenol phos-
phate was purchased from Fluka (Buchs, Switzerland). All
other chemicals were enzyme grade and purchased from
Fisher Scientific (Tustin, CA, USA) or Sigma Chemical Co.
(5)
age. This increase in bone mass closely correlates with an
increase in body weight and height. Puberty is initiated 2
years earlier in girls than in boys and during this period of
pubertal maturation, a sexual dimorphism is established.
The pubertal growth spurt lasts roughly 2 years longer in
boys than in girls, such that boys are taller, stronger, and
(6)
exhibit higher peak bone mass than do girls.
The most beneficial period for osteoporosis prevention
may be during the rapid growth phases that occur within the
(7)
first 20 years of life. The attainment of higher peak bone
mass during this period should lead to reduced risk for
fracture later in life by increasing peak bone mass.
(St. Louis, MO, USA).
In vivo work
In this study we are particularly interested in the different
rates of bone remodeling occurring before, during, and after
puberty. In studies comparing bone mass accumulation,
Slemenda et al. found that higher peak bone mass in black
girls is associated with lower bone turnover during puberty
Animals: Inbred C3H and B6 breeder mice were pur-
chased from The Jackson Laboratory (Bar Harbor, ME,
USA). On receipt, the mice were allowed to acclimate to the
new environment for 7 days and were then mated to produce
the pups used in this study.
Animal protocol: The female breeders were housed indi-
vidually several days before expected parturition and fed a
breeders’ diet (10% fat content; Harlan Teklad, Placentia,
CA, USA). The pups were weaned and weighed on postna-
tal day 21 using a digital scale (ACCULAB V-600) to the
nearest 0.1 g. At weaning, the pups were fed regular rodent
chow (4% fat content; Harlan Teklad) and housed 4–5 pups
per cage with cages of females alternating with cages of
males on each rack to eliminate uneven pheromone distri-
bution. Pubertal maturation in B6 female mice begins when
(8)
compared with that of white girls. During puberty, a
complex series of events occurs that impacts the changes in
bone turnover and attainment of peak bone mass, such as
periosteal expansion and endosteal contraction, longitudinal
(
9)
growth, modeling-dependent remodeling,
hormone–dependent reduction in bone turnover or remod-
eling.
and sex
(10)
With the development of assays for serum/urine
levels of bone formation and resorption markers, the mea-
surement of bone turnover in different conditions of growth
or disease becomes possible. During growth and in osteo-
(11–14)
lytic diseases, a high bone turnover rate exists.
serum estradiol increases on day 26 and vaginal opening
Studies using mice as a model system have shown that
mice, like humans, attain a peak bone density (at 4–6
months of age). Some strains experience age-related bone
(22)
occurs by day 31.
In humans, puberty starts roughly 2
years later in males than in females and ends later in males
(23)
than females.
We calculated that in mice, this difference
(15,16)
loss (1 year and older) leading to osteopenia.
Different
would constitute a 2–3 day delay for males compared with
females, so that puberty should be completed by day 35 in
males. Based on this data, we chose to collect pups at days
strains of inbred mice have been shown to exhibit different
(
17)
peak bone densities
ectomy.
and responded differently to ovari-
(18)
Similar to human studies using monozygotic
7
, 14, and 23 (before puberty starts); day 31 (at the end of
and dizygotic twins, different strains of inbred mice provide
genetically identical animals to dissect out the role of indi-
vidual genes and their products in the development of a trait
or disease. However, unlike the design restraints on sample
size and genetic diversity in humans, mice provide a rela-
tively unlimited number of genetically identical subjects.
Two inbred strains of mice, C3H/HeJ (C3H) and C57BL/6J
puberty for the females); day 35 (at the end of puberty for
the males); and on days 42, 49, and 56. Pups were killed by
CO inhalation (dry ice) followed by decapitation. Whole
2
blood was collected, allowed to clot, and then centrifuged.
The serum was skimmed off and stored at Ϫ70°C until
assayed. The left femur was dissected free of soft tissue
keeping the condyles and femoral neck intact. This bone
was used for the bone extract assay to determine bone-
specific alkaline phosphatase (ALP) activity and bone size
data. The right femur and proximal tibia were dissected free
and stored in 70% ethanol for volumetric bone density and
geometric parameter determination by peripheral quantita-
(B6), which have the same body size and weight, previously
have been shown to have very different peak bone
(17,19–21)
densities.
Therefore, we chose to compare these
two strains to study the following: (1) the relative contribu-
tion of postnatal (from birth until puberty), pubertal, and
postpubertal growth to the final peak bone density of each tive computed tomography (pQCT) using an XCT Research
strain; (2) the differences in bone remodeling between these M instrument (Norland Medical Systems, Inc., Fort Atkin-
two strains that may contribute to these different bone son, WI, USA). The experimental procedures performed in
densities; (3) the differences in serum level of insulin-like this study comply with the National Institutes of Health
growth factors (IGFs) between these two strains during (NIH) guide for the Care and Use of Laboratory Animals.
postnatal, pubertal, and postpubertal growth; and (4) to All animals studies were reviewed and monitored by the
determine if and when the sexual dimorphism in peak bone Animal Studies Subcommittee at the Jerry L. Pettis Veter-
mass seen in humans after puberty also is seen in these ans Administration Medical Center (Loma Linda, CA,
mice.
USA).

3
88
RICHMAN ET AL.
Bone densitometry
IGF-I RIA: IGF-I was measured by specific RIA as pre-
(27)
viously described.
The cross-reactivity of IGF-II in the
Volumetric bone density and geometric parameters were
determined by pQCT. Previously, pQCT has been validated
as a reliable method of determining volumetric bone density
IGF-I RIA was less than 0.5%. The sensitivity of the IGF-I
RIA was less than 50 ng/liter; the intra- and interassay CVs
were less than 10%.
(17)
and geometric parameters in mice.
Routine calibration
IGF-II RIA: IGF-II was measured by specific RIA as
was performed daily with a defined standard (cone phan-
tom) containing hydroxyapatite embedded in lucite (Nor-
land Medical Systems, Inc.). The voxel size was set at 0.07
mm and a half-millimeter-thick slice was scanned at the
middiaphysis. Analysis of the scans was performed using
the manufacturer-supplied software program (STRATEC
MEDIZINTECHNIK GMBH Bone Density Software, ver-
sion 5.40C; Norland Medical Systems). Volumetric bone
density and geometric parameters were estimated with Loop
analysis. For cortical analysis, the threshold was set at 350
(27)
previously described. The cross-reactivity of IGF-I in the
IGF-II RIA was less than 2%. The sensitivity of the IGF-II
RIA was 0.1 ␮g/liter; the intra- and interassay CVs were
less than 10%.
Statistics
Results are reported as mean Ϯ SD for 7–9 animals per
group and compared by two-tailed three-way analysis of
variance (ANOVA) and Newman–Keuls post hoc test using
commercially available statistical software (STATISTICA;
StatSoft, Inc., Tulsa, OK, USA). Results were considered
significantly different for p Ͻ 0.05.
3
mg/cm for the 23- to 56-day-old mice. The 7- and 14-day-
old mice had lower bone densities than did the 23- to
5
6-day-old mice and could not be measured at the 350
3
mg/cm threshold. Therefore, we used a threshold of 150
mg/cm for the 7- to 23-day-old mice in the cortical anal-
3
RESULTS
ysis. For cancellous analysis, the outer threshold was set at
3
3
2
50 mg/cm and the inner threshold was set at 300 mg/cm
The pattern of growth in body weight, femur weight, size,
for 23- to 56-day-old mice, and 125 mg/cm and 150 mg/ and mineral density have been determined for C3H and B6
cm for 7-to 14-day-old mice, respectively. Values for 7- mice for the postnatal, pubertal, and postpubertal growth
3
3
and 14-day-old mice may be less accurate than for the older periods. Although the general pattern of growth was similar
mice because of the lower threshold used, but we have for the two strains of mice, there were differences in femur
included them in this article because they show the rapid bone growth. These differences were statistically significant
growth that occurs during the postnatal period (days 7–23). as determined by three-way ANOVA (strain, sex, and day).
The results of further analysis with Newman–Keuls post
hoc testing are given in the tables and the figure legends.
Biochemical assays
Physical measurements
Femoral bone extraction: The freshly dissected left fem-
Body weight: Weight gain in all the mice was increased
ora were rinsed in 1ϫ phosphate-buffered saline (PBS) for
roughly 3-fold over day 7 during postnatal growth (days
2
4 h at 4°C and then transferred to 0.01% Triton X-100
7
–23). There was no difference in body weights between
(Sigma Chemicals) for 72 h at 4°C to extract the membrane
females of C3H and B6 mice at any time points. Male C3H
mice were heavier than B6 male mice at day 42 and day 56
bound ALP from the osteoblast cell surfaces. Bone-specific
ALP activity and total protein content were determined for
this extract. The femora were dehydrated in 75% ethanol
and then allowed to dry out completely at 37°C. Once dry,
the femoral length was measured to the nearest 0.01 mm
using a caliper (Dial Caliper; Mitutoyo Corp., Japan) and
the bones were weighed to the nearest 0.01 mg using a
balance (Millibalance model 7500, Electrobalance DTL;
Cahn Instruments, Inc., Cerritos, CA, USA).
ALP assay of bone extracts: ALP activity of the bone
extract was determined using a paranitrophenol phosphate
substrate as described previously.
Protein assay: The total protein content of the bone
extract was determined using Bradford assay (BioRad, Her-
cules, CA, USA)
Osteocalcin assay: Osteocalcin levels in serum were
measured by specific radioimmunoassay (RIA) that was
previously validated for mice. The mouse osteocalcin RIA
(Tables 1 and 2). Within each strain, males gained more
weight than females during pubertal growth (in B6, 161%
for males and 133% for females; in C3H, 158% for males
and 143% for females). Males were heavier than females at
day 31 and thereafter for both strains (Tables 1and 2).
Femoral bone weight: The fastest rate of femoral growth
was during the postnatal period (days 7–23) when dry
femoral weight increased 8-fold in B6 and 7-fold in C3H
mice. The C3H bones were significantly heavier than the B6
bones at day 42 and day 49 in male mice and at day 56 in
female mice (Tables 1 and 2).
Femoral bone length: Femoral bone length increased
more than 2-fold in both strains of mice during postnatal
period. Although femora of B6 mice were shorter on day 7,
the length increased more than in C3H mice during postna-
tal growth (days 7–23) so that there was no difference
between strains from day 23 (Tables 1 and 2). Within
strains, male femoral length was not different from that of
females, either during or after puberty.
(24)
(25)
had a CV of less than 8%.
Separation of IGF binding proteins from IGFs: Because
IGF binding proteins (IGFBPs) produce artifacts in IGF
RIAs, complete separation of the IGFBPs from the IGFs is
necessary in order for the IGF determinations to be valid.
We have used previously validated BioSpin protocol to
Middiaphysis geometric parameters
Area: Total area at the middiaphysis of the femur in-
(26)
separate IGFs from IGFBPs.
creased 2-fold during postnatal growth (days 7–23) in both

SKELETAL CHANGES IN MICE DURING PUBERTY
389

3
90
RICHMAN ET AL.

SKELETAL CHANGES IN MICE DURING PUBERTY
391
FIG. 2. Changes in periosteal circumference in C3H and B6 mice
from day 7 to day 56. At day 23 periosteal circumference was 5% (p ϭ
FIG. 1. Changes in medullary area in C3H and B6 mice from day 7
to day 56. Medullary area increased significantly between days 7 and 0.03) greater in B6 male mice and 7% (p Ͻ 0.01) in B6 female mice
3 in both strains. Medullary area was greater by 45% or more in B6
than in C3H male and female mice, respectively. On day 56, periosteal
2
mice compared with C3H mice for both sexes from day 14 onward. circumference was 8% (p Ͻ 0.001) greater in B6 male mice and 12%
Medullary area was not significantly different in males versus females (p Ͻ 0.01) greater in B6 female mice than in C3H male and female
for either strain. Values expressed as mean of 7–9 animals Ϯ SD; B6 mice, respectively. Within strain, B6 male mice had 5% (p ϭ 0.01)
male mice, closed triangles; B6 female mice, closed circles; C3H male
mice, open triangles; C3H female mice, open circles.
greater periosteal circumference than B6 female mice on day 56 and
C3H male mice had 8% (p Ͻ 0.001) greater periosteal circumference
than C3H female mice on day 42. Values expressed as mean of 7–9
animals Ϯ SD; B6 male mice, closed triangle; B6 female mice, closed
circles; C3H male mice, open triangles; C3H female mice, open circles;
B6 and C3H mice. There was no difference by strain or sex SD ϭ bars.
in total area on day 7 and day 14. Total area in B6 male mice
was 9% (p ϭ 0.028) greater than that of C3H males on day
4
9 and 17% (p Ͻ 0.001) greater on day 56. B6 female mice female mice had 12% (p Ͻ 0.001) greater periosteal cir-
had 13% (p ϭ 0.008) greater total area than did C3H cumference than that of C3H female mice at day 56, and B6
females from day 23 onward (Tables 1 and 2). The total area male mice had 8% (p Ͻ 0.001) greater periosteal circum-
at the middiaphysis in B6 males was 9% (p ϭ 0.005) greater ference than that of C3H male mice at day 56. Within
than that of B6 females on day 56 and the value for C3H strains, periosteal circumference in B6 male mice was 5%
males was 16% (p Ͻ 0.001) greater than that of C3H (p ϭ 0.01) greater than that of B6 female mice by day 56
females on day 49.
Cortical area: Cortical area at the middiaphysis of the C3H male mice than that of C3H female mice by day 42.
femur was not different by strain or sex during rapid post-
Endosteal circumference: Endosteal circumference of the
and periosteal circumference was 8% (p Ͻ 0.001) greater in
natal growth (days 7–23) but cortical area increased by 82% middiaphysis increased 65–90% during rapid postnatal
in B6 mice and 63% in C3H mice (Tables 1 and 2). C3H growth (days 7–23) and a total of 7–10% during pubertal
male mice had greater cortical area than did B6 male mice (days 23–31) and postpubertal growth (days 31–56) in B6
on day 31 (17%, p ϭ 0.02) through day 56 (13%, p ϭ mice (Fig. 3). In C3H mice, endosteal circumference in-
0
.005). C3H female mice had greater cortical area than that creased by 60–70% during rapid postnatal growth (days
of B6 female mice on day 31 (20%, p ϭ 0.007), day 42 7–23) but did not change significantly throughout the rest of
16%, p ϭ 0.03), and day 56 (15%, p ϭ 0.01). Within the study. Endosteal circumference was significantly greater
(
strains, cortical area was 13% (p ϭ 0.03) greater in B6 male in B6 mice than in C3H mice for the entire study. Within
mice than in B6 female mice by day 56, while cortical area strains, endosteal circumference did not differ across sex for
was 19% (p Ͻ 0.001) greater in C3H male mice than in C3H the entire study (Fig. 3).
female mice by day 42.
Cortical thickness: Cortical thickness at the middiaphysis
Medullary area: Medullary area increased dramatically dur- of the femur was roughly 30% (p Ͻ 0.01) greater in C3H
ing rapid postnatal growth (days 7–23) in both strains of mice mice than in B6 mice throughout the study. Within strains,
(
Fig. 1). Medullary area at the middiaphysis of the femur was cortical thickness was 9% (p Ͻ 0.05) greater in C3H male
significantly greater in B6 mice than in C3H mice from day 14 mice than in C3H female mice from day 42. But no signif-
45% in B6 male mice vs. C3H male mice [p ϭ 0.001] and icant differences across sex were evident in B6 mice (Fig.
5% in B6 female mice vs. C3H female mice [p Ͻ 0.001]). B6 4). The greatest increase in cortical thickness occurred dur-
(
5
mice had greater medullary area than did C3H mice for the ing rapid postnatal (days 7–23) and pubertal (days 23–31)
duration of the study. At day 56, B6 male and female mice growth, while the rate of change in cortical thickness de-
exhibited 149% and 108% greater medullary area than C3H creased dramatically after puberty.
male and female mice, respectively. Within strain, there was no
difference between sexes at any time during the study.
Periosteal circumference: Periosteal circumference at the
Bone densitometry
middiaphysis of the femur did not differ across strain or sex
Mineral content: Total mineral content of the femoral
during rapid postnatal growth (days 7–14) (Fig. 2). B6 middiaphysis increased 4-fold in both B6 mice and C3H

3
92
RICHMAN ET AL.
FIG. 3. Changes in endosteal circumference in C3H and B6 mice
from day 7 to day 56. Endosteal circumference was significantly greater
in B6 mice compared with C3H mice at all age groups except at day 7.
Endosteal circumference was not significantly different in male versus
female mice for either strain. Endosteal circumference increased sig-
nificantly between days 7 and 23 after which there was no significant
difference in endosteal circumference. Values expressed as mean of
FIG. 5. Changes in middiaphyseal total mineral content in C3H and
B6 mice from day 7 to day 56. On day 23, C3H male mice had 56%
(p Ͻ 0.001) and C3H female mice had 51% (p Ͻ 0.01) greater total
mineral content than B6 male and female mice, respectively. On day
3
3
5, C3H male mice had 42% (p Ͻ 0.001) and C3H female mice had
9% (p Ͻ 0.001) greater total mineral content than B6 male and female
mice, respectively. On day 56, C3H males had 36% (p Ͻ 0.001) and
C3H females had 40% (p Ͻ 0.001) greater total mineral content than
B6 male and female mice, respectively. Within strain, B6 male mice
had 25% (p ϭ 0.01) greater mineral content than B6 female mice by
day 42 and C3H male mice had 23% greater mineral content (p Ͻ
7
–9 animals Ϯ SD; B6 male mice, closed triangles; B6 female mice,
closed circles; C3H male mice, open triangles; C3H female mice, open
circles; SD ϭ bars.
0
7
.001) than C3H female mice on day 42. Values expressed as mean of
–9 animals Ϯ SD; B6 male mice, closed triangle; B6 female mice,
closed circles; C3H male mice, open triangles; C3H female mice, open
circles; SD ϭ bars.
(C3H vs. B6 female mice, 27%; C3H vs. B6 male mice,
4
1%; p Ͻ 0.05), and by day 56, this difference was even
greater (C3H vs. B6 female mice, 60%; C3H vs. B6 male
mice 62%; p Ͻ 0.001). Within strain, there was no signif-
icant difference in volumetric bone density between sexes
(
Tables 1 and 2). The greatest difference in rate of total
density accretion occurred during rapid postnatal (days
FIG. 4. Changes in cortical thickness in C3H and B6 mice from day 7–23) and pubertal (days 23–31) growth (Figs. 6 and 7). The
2
3 to day 56. Cortical thickness was significantly greater in C3H mice
rate of bone density gain in C3H mice during postnatal and
pubertal growth was twice that of B6 mice, while it was not
different after puberty.
compared with B6 mice at all time points studied. Cortical thickness
was significantly greater in male C3H mice compared with female C3H
mice at days 42, 49, and 56. In B6 mice, the cortical thickness was
significantly different in male mice compared with female mice. Cor-
tical thickness increased significantly during puberty in both strains of Biochemical assays
mice. Values expressed as mean of 7–9 animals Ϯ SD; B6 males mice,
Serum IGF-I: C3H female mice had significantly higher
closed triangles; B6 female mice, closed circles; C3H male mice, open
triangles; C3H female mice, open circles.
serum levels of IGF-I than did B6 mice on day 7 and day 14,
whereas C3H male mice exhibited higher serum IGF-I
levels at day 7 (Tables 1 and 2). In the pooled data from
male and female mice, C3H mice had greater serum IGF-I
mice during rapid postnatal growth (days 7–23; Fig. 5). levels than B6 mice at days 7, 14, 35, and 42 (data not
Total mineral content in C3H mice was roughly 40% (p Ͻ shown).
0
(
.01) greater than that of B6 mice throughout the study
44% on day 7, 54% on day 31, and 38% on day 56). Within not different by strain or sex. By day 23, serum IGF-II was
strain, total mineral content of B6 male mice was 25% (p ϭ nearly undetectable and remained at low levels for the
Serum IGF-II: On day 7 and day 14, IGF-II levels were
0
0
.01) greater than that of B6 female mice and 23% (p Ͻ duration of the study (data not shown).
.001) greater in C3H male mice than in C3H female mice Serum osteocalcin: Serum osteocalcin was significantly
higher in B6 mice compared with that of C3H mice from
on day 42.
Density: Total volumetric bone density of the middiaphy- day 7 (77%, p Ͻ 0.01) to 42 (33%, p Ͻ 0.01) in males and
sis increased more in C3H mice (females, 81% and males, from day 7 (94%, p Ͻ 0.01) to 31 (42%, p Ͻ 0.01) in
8
3%) than in B6 mice (females, 37% and males, 51%) females. Serum osteocalcin increased roughly 2-fold during
during rapid postnatal growth (days 7–23). Greater volu- postnatal growth (days 7–23) in male mice of both strains
metric bone density was present in C3H mice on day 7 and then decreased steadily until day 56. However, in fe-

SKELETAL CHANGES IN MICE DURING PUBERTY
393
FIG. 8. Changes in bone-specific ALP activity in bone extracts for
C3H and B6 mice from day 7 to day 56. On day 14, bone-specific ALP
activity in C3H male mice was 75% (p Ͻ 0.001) and in C3H female
mice was 95% (p Ͻ 0.001) greater than B6 male and female mice,
respectively. On day 23, C3H male mice had 21% (p Ͻ 0.01) greater
bone-specific ALP activity than B6 male mice, but C3H female mice
were not different from B6 female mice. Values expressed as mean of
FIG. 6. Changes in bone density in C3H and B6 mice from day 7 to
day 56. Because bone density was not different between male and
female mice of either strain, the data for both sexes were pooled (p Ͻ
0
3
7
.01). Total volumetric bone density of the middiaphysis was about
0% greater in C3H mice than in B6 mice at day 7. C3H mice exhibited
0% more bone density than did B6 mice after puberty (day 31). Bone
density was significantly higher in C3H mice than B6 mice at all time
points (p Ͻ 0.01). Bone density more than doubled during puberty in
both strains of mice. Values expressed as mean of 7–9 animals Ϯ SD;
B6 mice, closed triangle; C3H mice, open triangles; SD ϭ bars.
7
–9 animals Ϯ SD; B6 male mice, closed triangle; B6 female mice,
closed circles; C3H male mice, open triangles; C3H female mice, open
circles; SD ϭ bars.
mice (Fig. 8). A decrease of 30% in ALP activity occurred
during puberty (days 23–31). Within strains, there was no
difference in ALP activity between sexes at any time.
Correlations: In both strains, body weight was highly
correlated with all other physical, geometric, and densitom-
etry measurements throughout the study (data not shown).
Serum IGF-I levels were correlated positively with physical,
geometric, and densitometry measurements in both strains,
especially during postnatal growth but less so during puber-
tal and postpubertal growth (data not shown). In the pooled
data from C3H and B6 strains of mice belonging to both
sexes, multiple regression analyses were carried out using
strain, sex, age, IGF-I, and osteocalcin as independent vari-
ables (Table 3). The ␤-coefficients, which represent the
relative contribution of each independent variable in the
prediction of the dependent variable, reveal that age is the
most important predictor of various bone parameters, as
expected. Furthermore, strain contributed predominantly to
the differences in density, endosteal circumference, and
cortical thickness. In addition, serum levels of IGF-I and
FIG. 7. Changes in the rate of volumetric bone density accretion per
day in C3H and B6 mice from day 7 to day 56. C3H mice had a 2.8-fold
and a 1.6-fold higher rate of bone accretion from day 7 to day 14 and
from day 23 to day 31, respectively, compared with B6 mice. The gain
in bone density for each time period was calculated from the mean
value for each time period and was divided by the duration in days to
obtain the rate. Male and female mice were calculated separately and
used as replicate values for the calculation of significance. B6 mice, osteocalcin provided statistically significant contributions to
closed triangles; C3H mice, open triangles.
the prediction of various bone parameters (Table 3). To
further evaluate the relative contribution of the previously
mentioned five independent variables to predict the gender
male mice, serum osteocalcin levels doubled from days differences in bone size, we performed multiple regression
7
2
–14 and then decreased steadily until day 56 (Tables 1 and analyses using data from prepubertal (days 7–23), pubertal
). Serum osteocalcin levels in female mice of both strains (days 23–35), and postpubertal (days 35–56) periods. We
were 32% higher than those in male mice on day 14 (p Ͻ found that the relative contribution of gender to differences in
.01), but were not different at other time points. periosteal circumference was much greater during the postpu-
Bone extract ALP: Bone extract ALP normalized for total bertal period than the prepubertal period (␤-coefficient of 0.36
0
extractable protein showed a 4-fold increase in activity vs. 0.11). To determine if gender differences in periosteal
during the rapid-bone modeling associated with increased circumference can be explained based on IGF-I, we deter-
bone length and size during postnatal growth (days 7–23) in mined the relative contribution of strain, sex, and age to dif-
both strains, although bone extract from C3H mice con- ferences in IGF-I using data from prepubertal (days 7–23) and
tained 2-fold greater ALP activity on day 14 than that of B6 postpubertal periods (days 35–56). We found that strain, sex,

3
94
RICHMAN ET AL.
TABLE 3. MULTIPLE REGRESSION ANALYSIS IN THE POOLED DATA FROM MALES AND FEMALES OF C3H AND
B6 STRAINS OF MICE
␤-Coefficient for independent variable
Dependent variable
Strain
Sex
Age
IGF-I
Osteocalcin
Multiple r
Body weight
Bone length
Total content
Total density
0.04
0.16*
0.03
0.88*
0.95*
0.86*
0.79*
0.85*
0.87*
0.66*
0.58*
0.14*
0.09*
0.07*
0.01
0.14*
0.14*
0.09*
0.12*
0.11*
0.23*
0.05*
0.08*
0.16*
0.23*
0.26*
0.005
0.95*
0.94*
0.95*
0.95*
0.92*
0.91*
0.86*
0.88*
0.11*
0.34*
0.59*
0.18*
0.13*
0.45*
0.54*
0.14*
0.08*
0.11*
0.09*
0.02
Total area
Periosteal circumference
Endosteal circumference
Cortical thickness
0.16*
*
p Ͻ 0.05
and age nearly contributed equally (␤-coefficient of 0.39, 0.3, period (Fig. 3). Although neither the periosteal nor the
and 0.26, respectively) during the postpubertal period. How- endosteal circumference showed significant change during
ever, during the prepubertal period sex had no effect on serum postpubertal growth (days 35–56) in C3H mice, both the
IGF-I as expected (␤-coefficient ϭ 0.01).
periosteal and the endosteal circumference increased in B6
mice during this period, resulting in B6 mice having signif-
icantly greater periosteal circumference than C3H mice at
56 days of age. Consistent with these data, Sheng et al.
recently have shown, by histomorphometric analysis, that
(21)
DISCUSSION
In previous studies, two inbred strains of mice, C3H and B6, the total cross-sectional area of the femur at the middiaphy-
exhibited similar body weights but different volumetric peak sis is significantly greater in the B6 mice at 6 weeks of age
(17,19–21)
bone densities.
which this divergence in bone density becomes apparent, we
To identify the growth period(s) in compared with C3H mice.
The cellular mechanisms that contribute to the greater
evaluated temporal skeletal changes in C3H and B6 mice cortical thickness in C3H mice compared with B6 mice can
belonging to both sexes before, during, and after puberty. Our only be speculated at this time. In this regard, the findings
data show that although C3H mice exhibit slightly higher total that endosteal but not periosteal circumference is signifi-
bone density compared with B6 mice at 7 days of age (ap- cantly different between the two strains of mice during
proximately 25%), total bone density is 69% greater in C3H postnatal growth suggested that differences in endosteal
mice compared with B6 mice at 31 days of age, despite similar bone formation and resorption could, in part, contribute to
body size, body weight, and bone length. Accordingly, the gain the observed differences in cortical thickness. Consistent
in bone density in C3H mice during postnatal (days 7–23) and with this idea were the findings that (1) C3H mice exhibited
pubertal (days 23–31) growth is more than twice that of B6 significantly higher ALP activity in the femoral bone extract
mice. In addition, the findings that the gain in bone density than did B6 mice at the time when maximal differences in
increases by more than 70% during puberty (days 23–31) attest bone density between the two strains occur, (2) C3H mice
that rapid skeletal changes occur during this period. Therefore, exhibited greater endosteal bone formation rate than B6
(21)
the postnatal (days 7–23) and pubertal (days 23–31) growth mice by histomorphometric analysis,
and (3) there were
periods appear to be critical in the development of the differ- fewer osteoclast precursors in the marrow of C3H mice than
(28)
ence in bone density between C3H and B6 mice. The differ- in B6 mice.
ence in total bone density seen at day 35 between C3H and B6 relative contribution of endosteal bone formation versus
mice was maintained during the rest of the study period.
endosteal bone resorption in contributing to differences in
Further studies are needed to evaluate the
The differences in cortical thickness between the two cortical thickness between the two strains of mice during
strains of mice reflect the greater accumulation of bone postnatal growth.
mineral in C3H mice compared with B6 mice during post-
The divergence in total volumetric bone density at the
natal growth. This difference in cortical thickness between femoral middiaphysis between C3H and B6 strains during
C3H and B6 mice developed during the rapid growth of the postnatal growth may, in part, be caused by greater bone
postnatal (days 7–23) and pubertal (days 23–31) periods. area in the C3H mice than in the B6 mice because of
The difference in cortical thickness between C3H and B6 differences in cortical thickness. However, the possibility
mice appeared to be caused by significantly smaller en- that the bones of C3H mice during early postnatal growth
dosteal circumference in C3H mice (approximately 20% may be more mineralized or less porous compared with B6
less) compared with B6 mice. The increase in cortical mice cannot be ruled out because C3H mice exhibit approx-
thickness of approximately 25% during puberty (days 23– imately 30% greater bone density than B6 mice despite
3
5) in both strains of mice appeared mainly to be caused by similar bone area on day 7. Although the finding that C3H
increased periosteal circumference, because the endosteal mice exhibit greater ash weight and calcium content com-
(19)
circumference did not show significant change during this pared with the B6 mice at 9 weeks of age
are consistent

SKELETAL CHANGES IN MICE DURING PUBERTY
395
studies are needed to evaluate the cause and effect relation-
ship between changes in serum levels of IGF-I and skeletal
changes in C3H and B6 mice during postnatal growth.
In contrast to postnatal growth, the mineral accumulation
during puberty appears to be predominantly because of
decreased bone resorption caused by increased male and
female sex steroid hormones. There are a number of find-
ings that support this idea. First, the increase in total bone
density during puberty in both C3H and B6 mice was
associated with a decrease in biochemical markers of bone
remodeling in both serum and bone extracts. Second, we
and others have shown that the increase in mineral accu-
mulation during sexual development in girls is associated
(8,10,30)
with decreased bone turnover.
Third, Slemenda et
(8)
al. have shown that black children accumulated 10%
greater bone mass and had significantly reduced bone turn-
over, as measured by serum levels of osteocalcin and
tartrate-resistant acid phosphatase, compared with white
children during pubertal growth. Although sex steroid
FIG. 9. A model of the changes in bone size during postnatal,
pubertal, and postpubertal growth in C3H and B6 mice. On day 7, the
periosteal circumference of C3H and B6 femora was similar; as illus-
trated by the cross-sectional diagram of the femur at the middiaphysis,
B6 mice had a 20% greater endosteal circumference than did C3H
mice, and C3H mice had 34% greater volumetric bone density than did
(31–33)
hormone–induced changes in cytokine production
have been implicated in the reduced bone turnover that
B6 mice. Between days 7 and 23 periosteal circumference increased occurs during puberty, the cause and effect relationship
5
0% in C3H mice and 58% in B6 mice, while endosteal circumference between cytokine production and the reduction in bone
increased by 90% in both strains. On day 23, the periosteal circumfer- turnover during puberty remains to be established.
ence of C3H and B6 femora was similar; B6 mice had 16% greater
endosteal circumference than did C3H mice, and C3H mice had 69%
greater volumetric bone density than did B6 mice. On day 35, B6 mice
had about 4% greater periosteal circumference than did C3H mice, B6
was significantly higher in the C3H mice than in the B6 mice
had a 22% greater endosteal circumference than did C3H mice, and
during postnatal growth, we anticipated serum osteocalcin
C3H mice had 56% greater volumetric bone density than did B6 mice.
On day 56, B6 mice had 8–12% greater periosteal circumference than
did C3H mice, B6 had a 35% greater endosteal circumference than did
C3H mice, and C3H mice had 61% greater volumetric bone density is possible that osteocalcin expression in osteoblasts is higher
Surprisingly, we found that serum osteocalcin levels were
significantly lower in C3H mice compared with B6 mice
throughout the study. Because ALP activity in bone extract
levels also to be higher during this period in C3H mice. There
are a number of potential explanations for this discrepancy. It
than did B6 mice.
in C3H mice compared with B6 mice; however, a smaller
proportion of synthesized osteocalcin was released into the
extracellular fluid compared with the matrix compartment in
with this idea, further studies are needed to evaluate whether C3H mice compared with B6 mice, resulting in reduced levels
or not C3H bones contain less porosity or more mineraliza- in serum. Metabolic differences in serum osteocalcin (e.g.,
tion of the bone matrix compared with B6 mice during half-life and clearance rate) between C3H and B6 mice also
postnatal growth when the greatest difference in density was may contribute to the lower levels of serum osteocalcin in C3H
seen despite similarities in bone area.
mice compared with B6 mice. Alternatively, the higher level of
The rapid increase in total mineral content during post- osteocalcin produced in the B6 mice may act as an inhibitor of
natal growth (days 7–23) appeared mainly to be caused by bone formation during postnatal growth because bone forma-
increased bone size, which increased by more than 50% tion was increased at the endosteum in mice lacking osteo-
(34)
during this period (Fig. 9). This increase in bone size calcin.
appeared to be caused by increased periosteal bone forma- Our findings also show that sexual dimorphic skeletal
tion rate because both osteocalcin levels in the serum and changes occurred in mice during puberty as in the case of
(4)
ALP activity in bone extract increased during this period. In humans. This sexual dimorphism may be attributed to
terms of the individual factors that contribute to the in- greater bone size and cortical thickness in men than in
(35)
creased periosteal bone formation, we postulate that IGF-I women, which in turn may be caused by a later start and
(5)
may be a potential candidate because: (1) serum levels of a longer pubertal growth period in boys than girls. How-
IGF-I increased dramatically during this period (Tables 1 ever, volumetric bone mineral density (BMD) does not appear
(36–38)
and 2) and showed significant correlation with changes in to be different between the two sexes in humans.
Within
periosteal circumference and biochemical measurements of strains, we found that 8-week-old male mice had a greater total
bone formation during this period (Table 3); and (2) mice mineral content and a larger periosteal circumference than did
lacking IGF-I exhibit dramatic reduction in postnatal skel- female mice, despite similar endosteal circumference. Thus,
(29)
etal growth. Consistent with a role for IGF-I, our findings male mice of both strains exhibited increased bone size and
also show significantly greater serum IGF-I levels in C3H cortical thickness resulting in greater total mineral content
mice than in B6 mice on days 7, 14, 35, and 42 in the pooled compared with female mice. The molecular mechanisms that
data from male and female mice. Consistent with these data, are responsible for the increased periosteal bone formation
we have previously shown that the IGF-I differences are and/or reduction in periosteal bone resorption in the male mice
(20)
statistically significant up to 10 months of age.
Further compared with female mice have yet to be established.

3
96
RICHMAN ET AL.
In conclusion, the difference in bone density between 12. Ross PD, Knowlton W 1998 Rapid bone loss is associated with
increased levels of biochemical markers. J Bone Miner Res
C3H and B6 strains of mice is established early and main-
tained throughout the period of growth. Two growth peri-
ods, postnatal and pubertal, are particularly critical to the
development of peak BMD. Although the difference in bone
density is maintained throughout, there are differences in
the pattern of growth between strains that suggest that the
1
3:297–302.
1
3. Weaver CM, Peacock M, Martin BR, McCabe GP, Zhao J,
Smith DL, Wastney ME 1997 Quantification of biochemical
markers of bone turnover by kinetic measures of bone forma-
tion and resorption in young healthy females. J Bone Miner
Res 12:1714–1720.
genes that contribute to the difference in BMD during each 14. Garnero P, Delmas PD 1997 Bone markers. Baillieres Clin
Rheumatol 11:517–537.
period of growth may be different. This is consistent with
(39)
15. Kalu DN 1999 Animal models of the aging skeleton. In: Rosen
CJ, Glowacki J, Bilezikian JP (eds.) The Aging Skeleton.
Academic Press, San Diego, CA, USA, pp. 37–50.
6. Ornoy A, Katzburg S 1995 Osteoporosis: Animal models for
the human disease. In: Ornoy A (ed.) Animal Models of
Human Related Calcium Metabolic Disorders. CRC Press,
New York, NY, USA, pp. 105–126.
7. Beamer WG, Donahue LR, Rosen CJ, Baylink DJ 1996 Ge-
netic variability in adult bone density among inbred strains of
mice. Bone 18:397–403.
8. Bain SD, Bailey MC, Celino DL, Lantry MM, Edwards MW
1993 High-dose estrogen inhibits bone resorption and stimu-
lates bone formation in the ovariectomized mouse. J Bone
Miner Res 8:435–442.
the evidence from quantitative trait loci studies
that
multiple genes contribute to the difference in bone density.
The genes that are expressed differentially in the skeleton of
C3H and B6 mice during postnatal and pubertal growth may
provide clues regarding the genes that contribute to the
BMD differences between these two strains.
1
1
ACKNOWLEDGMENTS
1
The authors acknowledge the expert technical assistance
provided by Daniel Bruch and Joe Rung-Aroon in this
study. Support for this work was received from the NIH
grant AR31062, the National Medical Technology Test
Bed, and the U.S. Department of the Army.
1
9. Chen C, Kalu DN 1999 Strain differences in bone density and
calcium metabolism between C3H/HeJ and C57BL/6J mice.
Bone 25:413–420.
2
0. Rosen CJ, Dimai HP, Vereault D, Donahue LR, Beamer WG,
Farley J, Linkhart S, Linkhart T, Mohan S, Baylink DJ 1997
Circulating and skeletal insulin-like growth factor-I (IGF-I)
concentrations in two inbred strains of mice with different
bone mineral densities. Bone 21:217–223.
REFERENCES
1
. Lloyd T, Rollings N, Andon MB, Demers LM, Eggli DF,
Kieselhorst K, Kulin H, Landis JR, Martel JK, Orr G, Smith P
2
1. Sheng MH, Baylink DJ, Beamer WG, Donahue LR, Rosen CJ,
Lau KH, Wergedal JE 1999 Histomorphometric studies show
that bone formation and bone mineral apposition rates are
greater in C3H/HeJ (high-density) than C57BL/6J (low-
density) mice during growth. Bone 25:421–429.
2. Ahima RS, Dushay J, Flier SN, Prabakaran D, Flier JS 1997
Leptin accelerates the onset of puberty in normal female mice.
J Clin Invest 99:391–395.
3. Bonjour JP, Theintz G, Buchs B, Slosman D, Rizzoli R 1991
Critical years and stages of puberty for spinal and femoral
bone mass accumulation during adolescence. J Clin Endocri-
nol Metab 73:555–563.
4. Farley JR, Hall SL, Herring S, Tarbaux NM 1992 Two bio-
chemical indices of mouse bone formation are increased, in
vivo, in response to calcitonin. Calcif Tissue Int 50:67–73.
1
992 Determinants of bone density in young women. I. Rela-
tionships among pubertal development, total body bone mass,
and total body bone density in premenarchal females. J Clin
Endocrinol Metab 75:383–387.
. Theintz G, Buchs B, Rizzoli R, Slosman D, Clavien H, Si-
zonenko PC, Bonjour JP 1992 Longitudinal monitoring of
bone mass accumulation in healthy adolescents: Evidence for
a marked reduction after 16 years of age at the levels of lumbar
spine and femoral neck in female subjects. J Clin Endocrinol
Metab 75:1060–1065.
. Eisman JA, Kelly PJ, Morrison NA, Pocock NA, Yeoman R,
Birmingham J, Sambrook PN 1993 Peak bone mass and os-
teoporosis prevention. Osteoporos Int 3(Suppl 1):56–60.
. van der Meulen MCH, Carter DR 1999 Mechanical determi-
nants of peak bone mass. In: Rosen CJ, Glowacki J, Bilezikian
2
2
2
3
2
4
JP (eds.) The Aging Skeleton. Academic Press, San Diego, 25. Srivastava AK, Catillo G, Wergedal JE, Mohan S, Baylink DJ
CA, USA, pp. 105–114.
. Gertner JM 1999 Childhood and adolescence. In: Favus MJ
2000 Development and application of a synthetic peptide
based osteocalcin assay for the measurement of bone forma-
tion in mouse serum. Calcif Tissue Int 67:255–259.
5
(ed.) Primer on Metabolic Bone Diseases and Disorders of
Mineral Metabolism, 4th ed. Lippincott Williams & Wilkins, 26. Mohan S, Libanati C, Dony C, Lang K, Srinivasan N, Baylink
Philadelphia, PA, USA, pp. 45–49.
. Abbassi V 1998 Growth and normal puberty. Pediatrics 102(2
Pt 3):507–511.
DJ 1995 Development, validation, and application of a radio-
immunoassay for insulin-like growth factor binding protein-5
in human serum and other biological fluids. J Clin Endocrinol
Metab 80:2638–2645.
6
7
. Seeman E, Tsalamandris C, Formica C 1993 Peak bone mass,
a growing problem? Int J Fertil Menopausal Stud 38(Suppl 27. Mohan S, Bautista CM, Herring SJ, Linkhart TA, Baylink DJ
2
):77–82.
1990 Development of valid methods to measure insulin-like
growth factors-I and -II in bone cell-conditioned medium.
Endocrinology 126:2534–2542.
8
. Slemenda CW, Peacock M, Hui S, Zhou L, Johnston CC 1997
Reduced rates of skeletal remodeling are associated with in-
creased bone mineral density during the development of peak 28. Linkhart TA, Linkhart SG, Kodama Y, Farley JR, Dimai HP,
skeletal mass. J Bone Miner Res 12:676–682.
. Schiessl H, Frost HM, Jee WS 1998 Estrogen and bone-muscle
strength and mass relationships. Bone 22:1–6.
Wright KR, Wergedal JE, Sheng M, Beamer WG, Donahue
LR, Rosen CJ, Baylink DJ 1999 Osteoclast formation in bone
marrow cultures from two inbred strains of mice with different
bone densities. J Bone Miner Res 14:39–46.
9
1
1
0. Cadogan J, Blumsohn A, Barker ME, Eastell R 1998 A lon-
gitudinal study of bone gain in pubertal girls: Anthropometric 29. Baker J, Liu JP, Robertson EJ, Efstratiadis A 1993 Role of
and biochemical correlates. J Bone Miner Res 13:1602–1612.
1. Garnero P, Delmas PD 1998 Biochemical markers of bone
insulin-like growth factors in embryonic and postnatal growth.
Cell 75:73–82.
turnover. Applications for osteoporosis. Endocrinol Metab 30. Mauras N, Rogol AD, Haymond MW, Veldhuis JD 1996 Sex
Clin North Am 27:303–323. steroids, growth hormone, insulin-like growth factor-1: Neu-

SKELETAL CHANGES IN MICE DURING PUBERTY
397
roendocrine and metabolic regulation in puberty. Horm Res 37. Kalender WA, Felsenberg D, Louis O, Lopez P, Klotz E,
4
5:74–80.
Osteaux M, Fraga J 1989 Reference values for trabecular and
cortical vertebral bone density in single and dual-energy quan-
titative computed tomography. Eur J Radiol 9:75–80.
38. Gilsanz V, Gibbens DT, Roe TF, Carlson M, Senac MO,
Boechat MI, Huang HK, Schulz EE, Libanati CR, Cann CC
1988 Vertebral bone density in children: Effect of puberty.
Radiology 166:847–850.
3
3
1. Manolagas SC, Bellido T, Jilka RL 1995 Sex steroids, cyto-
kines and the bone marrow: New concepts on the pathogenesis
of osteoporosis. Ciba Found Symp 191:187–196.
2. Kimble RB, Srivastava S, Ross FP, Matayoshi A, Pacifici R
1
996 Estrogen deficiency increases the ability of stromal cells
to support murine osteoclastogenesis via an interleukin-1 and
tumor necrosis factor-mediated stimulation of macrophage 39. Beamer WG, Rosen CJ, Shultz KL, Pung AE, Churchill GA,
colony-stimulating factor production. J Biol Chem 271:
8890–28897.
Donahue LR, Baylink DJ 1999 Mice congenic for bone density
QTLs exhibit corresponding bone density changes. J Bone
Miner Res 14(Suppl 1):S174.
2
3
3
3. Hofbauer LC, Gori F, Riggs BL, Lacey DL, Dunstan CR,
Spelsberg TC, Khosla S 1999 Stimulation of osteoprotegerin
ligand and inhibition of osteoprotegerin production by glu-
cocorticoids in human osteoblastic lineage cells: Potential
paracrine mechanisms of glucocorticoid-induced osteoporosis.
Endocrinology 140:4382–4389.
Address reprint requests to:
Subburaman Mohan, Ph.D.
4. Ducy P, Desbois C, Boyce B, Pinero G, Story B, Dunstan C,
Smith E, Bonadio J, Goldstein S, Gundberg C, Bradley A,
Karsenty G 1996 Increased bone formation in osteocalcin-
deficient mice. Nature 382:448–452.
Musculoskeletal Disease Center
J.L. Pettis Veterans Administration Medical Center
1
1201 Benton Street (151)
Loma Linda, CA 92357, USA
3
3
5. Bonjour J-P, Rizzoli R 1995 Bone acquisition in adolescence.
In: Marcus R, Feldman D, Kelsey J (eds.) Osteoporosis. Ac-
ademic Press, San Diego, CA, USA, pp. 465–476.
6. Gilsanz V, Boechat MI, Roe TF, Loro ML, Sayre JW, Good-
man WG 1994 Gender differences in vertebral body sizes in Received in original form December 28, 1999; in revised form
children and adolescents. Radiology 190:673–677. June 13, 2000; accepted August 31, 2000.