Osteoporosis is characterized by low bone mass, increased bone fragility, and a greater risk for bone fracture. This disease is associated with significant morbidity and mortality and has become a major public health concern. Pharmacological intervention can aid in the prevention and treatment of osteoporosis, but such therapies are often accompanied by undesirable side effects.1, 2 Biophysical stimuli have been suggested as alternatives to drug-based therapies. One such promising stimulus is low-intensity ultrasound stimulation (LIUS).3–5
Ultrasound is an acoustic pressure wave with mechanical energy that can transmit into biological tissues. It has been used widely as an operative and diagnostic tool.6 It has therapeutic potential; in vivo animal studies demonstrated that LIUS accelerated the healing of bone fractures. LIUS enhances the mechanical properties of the healing callus,7–9 as well as bone bridging,9, 10 and is also effective for treating nonunion fractures11, 12 and distraction osteogenesis in animal models and clinical trials.13, 14 Ultrasound appears to affect healing by modulating bone structure. The microarchitecture of LIUS-treated, regenerated bone is enhanced in rats following osteodistraction.14 Similarly, improvements in mechanical strength, bone mineral density (BMD), and hard callus area in LIUS-treated rabbits were reported.13 BMD in the tibia callus at the location where LIUS is applied is also significantly higher in treated animals.15 Together, these results suggest that LIUS accelerates callus maturation and reduces healing time. These in vivo data are supported by in vitro cellular studies.3, 16–20 LIUS might be effective for depressing osteoclastic function (3) and for suppressing osteoclast formation,3, 17, 20 while also enhancing osteoblast formation and function.16, 18–20
Based on these findings, LIUS may be effective for preventing/treating osteoporotic bone loss. However, the effect of LIUS on osteoporotic bone remains controversial.1, 5, 21, 22 In some cases, LIUS has been an effective therapy in osteoporosis models,5, 22 while in others it has proven ineffective.1, 21 Warden et al.1 showed no beneficial effects of LIUS for preventing osteoporosis, because intact osteoporotic bone might be less sensitive to LIUS than isolated bone cells and defective bone, and it might be difficult to affect bone with LIUS.21 In contrast, Carvalho and Cliquet22 and Perry et al.5 observed more new bone formation and enhanced microarchitecture in LIUS-treated than nontreated groups.22
The reason for the controversy is unclear, but uncontrollable variables or intrinsic limitations of conventional experimental and analytic methods may be responsible. For example, existing studies largely ignored the effects of bone heterogeneity, individual variability, and location of the LIUS application. The trabecular bone network is heterogeneous. Thus, conventional histomorphometry used to determine the effects of LIUS on bone microarchitecture in most studies has intrinsic limitations due to the variability resulting from analysis of a few fields of view.23 3D bone microarchitecture imaging is a viable method for overcoming these limitations, but few studies have utilized this technology. Moreover, adaptation within a single bone can differ from location to location,24 but studies have yet to examine this variable with regard to the effects of LIUS. Longitudinal studies have not been widely performed, so existing data do not account for differences in baseline values between individuals.23 Finally, previous studies did not determine the effects of irradiation location/direction on the efficacy of LIUS application. Thus, our aim was to address these issues (bone heterogeneity, individual variability, locations of LIUS application, longitudinal study) and determine whether LIUS therapy can effectively prevent or treat osteoporotic bone loss induced by estrogen deficiency.
MATERIALS AND METHODS
All procedures were performed under a protocol approved by the Yonsei University Animal Care Committee (YWC-P107). Eight 14-week-old virginal ICR mice were ovariectomized (OVX) to induce bone loss, which was confirmed by changes in microarchitecture2 (52.2% decrease in bone volume fraction [BV/TV] with OVX).
Application of LIUS
Right tibiae were treated with LIUS (US group); left tibiae were untreated and served as an internal control (CON group). LIUS was administered at a repeated frequency of 1.0 kHz with a spatial-averaged temporal-averaged intensity of 30 mW/cm2 and a pulse width of 200 µs.1, 21 LIUS was applied for 20 min/day, 5 days/week, over a 6-week period.21 Before the experiment, LIUS output characteristics were measured by hydrophonic scanning. All mice were immobilized using a customized restrainer,25 and both tibiae were submerged in warm water (35–40°C) for LIUS administration (Fig. 1).1
Bone Structure Analysis
Both tibiae were scanned after 0, 3, and 6 weeks of LIUS treatment using in vivo micro-computed tomography (CT; Skyscan 1076, SKYSCAN N.V., Aartselaar, Belgium) at a voxel resolution of 18 µm × 18 µm × 18 µm. During the scan, animals received a combination of ketamine (1.5 ml/kg, Huons, Seoul, Korea) and xylazine (0.5 ml/kg, Bayer Korea, Seoul, Korea). Structural parameters for trabecular and cortical bone were analyzed using the images and CT-AN 1.8 software (Skyscan). The BV/TV (%), trabecular thickness (Tb.Th, mm), trabecular number (Tb.N, mm−1), trabecular separation (Tb.Sp, mm), structure model index (SMI), and trabecular bone pattern factor (Tb.Pf, mm−1) were examined for the entire trabecular bone. Also, 2D cross-sectional images of trabecular bone were subdivided into five regions of interest (ROIs) to evaluate the effects of the LIUS irradiation location/direction (Fig. 2a). The ROIs were 0.5 mm in diameter, corresponding to the maximum selectable diameter in the medullary cavity. For cortical bone, cross-sectional thickness (Cs.Th, mm) and mean polar moment of inertia (MMI, mm4) were examined. 2D cross-sectional images of the cortical bone were subdivided into four ROIs to evaluate the effect of the LIUS irradiation location/direction (Fig. 2b).
Animals were sacrificed after 6 weeks of LIUS. Both tibiae were extracted, and skin, muscle, and tendons were removed. The tibiae were fixed for 3 days in 10% neutral-buffered formalin, treated with 10% formic acid for 1 h, decalcified with a 10% ethylenediaminetetraacetic acid solution, and embedded in paraffin. Four-micrometer-thick sections were cut through the long axis in the sagittal plane with a microtome (Microm, Walldorf, Germany) and then visualized with Masson's trichrome stain. Analyses were performed using a microscope (Olympus BX50, Tokyo, Japan) to evaluate new bone formation (red: uncompleted mineralization, blue: mature mineralization) and osteocytes. The osteocytes in a 200 µm × 200 µm field of view were counted.
The structural parameters were compared using ANOVA with a mixed factorial design and repeated measures. A paired t-test was performed to compare the number of osteocytes between the US and CON groups. All descriptive data are represented as mean ± standard error. p Values <0.05 were considered significant.
No significance differences in the structural parameters of the trabecular or cortical bone were found in the US group during the study period (Tables 1–4). In the CON group, on weeks 3 and 6, the BV/TVs and Tb.Ns decreased significantly compared to week 0, whereas the Tb.Pfs and SMI increased significantly (p < 0.05). However, no significant differences for other structural parameters of trabecular and cortical bone were observed in the CON group during the study. Moreover, in both the US and CON groups, BV/TVs for most regions of trabecular bone, except to R2 and R3, and Cs.Ths for four regions of cortical bone did not change significantly over time.
Table 1. Descriptive Values of Structural Parameters for Trabecular Bone, n = 8
R, region; US, ultrasound group; CON, control group.
Significant difference compared to week 0 (p < 0.05).
0.76 ± 0.93
1.13 ± 1.27
0.23 ± 0.38
0.56 ± 0.60
0.27 ± 0.51
0.01 ± 0.02
1.42 ± 2.23
0.86 ± 1.79
0.41 ± 0.81
0.55 ± 1.11
0.04 ± 0.09
0.00 ± 0.01
Table 4. Descriptive Values of Cross-Sectional Thickness (Cs-Th, mm) for Four Regions of Cortical Bone, n = 8
0.38 ± 0.07
0.34 ± 0.16
0.36 ± 0.07
0.28 ± 0.04
0.24 ± 0.12
0.31 ± 0.08
0.39 ± 0.06
0.34 ± 0.15
0.37 ± 0.08
0.41 ± 0.06
0.32 ± 0.14
0.34 ± 0.05
0.34 ± 0.06
0.24 ± 0.10
0.29 ± 0.06
0.42 ± 0.09
0.37 ± 0.18
0.40 ± 0.10
R, region; US, ultrasound group; CON, control group.
0.27 ± 0.06
0.23 ± 0.10
0.24 ± .05
0.31 ± 0.07
0.27 ± 0.12
0.29 ± 0.04
Animals were examined for LIUS-induced structural changes after 3 and 6 weeks (Fig. 3). The increment and decrement rates were determined by calculating the degree of change relative to the value obtained at week 0. At week 3, the SMI increment rate in the US group (1/8 animals, 12.5%) was significantly smaller than that of the CON group (p < 0.05). However, no significant differences between the US and CON groups were observed for the other structural parameters (Fig. 3). After 6 weeks of LIUS treatment, the decrement rates of BV/TV (8/8 animals, 100%), Tb.N (6/8 animals, 75%), and Tb.Pf (7/8 animals, 87.5%) were significantly smaller than those in the CON group (p < 0.05). Although the Tb.Th decrement rate was small in the US group, it remained unchanged (4/8 animals, 50%, p > 0.05). The increment rates of the other parameters (Tb.Sp: 5/8, 62.5%; SMI: 5/8, 62.5%) did not differ significantly between the two groups. However, most trabecular bone structural parameters at week 6 tended to be enhanced in the US group.
No significant differences in BV/TV based on values calculated from the BV/TV at week 0 were found in any of the five ROIs between the US and CON group at week 3 (Fig. 4). However, by week 6, the US group showed an increase in the relative variation in the BV/TV within region 1 (7/8 animals, 87.5%, p < 0.05), although no significant differences were observed between the groups in other regions (R2: 2/8 animals, 25%, p > 0.05, R3: 5/8 animals, 62.5%, p > 0.05, R4: 4/8 animals, 50%, p > 0.05, R5: 5/8 animals, 62.5%).
LIUS-induced changes in the structural parameters of cortical bone were also observed (Fig. 5). No differences between the US and CON groups were found following 3 weeks of treatment. However, at 6 weeks, the relative variation in MMI (5/8, 62.5% animals) in the US group was significantly higher than in the CON group (p < 0.05), whereas the relative variation in Cs.Th (3/8, 37.5% animals) remained unchanged (p > 0.05). In the ROI analysis, at 3 weeks, the cortical bone thickness in region 2 (R2) increased significantly in the US group compared to the CON group (6/8 animals, 75%, p < 0.05, Fig. 6). However, most cortical bone structural parameters at week 6 tended to be enhanced in the US group. By week 6, cortical bone thickness in region 1 (R1) decreased significantly less in the US group than in the CON group (R1: 6/8 animals, 75%, p < 0.05), and increased significantly in region 2 (R2) compared to the CON group (R2: 7/8 animals, 88%, p < 0.05). Changes in the structural parameters at week 6 based on week 3 were calculated, but no difference were observed between the US and CON groups.
The structures of the trabecular and cortical bone in the US group were maintained relative to those in the CON group, and new bone formation and increases in bone thickness were generally greater in the US group (Fig. 7).
Increased levels of new bone formation, incomplete bone mineralization in the trabecular bone (Fig. 8), and more osteocytes in the cortical bone were observed in the US group (16.7 ± 2.1) relative to the CON group (7.3 ± 2.1, p < 0.05). Also, the trabecular bone was thicker in the US group, and the thickness of the endosteal cortical bone near the region directly treated with LIUS increased. These results are consistent with the structural parameter data and correspond well to the visual analysis (Fig. 7).
We demonstrated the quantitative and qualitative effectiveness of LIUS for treating or preventing osteoporotic bone loss through an analysis of bone structural characteristics. Bone generally responds to external mechanical loading via mechanosensory mechanisms, including mechanoreception and mechanotransduction,26, 27 suggesting that LIUS should have a positive effect on bone regeneration. However, conflicting reports exist about the ability of LIUS to modulate bone loss and its clinical use in treating osteoporosis.1, 5, 21, 22 Differences in LIUS efficacy may be due to uncontrollable factors or intrinsic limitations of conventional experimental and analytic methods. We attempted to overcome these factors using 3D bone microarchitecture analysis, examining the partial effects of LIUS at different sites within the same bone, conducting a longitudinal study to account for individual differences in baseline values, and assessing the effect of irradiation location and direction during LIUS application. Thus, our study was designed to address the potentially confounding factors of previous studies and to provide qualitative and quantitative examination of the effects of LIUS on the prevention and treatment of osteoporotic bone loss.
The structural parameter analysis showed that the decrement rates of BV/TV and Tb.N and the increment rate of Tb.Pf for trabecular bone in the US group were statistically different at week 6 relative to the CON group. Also, 6 weeks of LIUS treatment affected the increment rates of Tb.Sp, Tb.Th, and SMI for trabecular bone, indicating that the treatment is generally effective for preventing bone loss, although there were no significance differences between the US and CON group. Together, these data indicate that LIUS might prevent bone loss and bone disconnection and thereby restrain the continuous progress of bone loss that occurs in osteoporosis. For the cortical bone, the relative variation in MMI in the US group was significantly larger than that in the CON group at week 6. These analyses also suggested that the smaller decrement rates of Cs.Th for cortical bone in the US group were generally effective for preventing bone loss at 6 weeks. These findings suggest that LIUS may also restrain osteoporosis-associated progression of bone loss in the cortical bone. However, the magnitude of the effects for preventing bone loss was small, possibly due to an inability of the US to reach bone.21, 28 Also, local stimuli in partial bone cause systemic adaptations in multiple bones by neuronal regulation.29 Therefore, although bones in the CON group were not stimulated directly by LIUS, they may have adapted through neuronal regulation.
To evaluate site-specific bone adaptation and examine the influence of irradiation location/direction, we analyzed BV/TV in five ROIs in trabecular bone and evaluated cortical bone thickness in four ROIs. The relative variation in BV/TV for the trabecular bone and Cs.Th for the cortical bone in R1, the site of direct LIUS irradiation, were significantly larger and smaller, respectively, in the US group than in the CON group. However, in the US group relative little variation occurred in BV/TV for the trabecular bone and Cs.Th for the cortical bone in R3 and R4, the furthest site of direct LIUS irradiation, for treating or preventing osteoporotic bone loss compared to the CON group. These results suggest that bone adaptation could differ at distinct sites and that the irradiation location/direction of LIUS application could affect the treatment outcome. These data are consistent with previous reports that suggested that bone cells may respond to temporal and positional information in the form of local stimuli, such as strain-energy density and deformation of the bone matrix.24, 30–32 However, additional studies are needed to assess the effects of location, direction, frequency, and intensity on LIUS treatment outcomes.
The histological analysis revealed new bone formation in the endosteal cortical bone in the US group, especially near the region directly stimulated by LIUS (Fig. 7). New bone formation in trabecular bone and the thickness of the cortical bone also increased in the US group. These results indicate that LIUS might stimulate new bone formation, consistent with previous studies.5, 7, 22 In addition, more osteocytes were observed in the US group. Osteocytes regulate osteoblastic and osteoclastic activities and play a role in mechanotransduction.33 When osteocyte apoptosis increases, which also occurs during estrogen deficiency,34 bone fragility increases.35 Our data indicate that LIUS may prevent osteocyte apoptosis and bone weakness induced by estrogen deficiency, and affect mechanotransduction, thereby preventing bone loss.
We did not identify the LIUS cellular mechanism of action for the biological effects. Previous studies showed that LIUS regulates osteoblasts and osteoclasts. LIUS might depress osteoclastic function by affecting paracrine regulation and decreasing receptor activation of nuclear factor kappa B.3, 17 Specifically, LIUS suppresses osteoclast formation by inhibiting both nuclear factor kappa B ligand and macrophase colony stimulating factor activation.20 In parallel, LIUS enhances osteoblastic differentiation20 and increases alkaline phosphatase activity, preosteoblast cell mineralization,19 and prostaglandin E2 production via cyclooxygenase 2 in mouse osteoblastic cells.16 It also induces osteogenic transcription of osteoblasts via rapid activation of membrane-bound Gαi and cytosolic extracellular signal-regulated kinase (2). Finally, LIUS-stimulated bone marrow cells upregulate osteocalcin and insulin-like growth factor-I in a biphasic manner, suggesting that LIUS induces anabolic responses in osteoblasts.18 These results imply that LIUS might enhance bone formation and inhibit bone resorption. Our results may have been due to such cellular effects of LIUS.
In summary, our results show that LIUS may restrain the continuous progress of bone loss by enhancing bone formation and suppressing bone resorption. Therefore, LIUS may be effective for preventing bone loss and decreasing the risk of bone fracture.
This work was supported by a Korean Science and Engineering Foundation (KOSEF) grant funded by the Korean government (MEST) (R01-2008-000-11641-0).