Materials Transactions, Vol. 54, No. 8 (2013) pp. 1257 to 1261
Special Issue on New Functions and Properties of Engineering Materials Created by Designing and Processing II
EXPRESS
© 2013 The Japan Institute of Metals and Materials
REGULAR ARTICLE
Preferential Orientation of Collagen/Biological Apatite in Growing Rat Ulna
under an Artificial Loading Condition
Jun Wang+1, Takuya Ishimoto and Takayoshi Nakano+2
Division of Materials and Manufacturing Science, Graduate School of Engineering, Osaka University, Suita 565-0871, Japan
Mechanical loading plays a key role in altering macroscopic and microscopic bone structure through functional adaptation; however, the
anisotropic micro-organization of collagen and biological apatite (BAp) in adaptively created bone tissue is not well understood in spite of its
importance in the mechanical function of bone. In this study, we artificially applied axial compressive loading (15 N at 2 Hz) to a growing rat
ulna 10 min/day for 3 weeks to induce new bone under increased mechanical stimulus. Artificial loading induced marked increases in the
structural properties of the loaded ulna; the cortical bone area of the mid-shaft was 43.3% larger in the loaded ulna than the contralateral control
ulna. The newly formed bone was located mainly on the medial periosteal surface of the ulnar mid-shaft, which experienced the highest
compressive strain. The present study firstly clarified that new bone induced by an artificial load showed preferential orientation of collagen and
BAp c-axis along the ulnar long axis, which is similar to the pre-existing bone, although the degree of orientation and bone mineral density
(BMD) were still impaired after loading for 3 weeks. This anisotropic organization of collagen and BAp crystals corresponded to that of
osteocytes, implying that osteocytes are involved in the formation of anisotropic bone micro-organization which is important aspect of bone
mechanical function regarding material properties. [doi:10.2320/matertrans.ME201314]
(Received January 23, 2013; Accepted February 27, 2013; Published April 12, 2013)
Keywords: axial compressive loading, functional adaptation, biological apatite (BAp), preferential orientation of collagen and BAp, osteocytes
1.
Introduction
It is commonly accepted that bone adapts its mass and
architecture according to the prevailing mechanical loading
environment. The mechanisms involved in bone adaptation
have been the object of considerable research over the past
decades. To clarify the responses of bone to the mechanical
environment, many loading and unloading models have been
developed to modulate in vivo mechanical loads on bone.1­4)
Most of these studies focused on the change in mass and/or
geometry of bone along with the change in loading condition.
Moderately increased mechanical stimuli induce bone
formation,1,5,6) while unloaded bones experience bone
absorption,7­10) to adjust the bone mechanical functions to
the mechanical requirements. To our knowledge, however,
the bone microstructure, including the crystallographic
orientation of biological apatite (BAp) and collagen, induced
by artificial loading is poorly defined in spite of its impact on
mechanical function. The degree of the preferential BAp caxis orientation along the longitudinal axis of the long bone,
rather than bone mineral density (BMD), has been demonstrated to be a dominant determinant of Young’s modulus in
the same direction.11)
The preferential orientation of BAp might be closely
related to the applied load (stress) on bones. Nakano et al.12)
have reported that BAp orientation varies depending on bone
type, bone shape, and anatomical location in the body.
In vivo stress conditions, especially principal stress,13) appear
to affect the preferential BAp orientation. Demonstration of
the formation of anisotropic BAp organization is noteworthy
in bone biomechanics because BAp orientation effectively
modifies bone mechanical functions. This may also contribute to clarification of the bone adaptive response through an
+1
Graduate Student, Osaka University
Corresponding author, E-mail: [email protected]
+2
alteration of bone microstructure, which has not been well
documented. Moreover, osteocytes which inhabit in bone
matrix are of interest because they alter their morphology and
alignment in response to the stress applied on bone14) and are
believed to be involved in the regulation of bone adaptive
responses.15)
The aim of the present study was to clarify whether the
preferential orientation of the BAp c-axis and related collagen
fibers,16) which is similar to that of the pre-existing bone
grown under physiological mechanical conditions, forms in
the bone induced under artificial loading by utilizing a rat
forelimb compressive loading model.
2.
Materials and Methods
2.1 Animal
A skeletally immature male Sprague­Dawley rat (age, 7
weeks; body weight, 200 g) was purchased from Japan SLC,
Inc. (Shizuoka, Japan), which shows rapid growth17) and
unclosed ulnar epiphyseal line.18) The animal was housed in a
cage with a 12 h/12 h light/dark cycle and provided with a
standard diet and water ad libitum and acclimated for 1 week
before experimentation.
2.2 In vivo loading of the ulna
In vivo ulna loading was carried out according to the
method established by Torrance et al.19) The anesthetized rat
was placed on a mat, and the right forelimb was held in
rounded cups between the olecranon process of ulna and the
flexed carpus (Fig. 1). The right forelimb was cyclically
loaded in a compression mode with a peak force of 15 N at
2 Hz for 1200 cycles/day, 5 days/week, for 3 weeks using
a micro-material testing machine (MMT-101N; Shimadzu,
Kyoto, Japan). The load frequency of 2 Hz was used because
it corresponds to the stride frequency during natural
locomotion.20) The left forelimb received no treatment as a
1258
J. Wang, T. Ishimoto and T. Nakano
Fig. 1
Photograph of the rat ulna in vivo loading model.
control. After loading, the rat was demounted from the
loading device and returned to normal activity. The rat was
sacrificed on day 21; the left and right forelimbs were
harvested and kept in 10% formalin neutral buffered solution.
All animal procedures and protocols were approved by
the Animal Experiment Committee of Osaka University
Graduate School of Engineering (approval number: 22-1-0).
2.3 Assessment of bone mass
The bone mass was measured by micro-focused X-ray
computed tomography (µCT) (SMX-100CT; Shimadzu,
Kyoto, Japan). Images were taken at the ulnar mid-shaft
under 53 kV and 60 µA radiation with a spatial resolution of
12.45 µm. Tri/3D-BON software (Ratoc System Engineering, Tokyo, Japan) was used to quantitatively analyze the
bone structural properties of cortical bone area (mm2),
periosteal circumference (mm), and second moment of inertia
(I) (mm4) on the binalized images (IM-L and IC-C; subscripts
M-L and C-C represent neutral axis for bending specified for
the calculation). A 300-µm-thick region at the mid-shaft was
used for calculations.
2.4
Assessment of volumetric bone mineral density
(BMD)
The volumetric BMD (mg/cm3) was measured at the ulnar
mid-shaft by peripheral quantitative computed tomography
(pQCT) (XCT Research SA+; Stratec Medizintechnik
GmbH, Birkenfeld, Germany). The ulna was placed in a
plastic tube filled with 10% formalin. After the performance
of a scout view to enable scan localization, a cross-sectional
scan was performed at the ulnar mid-shaft using a resolution
of 70 © 70 © 460 µm.
Assessment of preferential orientation of BAp c-axis
and collagen
A 300-µm ulnar longitudinal section including the newly
formed portion was prepared by sectioning in the mediolateral planes with a diamond circular saw (BS-300CP; Exakt
Apparatebau, Norderstedt, Germany). The degree of prefer-
2.5
ential BAp c-axis orientation was quantitatively analyzed
using a microbeam X-ray diffractometer (µXRD) system (RAxis BQ; Rigaku, Tokyo, Japan) with a transmission optical
system as previously described.21,22) Mo-K¡ radiation was
generated at a tube voltage of 50 kV and a tube current of
90 mA. The incident beam was focused on a beam spot
100 µm in diameter by a double-pinhole metal collimator
and radiated vertically to the long bone axis on the anterior
surface with an exposure time of 3600 s. Two peaks of
specific crystallographic planes of BAp, (002) and (310),
were used for analyzing the degree of the preferential BAp caxis orientation as a relative ratio of (002) diffracted intensity
to (310) intensity.12,23) The orientation degree of randomly
oriented hydroxyapatite (NIST #2910: calcium hydroxyapatite) powder showed an intensity ratio of 0.8; therefore,
a value over 0.8 indicates preferential BAp c-axis orientation
in a specific direction. Measurements were carried out at 5
points, with 100-µm intervals, from the periosteal surface at
the mid-shaft of the ulna.
After the thickness was reduced to 70 µm, the ulnar
longitudinal section was observed using a polarized light
microscope (BZ100; Olympus, Tokyo, Japan) with polarizer
and analyzer in cross position.
2.6 Observation of osteocyte morphology and alignment
The morphology of osteocytes was observed with a
confocal laser scanning microscope (CLSM) (FV1000DIX81; Olympus, Tokyo, Japan) in both newly formed bone
and pre-existing cortical bone. The above-mentioned longitudinal section of 70-µm thickness was stained with
fluorescein isothiocyanate isomer I (FITC) (F7250; SigmaAldrich, St. Louis, Missouri, USA) according to a previously
described method.24­26) A micrograph was captured using an
UPlanSApo 60© oil immersion objective lens (numerical
aperture = 1.35).
3.
3.1
Results and Discussion
Loading-related adjustment in bone structural
properties
Axial compressive loading (15 N) of the forelimb for 3
weeks apparently increased bone mass compared with the
contralateral control ulna (Fig. 2, Table 1). The cortical bone
area and the periosteal circumference of the mid-shaft were
43.3 and 13.8% greater, respectively, in the loaded ulna than
control ulna.
Notably, the marked increase in bone formation induced by
mechanical loading was mainly observed on the medial
surface of the ulnar mid-shaft, principally on the periosteal
surface, while there was very little osteogenic response on the
lateral periosteal or endocortical surfaces (Fig. 2). Similarly,
artificial loading did not induce evident new bone formation
on cranial or caudal surfaces of the loaded ulna relative to the
control ulna (Fig. 2). Due to the natural curvature of the ulna,
an axial compressive force induced bending in the mediolateral direction, resulting in compression on the medial
surface and tension on the lateral surface,20,27) with the ratio
of compressive to tensile strain being around 1.5,28) whereas
there was little strain at the caudal and cranial surfaces
because the craniocaudal axis was the neutral axis for
Orientation of Biological Apatite in Bone Induced by Artificial Loading
(b)
(a)
(c)
Long axis
(a)
1259
(d)
500µm
(b)
(c)
310
211
002
Medial
2mm
Caudal
Cranial
Lateral
Fig. 2 µCT images of (a), (b) whole bone and (c), (d) mid-diaphysis crosssections in (a), (c) control and (b), (d) loaded ulnae. Arrows indicate ulnar
mid-shaft where the cross-sectional images were taken.
Table 1 Effects of artificial loading on ulnar structural properties.
Cortical
bone area
(mm2)
Periosteal
circumference
(mm)
IM-L
(mm4)
IC-C
(mm4)
Control
1.50
5.33
0.45
0.10
Loaded
2.15
6.07
0.62
0.29
Fig. 3 (a) Polarized light microscopy image of the longitudinal section of
the loaded ulna. Dotted line indicates border between new bone and preexisting bone. Bright contrast represents preferential collagen orientation
along the long axis of the ulna. Typical µXRD patterns of the (b) newly
formed and (c) pre-existing bones.
3.2.2
bending. Strain increases linearly with distance from the
neutral axis; therefore, the highest strain was experienced on
the periosteal surface. Because mechanically induced bone
formation is distributed in accordance with strain magnitude,5) new bone formation was highest on the medial
periosteal surface, which experienced the highest compressive strain.20) These results are consistent with those of other
studies showing that new bone is predominantly formed in
response to compressive strains than tensile strains.29,30)
As shown in Table 1, second moment of inertia was
markedly increased both across the medial­lateral (IM­L) axis
and caudal­cranial (IC­C) axis in the loaded right ulna; IC­C in
particular showed nearly a 3-fold increase. The new bone
formed on the medial periosteal surface largely contributed
to this drastic increase in the IC­C, indicating the great
enhancement of resistance to bending induced by artificial
loading. Bone adapted to the artificially increased load by
effectively altering its macroscopic structure.
3.2 Loading-related change in bone material properties
3.2.1 Loading-related change in BMD
The BMD of the newly formed bone (1002 mg/cm3) was
somewhat lower than that of the pre-existing cortical bone
(1246 mg/cm3), although the BMD of the newly formed
bone was over 690 mg/cm3, which is usually regarded as the
threshold of cortical bone. There were pores in the newly
formed bone [Fig. 2(c)], suggesting that the bone induced by
artificial loading was not fully consolidated and not a fully
mature bone after loading for 3 weeks. Loading for more than
approximately 2 months is required for the BMD of new
bone to reach a normal level.29)
Loading-related change in the preferential orientation of collagen and BAp c-axis
Another important aspect of bone material properties is the
preferential collagen and BAp c-axis orientation, which is
a more important determinant of Young’s modulus than
BMD.11) Figure 3 shows a polarized microscopy image and
representative microbeam X-ray diffraction patterns from the
newly formed and pre-existing bones at the ulnar mid-shaft.
Polarized microscopy revealed anisotropic collagen orientation along the bone axis in the new bone similar to that in
the pre-existing bone, although the homogeneity of the
orientation was slightly lower than that in the pre-existing
bone. The two XRD patterns were similar, showing 002 arc
and 310 arc along the vertical and horizontal directions,
respectively. The vertical direction in the diffraction pattern
corresponds to the longitudinal axis of the ulna; therefore,
the preferential orientation of the BAp c-axis along the ulnar
long axis was found both in the newly formed bone and the
pre-existing bone.
Figure 4 shows the degree of the preferential orientation of
the BAp c-axis [intensity ratio of (002)/(310)] along the long
bone axis in the ulnar mid-shaft. The newly formed bone
(points 1 and 2) showed high values for the degree of the
preferential BAp orientation, which were much higher than
the value of randomly oriented hydroxyapatite (NIST #2910:
calcium hydroxyapatite) powder, 0.8. However, the degree
of orientation in newly formed bone did not reach the levels
found in pre-existing cortical bone (points 4 and 5). Taken
together with the BMD result, the bone induced by artificial
loading had not reached an equilibrium state at 3 weeks under
the artificially increased mechanical stimulus; the new bone
was still undergoing maturation at this time. To determine the
equilibrium state with increased load, the loading period
should be prolonged. During maturation, the degree of BAp
1260
J. Wang, T. Ishimoto and T. Nakano
(a)
Long axis
L
1 2 3 4 5
Medial
Intensity ratio of (002)/(310)
14
12
Lateral
C
200µm
(b)
10
8
50µm
6
4
New bone
2
0
1
2
Pre-existing bone
3
Point
4
5
Fig. 4 (a) Representation of measured points for µXRD. (b) Distribution of
the degree of BAp c-axis orientation in the new bone and pre-existing
bone. Horizontal broken line indicates the value of the intensity ratio for
randomly oriented BAp.
c-axis orientation along the bone axis possibly progresses
along with mineralization according to the selective and
anisotropic crystal growth proposed using a rat tibia midshaft.31) Takano et al.32) demonstrated that the anisotropy of
acoustic velocity, corresponding to the anisotropy of elastic
modulus of new bone, increased with time as BMD of the
new bone increased, while the anisotropy in the demineralized bone (collagen matrix) did not change, possibly
reflecting anisotropic crystal growth. The new bone induced
in the present study could be a source of well-functioning
bone with highly oriented BAp.
The micro-organization of the bone induced by artificial
loading varies among literatures. In some cases, the new bone
adaptively formed against increased load was reported to
be a woven-type bone, which was poorly mineralized
with disorganized microstructure,30,33) while in other cases,
lamellar bone formation was reported.1,34,35) The microorganization of the new bone varied partly depending on
magnitude of in vivo strain5,36) and bone formation rate37)®
which are correlated to each other.30) According to the report
by Turner et al.,29) the disorganized collagen structure in the
bone induced by artificial loading was seen even after the
bone was densified. It is important to avoid the formation
of woven bone possessing disorganized microstructure and
impaired mechanical function38) because rat bone does not
generally undergo remodeling39) to replace woven bone with
lamellar bone. In the present study, the organization of
collagen and BAp crystals was highly anisotropic; therefore,
the loading condition might be optimal to finally form mature
bone with well-organized microstructure and related mechanical functions. Further study will aim at clarifying the
New bone
Pre-existing bone
Fig. 5 CLSM image of a longitudinal section of loaded ulna at the midshaft. L and C indicate osteocyte lacunae and canaliculi, respectively.
relationship between the amplitude of strain applied on the
bone and resultant collagen/BAp orientation.
3.3 Shape and alignment of osteocytes in the new bone
It is generally well accepted that osteocytes, cells
embedded within the mineralized matrix of bone, can sense
in vivo stress (strain) and transmit chemical signals to control
activation of osteoblasts and osteoclasts in response to the
change of loading condition.33) The shape and alignment
of osteocytes appear to change to effectively sense the
mechanical stimuli. Osteocytes in load-bearing tibias and
fibulas had a spindle shape and were aligned along the long
axis corresponding to the principal loading direction, whereas
those in less-loaded calvaria were more spherical.14,40) Thus,
the shape and alignment of osteocytes can be an indicator of
the stress applied on bone and osteocyte mechanosensing.
Figure 5 shows the CLSM image of the medial longitudinal section at the ulna mid-shaft. The osteocytes in the
new bone were elongated and aligned along the principal
loading direction, which corresponded to the direction of the
bone long axis, similar to those in the pre-existing bone.
Osteocytes appear to align themselves parallel to the
principal loading direction, which reflects the preferential
orientation of collagen and BAp c-axis, in response to
mechanical loading. Thus, we believe that osteocytes not
only mirror the orientation of collagen and BAp c-axis, but
also play an important role in the formation of the anisotropic
micro-organization composed of collagen and BAp crystals.
Although it presently remains uncertain how osteocytes
perceive mechanical loading and translate this into biochemical signals, the harmonized organization in the preferential
orientation of collagen and BAp c-axis and the osteocyte
long axis in response to mechanical loading during the bone
formation process indicates that osteocytes can be regarded
as an important factor in regulating bone microstructure, as
well as bone mass and BMD, in the functional adaptation
process of bone.
Orientation of Biological Apatite in Bone Induced by Artificial Loading
The findings in the present study are clinically important
from a viewpoint of biomechanics because mechanical
functions of bone change largely due to the alteration of
the degree of preferential BAp orientation which forms in
response to in vivo stress condition. In bone healing process,
for example, in vivo artificial loading might be useful to
enhance the formation of bone with preferential BAp
orientation and the resultant mechanical integrity, which is
usually not achievable even under the usage of tissue
engineering techniques.11,23)
4.
Conclusion
The present study first reported anisotropic micro-organization of collagen and BAp crystals as well as osteocytes
along the principal loading direction in the bone induced by
artificial loading. It is revealed that, in response to the
increased load, additional bone with a preferential orientation
of the collagen and BAp c-axis forms to adjust bone
mechanical functions according to increased mechanical
requirements. Bone with oriented collagen and BAp
possesses superior mechanical integrity, which contributes
to effective adaptation in terms of biomechanics.
Acknowledgments
This work was partly supported by the Funding Program
for Next Generation World-Leading Researchers (LR023)
from the Japan Society for the Promotion of Science (JSPS).
J. Wang received scholarship from the China Scholarship
Council (CSC).
REFERENCES
1) A. M. Weatherholt, R. K. Fuchs and S. J. Warden: Bone 52 (2013) 372­
379.
2) Q. Q. Zeng, W. S. S. Jee, A. E. Bigornia, J. G. King, Jr., S. M. D’Souza,
X. J. Li, Y. F. Ma and W. J. Wechter: Bone 19 (1996) 13­21.
3) J. M. LaMothe, N. H. Hamilton and R. F. Zernicke: Med. Eng. Phys. 27
(2005) 277­284.
4) B. A. Uthgenannt and M. J. Silva: J. Biomech. 40 (2007) 317­324.
5) Y. F. Hsieh, A. G. Robling, W. T. Ambrosius, D. B. Burr and C. H.
Turner: J. Bone Miner. Res. 16 (2001) 2291­2297.
6) A. G. Robling, F. M. Hinant, D. B. Burr and C. H. Turner: J. Bone
Miner. Res. 17 (2002) 1545­1554.
7) G. K. Wakley, B. L. Baum, K. S. Hannon and R. T. Turner: Calcif.
Tissue Int. 43 (1988) 383­388.
8) H. Yonezu, T. Ikata, S. Takata and A. Shibata: J. Bone Miner. Metab.
17 (1999) 259­265.
9) S. H. Lee, M. Todai, M. Tane, K. Hagihara, H. Nakajima and T.
1261
Nakano: J. Mech. Behav. Biomed. Mater. 14 (2012) 48­54.
10) M. Weinreb, G. A. Rodan and D. D. Thompson: Bone 10 (1989) 187­
194.
11) T. Ishimoto, T. Nakano, Y. Umakoshi, M. Yamamoto and Y. Tabata:
J. Bone Miner. Res. (2013), doi:10.1002/jbmr.1825.
12) T. Nakano, K. Kaibara, Y. Tabata, N. Nagata, S. Enomoto, E.
Marukawa and Y. Umakoshi: Bone 31 (2002) 479­487.
13) Y. Noyama, T. Nakano, T. Ishimoto, T. Sakai and H. Yoshikawa: Bone
52 (2013) 659­667.
14) A. Vatsa, R. G. Breuls, C. M. Semeins, P. L. Salmon, T. H. Smit and
J. Klein-Nulend: Bone 43 (2008) 452­458.
15) L. F. Bonewald and M. L. Johnson: Bone 42 (2008) 606­615.
16) W. J. Landis: Bone 16 (1995) 533­544.
17) S. J. Sample, M. Behan, L. Smith, W. E. Oldenhoff, M. D. Markel,
V. L. Kalscheur, Z. Hao, V. Miletic and P. Muir: J. Bone Miner. Res. 23
(2008) 1372­1381.
18) A. B. Dawson: Anat. Rec. 31 (1925) 1­17.
19) A. G. Torrance, J. R. Mosley, R. F. L. Suswillo and L. E. Lanyon:
Calcif. Tissue Int. 54 (1994) 241­247.
20) J. R. Mosley, B. M. March, J. Lynch and L. E. Lanyon: Bone 20 (1997)
191­198.
21) T. Nakano, T. Ishimoto, N. Ikeo and A. Matsugaki: Progress in
Advanced Structural and Functional Materials Design, ed. by T.
Kakeshita, (Springer Japan, Tokyo, 2012) pp. 155­167.
22) K. Sasaki, T. Nakano, J. D. Ferrara, J. W. Lee and T. Sasaki: Mater.
Trans. 49 (2008) 2129­2135.
23) T. Nakano, K. Kaibara, T. Ishimoto, Y. Tabata and Y. Umakoshi: Bone
51 (2012) 741­747.
24) C. Ciani, S. B. Doty and S. P. Fritton: Bone 44 (2009) 1015­1017.
25) H. Kamioka, T. Honjo and T. Takano-Yamamoto: Bone 28 (2001) 145­
149.
26) A. Matsugaki, G. Aramoto and T. Nakano: Biomaterials 33 (2012)
7327­7335.
27) S. J. Warden, J. A. Hurst, M. S. Sanders, C. H. Turner, D. B. Burr and
J. Li: J. Bone Miner. Res. 20 (2005) 809­816.
28) Y. F. Hsieh, T. Wang and C. H. Turner: Bone 25 (1999) 379­382.
29) C. H. Turner, T. A. Woltman and D. A. Belongia: Bone 13 (1992) 417­
422.
30) S. P. Kotha, Y. F. Hsieh, R. M. Strigel, R. Muller and M. J. Silva:
J. Biomech. 37 (2004) 541­548.
31) T. Nakano and T. Ishimoto: Kidney Metab. Bone Dis. 22 (2009) 215­
224 (in Japanese).
32) Y. Takano, C. H. Turner and D. B. Burr: J. Bone Miner. Res. 11 (1996)
1292­1301.
33) J. A. Lynch and J. S. Matthew: Bone 42 (2008) 942­949.
34) J. A. McKenzie and M. J. Silva: Bone 48 (2011) 250­258.
35) Y. F. Hsieh and C. H. Turner: J. Bone Miner. Res. 16 (2001) 918­
924.
36) C. H. Turner, M. R. Forwood, J. Y. Rho and T. Yoshikawa: J. Bone
Miner. Res. 9 (1994) 87­97.
37) M. Kashii, J. Hashimoto, T. Nakano, Y. Umakoshi and H. Yoshikawa:
J. Bone Miner. Metab. 26 (2008) 24­33.
38) T. Ishimoto, T. Nakano, M. Yamamoto and Y. Tabata: J. Mater. Sci.
Mater. Med. 22 (2011) 969­976.
39) D. H. Enlow: Am. J. Anat. 110 (1962) 79­101.
40) A. Himeno-Ando, Y. Izumi, A. Yamaguchi and T. Iimura: Biochem.
Biophys. Res. Commun. 417 (2012) 765­770.