Full text of DOI:10.1359/jbmr.2002.17.9.1639

JOURNAL OF BONE AND MINERAL RESEARCH
Volume 17, Number 9, 2002
©
2002 American Society for Bone and Mineral Research
Regional Alterations of Type I Collagen in Rat Tibia
Induced by Skeletal Unloading
1
2
3
1
MASASHI SHIIBA, SARA B. ARNAUD, HIDEKI TANZAWA, EIJI KITAMURA, and
1
MITSUO YAMAUCHI
ABSTRACT
Skeletal unloading induces loss of mineral density in weight-bearing bones that leads to inferior bone
mechanical strength. This appears to be caused by a failure of bone formation; however, its mechanisms still
are not well understood. The objective of this study was to characterize collagen, the predominant matrix
protein in bone, in various regions of tibia of rats that were subjected to skeletal unloading by 4 weeks tail
suspension. Sixteen male Sprague–Dawley rats (4 months old) were divided into tail suspension and ambu-
latory controls (eight rats each). After the tail suspension, tibias from each animal were collected and divided
into five regions and collagen was analyzed. The collagen cross-linking and the extent of lysine (Lys)
hydroxylation in unloaded bones were significantly altered in proximal epiphysis, diaphysis, and, in particular,
proximal metaphysis but not in distal regions. The pool of immature/nonmineralized collagen measured by its
extractability with a chaotropic solvent was significantly increased in proximal metaphysis. These results
suggest that skeletal unloading induced an accumulation of post-translationally altered nonmineralized
collagen and that these changes are bone region specific. These alterations might be caused by impaired
osteoblastic function/differentiation resulting in a mineralization defect. (J Bone Miner Res 2002;17:1639–1645)
Key words: type I collagen, cross-link, lysyl hydroxylation, bone region, tail suspension
INTRODUCTION
etal unloading may cause inhibitions of osteoprogenitor cell
functions and alterations in mRNA expression of growth
T IS commonly accepted that alterations in the loading factors and bone matrix proteins in the osteoblasts.
(9–12)
I
state of the skeletal system result in changes in bone mass, During spaceflight, the maturation of newly formed matrix
(1)
matrix composition, and mechanical properties. Bone loss and mineral moieties in weight-bearing bones seems to be
(13)
and subsequent impairment of bone mechanical properties impaired.
Thus, it appears that skeletal unloading in-
are well known changes that occur during long-term bed
rest, paralysis, immobilization, and spaceflight.
hindlimb elevation model of rat by tail suspension has
been widely used to simulate the bone changes induced by
weightlessness including a temporary cessation of bone
duces a shift of cellular populations toward a less mature
phenotype resulting in matrix alterations and eventually
bone loss.
(
2)
(3)
(4)
(5)
A
(
6)
Type I collagen is the predominant matrix molecule in
(14)
bone
and its physicochemical properties significantly
(
7)
(8)
formation and inferior mechanical properties. The skel-
differ from other tissues because of its specific post-
(15)
translational modifications. Hydroxylation of specific ly-
sine residues is an important post-translational modification
The authors have no conflict of interest.
1
Dental Research Center, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, USA.
National Aeronautics and Space Administration-Ames Research Center, Moffett Field, California, USA.
Department of Oral Surgery, School of Medicine, Chiba University, Chiba, Japan.
2
3
1
639

1
640
SHIIBA ET AL.
in defining cross-linking and in attaching carbohydrate moi-
(16–18)
eties that may play a number of structural roles.
Covalent intermolecular cross-linking is critical in provid-
(15,19)
ing the fibrils with tensile strength.
Studies have indi-
cated that these post-translational modifications of collagen
play crucial roles in bone formation, mineralization, and
(
15,20–24)
maturation.
Here, we hypothesize that bone cells synthesize post-
translationally altered collagen in response to skeletal un-
loading, which eventually may lead to a mineralization
defect. The objective of this study was to characterize
collagen in various regions of rat tibia after 4 weeks of
skeletal unloading by tail suspension.
MATERIALS AND METHODS
Rat tail suspension
Sixteen male Sprague–Dawley rats (4 months old) were
divided into eight control (ambulatory) and eight experi-
mental (tail suspended) groups. Tail suspension was per-
formed according to the method described by Wronski and
(
6)
Morey-Holton. Ambulatory rats were fed the average
amount eaten by suspended rats daily. All rats were killed
after 4 weeks when the tibias from both groups were col-
lected and stored at Ϫ20°C until processing for analysis. All
procedures were approved by the Animal Care and Use
Committee at the National Aeronautics and Space Admin-
istration (NASA) Ames Research Center.
FIG. 1. Scheme of bone dissection. The tibia was divided into five
regions: proximal epiphysis, proximal metaphysis, diaphysis, distal
metaphysis, and distal epiphysis.
droxyproline (Hyp) and expressed as micrograms of colla-
gen per milligrams of dried bone powder.
Preparation of pulverized bone sample
The fibula was removed and soft tissues including peri-
osteum were cleaned off the bone surface and each tibia was
divided into five regions: proximal epiphysis, proximal me-
taphysis, diaphysis, distal metaphysis, and distal epiphysis
Collagen cross-link analysis
An aliquot of pulverized bone sample was demineralized
with 0.5 M of EDTA (Fisher Scientific), 0.05 M of Tris-
HCl, pH 7.4, and washed with cold distilled water exhaus-
(Fig. 1). Bone marrow in the metaphysis and diaphysis was
removed by flushing with cold PBS, pH 7.4, and then all
segments were pulverized to a fine powder under liquid N2
(26)
tively and lyophilized.
Five milligrams of demineralized
using a Spex Freezer Mill (Spex, Inc., Metuchen, NJ, USA).
Pulverization was performed in the same way for all sam-
ples to avoid potential artifacts due to the powder size.
collagen then was suspended, reduced with standardized
3
NaB H , hydrolyzed with 6N constant boiling HCl, and
4
(25)
(18,26)
subjected to amino acid and cross-link analyses.
Pulverized samples were washed with cold 0.015 M of
N-trismethyl-2-aminoethanesulfonic acid (TES; Fisher Sci-
entific, Fair Lawn, NJ, USA), 0.005 M of Tris-HCl, pH 7.4,
and then with cold distilled water several times by centrif-
ugation (2700g) and lyophilized.
The cross-link precursor aldehyde 2-hydroxy-5-amino-5-
carboxypentanal (Hyl ) and the reducible cross-links
ald
dehydro-dihydroxylysinonorleucine (deH-DHLNL) and
dehydro-hydroxylysinonorleucine (deH-HLNL) and their
ketamines were analyzed as their reduced forms dihy-
droxynorleucine (DHNL), DHLNL, and HLNL respec-
tively. The reducible and nonreducible cross-links were
quantified and expressed as moles of cross-links per mole of
Analysis of amino acid composition and
collagen content
(26,27)
collagen as described.
Two milligrams of dried bone powder was hydrolyzed
with 200 ␮l of constant boiling 6N HCl (Pierce, Rockford,
Extraction and analysis of immature collagen
IL, USA) in vacuo, after flushing with N gas, for 22 h at
2
1
05°C. The hydrolysate was dried, reconstituted in 400 ␮l
Approximately 10 mg of pulverized bone sample was
of distilled water, filtered, and then an aliquot of hydrolysate suspended in 1 ml of 6 M of guanidine-HCl, 0.05 M of
was subjected to amino acid analysis as described previ- Tris-HCl, pH 7.4, containing a cocktail of protease inhibi-
(
18,26)
ously.
Amino acid composition of bone matrix protein tors, that is, 0.5 mM of phenylmethanesulfonylfluoride (Re-
was expressed as residues per 1000 total amino acids. The search Organics, Cleveland, OH, USA), 5 mM of benzami-
collagen content was calculated based on the value of hy- dine (Aldrich, Milwaukee, WI, USA), 1.5 ␮M of pepstatin

REGIONAL ALTERATION OF COLLAGEN IN UNLOADED BONE
1641
TABLE 1. AMINO ACID COMPOSITIONS OF BONE MATRIX
Proximal epiphysis
Proximal metaphysis
Diaphysis
Distal metaphysis
Distal epiphysis
Amb.
Susp.
Amb.
Susp.
Amb.
Susp.
Amb.
Susp.
Amb.
Susp.
†
†
Hyp
Asp
63.8 Ϯ 2.3
66.9 Ϯ 1.0
31.3 Ϯ 0.5
46.4 Ϯ 0.4
91.4 Ϯ 1.2
89.6 Ϯ 4.8
65.9 Ϯ 1.4
65.2 Ϯ 1.7
29.9 Ϯ 0.7*
45.9 Ϯ 1.3
91.2 Ϯ 1.8
75.2 Ϯ 2.9
60.1 Ϯ 1.9
26.5 Ϯ 2.0
43.4 Ϯ 1.4
85.1 Ϯ 2.4
82.4 Ϯ 4.2
56.8 Ϯ 0.8
80.6 Ϯ 5.9
57.3 Ϯ 2.2
25.4 Ϯ 1.4
44.7 Ϯ 1.0
88.6 Ϯ 0.9
85.5 Ϯ 2.7
54.2 Ϯ 0.8
23.5 Ϯ 0.3
43.4 Ϯ 0.4
87.7 Ϯ 0.6
82.5 Ϯ 4.3
56.5 Ϯ 1.7
24.1 Ϯ 0.9
44.1 Ϯ 0.7
87.1 Ϯ 0.6
82.7 Ϯ 1.7
55.6 Ϯ 0.8
23.6 Ϯ 0.6
44.4 Ϯ 0.5
88.0 Ϯ 0.8
85.1 Ϯ 1.6
54.1 Ϯ 1.3
23.1 Ϯ 0.5
43.7 Ϯ 0.5
86.4 Ϯ 0.7
86.0 Ϯ 2.8
54.3 Ϯ 0.7
23.2 Ϯ 0.3
43.8 Ϯ 0.4
86.5 Ϯ 0.6
†
†
†
a
Thr
24.2 Ϯ 0.5*
42.5 Ϯ 0.2
86.4 Ϯ 0.6
a
Ser
Glu
Pro
Gly
Ala
Val
Met
Ile
Leu
Tyr
Phe
His
Hyl
Lys
Arg
Total
91.6 Ϯ 4.5
108.5 Ϯ 10.0 107.9 Ϯ 4.7
253.0 Ϯ 18.0 283.5 Ϯ 3.6
103.9 Ϯ 6.1 111.0 Ϯ 5.6* 109.4 Ϯ 4.4 110.4 Ϯ 1.2 112.7 Ϯ 1.6 115.0 Ϯ 3.5
283.6 Ϯ 9.2 296.3 Ϯ 3.2
104.0 Ϯ 1.0 104.2 Ϯ 0.7
†
†
†
†
242.4 Ϯ 4.5 255.4 Ϯ 4.1
289.9 Ϯ 7.3 296.7 Ϯ 5.7 299.9 Ϯ 3.2 297.0 Ϯ 2.3
103.7 Ϯ 0.7 103.4 Ϯ 0.4 103.6 Ϯ 0.4 102.9 Ϯ 0.6
97.2 Ϯ 1.0
34.9 Ϯ 0.5
15.6 Ϯ 3.9
19.5 Ϯ 0.6
49.2 Ϯ 1.4
11.9 Ϯ 0.2
21.9 Ϯ 0.8
13.9 Ϯ 0.4
10.0 Ϯ 0.6
36.0 Ϯ 0.7
53.5 Ϯ 1.5
1000
98.0 Ϯ 0.9
33.1 Ϯ 0.9
15.6 Ϯ 3.7
99.2 Ϯ 1.9
34.5 Ϯ 3.7
11.1 Ϯ 4.6
18.6 Ϯ 3.4
43.6 Ϯ 6.2
10.2 Ϯ 1.8
22.0 Ϯ 3.8
12.3 Ϯ 2.9
9.7 Ϯ 1.2
33.0 Ϯ 5.7
51.2 Ϯ 3.9
1000
101.2 Ϯ 2.4
29.3 Ϯ 0.6
12.0 Ϯ 4.7
†
†
30.2 Ϯ 1.7
5.3 Ϯ 1.1
13.9 Ϯ 1.1
35.4 Ϯ 3.0
9.3 Ϯ 0.9
20.4 Ϯ 1.4
9.2 Ϯ 1.2
9.3 Ϯ 0.6
27.0 Ϯ 2.5
50.3 Ϯ 1.9
1000
27.5 Ϯ 0.3
5.2 Ϯ 0.7
12.5 Ϯ 0.2
31.3 Ϯ 0.7
8.1 Ϯ 0.2
18.3 Ϯ 0.3
7.3 Ϯ 0.4
9.8 Ϯ 0.5
23.4 Ϯ 0.9
49.3 Ϯ 1.5
1000
29.0 Ϯ 1.3
5.5 Ϯ 0.2
13.0 Ϯ 0.6
33.8 Ϯ 2.1
8.6 Ϯ 0.7
19.8 Ϯ 1.4
8.3 Ϯ 1.0
8.5 Ϯ 0.8
25.4 Ϯ 1.9
49.6 Ϯ 1.2
1000
27.7 Ϯ 0.6
3.6 Ϯ 1.0
12.6 Ϯ 0.3
32.4 Ϯ 1.1
8.6 Ϯ 0.3
19.2 Ϯ 0.8
7.7 Ϯ 0.3
8.6 Ϯ 0.3
24.1 Ϯ 0.9
49.6 Ϯ 1.7
1000
26.4 Ϯ 0.6
5.2 Ϯ 0.2
12.3 Ϯ 0.5
30.5 Ϯ 1.1
7.6 Ϯ 0.2
18.3 Ϯ 0.5
6.7 Ϯ 0.5
9.6 Ϯ 0.2
23.6 Ϯ 0.5
50.4 Ϯ 0.9
1000
26.8 Ϯ 0.4
5.4 Ϯ 0.1
12.5 Ϯ 0.3
30.8 Ϯ 0.7
7.7 Ϯ 0.4
18.0 Ϯ 0.8
6.6 Ϯ 0.5
9.4 Ϯ 0.5
23.8 Ϯ 0.8
49.5 Ϯ 1.0
1000
†
†
†
†
†
18.3 Ϯ 0.5*
45.7 Ϯ 1.6*
14.7 Ϯ 1.0*
†
34.5 Ϯ 0.6
†
10.5 Ϯ 0.5
8.5 Ϯ 0.2*
17.4 Ϯ 0.6*
8.9 Ϯ 0.2*
11.0 Ϯ 0.7*
25.9 Ϯ 1.0*
50.8 Ϯ 1.4
1000
20.6 Ϯ 0.8
12.6 Ϯ 0.7*
10.3 Ϯ 0.6
†
†
33.3 Ϯ 0.8
52.6 Ϯ 1.6
1000
The average values (n ϭ 8) and SDs are shown as relative amounts in 1000 total residues.
Amb., ambulatory; Susp., suspended.
a
Not corrected for hydrolysis.
†
Asterisk indicates statistic significance between ambulatory and suspended group, calculated with ANOVA (* p Ͻ 0.05; p Ͻ
0
.01).
A (Sigma, St. Louis, MO, USA), and 2.1 ␮M leupeptin Amino acid analysis
Sigma) in dimethylsulfoxide (Fisher Scientific), and ex-
(
Table 1 shows the amino acid compositions of five re-
gions of tibias obtained from ambulatory and suspended
animals expressed as residues per 1000 total amino acids.
Overall, the profiles of all samples showed a similar pattern
indicative of collagen-rich matrices; however, relative Hyp
amounts were smaller in proximal regions than in distal
regions, indicating that the amounts of noncollagenous pro-
tein in proximal regions were relatively greater than those in
distal regions. Significant differences in amino acid compo-
sitions between two groups were observed in proximal
epiphysis, proximal metaphysis, and diaphysis but not in
distal regions, suggesting that the response of tibias to
unloading is region specific. Relative Hyp amount was
significantly increased in proximal metaphysis (9.6%) of
suspended rats (Table 1). The extent of lysine (Lys) hy-
droxylation in the bone matrix (i.e., hydroxylysine [Hyl]/
Hyl ϩ Lys, was significantly increased in the proximal
epiphysis (9.0%), proximal metaphysis (30.3%), and diaph-
ysis (14.9%) in the suspended group (Table 2). However,
when calculated by Hyl/Hyp, although the proximal me-
taphysis showed a trend of increase, no significant differ-
ence was observed between the two groups. No significant
difference in amino acid compositions in distal epiphysis or
distal metaphysis of tibias between the two groups was
tracted for 3 days at 4°C with constant stirring. Then, the
supernatants were collected by centrifugation (16,000g) at
4
°C for 1 h and the extraction procedure was repeated two
more times. The supernatants were pooled, dialyzed with a
000- to 8000-Da molecular weight cut-off membrane
Spectrum Medical Industries, Inc., Los Angeles, CA, USA)
6
(
against cold distilled water exhaustively and lyophilized.
Mineral unbound immature collagen can be extracted with
this procedure, whereas the mature, mineralized collagen
cannot because it is protected by mineral. The dried sample
was weighed and an aliquot of each sample was hydrolyzed
as described previously and subjected to amino acid analy-
sis to determine the collagen content in the extract.
Statistical analysis
Statistical analysis was performed with one-factor
ANOVA.
RESULTS
Body weight
The body weights were measured just before death. The observed (Tables 1 and 2).
initial average (ϮSD) body weight of rats used in this study
was 420 g (Ϯ20 g). The mean (ϮSD) weight of ambulatory
rats was significantly greater (p Ͻ 0.05) than that of sus-
Collagen cross-linking analysis
pended rats (n ϭ 8 each) at 423 g (Ϯ23 g) and 385 g (Ϯ33
g), respectively.
Three reducible compounds (DHNL, DHLNL, and
HLNL) and two nonreducible cross-links (pyridinoline

1
642
SHIIBA ET AL.
TABLE 2. LYS/PRO HYDROXYLATION OF BONE MATRIX
Proximal metaphysis Diaphysis Distal metaphysis
Amb. Susp. Amb. Susp. Amb. Susp.
Proximal epiphysis
Amb. Susp.
Distal epiphysis
Amb.
Susp.
Hyl/Hyp
15.7 Ϯ 1.2 15.7 Ϯ 0.9 12.8 Ϯ 1.4 13.4 Ϯ 1.2 11.6 Ϯ 1.0 11.5 Ϯ 0.9 10.4 Ϯ 1.6 10.4 Ϯ 0.3 11.3 Ϯ 0.1 11.0 Ϯ 0.8
†
†
Hyl/(Hyl 21.7 Ϯ 0.9 23.7 Ϯ 1.0* 22.9 Ϯ 4.1 29.9 Ϯ 1.9 25.8 Ϯ 2.6 29.6 Ϯ 1.4 25.1 Ϯ 1.1 26.3 Ϯ 0.5 28.9 Ϯ 0.8 28.3 Ϯ 1.0
ϩ Lys)
Hyp/(Hyp 41.6 Ϯ 0.7 41.9 Ϯ 1.0 41.0 Ϯ 2.1 43.3 Ϯ 1.9 43.6 Ϯ 1.1 43.5 Ϯ 0.7 43.0 Ϯ 0.5 42.8 Ϯ 0.4 43.0 Ϯ 0.3 42.8 Ϯ 0.6
ϩ Pro)
The ratio of Hyl/Hyp and Hyp to the sum of Hyp and Pro, are calculated with the values in Table 1. The average values (n ϭ 8) and
SDs are shown as relative amounts in 1000 total residues.
Amb., ambulatory; Susp., suspended.
Asterisk indicates statistic significance between ambulatory and suspended group, calculated with ANOVA (* p Ͻ 0.05; p Ͻ
†
0
.01).
TABLE 3. COLLAGEN CROSS-LINKS IN THE REGIONS OF THE TIBIA
Proximal epiphysis Proximal metaphysis Diaphysis Distal metaphysis
Distal epiphysis
Amb. Susp.
Amb.
Susp.
Amb.
Susp.
Amb.
Susp.
Amb.
Susp.
Cross-link (moles/mole of collagen)
†
DHNL
DHLNL
HLNL
Pyr
ND
0.09 Ϯ 0.01 0.11 Ϯ 0.01 0.13 Ϯ 0.01 0.13 Ϯ 0.02 0.13 Ϯ 0.03 0.13 Ϯ 0.01 0.10 Ϯ 0.01 0.11 Ϯ 0.01
†
0.83 Ϯ 0.06 0.72 Ϯ 0.07* 1.03 Ϯ 0.07 1.26 Ϯ 0.05 1.14 Ϯ 0.03 1.23 Ϯ 0.09* 1.03 Ϯ 0.09 1.11 Ϯ 0.09 0.85 Ϯ 0.06 0.86 Ϯ 0.05
0.25 Ϯ 0.02 0.23 Ϯ 0.02 0.41 Ϯ 0.02 0.45 Ϯ 0.02 0.40 Ϯ 0.03 0.41 Ϯ 0.04 0.44 Ϯ 0.02 0.45 Ϯ 0.05 0.38 Ϯ 0.03 0.39 Ϯ 0.05
0.27 Ϯ 0.03 0.23 Ϯ 0.03 0.12 Ϯ 0.01 0.15 Ϯ 0.01 0.17 Ϯ 0.02 0.16 Ϯ 0.02 0.16 Ϯ 0.01 0.15 Ϯ 0.02 0.20 Ϯ 0.01 0.19 Ϯ 0.01
0.07 Ϯ 0.01 0.07 Ϯ 0.01 0.11 Ϯ 0.02 0.12 Ϯ 0.02 0.12 Ϯ 0.01 0.12 Ϯ 0.02 0.14 Ϯ 0.01 0.13 Ϯ 0.02 0.10 Ϯ 0.01 0.11 Ϯ 0.01
†
†
†
d-Pyr
Ratio
†
DHLNL/HLNL 3.39 Ϯ 0.17 3.13 Ϯ 0.17* 2.48 Ϯ 0.14 2.77 Ϯ 0.11 2.89 Ϯ 0.21 3.03 Ϯ 0.19 2.37 Ϯ 0.23 2.5 Ϯ 0.15 2.23 Ϯ 0.13 2.24 Ϯ 0.19
Pyr/d-Pyr
3.83 Ϯ 0.57 3.29 Ϯ 0.53 1.08 Ϯ 0.14 1.24 Ϯ 0.13* 1.44 Ϯ 0.14 1.37 Ϯ 0.22 1.14 Ϯ 0.05 1.12 Ϯ 0.08 1.93 Ϯ 0.22 1.73 Ϯ 0.20
The average values (n ϭ 8) and SDs are shown. The amounts of cross-links are expressed as moles/mole of collagen.
Amb., ambulatory; Susp., suspended.
Asterisk indicates statistic significance between ambulatory and suspended group, calculated with ANOVA (* p Ͻ 0.05; p Ͻ
†
0
.01).
[Pyr] and deoxy Pyr [d-Pyr]) were identified by collagen Analysis of immature/nonmineralized collagen
cross-link analysis. Overall profiles were similar to one
To examine whether the alterations in amino acid com-
position (Table 1) is caused by an increase in collagen or a
decrease in noncollagenous proteins, the collagen content
per dried weight of pulverized bone was determined. The
results indicated that the collagen content was significantly
increased by 10.2% in this region after 4 weeks of tail
suspension. In addition, although statistically not signifi-
cant, both relative Hyp amount and Hyp value per dried
weight tended to increase in the diaphysis and proximal
epiphysis.
To evaluate the level of immature/nonmineralized colla-
gen in these regions, bone powder was extracted with 6 M
of guanidine-HCl and the extract was analyzed. The weight
of the extracts (data not shown) and its collagen content in
the extracts in the proximal metaphysis obtained from the
suspended group was significantly greater (20.5%) than that
from ambulatory group and showed a tendency to increase
in the diaphysis (Table 4). Amino acid analysis of these
another showing a typical cross-linking pattern of rat bone
(
26)
collagen.
The values of collagen cross-links calculated
are summarized in Table 3. The data showed that the
cross-links were significantly altered in the proximal epiph-
ysis, proximal metaphysis, and diaphysis of tibias of sus-
pended rats when compared with those of ambulatory rats.
DHLNL was significantly increased in the proximal me-
taphysis and diaphysis by 22.6% and 8.1%, respectively,
and significantly decreased in the proximal epiphysis by
1
3.7% during tail suspension. HLNL was significantly in-
creased in the proximal metaphysis of suspended rats by
1
2
0.4%. In addition, Pyr was significantly increased by
5.3% in the proximal metaphysis but significantly de-
creased by 17.3% in the proximal epiphysis of suspended
rats. Both ratios of DHLNL/HLNL (DHLNL/HLNL), and
Pyr/d-Pyr (Pyr/d-Pyr) were significantly increased in the
proximal metaphysis of suspended rats (11.5% and 15.3%,
respectively). However, the ratio of DHLNL/HLNL in the extracts showed that the relative Hyp content per 1000
proximal epiphysis was significantly diminished (7.8%) and amino acid residues was comparable between two groups in
the ratio of Pyr/d-Pyr showed a tendency to decrease all regions examined (Table 5). The ratio of Hyl to the sum
(14.2%) in the proximal epiphysis of suspended rats.
of Hyl and Lys was significantly higher in the extract from

REGIONAL ALTERATION OF COLLAGEN IN UNLOADED BONE
1643
TABLE 4. COLLAGEN AMOUNT IN THE DRIED BONE POWDER AND GUANIDINE-HCL EXTRACTS (⌴G/MG OF BONE)
Collagen in bone powder
Guanidine-HCl–extracted collagen
Amb.
Susp.
Amb.
Susp.
Proximal epiphysis
Proximal metaphysis
Diaphysis
Distal metaphysis
Distal epiphysis
129.8 Ϯ 6.4
122.6 Ϯ 4.8
123.1 Ϯ 7.4
108.3 Ϯ 8.6
162.4 Ϯ 23.0
133.1 Ϯ 9.1
135.1 Ϯ 3.8
126.3 Ϯ 3.1
100.8 Ϯ 5.5
158.3 Ϯ 7.1
64.1 Ϯ 9.1
51.5 Ϯ 5.2
55.1 Ϯ 11.9
60.4 Ϯ 7.0
77.2 Ϯ 12.3
66.9 Ϯ 8.7
62.0 Ϯ 8.7*
62.6 Ϯ 7.3
65.2 Ϯ 8.4
71.2 Ϯ 6.7
†
The average values (n ϭ 8) and SDs are shown.
Amb., ambulatory; Susp., suspended.
Asterisk indicates statistic significance between ambulatory and suspended group, calculated with ANOVA (* p Ͻ 0.05; p Ͻ
†
0
.01).
TABLE 5. AMINO ACID COMPOSITIONS OF THE EXTRACTS
Proximal epiphysis
Amb. Susp.
Proximal metaphysis
Diaphysis
Amb. Susp.
Distal metaphysis
Amb. Susp.
Distal epiphysis
Amb. Susp.
Amb.
Susp.
Hyp
Asp
82.3 Ϯ 3.5 87.4 Ϯ 2.4 85.8 Ϯ 4.9
52.7 Ϯ 1.4 51.6 Ϯ 2.4 46.2 Ϯ 9.0
23.7 Ϯ 0.7 22.4 Ϯ 1.2 22.0 Ϯ 1.1
39.9 Ϯ 1.0 39.6 Ϯ 1.3 39.6 Ϯ 0.6
86.8 Ϯ 2.6 85.9 Ϯ 1.3 82.2 Ϯ 2.1
89.2 Ϯ 1.9
47.3 Ϯ 0.8
20.0 Ϯ 0.4
38.7 Ϯ 0.5* 39.2 Ϯ 0.8 38.4 Ϯ 0.3
79.8 Ϯ 1.1 80.2 Ϯ 1.1 79.5 Ϯ 0.4
85.3 Ϯ 4.0 89.3 Ϯ 1.9
49.5 Ϯ 2.2 47.4 Ϯ 0.5
21.0 Ϯ 1.3 19.7 Ϯ 0.2
90.4 Ϯ 3.4 93.8 Ϯ 2.7 96.3 Ϯ 4.6 94.7 Ϯ 5.2
45.6 Ϯ 1.2 45.7 Ϯ 0.3 45.7 Ϯ 1.7 45.8 Ϯ 1.4
19.2 Ϯ 0.3 18.9 Ϯ 0.2 19.0 Ϯ 0.3 19.1 Ϯ 0.6
38.0 Ϯ 0.4 37.8 Ϯ 0.2 37.7 Ϯ 0.7 38.1 Ϯ 1.2
79.2 Ϯ 0.8 78.8 Ϯ 0.6 78.3 Ϯ 0.8 79.7 Ϯ 2.2
a
†
Thr
a
Ser
Glu
Pro 108.0 Ϯ 3.9 108.1 Ϯ 2.4 105.4 Ϯ 2.4 108.4 Ϯ 2.5 103.9 Ϯ 3.4 105.9 Ϯ 1.9 122.9 Ϯ 2.7 125.4 Ϯ 1.8 126.8 Ϯ 8.3 123.9 Ϯ 9.5
Gly 307.8 Ϯ 5.8 313.5 Ϯ 8.6 321.4 Ϯ 10.1 336.2 Ϯ 7.3 327.4 Ϯ 7.6 336.0 Ϯ 4.7 324.0 Ϯ 2.9 320.8 Ϯ 1.6 320.7 Ϯ 5.8 324.3 Ϯ 7.4
Ala 104.0 Ϯ 1.0 104.0 Ϯ 1.0 108.9 Ϯ 1.7 109.2 Ϯ 0.6 109.2 Ϯ 1.0 109.8 Ϯ 0.6 108.5 Ϯ 1.1 107.6 Ϯ 1.2 106.6 Ϯ 0.9 107.8 Ϯ 1.6
Val
Met
Ile
27.3 Ϯ 1.3 26.4 Ϯ 1.3 24.7 Ϯ 1.6
5.5 Ϯ 0.6 5.9 Ϯ 0.7 5.6 Ϯ 0.3
13.0 Ϯ 0.8 12.3 Ϯ 1.1 12.1 Ϯ 1.0
31.8 Ϯ 1.9 30.5 Ϯ 2.7 29.9 Ϯ 2.2
22.5 Ϯ 0.5* 23.7 Ϯ 1.4 22.3 Ϯ 0.4
22.9 Ϯ 0.2 22.9 Ϯ 0.3 23.2 Ϯ 0.3 22.9 Ϯ 0.1
4.7 Ϯ 0.3 4.5 Ϯ 0.8 4.4 Ϯ 0.9 4.2 Ϯ 0.3
11.2 Ϯ 0.7 10.2 Ϯ 0.6* 10.3 Ϯ 0.0 10.1 Ϯ 0.3 10.6 Ϯ 0.2 10.3 Ϯ 0.2
25.2 Ϯ 0.3 25.0 Ϯ 0.3 25.6 Ϯ 0.2 25.3 Ϯ 0.3
4.3 Ϯ 0.3 4.1 Ϯ 0.2 4.2 Ϯ 0.5 4.2 Ϯ 0.2
13.0 Ϯ 0.2 13.3 Ϯ 0.4 13.5 Ϯ 0.1 13.5 Ϯ 0.1
3.8 Ϯ 0.2 3.8 Ϯ 0.2 3.7 Ϯ 0.3 3.7 Ϯ 0.3
11.0 Ϯ 0.5 10.8 Ϯ 0.2 12.0 Ϯ 0.5 11.6 Ϯ 0.3
5.3 Ϯ 0.3
10.1 Ϯ 0.7
5.6 Ϯ 0.6
5.4 Ϯ 0.2
†
Leu
Tyr
Phe
His
Hyl
Lys
Arg
Total
26.1 Ϯ 1.2* 28.5 Ϯ 2.4 26.3 Ϯ 0.8
5.7 Ϯ 0.7
16.2 Ϯ 0.9 15.4 Ϯ 0.7 15.7 Ϯ 1.5
5.7 Ϯ 0.7 5.1 Ϯ 0.9 5.2 Ϯ 0.7
5.2 Ϯ 0.8 5.6 Ϯ 0.8
3.2 Ϯ 1.9*
14.2 Ϯ 0.6
4.2 Ϯ 0.3*
12.7 Ϯ 0.4
5.2 Ϯ 0.7
15.9 Ϯ 1.1 14.7 Ϯ 0.4
5.2 Ϯ 1.0 4.1 Ϯ 0.2
11.9 Ϯ 0.5 12.4 Ϯ 0.3
4.5 Ϯ 0.3
12.5 Ϯ 0.4 12.7 Ϯ 1.0 12.0 Ϯ 1.0
26.4 Ϯ 1.9 23.9 Ϯ 0.6 24.5 Ϯ 2.0
50.4 Ϯ 1.6 49.9 Ϯ 2.1 51.2 Ϯ 2.1
21.9 Ϯ 0.7* 24.7 Ϯ 2.4 22.2 Ϯ 0.3* 21.2 Ϯ 0.3 21.5 Ϯ 0.5 20.7 Ϯ 0.6 21.4 Ϯ 1.1
50.9 Ϯ 0.9
1000
52.4 Ϯ 1.0 52.0 Ϯ 0.8
1000 1000
55.7 Ϯ 1.6 55.2 Ϯ 1.6 51.0 Ϯ 2.5 49.7 Ϯ 2.3
1000 1000 1000 1000
1000
1000
1000
The average values (n ϭ 8) and SDs are shown as relative amounts in 1000 total residues.
Amb., ambulatory; Susp., suspended.
a
Not corrected for hydrolysis.
†
Asterisk indicates statistic significance between ambulatory and suspended group, calculated with ANOVA (* p Ͻ 0.05; p Ͻ 0.01).
the proximal metaphysis (11.9%) and diaphysis (10.0%) of but not in the distal regions of the tibia. Of these, a trabec-
suspended rats (Table 6), although the ratios of Hyl/Hyp in ular bone-rich proximal metaphysis appeared to be the
these regions were comparable between the two groups. region most susceptible to unloading. This is consistent with
(28)
This indicates that the pool of immature/nonmineralized studies from the flight rats and fractal dimension analysis
(
29)
collagen was significantly increased in the proximal me- data from the tail-suspended rats.
The loss of trabecular
bone of this region induced by skeletal unloading apparently
taphysis during tail suspension, and that Lys hydroxylation
of collagen, except for cross-linking sites, in this pool may
be comparable.
(30)
is more pronounced
ing
and remained longer after reload-
than that of cortical bone. Likely, the structural
(31)
characteristics of trabecular bone, that is, its rich vascularity
and larger surface areas in contact with bone marrow cells,
would make the proximal metaphysis respond to skeletal
DISCUSSION
This study shows that alterations of type I collagen oc- unloading more readily.
curred in the tibia by skeletal unloading. Interestingly, the
Extracellular matrices are essential for providing the tis-
alteration was bone region specific and it was observed in sue with structure and integrity, and type I collagen, the
the proximal epiphysis, proximal metaphysis, and diaphysis predominant protein of bone extracellular matrix, plays a

1
644
SHIIBA ET AL.
TABLE 6. LYS/PRO HYDROXYLATION OF THE EXTRACTS
Proximal metaphysis Diaphysis Distal metaphysis
Amb. Susp. Amb. Susp. Amb. Susp.
Proximal epiphysis
Amb. Susp.
Distal epiphysis
Amb.
Susp.
Hyl/Hyp
15.3 Ϯ 1.1 14.6 Ϯ 1.1 14.0 Ϯ 1.0 14.3 Ϯ 0.7 13.9 Ϯ 0.3 13.9 Ϯ 0.4 12.2 Ϯ 0.9 11.5 Ϯ 0.5 12.6 Ϯ 1.2 12.3 Ϯ 0.8
Hyl/(Hyl 32.2 Ϯ 1.5 34.7 Ϯ 1.8 32.8 Ϯ 2.8 36.7 Ϯ 0.4* 32.6 Ϯ 2.9 35.9 Ϯ 0.8* 34.0 Ϯ 0.8 33.4 Ϯ 0.6 36.5 Ϯ 1.4 35.1 Ϯ 1.5
ϩ Lys)
Hyp/(Hyp 43.2 Ϯ 0.5 44.7 Ϯ 0.3* 44.8 Ϯ 1.1 45.1 Ϯ 0.3 45.1 Ϯ 0.6 45.7 Ϯ 0.4 42.4 Ϯ 0.9 42.8 Ϯ 0.4 43.2 Ϯ 0.8 43.4 Ϯ 0.8
ϩ Pro)
The ratio of Hyl/Hyp and Hyp to the sum of Hyp and Pro are calculated with the values in Table 5. The average values (n ϭ 8) and
SDs are shown as relative amounts in 1000 total residues.
Amb., ambulatory; Susp., suspended.
Asterisk indicates statistic significance between ambulatory and suspended group, calculated with ANOVA (* p Ͻ 0.05).
(26)
crucial role in defining and modulating the 3-dimensional this study.
This may be due to the changes in the popu-
space for mineralization. Its specific molecular packing lation of post-translationally altered collagen during the
structure and post-translational chemistries appear to be prolonged tail suspension.
(15,27)
important for this function.
Suarez and coworkers ex-
In addition to the alteration of post-translational modifi-
amined the type I collagen chemistry in trabecular and cations of collagen, quantitative change in bone collagen
cortical bones obtained from tibias and femurs of rats and also appeared to occur after 4 weeks of unloading. Both Hyp
showed that the collagen molecule of trabecular bone con- per 1000 total amino acid residues and the collagen content
tained a higher number of glycosylated Hyl but a lower per dried bone (wt/wt) were significantly increased in the
(32)
number of Pyr cross-links than those of cortical bone,
proximal metaphysis in the suspended group. Furthermore,
consistent with our results (Tables 2 and 3). They suggested the collagen contents in the guanidine-HCl extracts of bone
that the lower number of Pyr cross-links in trabecular bone powder obtained from the suspended group were signifi-
collagen make it less stable than in cortical bone.
cantly greater in the proximal metaphysis (20.5%) than that
The extent of Lys hydroxylation calculated by Hyl/Hyl ϩ of the ambulatory group (Table 4). The amino acid analysis
Lys in bone matrix (Table 2) and guanidine-HCl extract revealed that the extracts obtained from all five regions were
(Table 6) significantly increased in the proximal metaphy- composed mainly of collagen. Because only immature non-
sis, although it showed no significant difference when cal- mineralized collagen can be extracted with guanidine-HCl
(24)
culated by Hyl/Hyp. This suggests that the extent of Lys from the undemineralized samples,
the data suggest that
hydroxylation of collagen, except for cross-linking sites, skeletal unloading induced an increased pool of such colla-
may not be altered by skeletal unloading and that the in- gen in the proximal metaphysis. A reduced mRNA expres-
crease of the ratio, that is, Hyl/Hyl ϩ Lys, could be caused sion of type I collagen has been observed in spaceflight,
by the relative increase of collagen. Slight but significant using an in vitro cell culture system with a human osteo-
(33)
increases in Gly in the proximal epiphysis, proximal me- sarcoma cell line ; in osteoblasts derived from embryonic
(34)
taphysis, and diaphysis support this speculation. However, chicken calvaria ; and in femur metaphysis of tail-
(35)
this needs to be confirmed by analyzing Lys hydroxylation suspended rats.
Although this study does not seem to
on the purified collagen molecule/chains because unloading agree with these reports with regard to collagen synthesis,
also may alter the extent of Pro hydroxylation of collagen. these findings clearly indicate that skeletal unloading is one
Significant alterations in collagen cross-linking were ob- of the key factors that regulate collagen metabolism, and the
served in the proximal regions (Table 4). The ratio of discrepancy could be caused by the differences of subject
DHLNL/HLNL was significantly increased in the proximal (i.e., mRNA vs. protein), age of the animals, sampling sites
metaphysis and decreased in the proximal epiphysis, and the and timing, etc. Further study is warranted to compare both
ratio of Pyr/d-Pyr was significantly increased in the proxi- mRNA expression and protein levels of type I collagen.
mal metaphysis. Thus, the level of hydroxylation of specific
In conclusion, this study showed that skeletal unloading
Lys residues involved in cross-linking appears to be in- induced accumulation of post-translationally altered and
creased in the proximal metaphysis but decreased in the nonmineralized collagen in the tibia and that these changes
proximal epiphysis. Although the significance of these al- were pronounced in the proximal regions of the bone. To
terations is not clear at this point, they could affect the define the alterations in the collagen molecules and their
stability of the fibril because of the altered cross-linking consequences on bone mineralization, further characteriza-
and, possibly, the fibril diameter through varied levels of tion of these changes together with functional studies at
glycosylation (i.e., the helical Hyl residues involved in cellular and molecular levels are warranted.
(18)
cross-linking are also glycosylation sites).
Our recent
study revealed that DHLNL and HLNL were significantly
increased during 8 weeks of tail suspension in the diaphyses
of tibia and femur and the ratios (i.e., DHLNL/HLNL or
ACKNOWLEDGMENTS
This study was supported by NASA grant NAG 2–1188
Pyr/d-Pyr) were decreased while they show no difference in and NIH grant DE10489.

REGIONAL ALTERATION OF COLLAGEN IN UNLOADED BONE
1645
REFERENCES
20. Fujii K, Tanzer ML 1977 Osteogenesis imperfecta: Biochem-
ical studies of bone collagen. Clin Orthop 82:271–277.
. Morey-Holton ER, Whalen RT, Arnaud SB, Van der Meulun 21. Wojtowicz A, Dziedzic-Goclawska A, Kaminski A, Stachow-
1
MC 1996 The skeleton and its adaptation to gravity. In: Fregly
M, Blatteis C (eds.) Handbook of Physiology, section 4, vol. I.
Oxford University Press, New York, NY, USA, pp. 691–791.
. Fiore CE, Pennisi P, Ciffo F, Scebba C, Amico A, Di Fazzio
icz W, Wojtowicz K, Marks SC, Yamauchi M 1997 Alteration
of mineral crystallinity and collagen cross-linking of bones in
osteopetrotic toothless (tl/tl) rats and their improvement after
treatment with colony stimulating factor-1. Bone 20:127–132.
2
S 1999 Immobilization-dependent bone collagen breakdown 22. Bank RA, Robins SP, Wijmenga C, Breslau-Siderius LJ, Bar-
appears to increase with time: Evidence for a lack of new bone
equilibrium in response to reduced load during prolonged bed
rest. Horm Metab Res 31:31–36.
. Elias AN, Gwinup G 1992 Immobilization osteoporosis in
paraplegia. J Am Paraplegia Soc 15:163–170.
. Marchetti ME, Houde JP, Steinberg GG, Crane GK, Goss TP,
Baran DT 1996 Humeral bone density losses after shoulder
surgery and immobilization. J Shoulder Elbow Surg 5:471–
doel AF, van der Sluijs HA, Pruijs HE, TeKoppele JM 1999
Defective collagen crosslinking in bone, but not in ligament or
cartilage, in Bruck syndrome: Indications for a bone-specific
telopeptide lysyl hydroxylase on chromosome 17. Proc Natl
Acad Sci USA 96:1054–1058.
3
4
2
2
3. Bailey AJ, Wotton SF, Sims TJ, Thompson PW 1992 Post-
translational modifications in the collagen of human osteopo-
rotic femoral head. Biochem Biophys Res Commun 185:801–
8
05.
4
76.
4. Wassen MH, Lammens J, Tekoppele JM, Sakkers RJ, Liu Z,
Verbout AJ, Bank RA 2000 Collagen structure regulates fibril
mineralization in osteogenesis as revealed by cross-link pat-
terns in calcifying callus. J Bone Miner Res 15:1776–1785.
5. Wu JJ, Eyre DR 1988 Fine powdering exposes the mineral-
protected collagen of bone to protease digestion. Calcif Tissue
Int 42:243–247.
5
6
. Morey ER, Baylink DJ 1978 Inhibition of bone formation
during space flight. Science 201:1138–1141.
. Wronski TJ, Morey-Holton ER 1987 Skeletal response to
simulated weightlessness: A comparison of suspension tech-
niques. Aviat Space Environ Med 58:63–68.
. Globus RK, Bikle DD, Morey-Holton E 1986 The temporal
response of bone to unloading. Endocrinology 118:733–742.
. Martin RB 1990 Effects of simulated weightlessness on bone
properties in rats. J Biomech 23:1021–1029.
. Kostenuik PJ, Halloran BP, Morey-Holton ER, Bikle DD 1997
Skeletal unloading inhibits the in vitro proliferation and dif-
ferentiation of rat osteoprogenitor cells. Am J Physiol 273:
E1133–E1139.
0. Zhang R, Supowit SC, Klein GL, Lu Z, Christensen MD,
Lozano R, Simmons DJ 1995 Rat tail suspension reduces
messenger RNA level for growth factors and osteopontin and
decreases the osteoblastic differentiation of bone marrow stro-
mal cells. J Bone Miner Res 10:415–423.
2
2
2
2
2
7
8
9
6. Shiiba M, Arnaud SB, Tanzawa H, Uzawa K, Yamauchi M
2
001 Alterations of collagen matrix in weight-bearing bones
during skeletal unloading. Connect Tissue Res 42:303–311.
7. Yamauchi M, Katz EP 1993 The post-translational chemistry
and molecular packing of mineralizing tendon collagen. Con-
nect Tissue Res 29:81–98.
8. Jee WS, Wronski TJ, Morey ER, Kimmel DB 1983 Effects of
spaceflight on trabecular bone in rats. Am J Physiol 244:
R310–R314.
9. Pornprasertsuk S, Ludlow JB, Webber RL, Tyndall DA, San-
hueza AI, Yamauchi M 2001 Fractal dimension analysis of
weight-bearing bones of rats during skeletal unloading. Bone
1
1
1. Bikle DD, Harris J, Halloran BP, Morey-Holton E 1994 Al-
tered skeletal pattern of gene expression in response to space-
flight and hindlimb elevation. Am J Physiol 267:E822–E827.
2. Zhang R, Supowit SC, Hou X, Simmons DJ 1999 Transform-
ing growth factor-beta2 mRNA level in unloaded bone ana-
lyzed by quantitative in situ hybridization. Calcif Tissue Int
2
9:180–184.
3
0. Laib A, Barou O, Vico L, Lafage-Proust MH, Alexandre C,
Rugsegger P 2000 3D micro-computed tomography of trabec-
ular and cortical bone architecture with application to a rat
model of immobilisation osteoporosis Med Biol Eng Comput
1
3
8:326–332.
6
4:522–526.
3
1. Collet P, Uebelhart D, Vico L, Moro L, Hartmann D, Roth M,
Alexandre C 1997 Effects of 1- and 6-month spaceflight on
bone mass and biochemistry in two humans. Bone 20:547–
551.
1
1
3. Simmons DJ, Grynpas MD, Rosenberg GD 1990 Maturation
of bone and dentin matrices in rats flown on the Soviet bio-
satellite Cosmos 1887. FASEB J 4:29–33.
4. Glimcher MJ 1976 Composition, structure, and organization of 32. Suarez KN, Romanello M, Bettica P, Moro L 1996 Collagen
bone and other mineralized tissues and the mechanism of
calcification. In: Greep RO, Astwood EB (eds.) Handbook of
type I of rat cortical and trabecular bone differs in the extent of
posttranslational modifications. Calcif Tissue Int 58:65–69.
Physiology, vol. 7 Endocrinology. American Physiological 33. Carmeliet G, Nys G, Bouillon R 1997 Microgravity reduces
Society, Washington DC, USA, pp. 25–116.
the differentiation of human osteoblastic MG-63 cells. J Bone
1
1
5. Yamauchi M 1995 Collagen: The major matrix molecule in
Miner Res 12:786–794.
mineralized tissues. In: Anderson JJB, Garner SC (eds.) Cal- 34. Landis WJ, Hodgens KJ, Block D, Toma CD, Gerstenfeld LC
cium and Phosphorous in Health and Disease. CRC Press,
Boca Raton, FL, USA, pp. 127–145.
2000 Spaceflight effects on cultured embryonic chick bone
cells. J Bone Miner Res 15:1099–1112.
6. Notbohm H, Nokelainen M, Myllyharju J, Fietzek PP, Muller 35. Drissi H, Lomri A, Lasmoles F, Holy X, Zerath E, Marie PJ
PK, Kivirikko KI 1999 Recombinant human type II collagens
with low and high levels of hydroxylysine and its glycosylated
forms show marked differences in fibrillogenesis in vitro.
J Biol Chem 274:8988–8992.
1999 Skeletal unloading induces biphasic changes in insulin-
like growth factor-I mRNA levels and osteoblast activity. Exp
Cell Res 251:275–284.
1
1
7. Eyre DR, Glimcher MJ 1973 Collagen cross-linking. Isolation
of cross-linked peptides from collagen of chicken bone. Bio-
chem J 135:393–403.
8. Yamauchi M, Katz EP, Mechanic GL 1986 Intermolecular
cross-linking and stereospecific molecular packing in type I
collagen fibrils of the periodontal ligament. Biochemistry 25:
Address reprint requests to:
Mitsuo Yamauchi, D.D.S., Ph.D.
Dental Research Center
University of North Carolina at Chapel Hill CB 7455
Chapel Hill, NC 27599–74553, USA
4
907–4913.
9. Eyre DR, Paz MA, Gallop PM 1984 Cross-linking in collagen Received in original form October 4, 2001; in revised form Feb-
and elastin. Annu Rev Biochem 53:717–748. ruary 12, 2002; accepted March 12, 2002.
1