Full text of DOI:10.1359/jbmr.2002.17.7.1264

JOURNAL OF BONE AND MINERAL RESEARCH
Volume 17, Number 7, 2002
© 2002 American Society for Bone and Mineral Research
Dietary Essential Amino Acid Supplements Increase Bone
Strength by Influencing Bone Mass and Bone
Microarchitecture in Ovariectomized Adult Rats Fed an
Isocaloric Low-Protein Diet
1
2
1
3
2
1
¨
P. AMMANN, A. LAIB, J.-P. BONJOUR, J.M. MEYER, P. RUEGSEGGER, and R. RIZZOLI
ABSTRACT
This study was designed to investigate whether the administration of dietary essential amino acid supplements
in adult rats made osteoporotic by estrogen deficiency and reduced protein intake could reverse the deleterious
effects caused by these maneuvers. This animal model was selected to mimic the situation observed in elderly
women in whom estrogen deficiency and/or low-protein intake (but also calcium and vitamin D deficiency) are
known to contribute to the pathogenesis of osteoporosis. Six-month-old rats were ovariectomized (OVX) and
fed an isocaloric 2.5% casein diet for 10 weeks or sham-operated (SHAM) and fed an isocaloric 15% casein
diet. The animals fed the 2.5% casein diet were given isocaloric supplements of essential amino acids in similar
relative proportion to that of casein at doses of 2.5% or 5% of total diet for an additional 16 weeks. Vertebrae,
femur, and tibia bone mineral density (BMD); ultimate strength; and microtomographic histomorphometry
were evaluated before and after dietary essential amino acid supplements. Essential amino acid supplements
increased vertebrae, femur, and tibia bone strength in OVX rats fed a low-protein diet. The mechanical
changes induced by this dietary isocaloric supplement were associated with the prevention of a further BMD
decrease or even with some increases and changes in microarchitecture such as from a rod to a plate
trabecular spacial configuration and increased cortical thickness. Higher insulin-like growth factor (IGF) I
levels, as well as greater bone formation and reduced bone resorption as assessed by biochemical markers of
bone remodeling, were found in rats receiving essential amino acid supplements. In conclusion, dietary
essential amino acid supplements increased bone strength through modifications of BMD, trabecular archi-
tecture, and cortical thickness possibly by an IGF-I–mediated process. (J Bone Miner Res 2002;17:1264–1272)
Key words: elderly, nutrition, osteoporosis, insulin-like growth factor I, ovariectomy
with decreased bone mineral density (BMD) and/or an in-
creased risk of osteoporotic fracture.(6,7) We previously estab-
lished the selective effects of reduced protein intake in rats fed
an isocaloric low-casein diet.(8) In this model, a decrease of
BMD and bone strength was observed at all skeletal sites
tested, including parts of the skeleton containing mainly cor-
tical bone. This was associated with an early depressed bone
INTRODUCTION
OW DIETARY intake of protein, calcium, or vitamin D con-
tributes to the pathogenesis of senile osteoporosis.(1–5)
Indeed, low intake of calcium and/or protein can be associated
L
The authors have no conflict of interest.
1Division of Bone Diseases (WHO Collaborating Center for Osteoporosis and Bone Diseases), Department of Internal Medicine,
University Hospital, Geneva, Switzerland.
2Institute of Biomedical Engineering, ETHZ, CH-Zu¨rich, Switzerland.
3School of Dentistry, University Hospital, Geneva, Switzerland.
1264

ESSENTIAL AMINO ACIDS INCREASE BONE STRENGTH IN RATS
1265
formation and late increased bone resorption.(8,9) One potential
link between dietary protein and bone homeostasis is insulin-
like growth factor (IGF) I, and a low level of IGF-I is a risk
factor for fracture.(1,10) Production and action of this growth
factor are impaired under protein deficiency.(1,8,9,11–17) IGF-I
administration to experimental animals increases BMD, mass,
and strength.(18,19) However, the exogenous administration of
IGF-I or growth hormone (GH) was unable to restore a normal
growth in rats fed a low-protein diet.(14) The treatment with
IGF-I/IGF-binding protein complex also failed to stimulate
bone formation in adult rats fed an isocaloric low-protein
diet.(9)
An apparent resistance to IGF-I at the level of bone
provided the rationale to try to modulate the endogenous
production of IGF-I and to attenuate the resistance to IGF-I
by administering essential amino acid supplements. This
was supported by the observation that protein supplements
in elderly patients with hip fracture attenuated bone loss(1)
while increasing circulating IGF-I. The model of adult fe-
male rats that had been made osteoporotic by ovariectomy
(OVX) and by an isocaloric low-protein diet(8) was selected
to test this hypothesis and to mimic the situation of elderly
women with protein malnutrition and menopause. We as-
sessed the role of protein replenishment on both IGF-I
levels and bone response, BMD, architecture, and strength
changes. The results indicate that essential amino acid sup-
plements increase BMD and strength, the latter mainly
through greater cortical thickness.
MATERIALS AND METHODS
Animals and treatments
All experimental designs and procedures were approved
by the Animal Ethics Committee of the Geneva University
Faculty of Medicine. Female Sprague–Dawley rats (RCC,
Ltd., Fu¨llinsdorf, Switzerland), housed individually at 25°C
with a 12:12 h light-dark cycle, were strictly pair-fed a
laboratory diet throughout the experimental period provided
by Novartis Nutrition (Berne, Switzerland) containing var-
FIG. 1. Effect of isocaloric essential amino acid supplements on
ious amounts of casein, 0.8% phosphorus, 1.1% calcium, BMD in adult female rats made osteoporotic by OVX and a low-protein
diet. BMD (g/cm2, n ϭ 8–9, and means Ϯ SEM) was measured in vivo
in OVX rats fed a low-protein diet (open circles) treated with essential
amino acid supplements at a dose of 2.5% (open triangle) or 5% (closed
triangle), as well as in control rats SHAM fed a 15% casein diet (open
square). The treatment period is represented by a stippled area. Using
ANOVA, * indicates a significant difference (p Ͻ 0.01) from OVX rats
fed a low-protein diet.
0.2% magnesium, 70–80% carbohydrates, and 5% fat. Iso-
caloric diets were produced by the addition of corn carbo-
hydrate, thus providing the same energy intake. The animals
also were given a daily dose of vitamin D dissolved in
peanut oil (100 IU/kg body weight). After 2 weeks of
adaptation to a 15% casein-containing diet, 6-month-old
female rats underwent transabdominal OVX or a sham
operation (SHAM) under anesthesia with intraperitoneal
ketamine hydrochloride (100 mg/kg body weight). The ef-
fectiveness of OVX was verified at the end of the experi- line, DL-methionine, arginine, threonine, histidine, and tryp-
ment by visualizing the atrophy of the uterus and absence of tophane) were in a proportion similar to that found in casein
ovarian tissue. SHAM rats were maintained on the 15% (24.0, 15.0, 11.3, 9.1, 7.4, 6.8, 6.6, 6.5, and 2.4%, respec-
casein-containing diet, whereas OVX rats received a 2.5% tively). Before and after the treatment period, blood was
casein isocaloric diet. Ten weeks after OVX and low- sampled from the tip of the tail. Urine was collected in rats
protein diet, rats were randomly allocated to three groups maintained in metabolic cages over 24 h for the determina-
receiving diets containing either 2.5% casein, 2.5% casein tion of total deoxypyridinoline excretion. Before the essen-
and 2.5% essential amino acids, or 2.5% casein and 5.0% tial amino acid supplementation period, a group of OVX
essential amino acids. All the diets were isocaloric. The rats fed a 2.5% casein diet was killed for lumbar vertebrae,
amino acids (leucine, lysine, isoleucine, phenylalanine, va- tibia, and femur mechanical testing (baseline). At the end of

1266
AMMANN ET AL.
TABLE 1. EFFECT OF ISOCALORIC EAA SUPPLEMENTS ON BMD CHANGES
Changes from baseline (%)
OVX
OVX
OVX
SHAM
Casein
EAA
2.5
—
2.5
2.5
2.5
5.0
15
—
No. of rats
Lumbar spine
Proximal tibia
Midshaft tibia
Femoral neck
8
9
9
8
Ϫ13.6 Ϯ 2.1
Ϫ12.9 Ϯ 1.3
Ϫ5.7 Ϯ 0.6
Ϫ12.3 Ϯ 3.8
2.5 Ϯ 5.3*
1.9 Ϯ 0.6*
0.2 Ϯ 0.3*
Ϫ5.7 Ϯ 5.6
3.8 Ϯ 3.5*
4.0 Ϯ 1.2*
2.3 Ϯ 0.8*
Ϫ1.6 Ϯ 1.3*
3.4 Ϯ 3.4*
Ϫ3.6 Ϯ 0.9*
Ϫ0.2 Ϯ 1.2*
3.9 Ϯ 2.8*
Values are means Ϯ SEM expressed in percent of pretreatment values. Treatment was started 10 weeks after the OVX and reduced
protein intake. EAA supplements were administered for 16 weeks. Using ANOVA, * indicates a significant difference from OVX rats
fed a low-protein diet.
EAA, essential amino acid.
the study, all groups were killed by an overdose of ketamine tibia then was placed in the material testing machine on two
hydrochloride.
supports separated by a distance of 20 mm and load was
applied to the middle of the shaft, thus creating a three-point
bending test. Compression testing of the proximal tibia was
performed by axial compression of the tibia plateau with the
shaft being fixed in methylmethacrylate cement.(8,21) Fem-
oral neck testing was performed by maintaining the femur in
a vertical position and embedding the shaft in methyl-
methacrylate cement up to the lesser trochanter before ap-
plication of a vertical load on the femoral head.(19) The
mechanical resistance to failure was tested using a servo-
controlled electromechanical system (Instron 1114; Instron
Corp., High Wycombe, UK) with the actuator displaced at
2 mm/minute. Both displacement and load were recorded.
Ultimate strength (maximal load, newton [N]), stiffness
(slope of the linear part of the curve, representing the elastic
deformation, N/mm), and energy (surface under the curve,
N ϫ mm) were calculated. Reproducibility was 4.8% for
vertebrae, 5.8% for proximal tibia, 3.3% for midshaft tibia,
and 4.5% for femoral neck, and it was evaluated as the CV
of pair sample measurements (left/right, L3–L4).
BMD measurements
BMD was measured by DXA using a Hologic QDR-1000
instrument adapted to small animals (Hologic, Inc.,
Waltham, MA, USA).(20) An ultra–high-resolution mode
(line spacing, 0.254 mm, and resolution, 0.127 mm) was
used with a 0.9-mm-diameter collimator. During the exam-
ination, the animals were anesthetized with ketamine hy-
drochloride (100 mg/kg body weight). BMD, bone mineral
content (BMC), and scanned area were recorded in vivo at
the level of lumbar vertebrae, tibia, and femur, as previously
described.(19) Ex vivo measurements were performed at the
level of the isolated vertebral body and femur in 2.5 cm of
saline solution. The in vivo reproducibility was evaluated
by the CV of repeated measurements with repositioning and
was Ͻ1.8%. The stability of the instrument was controlled
by scanning a phantom six times weekly.
Bone mechanical testing
Microtomographic histomorphometry by
microcomputed tomography
The lumbar spine, tibia, and femur were excised imme-
diately after death and frozen at Ϫ20°C in plastic bags. The
night before mechanical testing, the bones were thawed
Parameters of mass and architecture of the vertebrae were
slowly at 7°C and then maintained at room temperature. The investigated using microtomographic histomorphometry
L4 vertebrae were isolated from the lumbar spine at the with a high-resolution microcomputed tomography (␮CT)
level of the intervertebral discs. The vertebral pedicules system (␮CT 20; Scanco Medical AG, Bassersdorf, Swit-
were dissected out carefully to avoid any damage to the zerland) as previously described.(22,23) In summary, three-
cortical shell. Because the caudal and cranial surfaces of the dimensional (3D) images of a vertebra were acquired with a
rat vertebral body are not parallel, 1 mm of the caudal and voxel size of 20 ␮m in all spatial directions. No sample
cranial parts of each vertebral body was embedded in meth- preparation was needed, and the vertebrae were secured in
ylmethacrylate cement (Technovit 4071; Heraeus Kulzer a cylindrical sample holder in air. The resulting gray-scale
GmbH, Wehrheim, Germany) to ensure regular distribution images were segmented using a low-pass filter to remove
of the compressive forces. Between the different steps of noise and a fixed threshold to extract the mineralized bone
preparation, each specimen was kept immersed in physio- phase. The trabecular and cortical parts of the vertebral
logical solution. The fibula was removed and the length of bodies were separated with semiautomatically drawn con-
the tibia (distance from intermalleolar to intercondylar re- tours.
gion) was measured using a caliper with electronic digital
From the binarized images, structural indices were as-
display and the middle of the shaft was determined. The sessed. Relative bone volume (BV/TV), trabecular number

ESSENTIAL AMINO ACIDS INCREASE BONE STRENGTH IN RATS
1267
TABLE 2. EFFECT OF ISOCALORIC EAA SUPPLEMENTS ON BMD AND BONE MECHANICAL PROPERTIES
OVX
baseline
OVX
OVX
OVX
SHAM
Casein (%)
EAA (%)
No. of rats
2.5
—
6
2.5
—
8
2.5
2.5
9
2.5
5.0
9
15
—
8
Lumbar spine
BMD (mg/cm2)
US (N)
Stiffness (N/mm)
Energy (N ϫ mm)
Proximal tibia
BMD (mg/cm2)
US (N)
Stiffness (N/mm)
Energy (N ϫ mm)
Midshaft tibia
BMD (mg/cm2)
US (N)
Stiffness (N/mm)
Energy (N ϫ mm)
Femoral neck
BMD (mg/cm2)
US (N)
103.4 Ϯ 6.3
95.08 Ϯ 7.2
401 Ϯ 39
75.6 Ϯ 4.5*
73.5 Ϯ 9.2
282 Ϯ 42
102.8 Ϯ 4.8°
105.4 Ϯ 14.5
373 Ϯ 48
118.1 Ϯ 3.3*,°
157.7 Ϯ 15.0*,°
442 Ϯ 58
143.4 Ϯ 4.3*,°
165.3 Ϯ 11.9*,°
475 Ϯ 51
14.9 Ϯ 1.1
13.2 Ϯ 2.2
18.4 Ϯ 3.3
35.4 Ϯ 5.5*,°
34.1 Ϯ 4.5*,°
247.4 Ϯ 3.8
77.1 Ϯ 4.4
286 Ϯ 56
220.7 Ϯ 4*
63.1 Ϯ 9.0
366 Ϯ 47
251.3 Ϯ 2.1°
99.0 Ϯ 6.5*,°
283 Ϯ 59
261.7 Ϯ 1.7*,°
118.8 Ϯ 3.0*,°
364 Ϯ 29
288.9 Ϯ 3.2*,°
159.2 Ϯ 8.5*,°
390 Ϯ 37
24.4 Ϯ 5.7
13.8 Ϯ 3.5
37.7 Ϯ 3.5*,°
37.2 Ϯ 3.3*,°
48.5 Ϯ 4.0*,°
236.2 Ϯ 3.4
73.9 Ϯ 4.7
184 Ϯ 9
210.8 Ϯ 4.7*
55.5 Ϯ 3.0*
190 Ϯ 11
233.1 Ϯ 1.1°
81.5 Ϯ 4.4°
216 Ϯ 19
240.3 Ϯ 3.2°
77.6 Ϯ 2.5°
201 Ϯ 11
251.4 Ϯ 3.1*,°
87.5 Ϯ 5.3*,°
213 Ϯ 17
20.5 Ϯ 1.6
12.1 Ϯ 1.2*
20.37 Ϯ 1.8°
18.73 Ϯ 1.2°
20.2 Ϯ 2.0°
198.3 Ϯ 4.0
56.33 Ϯ 3.1
265 Ϯ 31
167.1 Ϯ 5.3*
55.3 Ϯ 2.8
225 Ϯ 31
207.9 Ϯ 2.6°
74.6 Ϯ 3.2*,°
274 Ϯ 32
218.2 Ϯ 2.2*,°
75.4 Ϯ 3.7*,°
275 Ϯ 27
253.7 Ϯ 2.4*,°
80.4 Ϯ 5.2*,°
390 Ϯ 23*,°
9.1 Ϯ 0.8
Stiffness (N/mm)
Energy (N ϫ mm)
8.4 Ϯ 0.8
9.0 Ϯ 1.2
12.8 Ϯ 2.4
13.5 Ϯ 1.3*,°
Values are means Ϯ SEM. The EAA supplements were started 10 weeks after the OVX and reduced protein intake. They were
administered for 16 weeks.
EAA, essential amino acid; US, ultimate strength (N).
* p Ͻ 0.05 vs. baseline and ° vs. OVX low-protein group as evaluated by ANOVA.
(Tb.N), trabecular thickness (Tb.Th), and trabecular sepa- nation of bone strength were evaluated by stepwise multiple
ration (Tb.Sp) were calculated by measuring directly the 3D regression analysis.
distances(23,24) in the trabecular network. Connectivity den-
sity based on Euler number(25) and the structure model
index (SMI) were calculated. The SMI quantifies the plate
RESULTS
versus rod characteristics of trabecular bone,(24) where an
Effect of essential amino acid supplements on BMD
SMI of 0 indicates a purely plate-shaped bone, an SMI of
3 a rodlike bone, and values in-between stand for a mixture
of plates and rods.
The combination of isocaloric low-protein intake and
OVX induced a marked decrease of BMD at all the inves-
tigated skeletal sites (Fig. 1). At the levels of the lumbar
spine, proximal tibia, and femoral neck, which are sites
composed of both cancellous and cortical bone, essential
amino acid supplements prevented further bone loss when
introduced 10 weeks after beginning the low-protein diet
(Table 1 and Fig. 1). However, BMD remained significantly
lower than in SHAM controls fed a 15% casein-containing
diet. Similar effects were observed when the BMD of iso-
lated bone was measured ex vivo (Table 2). At the level of
the midshaft tibia, mainly constituted of cortical bone, bone
loss was delayed as compared with other sites.
Biochemical determinations
Plasma osteocalcin and IGF-I were measured by radio-
immunoassay using reagents from Biomedical Technolo-
gies (Stoughton, MA, USA) for the former, and a kit from
Nichols Institute (San Juan Capistrano, CA, USA) after
extraction by acid-ethanol and cryoprecipitation, for the
latter. Total urinary deoxypyridinoline was determined after
acid hydrolysis using a kit from Quidel (Mountain View,
CA, USA).
Statistical analysis
Effect of essential amino acid supplements on bone
strength
All results are expressed as mean Ϯ SEM. Significance of
difference between groups was evaluated with a one-way
ANOVA followed by a Fisher’s test. Possible interactions
The BMD decrease induced by the combination of iso-
between bone density and bone dimensions in the determi- caloric low-protein intake and OVX was associated with

1268
AMMANN ET AL.
the midshaft tibia, mainly constituted of cortical bone, the
treatment prevented the alteration of mechanical properties
(Table 2 and Fig. 2). The linear regression between vertebra
ultimate strength and BMD was steeper in animals receiving
essential amino acid supplements (slope ϭ 2.60, r2 ϭ 0.48,
and p ϭ 0.004) than in control animals (slope ϭ 1.34, r2 ϭ
0.79, and p ϭ 0.0001). A similar trend was observed when
energy was analyzed for different skeletal sites studied.
Effect of essential amino acid supplements on
microtomographic histomorphometry
To further investigate the relation between a marked
improvement in bone strength and small changes in BMD
induced by essential amino acid supplements, we analyzed
bone microarchitecture by microtomographic histomor-
phometry, a variable that is not captured by DXA measure-
ment.
Trabecular bone architecture changes such as decreased
Tb.N, spacing connectivity, and structure (rods instead of
plates) were observed in OVX rats fed a low-protein diet
(Table 3 and Fig. 3). The 5% essential amino acid supple-
ments increased trabecular BV/TV and Tb.N, and reduced
trabecular spacing as compared with OVX rats fed an iso-
caloric low-protein diet (Table 3 and Fig. 3). This was
associated with a partial recovery of the indices of connec-
tivity and a shift of trabecular structure from rods to mixed
plates and rod forms. However, trabecular BV/TV was still
significantly lower in the 5% essential amino acid supple-
ments group as compared with the SHAM group. On the
other hand, a significant thickening of the cortices was
observed at the level of the vertebrae in rats treated with the
2.5% and 5% essential amino acid supplements (Table 3),
with values close to those of SHAM animals (p ϭ 0.24 and
p ϭ 0.42 for 2.5% and 5% essential amino acid supplements
vs. SHAM, respectively). This effect on cortical bone was
observed also at all the other investigated sites. Both BV/TV
and outer diameter were increased by the dietary supple-
ments (Table 3).
FIG. 2. Effect of isocaloric essential amino acid supplements on bone
strength in adult female rats made osteoporotic by OVX and a low-
protein diet. Ultimate strength (N; means Ϯ SEM) was measured at the
level of the vertebral body and proximal tibia by an axial compression
test and a three-point bending test at the level of the midshaft tibia. The
measurements were performed 10 weeks after OVX and of feeding a
low-protein diet (baseline) and after 16 weeks of essential amino acid
supplements. SHAM refers to sham-operated animals fed a 15% casein
diet. Using ANOVA, * indicates a significant difference (p Ͻ 0.01)
from baseline and ° from OVX rats fed a low-protein diet.
Effect of essential amino acid supplements on bone
remodeling
As compared with SHAM animals, OVX rats fed a low-
protein diet had an increase in urinary deoxypyridinoline
excretion, reflecting increased bone resorption. The serum
level of the bone formation marker osteocalcin was not
affected (Table 4). With essential amino acid supplements,
urinary deoxypyridinoline excretion decreased and serum
osteocalcin increased (Table 4) but did not reach the values
of SHAM animals.
significant alteration of bone mechanical properties (ulti-
mate strength, stiffness, and energy absorbed) at all the
investigated sites (Table 2 and Fig. 2). At the levels of the
Effect of essential amino acid supplements on IGF-I
As previously shown,(8) isocaloric low-protein diet de-
lumbar spine, proximal tibia, and femoral neck, essential creased plasma IGF-I (Fig. 4). In this condition, OVX was
amino acid supplements dose-dependently increased bone not associated, as expected, with an increment of plasma
strength (Table 2 and Fig. 2). A full recovery was observed IGF-I. Essential amino acid supplements increased plasma
at the level of the lumbar spine in rats receiving the sup- IGF-I to levels significantly higher than in SHAM animals
plement containing 5% essential amino acids. At the level of fed the 15% casein diet (Fig. 4). Despite the strict pair-

ESSENTIAL AMINO ACIDS INCREASE BONE STRENGTH IN RATS
1269
TABLE 3. EFFECTS OF ISOCALORIC EAA SUPPLEMENTS ON BONE ARCHITECTURE
OVX baseline
OVX
OVX
OVX
SHAM
Casein (%)
EAA (%)
Tb.N
Tb.Th (␮m)
Tb.Sp (mm)
SMI
2.5
—
2.5
—
2.5
2.5
2.5
5.0
15
—
2.86 Ϯ 0.45
57.7 Ϯ 1.0
0.468 Ϯ 0.149
2.53 Ϯ 0.11
1.13 Ϯ 0.09
67.4 Ϯ 3.2
0.944 Ϯ 0.079
2.19 Ϯ 0.20
1.29 Ϯ 0.12
66.4 Ϯ 1.6
0.836 Ϯ 0.091
2.25 Ϯ 0.10
2.37 Ϯ 0.24*
67.8 Ϯ 1.3
0.517 Ϯ 0.111
1.98 Ϯ 0.07°
3.96 Ϯ 0.16*
79.9 Ϯ 6.4
0.256 Ϯ 0.011
0.87 Ϯ 0.37°
Values are means Ϯ SEM. Results are Tb.N, Tb.Th, Tb.Sp, and SMI. SMI is an estimate of the plate vs. the rod characteristics of
trabecular bone (0 value ϭ purely plate-shaped bone, 3 value ϭ purely rodlike bone). The EAAs were started 10 weeks after the OVX
and reduced protein intake. They were administered for 16 weeks.
EAA, essential amino acid.
* p Ͻ 0.05 vs. baseline and ° vs. OVX low-protein group as evaluated by ANOVA.
feeding and identical energy intake, a low-protein diet was
associated with decreased body weight, which then was
dose-dependently corrected by essential amino acid supple-
ments. Body weight was 238.9 Ϯ 5.1 g, 176.4 Ϯ 4.8 g,
256.3 Ϯ 8.7 g, and 266.7 Ϯ 2.4 g in SHAM 15% casein,
OVX 2.5% casein, and 2.5% and 5% essential amino acid
treatment, respectively. Similar effects on muscle weight of
the lower limbs were observed (4.25 Ϯ 0.18 g, 3.06 Ϯ
0.07 g, 4.15 Ϯ 0.32 g, and 4.43 Ϯ 0.20 g for the corre-
sponding groups).
DISCUSSION
This study shows that nutritional intervention with essen-
tial amino acids is able to improve bone mechanical prop-
erties in female rats made osteoporotic by OVX and a
low-protein diet. Microarchitecture assessment indicates a
thickening of bone cortices. This effect is observed in as-
sociation with an increase in plasma IGF-I. Analysis of
biochemical markers suggests a decrease in bone resorption
and a stimulation of bone formation.
To mimic the situation observed in the elderly patient
with osteoporosis and a hip fracture,(1,5,8,26) adult rats were
OVX and received a low-protein diet. It is well recognized
that OVX results in alteration of BMD and mechanical
properties, essentially at skeletal sites containing cancellous
bone.(21,27–29) Skeletal sites containing mainly cortical bone
are not significantly altered by estrogen deficiency, at least
in rodents.(21,28,29) A reduced protein intake is associated
with a marked decrease of BMD, also at mainly cortical
sites, despite an identical energy intake in all the different
groups. A possible explanation could be that IGF-I did not
increase as usually observed after OVX.(8,21,30) Further-
FIG. 3. Effect of isocaloric essential amino acid supplements on
trabecular BV/TV and cortical bone thickness in lumbar vertebral body
of adult female rats made osteoporotic by OVX and a low-protein diet.
BV/TV (%; means Ϯ SEM) and cortical thickness (Cort Th, mm) were
measured by microtomographic histomorphometry (␮CT) at the level
of the L4 vertebra. The measurements were performed 10 weeks after
OVX and of feeding a low-protein diet (baseline) and after 16 weeks of
more, an additional increment of cortical bone turnover
could be expected in conditions of reduced protein intake.
Thus, this model seems to mimic the situation of elderly
women with senile osteoporosis. Previous studies indicated
that no modification of BMD, strength, and turnover, and of
somatotrope or gonadotrope axis could be observed in rats
fed 5, 10, or 15% casein. This effect was only observed in
rats fed the 2.5% casein isocaloric diet corresponding to a
reduction of 50% of the minimal protein requirement in the
essential amino acid supplements. SHAM refers to sham-operated
animals fed a 15% casein diet. All groups were fed isocaloric diets.
Using ANOVA, * indicates a significant difference (p Ͻ 0.01) from
baseline and ° from OVX rats fed a low-protein diet.

1270
AMMANN ET AL.
TABLE 4. EFFECTS OF ISOCALORIC EAA SUPPLEMENTS ON BIOCHEMICAL MARKERS OF BONE REMODELING
OVX
baseline
OVX
OVX
OVX
SHAM
Casein (%)
EAA (%)
2.5
—
2.5
—
2.5
2.5
2.5
5.0
15
—
No. of Rats
Plasma osteocalcin (nmol/liter)
Urinary deoxypyridinoline (nmol/24 h)
8
8
9
9
8
1.76 Ϯ 0.13
NA
2.14 Ϯ 0.13
3.1 Ϯ 0.2
2.26 Ϯ 0.07
2.69 Ϯ 0.08*,°
2.6 Ϯ 0.7*
2.03 Ϯ 0.07
0.9 Ϯ 0.2
2.3 Ϯ 0.4*,°
Values are means Ϯ SEM. Treatment was started 10 weeks after the OVX and reduced protein intake. Amino acid supplements were
administered for 16 weeks. Determinations were performed before (baseline) and 16 weeks later. Baseline urinary deoxypyridinoline was
not assessed (NA).
EAA, essential amino acid.
* p Ͻ 0.05 vs. OVX rats fed 2.5% casein diet and ° vs. SHAM rats as evaluated by ANOVA.
properties were evaluated by a compression test of the intact
vertebral body. This kind of sample preparation respects the
cortex but also the microarchitecture of the trabeculae and
their connection with the cortical shell and ensures the full
integration of all the determinants of bone strength.
One of the major determinants of bone strength is BMD,
which predicts up to 60% of the variance of bone
strength.(19,21) The essential amino acid supplements pre-
vented further bone loss and even increased BMD at the
level of the spine, proximal tibia, and femoral neck. Nev-
ertheless, there was a discrepancy between the moderate
influence of essential amino acid supplements on BMD and
the marked effect on bone strength, which resulted in a full
recovery at the level of the vertebrae. It is not excluded that
areal BMD measurement could preclude the detection of
microarchitecture modification.
Microarchitecture determinants were investigated at the
level of the vertebrae using microtomographic histomor-
FIG. 4. Effect of isocaloric essential amino acid supplements on
plasma IGF-I in adult female rats made osteoporotic by OVX and a
low-protein diet. Plasma IGF-I (nmol/liter, n ϭ 6–9, and means Ϯ
SEM) was measured in OVX rats fed a low-protein diet (open circles)
phometry and structural assessment. Their accuracy and
precision have been established previously.(22,23) This tech-
treated with essential amino acid supplements at a dose of 2.5% (open nique provides information regarding the classical parame-
triangle) or 5% (closed triangle), as well as in control rats SHAM fed
a 15% casein diet (open square). The treatment period is represented by
a stippled area. Using ANOVA, * indicates a significant difference
(p Ͻ 0.01) from OVX rats fed a low-protein diet and ° from SHAM
with 5% essential amino acids as compared with the time
controls.
ters of trabecular bone mass as evaluated by static histo-
morphometry. The results indicate higher trabecular BV/
TV, Tb.N, and decreased trabecular spacing in rats treated
controls, as well as a prevention of further deterioration
compared with baseline values. SMI distinguishes between
rods and plates of bony trabeculae. Our results suggest a
moderate but significant evolution of trabecular structures
rat.(8) This is real so long as the energy intake is normal and
back from rods to a mixed plates and rods form in those
animals treated with 5% essential amino acids as compared
with baseline values. However, the most impressive effect
concerns cortical thickness. The latter was markedly in-
creased in both groups receiving essential amino acid sup-
plements. An integration of trabeculae lying along the en-
dosteal cortical surface cannot be excluded. This increase of
cortical thickness, together with modification of trabecular
architecture, could account for the increase of bone strength
in rats fed 5% essential amino acid supplements.
the rats have stopped their growth.
The ability of nutritional intervention to restore bone
strength was then investigated. The type of nutritional sup-
plement was selected based on preliminary studies indicat-
ing that only essential amino acids (but not selected essen-
tial amino acids or nonessential amino acids) increased
circulating plasma IGF-I in rats fed an isocaloric low-
protein diet. The essential amino acid supplements in-
creased bone strength at skeletal sites formed by both cor-
tical and cancellous bone such as the vertebral body,
proximal femur, and proximal tibia and also at the level of
OVX induces an increment of IGF-I under normal casein
sites formed mainly by cortical bone such as the midshaft of intake. In rats fed a low-protein diet, this does not occur.(8)
long bones. A full restoration of mechanical properties was Amino acid supplements increased plasma IGF-I to values
even observed at the level of the vertebrae. The mechanical even 30% higher than SHAM receiving the 15% casein diet.

ESSENTIAL AMINO ACIDS INCREASE BONE STRENGTH IN RATS
1271
mur: An injury response to altered metabolism? Lancet 1:143–
146.
4. Tinetti ME, Speechley M, Ginter SF 1988 Risk factors for falls
among elderly persons living in the community. N Engl J Med
319:1701–1707.
5. Geinoz G, Rapin CH, Rizzoli R, Kraemer R, Buchs B, Slos-
man D, Michel JP, Bonjour JP 1993 Relationship between
bone mineral density and dietary intakes in the elderly. Osteo-
poros Int 3:242–248.
6. Rizzoli R, Ammann P, Bourrin S, Chevalley T, Bonjour JP
2001 Protein intake and bone homeostasis. In: Burckhardt P,
Dawson-Hughes B, Heaney RP (eds.) Nutritional Aspects of
Osteoporosis. Academic Press, San Diego, CA, USA, pp.
219–235.
7. Heaney RP 2000 Calcium, dairy products and osteoporosis
J Am Coll Nutr 19:83S–99S.
8. Ammann P, Bourrin S, Bonjour JP, Meyer JM, Rizzoli R 2000
Protein undernutrition-induced bone loss is associated with
decreased IGF-I levels and estrogen deficiency. J Bone Miner
Res 15:683–690.
9. Bourrin S, Ammann P, Bonjour JP, Rizzoli R 2000 Dietary
protein restriction lowers plasma insulin-like growth factor I
(IGF-I), impairs cortical bone formation, and induces osteo-
blastic resistance to IGF-I in adult female rats. Endocrinology
141:3149–3155.
Because estrogen could down-regulate the transcription of
IGF-I,(31,32) this overshot of IGF-I secondary to essential
amino acid supplements could possibly be explained by the
lack of estrogen repression on IGF-I production. Plasma
IGF-I could be responsible for the bone anabolic effect of
essential amino acid supplements. Indeed, administration of
IGF-I increases BMD and strength.(18,19)
To address the issue of the mechanisms involved, mark-
ers of bone remodeling such as deoxypyridinoline and os-
teocalcin were measured. An increase of bone resorption, as
evaluated by urinary deoxypyridinoline excretion without a
commensurate change of the bone formation marker osteo-
calcin, is found in OVX rats fed a low-protein diet. This can
be interpreted as an uncoupling between resorption and
formation, leading to accelerated bone loss.(8,9,17) Essential
amino acid supplements may attenuate bone resorption and
stimulate bone formation as indicated by changes in bone-
remodeling markers, compatible with a recoupling of the
two processes. Thus, a positive coupling could have led to
a positive bone balance and to architectural modifications of
both cortical and cancellous bone.
Despite an identical caloric intake, body weight de-
creased under a low-protein diet but was fully corrected by 10. Garnero P, Sornay-Rendu E, Delmas PD 2000 Low serum
IGF-1 and occurrence of osteoporotic fractures in postmeno-
pausal women. Lancet 355:898–899.
11. Canalis E, Agnusdei D 1996 Insulin-like growth factors and
their role in osteoporosis. Calcif Tissue Int 58:133–134.
12. Rosen CJ 1994 Growth hormone, insulin-like growth factors,
and the senescent skeleton: Ponce de Leon’s Fountain revis-
ited? J Cell Biochem 56:348–356.
13. Isley WL, Underwood LE, Clemmons DR 1983 Dietary com-
ponents that regulate serum somatomedin-C concentrations in
humans. J Clin Invest 71:175–182.
14. Thissen JP, Ketelslegers JM, Underwood LE 1994 Nutritional
regulation of the insulin-like growth factors. Endocr Rev 15:
80–101.
15. VandeHaar MJ, Moats-Staats BM, Davenport ML, Walker JL,
Ketelslegers JM, Sharma BK, Underwood LE 1991 Reduced
serum concentrations of insulin-like growth factor-I (IGF-I) in
protein-restricted growing rats are accompanied by reduced
IGF-I mRNA levels in liver and skeletal muscle. J Endocrinol
130:305–312.
16. Straus DS, Takemoto CD 1990 Effect of dietary protein de-
privation on insulin-like growth factor (IGF)-I and -II, IGF
binding protein-2, and serum albumin gene expression in rat.
Endocrinology 127:1849–1860.
17. Bourrin S, Toromanoff A, Ammann P, Bonjour JP, Rizzoli R
2000 Dietary protein deficiency induces osteoporosis in aged
male rats. J Bone Miner Res 15:1555–1563.
18. Ammann P, Rizzoli R, Muller K, Slosman D, Bonjour JP 1993
IGF-I and pamidronate increase bone mineral density in ovari-
ectomized adult rats. Am J Physiol 265:E770–E776.
19. Ammann P, Rizzoli R, Meyer JM, Bonjour JP 1996 Bone
density and shape as determinants of bone strength in IGF-I
and/or pamidronate-treated ovariectomized rats. Osteoporos
Int 6:219–227.
the amino acid supplements. Changes in hindlimb muscle
weight followed those of body weight, with an increase
under amino acid supplements.
In conclusion, nutritional intervention with essential
amino acid supplements is able to increase bone strength in
adult female rats made osteoporotic by OVX and a low-
protein diet. The effects of such supplements in animals
with a normal protein intake remain to be established. This
was associated with changes in bone microarchitecture.
These observations provide the preclinical basis for evalu-
ating the effects of essential amino acid supplements in the
elderly who are frail and/or osteoporotic.
ACKNOWLEDGMENTS
We thank I. Badoud for biomechanical testing, S. Cle´m-
ent for animal management, M. Linder for biochemical
determinations, C. Godin for technical assistance, and M.
Perez for typing the manuscript. Dr. R. Stern is acknowl-
edged for reading the manuscript. This project was sup-
ported by grants from the Swiss National Science Research
Foundation (grants 32–49757.96 and 32–58880.99) and
from Novartis Nutrition, Ltd. (Berne, Switzerland).
REFERENCES
1. Schurch MA, Rizzoli R, Slosman D, Vadas L, Vergnaud P,
Bonjour JP 1998 Protein supplements increase serum insulin-
like growth factor-I levels and attenuate proximal femur bone
loss in patients with recent hip fracture. A randomized, double-
blind, placebo-controlled trial. Ann Intern Med 128:801–809.
2. Wootton R, Brereton PJ, Clark MB, Hesp R, Hodkinson HM,
Klenerman L, Reeve J, Slavin G, Tellez-Yudilevich M 1979
Fractured neck of femur in the elderly: An attempt to identify
patients at risk. Clin Sci (Colch) 57:93–101.
20. Ammann P, Rizzoli R, Slosman D, Bonjour JP 1992 Sequen-
tial and precise in vivo measurement of bone mineral density
in rats using dual-energy X-ray absorptiometry. J Bone Miner
Res 7:311–316.
21. Ammann P, Bourrin S, Bonjour JP, Brunner F, Meyer JM,
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.
3. Bastow MD, Rawlings J, Allison SP 1983 Undernutrition,
hypothermia, and injury in elderly women with fractured fe-

1272
AMMANN ET AL.
22. Muller R, Koller B, Hildebrand T, Laib A, Gianolini S, Rueg- 29. Kalu DN 1991 The ovariectomized rat model of postmeno-
pausal bone loss. Bone Miner 15:175–191.
segger P 1996 Resolution dependency of microstructural prop-
erties of cancellous bone based on three-dimensional mu-
tomography. Technol Health Care 4:113–119.
30. Kalu DN, Arjmandi BH, Liu CC, Salih MA, Birnbaum RS
1994 Effects of ovariectomy and estrogen on the serum levels
of insulin-like growth factor-I and insulin-like growth factor
binding protein-3. Bone Miner 25:135–148.
31. Borski RJ, Tsai W, DeMott-Friberg R, Barkan AL 1996 Reg-
ulation of somatic growth and the somatotropic axis by go-
nadal steroids: Primary effect on insulin-like growth factor I
gene expression and secretion. Endocrinology 137:3253–
3259.
23. Laib A, Barou O, Vico L, Lafage-Proust MH, Alexandre C,
Ruegsegger P 2000 3D micro-computed tomography of tra-
becular and cortical bone architecture with application to a rat
model of immobilisation osteoporosis. Med Biol Eng Comput
38:326–332.
24. Hildebrand T, Ruegsegger P 1997 Quantification of Bone
Microarchitecture with the Structure Model Index. Comput
Methods Biomech Biomed Engin 1:15–23.
25. Odgaard A 1997 Three-dimensional methods for quantifica-
tion of cancellous bone architecture. Bone 20:315–328.
26. Day SM, DeHeer DH 2001 Reversal of the detrimental effects
of chronic protein malnutrition on long bone fracture healing.
J Orthop Trauma 15:47–53.
32. Ho KK, O’Sullivan AJ, Weissberger AJ, Kelly JJ 1996 Sex
steroid regulation of growth hormone secretion and action.
Horm Res 45:67–73.
Address reprint requests to:
P. Ammann, M.D.
27. Ammann P, Rizzoli R, Bonjour JP, Bourrin S, Meyer JM,
Vassalli P, Garcia I 1997 Transgenic mice expressing soluble
tumor necrosis factor-receptor are protected against bone loss
caused by estrogen deficiency. J Clin Invest 99:1699–1703.
28. Bonjour JP, Ammann P, Rizzoli R 1999 Importance of pre-
clinical studies in the development of drugs for treatment of
osteoporosis: A review related to the 1998 WHO guidelines.
Osteoporos Int 9:379–393.
Division of Bone Diseases
Department of Internal Medicine
University Hospital
CH-1211 Geneva 14, Switzerland
Received in original form May 27, 2001; in revised form January
29, 2002; accepted March 8, 2002.