Unloading Induces Osteoblastic Cell Suppression and Osteoclastic Cell Activation to Lead to Bone Loss via Sympathetic Nervous System*

Osteoporosis is one of the major health problems in our modern world. Especially, disuse (unloading) osteoporosis occurs commonly in bedridden patients, a population that is rapidly increasing due to aging-associated diseases. However, the mechanisms underlying such unloading-induced pathological bone loss have not yet been fully understood. Since sympathetic nervous system could control bone mass, we examined whether unloading-induced bone loss is controlled by sympathetic nervous tone. Treatment with β-blocker, propranolol, suppressed the unloading-induced reduction in bone mass. Conversely, β-agonist, isoproterenol, reduced bone mass in loaded mice, and under such conditions, unloading no longer further reduced bone mass. Analyses on the cellular bases indicated that unloading-induced reduction in the levels of osteoblastic cell activities, including mineral apposition rate, mineralizing surface, and bone formation rate, was suppressed by propranolol treatment and that isoproterenol-induced reduction in these levels of bone formation parameters was no longer suppressed by unloading. Unloading-induced reduction in the levels of mineralized nodule formation in bone marrow cell cultures was suppressed by propranolol treatment in vivo. In addition, loss of a half-dosage in the dopamine β-hydroxylase gene suppressed the unloading-induced bone loss and reduction in mineralized nodule formation. Unloading-induced increase in the levels of osteoclastic activities such as osteoclast number and surface as well as urinary deoxypyridinoline was all suppressed by the treatment with propranolol. These observations indicated that sympathetic nervous tone mediates unloading-induced bone loss through suppression of bone formation by osteoblasts and enhancement of resorption by osteoclasts.

Osteoporosis is one of the major age-related diseases in our modern world (1)(2)(3)(4). Especially, high fracture risk in osteoporosis patients results in not only loss of quality of life but also loss of life in a certain fraction of aged patients. The number of osteoporosis patients is estimated to be close to 10% of the whole population in many advanced countries. Among this patient population, a significant number of patients have disuse osteoporosis based on bedridden conditions caused by aging-related cardiovasucular as well as cerebrovascular diseases (5).
Bone has been known to be lost upon the removal of the mechanical stimuli, and prolonged lack of mechanical stimuli leads to disuse osteoporosis (5)(6)(7)(8)(9). However, the mechanisms underlying such disuse osteoporosis are largely unknown (2,5,6). Bone mass is determined by the actions of osteoblasts, which make bone, and those of osteoclasts, which resorb bone (2,(12)(13)(14)(15). The balance between the two activities is under the control of hormones and cytokines. Usually, simultaneous changes in the two activities are considered to be coupled to compensate bone loss. However, in the case of disuse osteoporosis, bone formation activities are not enhanced even in the presence of enhanced bone resorption. Rather, bone formation is significantly suppressed in disuse osteoporosis (16,17). Therefore, disuse osteoporosis is a critical pathological situation where bone mass is continuously lost without having any compensatory activity against the reduction of bone. However, how such a critical reduction in bone formation occurs in unloading-induced pathological bone loss is not yet known.
Bone formation and bone resorption are under the control of the systemic hormones and local cytokines (2,6). However, none of these factors have been proven to be the major cause of the disuse osteoporosis. In addition to the bedridden patients, astronauts under gravity conditions also lose bone due to the loss of mechanical stress. Analysis of such astronauts returning from space indicated that sympathetic nervous tone is enhanced in their muscle (18,19). Sympathetic nervous system would regulate bone mass via bone formation by osteoblasts systemically (20). However, nothing has been known about how this system is related to pathophysiology in bone metabolism in the body. Therefore, we examined whether sympathetic nervous tone is involved in reduction of bone mass in a disuse osteoporosis model via osteoblastic and osteoclastic cells using hind limb unloading.

MATERIALS AND METHODS
Animals-Male 129 or C57BL/6J mice (10 -14 weeks old) were used for the experiments. Mice were housed for at least 1 week prior to the study. * This work was supported by grants-in-aid received from the Japanese Ministry of Education (21st Century Center of Excellence Program, Molecular Destruction and Reconstitution of Tooth and Bone, Grants 14207056, 16659405, 16027215, and 16022221) and grants from the Japan Space forum, NASDA, and the Japan Society for Promotion of Science (JSPS Core to Core Program, Research for the Future Program, Genome Science). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
The mice were subjected to either intraperitoneal injections of propranolol (20 g/g of body weight/day) (21), continuous administration of guanethidine via an osmotic minipump (20 g/g of body weight/day) (22), or intraperitoneal injections of isoproterenol (6 g/g of body weight/day) (23). In the case of osmotic minipump implantation, the pump was implanted 12 h before the start of hind limb unloading. Osmotic minipumps were implanted into the subcutaneous tissue in the back of the animals according to the manufacturer's instruction. For dopamine ␤-hydroxylase (DBH) 1 gene deletion experiments, heterozygous knockout mice with a C57BL6/129sv F2 background and wild type litter mate mice were used (13-week-old females) (24). All of the mice were injected intraperitoneally with calcein at 4 mg/kg at 4 and 2 days before sacrifice. After treatment for 10 or 14 days, mice were anesthetized with tribromoethanol at 200 mg/kg and were sacrificed by cervical dislocation.
Hind Limb Unloading Model-Hind limb unloading was conducted by applying a tape to the surface of the hind limb to set a metal clip (10,16). The end of the clip was fixed to an overhead bar. The height of the bar was adjusted to maintain the mice at an ϳ30°head down tilt with the hind limbs elevated above the floor of the cage. The mice were subjected to hind limb unloading for 10 or 14 days. Loaded control mice were also housed individually under the same conditions except for hind limb unloading for the same duration.
Body Weight-The body weight of the mice was monitored during the experimental period. There were no significant changes in body weight in any of the groups during the course of the study. This confirmed that stress could be considered minimal in our experiments as previously described (16,21).
Two-dimensional Micro-CT Analysis of Bone-Bone volume/tissue volume (BV/TV) was determined based on two-dimensional micro-CT analyses using a micro-CT apparatus (Musashi, Nittetsu-ELEX Co., Kita-Kyushu City, Japan). The data were quantified by using automated image analysis system (Luzex-F, Nireco). The fractional bone volume (BV/TV) was obtained in an area of 0.47 mm 2 with its closest and furthest edges at 0.28 and 0.84 mm, respectively, distal to the growth plate of the proximal ends of the tibiae. The threshold level for the measurements was set at 110 for the analyses (16,21).
Histomorphometric Analysis of Bone-At the end of the experiments, the right femora of each mouse were removed and fixed in 70% ethanol. Serial 3-m-thick sagittal sections were made as undecalcified sections (right femora). For bone formation rate (BFR), metaphyseal cancellous bone in the femora was used to obtain bone fraction in a rectangular area of 0.34 mm 2 (0.5 ϫ 0.67 mm) with its closest and furthest edges at 0.3 and 0.8 mm distal to the growth plate, respectively. For decalcified sections, the left tibiae of the mice were removed at the end of the experiments and fixed in 4% paraformaldehyde and then decalcified in 20% EDTA. Serial 5-mm-thick sagittal sections were made using a microtome and stained for tartrate-resistant acid phosphatase (TRAP). TRAP-positive multinucleated cells attached to bone were scored as osteoclasts. Measurements were made within an area of 0.24 mm 2 (0.6 ϫ 0.4 mm), with its closest and furthest edges at 0.3 and 0.7 mm distal to the growth plate of the proximal ends of the tibiae. Histomorphometry was conducted to quantify the number of osteoclasts (Oc.N/ BS) and osteoclast surface (Oc.S/BS) as defined by Parfitt et al. (11).
Nodule Formation Analysis-We cultured the cells obtained from the bone marrow of the animals that were subjected to hind limb unloading (12) or loading in combination with the treatment with adrenergic modulators. The cells were cultured in the presence of ascorbic acid (50 g/ml) and ␤-glycerophosphate (10 mM) in ␣-minimal essential medium supplemented with 10% FBS, 1% antibiotics. After 3 weeks in culture, alizarin red staining was conducted. This staining was used to visualize the calcified materials formed in vitro. Briefly, after the cultures were terminated, the cells were fixed in 100% ethanol and then were stained in alizarin red solution (1%) for 1 min. The cultures were then rinsed several times with water. The area of alizarin red positive nodules was measured by using an image analyzer.
Urinary Deoxypyridinoline-Deoxypyridinoline levels in urine at the end of the hind limb unloading were measured by enzyme-linked immunosorbent assay (Metra Biosystems) (6). Urine samples were collected from five mice per group, which were housed in a metabolic cage during the last 24 h and analyzed.
Statistical Analysis-Data were expressed as means Ϯ S.D., and statistical evaluation was performed based on analysis of variance, using a statistical software package for Windows, Statview version 5.0 (SAS Institute). A p value less than 0.05 was considered to be statistically significant.

RESULTS
Hind limb unloading reduced bone volume in vehicle-treated control mice as reported previously (16) (Fig. 1, a and b, column 1 versus column 2). In contrast, treatment with propranolol, a ␤-adrenergic blocker acting at the receptor levels, suppressed hind limb unloading-induced reduction in bone mass expressed as BV/TV ( Fig. 1, a and b; no significant difference between columns 3 and 4). With regard to the drug effect in unloaded mice, propranolol treatment resumed the bone loss induced by unloading (Fig. 1b, column 2 versus column 4). Control-loaded mice treated with propranolol revealed significant difference in bone mass compared with unloaded vehicle-treated mice (Fig.  1b, column 2 versus column 3). If hind limb unloading suppressed the levels of bone mass and bone formation through sympathetic tone, not only the actions of the blockers for sympathetic signaling, which act at the receptor levels, but also the depletion of the presynaptic transmitter reservoir should affect the unloading-induced reduction in bone mass. Therefore, guanethidine sulfate was administrated to deplete norepinephrine at the presynaptic nerve ending levels in the animals that were subjected to hind limb unloading. As shown in Fig. 2, guanethidine treatment suppressed unloading-induced bone loss (Fig. 2, a and b). The pattern of the levels in BV/TV was similar to that in the case of propranolol (Fig. 1). Body weight was not altered significantly due to unloading or drug treatment in all of our experiments (Fig. 2, c-e).
To further test whether unloading signaling is exerted in the line of sympathetic tone, we examined the effects of activation by isoproterenol of sympathetic tone. This was in order to determine whether isoproterenol may mask the unloadinginduced signals. Isoproterenol treatment reduced bone mass in control-loaded mice as reported previously (Fig. 3, a and  unloading no longer reduced bone mass (Fig. 3, a and b; compare columns 3 and 4). Notably, the levels of bone mass reduction by isoproterenol treatment were similar to those in unloaded mice (Fig. 3b, column 3 versus column 2). Thus, the data of these three sets of experiments are in accordance with the notion that unloading signaling is suppressed by blockers for sympathetic signaling, and such signaling is no longer active in the presence of the action of ␤-adrenergic agonist.
In order to examine the mechanisms of ␤-adrenergic actions in unloading-induced bone loss, we conducted dynamic histomorphometry. Hind limb unloading induced reduction in the levels of bone mineral apposition rate (MAR)), mineralizing surface (MS), and BFR ( Fig. 4, a-d) after 10 days of hind limb unloading and propranolol treatment suppressed the unloading-induced decrease in MAR, MS, and BFR ( Fig. 4, a-d). The patterns of the columns were similar to those in the case of bone mass (Fig. 1). Guanethidine treatment also suppressed the hind limb unloading-induced reduction in MAR, MS, and BFR (Fig. 5, a-d) in the mice subjected to hind limb unloading for 2 weeks. Again, the patterns of the columns were similar to those in the case of bone mass (Fig. 1). Thus, depletion of sympathetic neurotransmitter has effects similar to the block of ␤-adrenergic receptor with respect to bone formation activity in vivo. Isoproterenol treatment alone suppressed the levels of mineral apposition rate, mineralizing surface, and bone formation rate, whereas in the presence of isoproterenol treatment, 10 days of hind limb unloading failed to suppress all of these parameters (Fig. 6, a-d). It is again notable that the patterns of the columns in Fig. 6 were similar to those observed in the effects of isoproterenol on bone mass (Fig. 3). These three lines of evidence based on dynamic histomorphometry indicated that bone formation is the target of sympathetic tone in mice subjected to hind limb unloading. a, two-dimensional micro-CT pictures of the midsagittal planes of the proximal regions of the tibiae after 14 days of hind limb unloading (Unload) or loading (Control Load) in vehicle-treated or guanethidine-treated 129 mice. Two-dimensional micro-CT analyses were conducted as described under "Materials and Methods." b, fractional trabecular BV/TV was quantified based on the image analysis of two-dimensional micro-CT pictures of the tibiae after 14 days of either hind limb unloading (Unload) or loading (Load) in vehicle-treated or guanethidine-treated mice. Data are expressed as means and S.D. *, statistically significant difference (p Ͻ 0.05). #, statistically significant difference between the vehicle group and drug-treated group (p Ͻ 0.05) (either control-loaded or unloaded groups). §, statistically significant difference between vehicle-treated unloaded group and drug-treated control-loaded group (p Ͻ 0.05). c-e, body weight was taken during the experiments for propranolol (c), guanethidine (d), and isoproterenol (e). No major alteration was observed. to the growth plate in the metaphyseal region were measured as described under "Materials and Methods." The mice were injected intraperitoneally with calcein at 4 mg/kg 4 and 2 days before sacrifice at 10 days. Data are expressed as means and S.D. of five bones from each of the four groups: vehicle-and propranolol-treated mice loaded and unloaded groups. *, statistically significant difference (p Ͻ 0.05). #, statistically significant difference between the vehicle group and drugtreated group (p Ͻ 0.05) (either control-loaded or unloaded groups). §, statistically significant difference between vehicle-treated unloaded group and drug-treated control-loaded group (p Ͻ 0.05).
We further examined whether our observations can be detected at cell levels in culture. For this purpose, a nodule formation assay was conducted by using bone marrow cells obtained from the bone of the mice after they were subjected to hind limb unloading or control loading in the presence or the absence of the treatment with pharmacological agents. Hind limb unloading reduced nodule formation in the cultures of cells obtained from the animals (Fig. 7, a and b, column 1  versus column 2). Propranolol treatment in vivo suppressed the unloading-induced reduction in the mineralized nodule formation in culture (Fig. 7, a and b, column 3 versus column 4). Isoproterenol treatment suppressed the levels of nodule formation in loaded control mice (Fig. 8, a and b, column 1 versus  column 3), and in the presence of isoproterenol treatment in vivo, hind limb unloading failed to further reduce the levels of nodule formation in bone marrow cells in culture (Fig. 8, a and  b, column 3 versus column 4). Thus, these in vitro experiments indicated that ␤-adrenergic sympathetic tone mediates unloading-induced reduction in mineralization of bone marrow cell cultures.
In order to examine the effects of the sympathetic nervous tone on unloading-induced reduction in bone mass and bone formation in a genetic model rather than pharmacological modulation, DBH (dopamine ␤-hydroxylase) knockout mice were subjected to hind limb unloading. Hind limb unloading reduced bone mass by about 68.6% in wild type littermate mice (Fig. 9a,  column 1 versus column 2). Heterozygous loss for the dopamine ␤-hydroxylase gene attenuated the reduction in bone loss by about 29.7% after hind limb unloading (Fig. 9a, column 3

FIG. 5. Guanethidine treatment suppressed unloading-induced reduction in bone formation in vivo.
Mice were treated with guanethidine as described in the legend to Fig. 3. The number of mice in each group is indicated as N. a, calcein double-labeled surfaces of the bones at the ends of the femora after hind limb unloading (Unload) or loading (Load) in vehicle-or guanethidine-treated mice. The arrows indicate the lines of calcein labeling (light green) used to obtain data shown in b-d. b-d, in the undecalcified sections of the distal ends of the femora, MAR (b), MS (c), and BFR (d) were measured in all areas at 0.3-0.8 mm distal to the growth plate in the metaphyseal region as described under "Materials and Methods." The mice were injected intraperitoneally with calcein at 4 mg/kg 4 and 2 days before sacrifice at 2 weeks. Data are expressed as means and S.D. for five bones from each of the vehicle-and guanethidine-treated mice groups. *, statistically significant difference (p Ͻ 0.05). #, statistically significant difference between the vehicle group and drug-treated group (p Ͻ 0.05) (either control-loaded or unloaded groups). §, statistically significant difference between vehicle-treated unloaded group and drug-treated controlloaded group (p Ͻ 0.05).
FIG. 6. Isoproterenol treatment suppressed the level of bone formation parameters, and unloading did not further suppress these parameters in vivo. Mice were treated with isoproterenol as described in the legend to Fig. 3. The number of mice in each group is indicated as N. a, calcein double-labeled surfaces of the bones at the ends of the femora after hind limb unloading (Unload) or loading (Load) in vehicle or isoproterenol-treated mice. The arrows indicate the lines of calcein labeling (light green) used to obtain data shown in b-d. b-d, in the undecalcified sections of the distal ends of the femora, MAR (b), MS (c), and BFR (d) were measured in all areas at 0.3-0.8 mm distal to the growth plate in the metaphyseal region as described under "Materials and Methods." The mice were injected intraperitoneally with calcein at 4 mg/kg 4 and 2 days before sacrifice at 10 days. Data are expressed as means and S.D. for five bones from each of the vehicle-and isoproterenoltreated mouse groups. *, statistically significant difference (p Ͻ 0.05). #, statistically significant difference between the vehicle group and drugtreated group (p Ͻ 0.05) (either control-loaded or unloaded groups).
versus column 4). The rate of bone loss due to unloading (calculated as (control load Ϫ unload)/(control load ϫ 100%) was significantly reduced from 68 Ϯ 7% in DBHϩ/ϩ (wild type) to 30 Ϯ 24% in DBHϩ/Ϫ (Fig. 9c) (p Ͻ 0.05). The nodule formation in cultures of bone marrow cells of the wild type littermate was reduced by unloading (Fig. 9b, column 1 versus column 2). The marrow cells obtained from DBH gene heterozygous knockout mice indicated suppression of hind limb unloadinginduced reduction in nodule formation (Fig. 9b, column 3

versus column 4).
During the course of unloading-induced bone loss, bone resorption also occurs as critical events to reduce bone mass. Unloading in tail-suspended mice caused an increase in osteoclast number (Oc.N/BS) and osteoclast surface (Oc.S/BS) based on histomorphometry in vivo as reported previously (Fig. 10, b-g, column 1 versus column 2). In contrast, inhibi-tion of sympathetic tone by treatment with propranolol or guanethidine, suppressed such unloading-induced increase in osteoclast number (Oc.N/BS) (Fig. 10, a and b for propranolol, d for guanethidine; column 3 versus column 4) and osteoclast surface (Oc.S/BS) (Fig. 10, c for propranolol, e for guanethidine; column 3 versus column 4). The levels of Oc.N/BS and Oc.S/BS were enhanced by either unloading or isoproterenol treatment alone to similar levels (Fig. 10, f and  g, column 2 versus column 3). The simultaneous presence of unloading conditions and isoproterenol treatment resulted in similar levels in the increase in osteoclast number (Oc.N/BS) and surface (Oc.S/BS) to those in mice subjected to either one of the two conditions alone (Fig. 10, f and g, column 4 versus  columns 2 and 3). Furthermore, unloading-induced increase in deoxypyridinoline excretion into urine (Fig. 11, column 1  versus column 2) was also suppressed by guanethidine treatment (Fig. 11, column 3 versus column 4).

FIG. 8. Isoproterenol treatment in vivo suppressed the levels of nodule formation, and unloading conditions failed to further reduce the mineralization levels in vitro.
Mice were treated with isoproterenol as described in legend to Fig. 6. The number of mice in each group is indicated as N. The cells obtained from the bone marrow of the animals subjected to hind limb unloading with the isoproterenol or vehicle treatment were cultured in the presence of ascorbic acid and ␤-glycelophosphare. After 3 weeks, alizarin red staining was conducted, and the area of the alizarin red-positive nodule was quantified. *, statistically significant difference (p Ͻ 0.05). #, statistically significant difference between the vehicle group and drug-treated group (p Ͻ 0.05) (either control-loaded or unloaded groups).

FIG. 7. Propranolol treatment in vivo suppressed unloadinginduced reduction in nodule formation in vitro.
Mice were treated with propranolol as described in the legend to Fig. 1. The cells obtained from the bone marrow of the animals subjected to hind limb unloading with the propranolol or vehicle treatment were cultured in the presence of ascorbic acid and ␤-glycelophosphare. After 3 weeks, alizarin red staining was conducted, and the area of the alizarin red-positive nodule was quantified. The number of wells in each group is indicated as N. *, statistically significant difference (p Ͻ 0.05). #, statistically significant difference between the vehicle group and drug-treated group (p Ͻ 0.05) (either control-loaded or unloaded groups). §, statistically significant difference between vehicle-treated unloaded group and drug-treated control-loaded group (p Ͻ 0.05).
FIG. 9. Bone loss and reduction in the levels of nodule formation was suppressed in DBH؉/؊ mice and wild-type mice. a, fractional trabecular BV/TV was obtained by quantification of micro-CT pictures of the tibiae after 2 weeks of either hind limb unloading (Unload) or loading (Load) in wild type or DBHϩ/Ϫ mice. Each of the four groups consisted of six mice. Data are expressed as means and S.D. The absolute value for the bone volume is different from those in Fig. 1 due to the mouse strain. b, mineralized nodule formation. Data are expressed as means and S.D. The number of mice in each group is indicated as N. *, statistically significant difference (p Ͻ 0.05). #, statistically significant difference between the vehicle group and drugtreated group (p Ͻ 0.05) (either control-loaded or unloaded groups). §, statistically significant difference between wild type unloaded group and drug-treated control-loaded group (p Ͻ 0.05). c, bone loss rate was calculated as follows: bone loss rate ϭ (control load Ϫ unload)/(control load) ϫ 100%. Bone loss rate in wild type mice was more than 2-fold the bone loss rate in DBHϩ/Ϫ mice. *, statistically significant difference (p Ͻ 0.05).

DISCUSSION
Our data reveal that sympathetic nervous tone is mediating unloading-induced bone loss via reduction in osteoblastic cell activity as well as enhancement in osteoclastic cell activity. This is the first report that sympathetic control of the bone mass is involved in the unloading-induced bone loss by control- FIG. 10. Pharmacological inhibition or activation of sympathetic tone suppresses or enhances unloading-induced increase in bone resorption, respectively. Mice were subjected to hind limb unloading as described under "Materials and Methods." To inhibit sympathetic tone, some of these mice were treated with propranolol or guanethidine. For activation of sympathetic tone, some of these mice were treated with isoproterenol. After unloading for 10 or 14 days, decalcified sections of the hind limb bones were prepared to determine osteoclast number and osteoclast surface. The number of mice in each group is indicated as N. a, histological sections showing the osteoclasts stained for TRAP (red). b-g, sympathetic tone inhibition by treatment with propranolol (b and c) or guanethidine (d and e) suppressed unloading-induced increase in osteoclast number (b and d) and osteoclast surface (c and e). Sympathetic tone activation by the treatment with isopreoterenol (f and g) enhanced unloading-induced increase in osteoclast number (f) and osteoclast surface (g). *, statistically significant difference (p Ͻ 0.05). Each group consisted of five bones from five mice. #, statistically significant difference between the vehicle group and drug-treated group (p Ͻ 0.05) (either control-loaded (column 1 versus column 3) or unloaded groups (column 2 versus column 4). §, statistically significant difference between vehicle-treated unloaded group (column 2) and drug-treated control-loaded group (column 3) (p Ͻ 0.05). ling osteoblasts. Unloading-induced bone loss was suppressed by the treatment of the animals with propranolol, a receptor antagonist, suggesting that ␤-adrenergic receptors in the sympathetic nervous system are the target of unloading-induced bone loss. The peripheral sympathetic nervous targets receive signals from the proximal upper nervous ending, which releases noradrenalin as a neurotransmitter into the synaptic gap. We observed that unloading-induced bone loss was again suppressed by the treatment of the animals with guanethidine. These data indicate that the depletion of noradrenalin in the proximal ending of the synaptic gap could suppress the unloading-induced osteoporosis. Furthermore, unloading failed to further suppress the bone volume that was reduced by a ␤-adrenergic agonist, isoproterenol. These three series of observations further indicate that sympathetic nervous tone is involved in unloading-induced pathological bone loss.
We also examined the effects of heterozygous deletion of the DBH gene. DBH is required for the sympathetic nervous tone. Therefore, we subjected the heterozygous knockout mice to hind limb unloading. Unloading-induced bone loss was attenuated by the absence of the half-dosage of dopamine ␤-hydroxylase gene, indicating that the presence of a full dosage of DBH gene in the animals is necessary for the complete effects of the unloading-induced bone loss. Furthermore, DBH data excluded the possibility that pharmacological experiments might be influenced by possible artifacts due to the systemic drug administration. Thus, both pharmacological and genetic interventions of sympathetic signals supported the idea that the sympathetic nervous system causes pathological loss of bone in unloading-induced osteopenia.
Since bone formation is the critical activity to determine the levels of bone loss due to unloading, it is the major target to elucidate the mechanisms required for unloading-induced loss of bone mass. Dynamic histomorphometric analyses on osteoblastic cell activity in vivo revealed that the reduction in bone formation activity in vivo due to unloading was suppressed by a series of pharmacological agents including propranolol and guanethidine. Furthermore, bone cell culture experiments us-ing the bone marrow cells taken from the animals subjected to either pharmacological or genetic interventions of the sympathetic nervous tone indicate that these interventions suppressed unloading-induced reduction in mineralized nodule formation. The interpretation of the reduction in the formation of osteoblastic bone nodules after 3 weeks in culture would be that progenitor cell populations for osteoblastic cell lineage could be reduced at the point of harvesting the cells from animals at the end of unloading. This suppression was blocked by the treatment with propranolol. We also carried out 3-week culture experiments to form bone nodules in the presence or absence of isoproterenol or propranolol and guanethidine in culture using bone marrow cells that were taken from wild type (untreated) C57Bl6 mice. There was no effect of these agents in culture to modulate the nodule formation in the bone marrow cells from untreated mice (data not shown). These data support the idea that the progenitor population in bone marrow in vivo would be reduced by the unloading condition or by treatment with drugs such as propranolol, guanethidine, and isoproterenol during the periods of the tail suspension experiments. Therefore, the action of sympathetic nervous tone during unloading-induced bone loss renders cell level effects on osteoblastic cell activity.
It is known that propranolol at doses of 10 g/g body weight could reduce blood pressure in mice. It is also known that such blood pressure change could be enhanced by the treatment with isoproterenol. However, there has yet been no clear evidence in clinical or experimental settings in terms of a direct relationship between blood pressure and bone mass. However, at this point, we cannot fully exclude the possibility that propranolol or isoproterenol used in our experiments may have affected bone mass via regulation of blood pressure, and these points have to be elucidated in the future.
It is intriguing to consider the differences in the time courses of the ␤-adrenergic receptor modulators. Propranolol has been known to reduce blood pressure in days, whereas isoproterenol and guanethidine alter blood pressure immediately. It is not known whether such time course differences seen in their effects on blood pressure may also affect their modulation of the bone mass. However, both propranolol and guanethidine blocked unloading-induced bone loss similarly in our experiments. This may partially be due to the nature of bone where the biological read out (i.e. bone mass) could be detected in a relatively slow manner compared with blood pressure.
If these types of drugs could be proven to be effective in the treatment of unloading-induced osteoporosis, it has to be considered that side effects might occur in the treatment of the patients. In the future, we may have to identify certain windows of the dosages by which bone effects may be obtained without affecting the blood pressure, or these drugs may be used only for those patients whose side effects could be predicted to be less based on the individual genomic data.
Our data indicated that sympathetic tones regulate unloading-induced enhancement in bone resorption. This was evidenced by the observations that inhibition of sympathetic tone by propranolol and guanethidine, suppressed unloading-induced bone resorption, and this leads to suppression in bone loss. Histomorphometric analyses indicated that unloading activates bone resorption through the increase in osteoclast number and osteoclast surface. These increases in the bone resorption parameters due to unloading were suppressed by treatment with propranolol and guanethidine. Such observations regarding the inhibitors for sympathetic tone effects on unloading-induced bone resorption were not limited to particular bone, since suppression of sympathetic tone by the treatment with guanethidine also suppressed unloading-induced FIG. 11. Pharmacological inhibition of sympathetic tone suppresses unloading-induced systemic bone loss assessed by the levels of urinary deoxypyridinoline excretion. Mice were subjected to hind limb unloading and guanethidine treatment as described under "Materials and Methods." After 14 days, urine samples of the mice were collected, and the amount of deoxypyridinoline in the urine was measured as described under "Materials and Methods." *, statistically significant difference (p Ͻ 0.05). The number of mice in each group is indicated as N. #, statistically significant difference between the vehicle group and drug-treated group (p Ͻ 0.05) (either control-loaded or unloaded groups). §, statistically significant difference between vehicle-treated unloaded group and drug-treated control-loaded group (p Ͻ 0.05).
increase in deoxypyridinoline excretion into urine, which is a systemic bone resorption maker. These data revealed that sympathetic tone regulates unloading-induced bone resorption as well.
Involvement of sympathetic tone in the induction of bone resorption after unloading was further supported by the analysis on the mice subjected to simultaneous unloading and isoproterenol treatment. Either isoproterenol treatment or unloading alone could cause an equivalent increase in the levels of bone loss as well as an increase in the levels of bone resorption parameters (osteoclast number and osteoclast surface). The simultaneous presence of unloading conditions and isoproterenol treatment resulted in an increase in bone loss as well as an increase in osteoclast number and osteoclast surface to levels equivalent to those in the cases of either one of the two conditions alone. These data further support the notion that sympathetic nervous tone and the unloading condition would share signaling pathways.
Unloading induces rapid bone loss and increases fracture risk significantly, especially in elderly bedridden patients. In fact, deoxypyridinoline excretion into urine was reported to increase in astronauts within a few hours after they are exposed to microgravity conditions in space. Such rapid bone loss is caused by an immediate response of osteoclastic activity when the body is subjected to unloading conditions. Although this phenomenon is so clearly observed in human (bedridden patients and astronauts) and various animal models, the mechanism for such a response of bone resorbing activity has not been identified. Since the nervous system could elicit signals at a relatively fast speed, our identification of the sympathetic tone as responsible for the unloading-induced increase in bone resorption and loss of bone mass would explain the rapid response of osteoclasts to unloading.
Although we have used DBH knockout mice to see the effects of such a deletion of the gene on the bone loss due to tail suspension, this enzyme is also required to produce epinephrine in the adrenals as well as norepinephrine. The enzyme required to produce epinephrine from norepinephrine, phenylethanolamine-N-methyltransferase, has been suggested in several reports to be subjected to induction by immobilizationinduced stress. As a result, the DBH heterozygous knock-out mice results may not represent conclusive proof of a prominent role for the sympathetic nervous system here. However, our data on the effects of unloading on bone in DBH knockout mice is at least in part in accordance with the idea that sympathetic tone is involved in the bone loss due to unloading.
In conclusion, our data indicate that sympathetic tone is in charge of the pathological reduction in bone mass upon unloading by suppressing osteoblastic cell actions and enhancing osteoclastic actions. These data predict that identification of the involvement of systemic modulation in the bone loss in unloading conditions could give a clue to appropriate measures to treat patients with disuse osteoporosis (9).