The authors have no conflict of interest.
Sympathetic Nervous System Does Not Mediate the Load-Induced Cortical New Bone Formation†
Article first published online: 8 AUG 2005
Copyright © 2005 ASBMR
Journal of Bone and Mineral Research
Volume 20, Issue 12, pages 2159–2168, December 2005
How to Cite
de Souza, R. L., Pitsillides, A. A., Lanyon, L. E., Skerry, T. M. and Chenu, C. (2005), Sympathetic Nervous System Does Not Mediate the Load-Induced Cortical New Bone Formation. J Bone Miner Res, 20: 2159–2168. doi: 10.1359/JBMR.050812
- Issue published online: 4 DEC 2009
- Article first published online: 8 AUG 2005
- Manuscript Accepted: 4 AUG 2005
- Manuscript Revised: 14 JUL 2005
- Manuscript Received: 5 APR 2005
- sympathetic nervous system;
- mechanical loading;
- bone formation
The contribution of the SNS to bone's response to mechanical loading is unclear. Using a noninvasive model of axial loading of the murine tibia, we found that sciatic neurectomy enhances load-induced new cortical bone formation and that pharmacological blockade of the SNS does not affect such responses, indicating that the SNS does not mediate the osteogenic effects of loading in cortical bone.
Introduction: There is increasing evidence that the sympathetic nervous system (SNS) contributes to the regulation of bone mass and may influence remodeling by modulating bones' response to mechanical load-bearing. The aim of this study was to examine the effect of sciatic neurectomy (SN) on the changes in cortical bone formation induced in response to mechanical loading and to investigate whether the SNS is directly involved in such load-induced responses.
Materials and Methods: Accordingly, load-induced responses were compared in tibias of growing and adult control C57Bl/J6 mice and in mice submitted to unilateral SN; noninvasive axial loading that induced 2000 μstrain on the tibia lateral midshaft cortex was applied cyclically, 5 or 100 days after surgery, for 7 minutes, 3 days/week for 2 weeks, and mice received calcein on the third and last days of loading. Tibias were processed for histomorphometry, and transverse confocal images from diaphyseal sites were analyzed to quantify new cortical bone formation. Chemical SNS inactivation was achieved by prolonged daily treatment with guanethidine sulfate (GS) or by the introduction of propranolol in drinking water.
Results: Our results show that new cortical bone formation is enhanced by loading in all tibial sites examined and that load-induced periosteal and endosteal new bone formation was greater in the SN groups compared with sham-operated controls. This SN-related enhancement in load-induced cortical bone formation in tibias was more pronounced 100 days after neurectomy than after 5 days, suggesting that longer periods of immobilization promote a greater sensitivity to loading. In contrast, the increases in new bone formation induced in response to mechanical loading were similar in mice treated with either GS or propranolol compared with controls, indicating that inactivation of the SNS has no effect on load-induced cortical new bone formation.
Conclusions: This study shows that SN, or the absence of loading function it entails, enhances loading-related new cortical bone formation in the tibia independently of the SNS.
IT IS WELL documented that both sympathetic and sensory nerve fibers are present in periosteum and bone(1–4) and form dense parallel networks around blood vessels adjacent to bone trabeculae, in close contact to bone cells.(5–7) This distribution highlights a possibility that sensory and sympathetic innervation contribute to regulating bone cell activity and, thereby, the balance of (re)modeling. Indeed, evidence for this is accumulating.(8–13) It is known that both surgical and chemical sympathectomy can modulate bone cell function; however, both anabolic and catabolic effects of the sympathetic nervous system (SNS) have been observed,(8–12) and therefore, its role in regulating bone remodeling remains controversial.
It has been shown, nevertheless, that dopamine-β-hydroxylase deficient-mice that are unable to synthesize catecholamines exhibit a high bone mass phenotype, whereas mice in which the dopamine transporter has been deleted have low bone mass.(14,15) The presence of axons containing catecholamines in the vicinity of osteoblasts has been established in vivo.(7,14) Furthermore, both osteoblasts and osteoclasts also express adrenergic receptors,(14,16) and in vitro and in vivo studies have shown that catecholamine affects their biological functions.(4,17–19) It has also been shown that treatment of mice with the β-adrenergic receptor agonist isoproterenol decreases bone mass, whereas mice treated with the β-blocker propranolol have increased bone mass.(14) These animal studies prompted more recent clinical studies in which the effects of β-adrenergic blockers on BMD and fracture risk were assessed. These studies have only accentuated the contentious role of the SNS in regulating bone remodeling, showing both that β-blockers can reduce the risk of fractures by increasing BMD and that β-blocker use and BMD are unrelated.(20–23)
The profound influence of the nervous system on bone formation was recently shown in leptin-deficient (ob/ob) mice.(24)Ob/ob mice exhibit increased trabecular bone volume, leading to high bone mass. These studies have shown that the effects of leptin on bone mass are mediated through binding to its hypothalamic receptor and that this is sufficient to induce bone loss by SNS-mediated decrease in osteoblastic function.(14,24) This proposed central control of bone mass is strengthened by the observation that neuropeptide Y2 receptor-deficient mice have a 2-fold elevation in their trabecular bone volume compared with control mice and by the finding that intracerebroventricular administration of neuropeptide Y (NPY) also causes trabecular bone loss.(25) Very recently, it was established that leptin also controls bone resorption through the SNS.(26) These findings suggest that the SNS-mediated control of both aspects of bone remodeling is a major influence regulating bone mass and architecture. The increased bone formation in ob/ob mice and NPY2 receptor-deficient mice was, however, only observed in trabecular bone and not studied in cortical bone in those studies. Previous work has shown that mice lacking a functional leptin receptor and leptin-resistant rats exhibit both lower cortical bone thickness and area compared with the controls,(27,28) indicating that in contrast to trabecular bone, the absence of leptin signaling inhibits cortical bone formation. Those conflicts regarding the effects of leptin on bone mass have been resolved in a recent study showing regional differences in the regulation of bone mass in leptin deficiency.(29) Similarly, although propranolol treatment increases trabecular bone mass, its effect on cortical bone formation has not been studied. Moreover, bone mass recovery induced by propranolol in tail-suspended rats is not observed in cortical bone,(30) suggesting that the SNS may have little influence on cortical bone.
Another key regulatory influence on normal bone remodeling is the skeleton's adaptation to the mechanical load-bearing environment.(31–33) It has been hypothesized that the nervous system influences bone remodeling through a regulation of bones' response to mechanical load-induced strain. The classical view, however, is that bone adaptation to mechanical loading is not centrally controlled; Hert et al.(34) showed that innervated and denervated limbs react identically to controlled loading. Our group and others have provided additional support for a lack of central control by showing that bone explants and isolated bone cells are sensitive to mechanical stimuli.(35,36) This suggests that if the nervous system is involved, it is as a modulator rather than a primary influence. Consistent with this modulatory influence are the findings that the β-adrenergic pathway of the SNS is a mediator of the trabecular bone loss induced by the absence of mechanical loading in tail-suspended rat and mouse models.(30,37) The SNS may therefore have both systemic osteotrophic effects and local effects because of changes in mechanical loading. Systemic effects of the SNS are supported by studies showing that propranolol increases BMD over the whole body,(14) by the demonstration that sciatic neurectomy (SN) induces an expected bone loss in disused neurectomized limbs, but also in the nondisused contralateral limbs,(38) and that stroke patients who are immobilized retain the difference in BMD between paretic and nonparetic arms.(39)
The aim of this study was to examine the effect of SN on the changes in cortical bone formation induced in response to mechanical loading and to investigate whether the SNS is directly involved in such load-induced responses. We have previously shown that, in addition to causing limb immobilization, SN is associated with a decreased density of nerve fibers in the tibias and can therefore be used as a model of limb denervation.(40) We used a novel, noninvasive model of axial loading of the tibia, which facilitates study of the osteogenic effects of loading in murine cortical bone,(41) and found that SN enhances load-related new cortical bone formation in the tibia and that pharmacological blockade of the SNS does not affect such responses. This indicates that the absence of loading function promotes osteogenic responses to applied artificial loads and that the SNS does not mediate such load-induced increases in new cortical bone formation.
MATERIALS AND METHODS
All experimental procedures used female C57BL/J6 animals 10 or 20 weeks of age purchased from Charles River Company. Mice were allocated a 1-week rest period after transport before the start of experiments. Animals were housed in groups of five in polypropylene cages and subjected to a 12-h light/dark cycle, with room temperature maintained at 21°C, and they were fed ad libitum with a maintenance mouse diet (Special Diet Services, Witham, UK). All procedures complied with the Animals (Scientific Procedures) Act 1986 and were reviewed and passed by the ethics committee of the Royal Veterinary College.
In the first experiment, designed to study the effect of short-term SN on the bone response to mechanical loading in growing mice, 40 female C57Bl/J6 mice 10 weeks of age were randomly divided into two groups: one sham-operated group on the right leg (n = 20) and one SN group (SN) on the right leg (n = 20). Briefly, mice were anesthetized with oxygen and halothane (Rhone Merieux, Essex, UK) and premedicated with buprenorphine (0.1 mg/kg) by subcutaneous injection (Vetergesic; Animalcare, York, UK), and SN was accomplished by resecting a 3- to 4-mm segment of the sciatic nerve posterior to the hip joint. The neurectomized animals were able to move around in the cage and gained access to food and water without difficulties. On the fifth day after surgery, 10 mice per group were loaded for 2 weeks, whereas the other 10 mice/group were anesthetized to act as controls. Mice were killed 18 days after surgery by intraperitoneal injection of 0.5 ml sodium pentobarbitone (Euthesate; Willows Francis Veterinary, Crawley, UK). Right tibias were dissected and processed for bone histomorphometry analysis.
To compare the effects of short-term and long-term SN on load-induced bone response in adult mice, we performed two further experiments. In the first additional experiment, female C57Bl/J6 mice were divided into three groups (n = 7/group). Two groups were neurectomized (SN) at 7 weeks of age, whereas the control group was sham-operated at the same age. Thirteen weeks after surgery (at 20 weeks), one SN group and the sham group were loaded for 2 weeks, whereas the second SN group was anesthetized without loading for control. Mice were killed at 22 weeks. In the second additional experiment, two groups of female C57Bl/J6 mice 20 weeks of age (n = 7/group) were neurectomized, whereas the control group was sham-operated at the same age. Five days after surgery, one SN group and the sham group were loaded for 2 weeks, whereas the remaining SN group was anesthetized alone for control. Mice were killed at 22 weeks, and right and left (used as contralateral controls) tibias were dissected and processed for bone histomorphometry.
Chemical sympathectomy with guanethidine sulfate
To establish the contribution made by the SNS to the bone response to mechanical loading, 40 female C57Bl/J6 mice 10 weeks of age were randomly divided into two groups. One group received a daily intraperitoneal injection of guanethidine sulfate (GS; 40 mg/kg/day; Sigma) for 3 weeks; this treatment chemically inactivates the SNS. Mice treated with GS rapidly showed the presence of ptosis, a clinical sign that affects the eyelids and that is an indicator of successful sympathetic denervation. The mice did not exhibit any other physical signs. The sham group received a daily intraperitoneal injection of 0.9% saline (5 ml/kg/day). Two weeks after the start of injections, one-half of the mice in each group were loaded for 2 weeks, whereas the other one-half were just anesthetized. Mice were killed 4 weeks after the start of the experimental protocol by intraperitoneal injection of 0.5 ml sodium pentobarbitone (Euthesate). Right tibias were dissected and processed for bone histomorphometry analysis.
Chemical sympathectomy with propranolol treatment
To further examine the involvement of the β-adrenergic pathway in the response to loading, 20 female C57Bl/J6 mice 10 weeks of age were randomly divided into two groups: a control group and a treated group. The treated group received propranolol (Sigma), an antagonist of the β-adrenergic receptors, diluted in their drinking water at a concentration of 0.5 g/liter for a period of 5 weeks. All mice were fed a normal diet and were allowed to drink water in unlimited amounts for the 5 weeks of the treatment. Four weeks after the start of treatment, control and treated mice were loaded for 2 weeks. Mice were killed 6 weeks after the start of treatment by intraperitoneal injection of 0.5 ml sodium pentobarbitone (Euthesate). All tibias were dissected (including the left contralateral controls) and processed for bone histomorphometry analysis. The efficacy of β-adrenergic receptor antagonism in the propranolol-treated mice was checked by measuring heart rate.(42) Bipolar electrodes were placed on the right and left upper limbs of the anesthetized mouse, and electrical activity of the heart was monitored using the Powerlab system (ADInstruments).
In vivo mechanical loading
The loading apparatus that we used was specifically adapted for the tibia, as previously described.(41) The cups are aligned vertically and positioned within a servo-hydraulic materials testing machine (Model HC10; Dartec, Stourbridge, UK). The upper cup into which the knee is positioned is attached to the actuator, and the lower cup is located on the “load cell.” The actuator was operated in load control to apply dynamic compression to the tibia. The loading regimen was calibrated ex vivo, as previously reported.(41) The right tibia of each mouse was held in the padded cups, and dynamic (2 Hz) axial loads (peak load, 12N; maximum loading rates, 28–44 N/s, 40 cycles/day) were applied through the knee joint for a period of 7 minutes on 3 alternate days per week for 2 weeks to engender peak strain of 2000 με on the lateral midshaft cortex. Because mice submitted to long-term disuse are known to exhibit decrement bone mass, we used an ex vivo calibration procedure to determine the loads required to generate similar peak strains of 2000 με on the lateral midshaft cortex of neurectomized tibias. We found that 10N were sufficient to engender 2000 με in neurectomized tibias. As a consequence, peak loads of 10N (instead of 12N for the controls) were applied to long-term neurectomized mice to ensure that responses to similar applied peak strains magnitudes were examined. For loading, mice were anesthetized by intraperitoneal injection with a mixture of hypnorm (Janssen Animal Health, Bucks, UK), midazolam (Roche, Derbyshire, UK), and sterile water for injection. Total duration of anesthesia lasted between 60 and 90 minutes. None of the mice showed signs of lameness or decreased activity levels after recovery from anesthesia. To measure new cortical bone formation, calcein (7.5 mg/kg; Sigma Chemical) was given on the third and last days of loading.
Analysis of cortical new bone formation
Tibias were dissected, fixed in buffered formal saline (BFS) for 48 h, washed, and dehydrated through graded alcohol concentrations. Bones were infiltrated and embedded using a flat polymethylmethacrylate (PMMA; British Drug House) bed technique.(43) After polymerization at 42°C, the PMMA blocks were trimmed and cut with an annular diamond saw into 500-μm-thick, serial planar parallel segments along the shaft length. Calcein labels were observed using a laser scanning confocal microscope (Carl Zeiss, Herts, UK). Transverse confocal images from three diaphyseal sites were analyzed, as shown in Fig. 1A. Measurements were made from the outside edges of periosteal and inside edges of endosteal labels (Fig. 1B). Periosteal and endosteal interlabel areas (μm2) represent new cortical bone formation.
Statistical analyses were performed using Instat software (Graphpad, San Diego, CA, USA). The significance of differences in periosteal and endosteal interlabel areas between control and treated groups were determined based on nonparametric Wilcoxon's or Mann-Whitney U-tests.
New cortical bone formation is enhanced at all tibial sites after loading
Changes in cortical bone formation induced by 12N of applied load, which engenders 2000 με on the lateral midshaft cortex of the tibia, were measured at three diaphyseal sites (site 5, 1.5 mm proximal to the midshaft; site 8, at the midshaft; site 12, 1.5 mm distal to the midshaft; Fig. 1A). We found that there was an increased bone formation in loaded compared with nonloaded tibias at each of these three sites (Fig. 1C). Periosteal new bone formation was significantly increased at all three sites in loaded compared with control tibias. Endosteal new bone formation was also significantly increased after loading, but only at, and distally to, the midshaft (Fig. 1C). Whereas the increase in periosteal new bone formation was more marked at the medial-posterior margin, endosteal bone formation was often more pronounced at the lateral-anterior surface (Fig. 1B). Confocal images from the tibias did not show any evidence of intracortical remodeling indicative of microdamage-induced osteoclast activation.
Sciatic denervation increases load-induced cortical new bone formation
Decreases in both periosteal and more markedly in endosteal bone formation were evident in response to 5-day SN in growing (10 week old) mice (Fig. 2). However, significant neurectomy-induced decreases in periosteal bone formation were only observed at the distal tibial site (Fig. 2E, site 12); the effects of neurectomy were modest in the proximal and middiaphyseal periosteal sites (Figs. 2A and 2C). Five days of neurectomy also induced decreases in total cortical bone areas in adult mice (20 week old; Fig. 3). The neurectomy-induced decreases in total cortical bone area were observed at all tibial sites in adult mice (Figs. 3A-3F) and were similar after short-term (5 days) and long-term (100 days) neurectomy, suggesting that neurectomy-related loss of bone occurred rapidly and was stable thereafter (Fig. 3). Because the basal levels of new bone formation are insignificant in these 20-week-old nonloaded adult mice, changes in bone formation are therefore expressed as the percentage of the total cortical bone area measured in contralateral control tibias.
The application of loading significantly increased new cortical bone formation at all tibial sites in both sham and neurectomized groups of mice. This was shown in growing mice loaded 5 days after neurectomy (Fig. 2) and in adult mice when loaded after either 5 or 100 days after neurectomy (Fig. 3). These data confirm that tibias are still able to respond to external mechanical loading even after long-term neurectomy.
Indeed, our data show that load-induced new bone formation was higher in the SN groups compared with sham-operated controls in growing and adult mice. In growing mice (Figs. 4A-4C), load-induced periosteal and endosteal new bone formation was higher in the SN groups compared with the sham, but only at certain sites. The increased load-induced new bone formation in SN groups only reached statistical significance at site 12 (distal) for periosteal new bone formation (Fig. 4C) and at site 5 (proximal) for endosteal new bone formation (Fig. 4A). Similarly, neurectomized tibias from adult mice showed significant enhancement in load-induced increases in total cortical new bone formation (Figs. 4D-4F). Short-term neurectomy (5 days) in adult mice provoked significant increased levels of load-induced cortical new bone formation at the proximal and mid-diaphyseal parts of the tibias (Figs. 4D and 4E), whereas it had no effect at the distal site (Fig. 4F). Long-term neurectomy (100 days) had more pronounced effects at proximal and diaphyseal sites (Figs. 4D and 4E), although the distal site was significantly affected. New bone formation in 20-week-old mice is expressed in absolute values because basal levels of bone formation in those mice are negligible. Overall, the neurectomy-related enhancement in load-induced total cortical new bone formation was more pronounced in tibias loaded 100 days after neurectomy than in those loaded after only 5 days. This suggests a greater sensitivity of bone to mechanical loading when immobilized for a longer duration.
GS treatment has no effect on load-induced cortical new bone formation
The sciatic nerve is a mixed nerve containing motor, sympathetic, and sensory nerve fibers.(44) To establish whether the effect of SN on load-induced cortical bone formation could be attributed to a decreased sympathetic innervation in the tibias, we treated 10-week-old mice for 3 weeks with GS, which chemically inactivates the SNS. Two weeks after the start of injections, control and GS-treated mice were loaded in vivo, using the same protocol as previously described. We showed that GS alone did not affect basal cortical new bone formation (Fig. 5). Furthermore, we found that control and GS-treated mice exhibited similar increases in periosteal and endosteal new bone formation in response to mechanical loading (Fig. 5), indicating that GS has no effect on load-induced cortical new bone formation.
Propranolol treatment has no effect on load-induced cortical new bone formation
To further confirm that the SNS does not influence the load-induced cortical new bone formation, another group of 10-week-old mice was also treated for 5 weeks with propranolol, a β-adrenergic blocker. Osteoblasts express β-adrenergic receptors,(14,16) and this pathway has been shown to be involved in unloading-induced trabecular bone loss.(30) Four weeks after the beginning of propranolol treatment, mice were loaded for 2 weeks. The efficacy of propranolol treatment was shown by a decrease in heart rate (data not shown). Control and propranolol-treated groups, however, exhibited similar basal and load-induced increases in periosteal and endosteal new bone formation at all tibial sites (Figs. 6A-6F). This shows that propranolol has no effect on basal or load-induced cortical new bone formation. Figure 7 shows total (endosteal and periosteal) cortical new bone formation (normalized against the levels of new bone formation in controls), and clearly shows that pharmacological inactivation of the SNS either by GS or propranolol does not modify load-induced cortical bone formation. This indicates that the SNS is not involved in the response of cortical bone to applied mechanical loading.
This study provides evidence that SN, or the absence of loading function it entails, enhances loading-related new bone formation in the tibia and that this cannot be attributed to SNS blockade. Using a recently validated model for applying compressive axial loads to the mouse tibia, in which resultant changes in cortical bone can be measured,(41) we showed induction of new periosteal bone formation in response to mechanical loading at all sites of the tibia, whereas endosteal bone formation was only increased distally to the midshaft. This noninvasive in vivo tibial loading induces a significant osteogenic response both in growing 10-week-old mice and in adult 20-week-old mice, consistent with our previous results and with those from another noninvasive murine tibia loading model that places the tibia in cantilever bending.(41,45)
Surgical SN is widely used to examine the bone loss induced by decreased mechanical loading.(46) As expected, short-term neurectomy in growing female mice induced a decreased cortical new bone formation, which was more pronounced distally from the midshaft, close to the tibiofibular junction. These results are in agreement with previous studies indicating that SN inhibits age-related bone growth in cortical bone by decreasing bone formation on the periosteal surface, while decreasing bone formation and increasing bone resorption on the endocortical surfaces.(47,48) We observed similar changes in the cortical bone of 10- and 20-week-old mice after short-term neurectomy, suggesting that these effects of SN on cortical bone mass do not depend on the age of mice. Our results also show similar decreases in cortical bone area after long-term neurectomy, confirming that there is a slow progression or a stabilization of the bone loss with time after SN.(38,48)
Our findings also show that both control and neurectomized limbs exhibit increased cortical bone formation in response to mechanical loading. This suggests that bones in innervated and denervated limbs react similarly to loading, as previously suggested by Hert et al.(34) However, we showed for the first time that SN enhances loading-related new cortical bone formation in the tibia, both in growing and adult mice. This was only observed at certain sites. This may be explained by bones responding to loading as whole structures, and our analyses only sample this structural response by measurement of new bone formation at selected sites. Mice that had been loaded for 2 weeks after 5 days of neurectomy had similar levels of total new bone to those seen in control mice after loading, suggesting an active adaptive osteogenic response that acts to rescue the neurectomy-induced bone loss. Our findings are consistent with an increased sensitivity of cortical bone to load after a period of unloading. Previous studies have shown recovery of immobilization-induced bone loss in growing and adult rats after remobilization, depending on the duration of the immobilization period.(49,50) However, the concept that bone is more sensitive to mechanical loading during immobilization is more novel and has not been extensively studied. Previous work using the avian ulna model has nevertheless suggested that the stimulus for an osteogenic response originates from the altered distribution of strain and a lack of averaging with the strains of normal activity.(51) This idea of strain averaging is consistent with the suggestion(52) that the predominant mechanically derived stimulus to the skeleton is derived from short periods of a particularly osteogenic activity rather than many repetitions of the strain situation experienced during the predominant activity.
We recently also used SN as a model of denervation in the rat tibia and observed a marked decrease in innervation in the tibial metaphysis of neurectomized limbs.(40) However, complete denervation of the long bones was not achieved after sciatic neurectomy, because, in addition to sciatic nerve, other nerves and their branches supply nerve fibers to the hind limbs.(44) The sciatic nerve is a mixed nerve, which contains sympathetic and sensory nerve fibers that innervate bone.(7) Recent studies have provided evidence that the SNS acts to decrease bone mass by regulating osteoblast activity through β-adrenergic receptors present in these cells.(14,16) Therefore, in addition to inducing immobilization, it is possible that SN increases the amount of new bone formation stimulated by mechanical loading by decreasing sympathetic innervation in the tibia. To study this possibility, we used two different techniques to chemically inactivate the SNS. GS treatment has been widely used to induce functional sympathectomy and acts by depleting neuronal noradrenaline stores.(53) Propranolol, in contrast, inactivates the SNS by blocking β-adrenergic receptors.(14) We did not find any effect of GS or propranolol on either basal or load-induced new cortical bone formation. Previous studies conducted to examine the effects of GS and propranolol treatments on basal bone formation have provided conflicting results. Hill et al.(9) have shown that GS treatment had no effect on cortical area, medullary area, or periosteal apposition rate in neonatal rats. Kondo et al.(37) reported that GS and propranolol alone did not affect cancellous bone volume, whereas Takeda et al.(14) showed that mice treated with propranolol have an increase in trabecular bone volume. These two most recent studies did not, however, investigate the effect of sympathectomy in cortical bone, and it is possible that cortical and trabecular bone compartments have differences in sympathetic innervation and/or disparate responses to sympathectomy. Indeed, the site-specific influence of the SNS on the skeleton is supported by recent work indicating that leptin deficiency produces contrasting phenotypes in bones of the limb and spine,(29) and that sympathectomy stimulates bone remodeling in membranous bones, whereas it has no effect on endochondral long bones.(10) The ability of the SNS to regulate trabecular bone mass while having no effect on cortical bone may be explained by the different turnover, the targeting of different subsets of osteoblasts, and/or by distinctive SNS interactions with hormones and mechanical loading signaling pathways in these two bone compartments.(40,54) It is possible that, whereas cortical bone is mainly controlled locally by mechanical loading, trabecular bone is mostly regulated by hormonal and neuronal pathways.
Our results do not support the hypothesis that nerve fibers, which are very abundant in the periosteum and trabecular surfaces,(7) could act as mechanoreceptors and transmit mechanical loading in bone.(55,56) Indeed, they indicate that the SNS has no regulatory influence on the response of cortical bone to mechanical loading. The discovery that control of osteoblast function is influenced by the CNS through the SNS has, however, led to the hypothesis that this SNS component of the nervous system controls the response of bone to mechanical loading. Indeed, recent work using the tail-suspended rat model indicates that the β-adrenergic pathway of the SNS acts as a mediator of the trabecular bone loss in this model.(30,37) Our results show no effects of GS and propranolol on the cortical bone response to loading, suggesting that the mechanisms involved in bones' response to neurectomy-induced unloading are different from those participating in the unloading induced by tail suspension. Tail suspension is a model that creates hypokinesia and hypodynamia at the hind limbs, which generates alterations in the autonomic nervous system and in neuromuscular function, similar to that seen after space flight.(57) It induces physiological stress in animals, nonphysiological blood distribution, reductions in plasma volume, and perturbations in the arterial vascular tone, events that may all contribute to changes in the SNS activity and in the bone responses to propranolol.(58,59) Alternatively, as mentioned earlier, the lack of effect of chemical sympathectomy on cortical bone response to externally applied mechanical loading may be the result of differences in the density of sympathetic innervation or changes in the expression of adrenergic receptors in trabecular and cortical bone,(60) because no significant effects of propranolol were reported in cortical bone in response to tail suspension.(30) We cannot, however, rule out a major contribution from sensory innervation to these mechanical load-induced responses.
Our findings show that sciatic denervation increases load-induced cortical new bone formation independently of the SNS. Because we could not ascribe any of the effects observed after neurectomy to sympathetic innervation, our data suggest that the lack of functional loading resulting from reduced muscle activity is responsible for this effect.
The authors thank Shabaz Hamid for performing heart rate measurements in mice. This work was funded by the Arthritis Research Campaign and The Wellcome Trust. We also thank CAPES, Ministry of Education, Brazil, for support.
- 11991 Distribution of CGRP-, VIP-, DßH-, SP-, and NPY-immunoreactive nerves in the periosteum of the rat. Cell Tissue Res 264: 469–480.,
- 21986 Innervation of periosteum and bone by sympathetic vasoactive intestinal peptide-containing nerve fibers. Science 232: 869–887., , , ,
- 31996 Neuropeptides and the puzzle of bone remodeling. Acta Orthop Scand 67: 632–639., ,
- 42002 Kinins and neuro-osteogenic factors. In: BilezikianJP, RaiszLG, RodanGA (eds.) Principles of Bone Biology, 2nd ed. Academic Press, San Diego, CA, USA, pp. 773–799.,
- 51998 Light- and electron-microscopic study of the distribution of axons containing substance P and the localization of neurokinin-1 receptor in bone. Cell Tissue Res 293: 87–93., , ,
- 61996 Immunohistochemical and ultrastructural localization of CGRP-positive nerve fibers at the epiphyseal trabecules facing the growth plate of rat femurs. Bone 18: 29–39., ,
- 71999 Evidence for a dense and intimate innervation of the bone tissue, including glutamate-containing fibers. Bone 25: 623–629., , ,
- 81999 Chemical sympathectomy impairs bone resorption in rats: A role for the sympathetic system on bone metabolism. Bone 25: 545–551., , ,
- 91991 Effects of neonatal sympathectomy and capsaicin treatment on bone remodeling in rats. Neuroscience 44: 747–755., ,
- 102000 Sympathectomy, which induces membranous bone remodeling, has no effect on endochondral long bone remodelling in vivo. J Bone Miner Res 15: 1354–1360.,
- 111990 The early effects of surgical sympathectomy on bone resorption in the rat incisor socket. Arch Oral Biol 35: 1003–1007., , ,
- 122003 Sympathectomy causes increased root resorption after orthodontic tooth movement in rats: Immunohistochemical study. Cell Tissue Res 313: 167–175., , ,
- 132005 Capsaicin-sensitive sensory neurons contribute to the maintenance of trabecular bone integrity. J Bone Miner Res 20: 257–267., , , , , , , , ,
- 142002 Leptin regulates bone formation via the sympathetic nervous system. Cell 111: 305–317., , , , , , , ,
- 152000 Bone histomorphometric and biomechanical abnormalities in mice homozygous for deletion of the dopamine transporter gene. Bone 26: 15–19., , , , ,
- 162002 Adrenergic regulation of bone metabolism: Possible involvement of sympathetic innervation of osteoblastic and osteoclastic cells. Microsc Res Tech 58: 77–84.
- 172003 In vivo stimulation of sympathetic nervous system modulates osteoblastic activity in mouse calvaria. Am J Physiol Endocrinol Metab 285: E661–E667.,
- 181998 Cathecolamines stimulate the proliferation and alkaline phosphatase activity of MC3T3-E1 osteoblast-like cells. Bone 23: 197–203., , ,
- 192000 Adrenergic stimulation of osteoclastogenesis mediated by expression of osteoclast differentiation factor in MC3T3-E1 osteoblast-like cells. Biochem Pharmacol 61: 579–586., , ,
- 202004 β-adrenergic blockers reduce the risk of fractures partly by increasing bone mineral density: Geelong osteoporosis study. J Bone Miner Res 19: 19–24., , , , ,
- 212004 Use of β-blockers and risk of fractures. JAMA 292: 1326–1332., , ,
- 222004 Beta-blocker use, bone mineral density and fracture risk in older women: Results from the EPIDOS prospective study. J Bone Miner Res 19: S455., , , ,
- 232005 Beta-blocker use, BMD and fractures in the study of osteoporotic fractures. J Bone Miner Res 20: 613–618., , , , , ,
- 242000 Leptin inhibits bone formation through a hypothalamic relay: A central control of bone mass. Cell 100: 197–207., , , , , , , , ,
- 252002 Hypothalamic Y2 receptors regulate bone formation. J Clin Invest 109: 915–921., , , , , ,
- 262005 Leptin regulation of bone resorption by the sympathetic nervous system and CART. Nature 434: 514–520., , , , , , , , , , , ,
- 271986 Osteopenia in mice with genetic diabetes. Diabetes Res Clin Pract 2: 157–163., ,
- 282002 Bone mass in obese diabetic Zucker rats: Influence of treadmill running. Calcif Tissue Int 70: 305–311., , , , , , , , ,
- 292004 Leptin deficiency produces contrasting phenotypes in bones of the limb and spine. Bone 34: 376–383., , , ,
- 302003 Sympathetic nervous system as a transmitter of mechanical loading in bone. Joint Bone Spine 70: 515–519., , , , ,
- 311988 Vital biomechanics: Proposed general concepts for skeletal adaptations to mechanical usage. Calcif Tissue Int 42: 145–156.
- 321992 Control of bone architecture by functional load bearing. J Bone Miner Res 7: S369–S375.
- 331996 Physical loading, exercise and bone. Bone 18: S1–S3., ,
- 341971 Effect of intermittent stress on the rabbit tibia after resection of the peripheral nerves. Folia Morphol (Prague) XIX: 378–387., ,
- 351997 Enhancement by sex hormones of the osteoregulatory effects of mechanical loading and prostaglandins in explants of rat ulnae. J Bone Miner Res 12: 1424–1430., , , , ,
- 361995 Mechanical strain-induced NO production by bone cells: A possible role in adaptive bone remodeling? FASEB J 9: 1609–1614., , , , ,
- 372003 Unloading-induced bone loss occurs through the central control via sympathetic nervous system. J Bone Miner Res 18: S45., , , , , , ,
- 382003 A substance P receptor (NK1) antagonist enhances the widespread osteoporotic effects of sciatic nerve section. Bone 33: 927–936., , , , ,
- 391999 Progressive hemiosteoporosis on the paretic side and increased bone mineral density in the nonparetic arm the first year after severe stroke. Osteoporos Int 9: 269–275., , , ,
- 402005 Dramatic decrease of bone innervation density after ovariectomy. Endocrinology 146: 503–510., , , , , , ,
- 412005 Non-invasive axial loading of the mouse tibia discloses load-induced increases in cortical bone formation and modification of trabecular bone organization. Bone (in press), , , , ,
- 421996 Response to cardiac sympathetic activation in transgenic mice overexpressing beta 2-adrenergic receptor. Am J Physiol 27: H630–H636., , , , ,
- 431983 Processing of uncalcified bone specimen for bone histomorphometry. In: ReckerRR (ed.) Bone Histomorphometry: Techniques and Interpretation. CRC Press, Boco Raton, FL, USA, pp. 13–35., , ,
- 441995 Effect of sciatic nerve section on neural ingrowth into the rat tibial fracture callus. Clin Orthop Relat Res 311: 247–257., , , , , , , ,
- 452002 Noninvasive loading of the murine tibia: An in vivo model for the study of mechanotransduction. J Bone Miner Res 17: 493–501., , , ,
- 461991 Depression of osteoblastic activity in immobilized limbs of suckling rats. J Bone Miner Res 6: 725–731., ,
- 471999 Cortical tibial bone volume in two strains of mice: Effects of sciatic neurectomy and genetic regulation of bone response to mechanical loading. Bone 25: 183–190., , , , , , , , ,
- 481996 Time responses of cancellous and cortical bones to sciatic neurectomy in growing female rats. Bone 19: 13–21., , , , , , ,
- 491991 Changes induced in growing rat bone by immobilization and remobilisation. Bone 12: 113–118., , , , ,
- 501996 Effects of immobilization, three forms of remobilisation, and subsequent deconditioning on bone mineral content and density in rat femora. J Bone Miner Res 11: 1339–1346., , , , , , , ,
- 511984 Regulation of bone formation by applied dynamic loads. J Bone Joint Surg Am 66: 397–402.,
- 521997 Strain magnitude related changes in whole bone architecture in growing rats. Bone 20: 191–198., , ,
- 532001 Novel method for localized, functional sympathetic nervous system denervation of peripheral tissue using guanethidine. J Neurosci Methods 112: 21–28.,
- 542004 Mechanical loading and bone formation. Available online at http://www.bonekey_ibms.org/cgi/content/full/ibmske.,
- 552001 Effect of loading on the development of nerve fibers around oral implants in the dog mandible. Clin Oral Impl Res 12: 219–224., , , , , , ,
- 561990 Periodontal mechanoreceptors and their functions. In: TaylorA (ed.) Neurophysiology of the Jaws and Teeth. Macmillan Press, London, UK, pp. 52–95.
- 572002 Cardiovascular and sympathetic neural responses to handgrip and col pressor stimuli in humans before, during and after space flight. J Physiol 544: 653–664., , , , , , , , , , , , , , , , , , , , , , ,
- 582003 Enhanced sympathoinhibitory response to volume expansion in conscious hindlimb-unloaded rats. J Appl Physiol 94: 1806–1812.,
- 591981 Plasma cathecolamines in rats: Daily variations in basal levels and incrementsin response to stress. Physiol Behav 26: 27–31., ,
- 602002 Origins of skeletal pain: Sensory and sympathetic innervation of the mouse femur. Neuroscience 113: 155–166., , , , , , , , , ,