The neuro-osteogenic network: The sympathetic regulation of bone resorption

Japanese Dental Science Review (2012) 48, 61—70

Available online at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/jdsr

Review Article

The neuro-osteogenic network: The sympathetic regulation of bone resorption Akifumi Togari *, Michitsugu Arai, Hisataka Kondo, Daisuke Kodama, Yuka Niwa Department of Pharmacology, School of Dentistry, Aichi-Gakuin University, 1-100 Kusumoto-cho, Chikusa-ku, Nagoya 464-8650, Japan Received 4 November 2011; received in revised form 1 December 2011; accepted 3 December 2011

KEYWORDS Osteoblast; Osteoclast; Neurite; Co-culture; Sympathetic neuron; Sensory neuron

Summary Bone is innervated by sympathetic and sensory neurons, which play important roles in bone remodeling. Direct neuro-osteogenic cross-talk has been demonstrated using an in vitro co-culture model comprising osteoblastic or osteoclastic cells, and neurite-spouting mouse superior cervical ganglia, suggesting that these cells are directly regulated by sympathetic neurons. The increase in sympathetic nervous activity causes bone loss through increased bone resorption and decreased bone formation, associated with b2-adrenergic activity toward both osteoblastic and osteoclastic cells. The increased bone resorption is based on the stimulation of both osteoclastogenesis and osteoclastic activity. These studies suggest b-blockers to be effective against osteoporosis, in which case there is increased sympathetic activity. Epidemiological studies have demonstrated high blood pressure to be associated with increased bone loss and bblockers to be potential candidates for drugs to treat osteoporosis and fractures. In animal experiments, a lower dose of b-blocker also improved bone mass and biomechanical fragility in a hypertensive model. Although the interplay between sympathetic and sensory neurons is poorly understood, interestingly, sensory denervated rats have been demonstrated to show similar bone metabolism to rats with hyperactivity of sympathetic tone. This review summarized both in vitro and in vivo evidence implicating sympathetic neuronal activity in bone resorption. # 2011 Japanese Association for Dental Science. Published by Elsevier Ltd. All rights reserved.

Contents 1. 2.

Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . Nerve-bone cell interplay . . . . . . . . . . . . . . . 2.1. Adrenergic receptors and neuropeptide 2.2. Diffusible axon guidance molecules . . . 2.3. Nerve-bone cell communication . . . . . .

........ ........ receptors ........ ........

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

* Corresponding author. Tel.: +81 52 757 6742; fax: +81 52 752 5988. E-mail address: [email protected] (A. Togari). 1882-7616/$ — see front matter # 2011 Japanese Association for Dental Science. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.jdsr.2011.12.002

. . . . .

. . . . .

. . . . .

. . . . .

62 62 62 62 63

62 3.

4.

5.

A. Togari et al. Sympathetic effect on bone resorption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Effect of adrenergic agonists on osteoclastogenesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. Effect of adrenergic agonists on osteoclastic activity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3. Sympathetic regulation of bone resorption in vivo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4. Experimental and clinical evaluation of the effects of b-AR antagonists (b-blockers) on bone fracture risk . . . . Physiological modification of sympathetic effect on bone resorption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. Neuropeptides modifying adrenergic osteoclastogenesis in vitro. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2. Relationship between the sympathetic and sensory nervous systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Concluding remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1. Introduction Neural regulation of the bone metabolism mediated by osteoblastic and osteoclastic cells has been demonstrated [1—7]. Previous studies showed that human osteoblastic and osteoclastic cells not only possess adrenergic receptors (ARs) and neuropeptide receptors but also constitutively express axon guidance molecules for growing nerve fibers [8,9]. These findings together with the immunohistochemical evidence that mammalian bones are widely innervated by sympathetic and sensory nerves suggest that the extension of the axons of sympathetic and peripheral sensory nerves to osteoblastic and osteoclastic cells is required for the dynamic neural regulation of local bone metabolism. Direct nerveosteoblastic cell communication was revealed using an in vitro co-culture model comprising osteoblastic cells, and neurite-spouting superior cervical ganglia [10,11]. Recent bulk experimental studies showed that the sympathetic nervous system was involved in increasing bone resorption and decreasing bone formation, and that b-AR antagonists were effective against osteoporosis attributed to increased sympathetic nervous activity. Then, although neuropeptides are known to have significant osteotropic effects on bone metabolism [12,13], neuropeptide Y (NPY) and calcitonin generelated peptide (CGRP) have been a focus of our research, because of their modifying effect on osteoclastogenesis elicited by adrenergic stimulation [14,15]. The present article reviews our current understanding of the neuro-osteogenic network and sympathetic effects on bone resorption based on a variety of studies in vivo and in vitro. It also covers the physiological modification of sympathetic effects on bone resorption and discusses the role of neuropeptides in the modulation of adrenergic bone resorption.

2. Nerve-bone cell interplay 2.1. Adrenergic receptors and neuropeptide receptors Histochemical approaches have revealed the presence of vasoactive intestinal peptide (VIP), CGRP, substance P (SP), NPY, and noradrenaline (NA) in osseal nerve fibers [16,17]. Pharmacological evidence also shows that both osteoblastic cells and osteoclastic cells possess receptors for neuropeptides and NA. These observations suggest that the expression of these receptors physiologically regulates bone cell activities. In 1997, we demonstrated that a1B-

64 64 64 65 66 67 67 68 68 68 69

AR, a2B-AR, b2-AR, CGRP-R, NPY-R, and VIP-1R, but not a1A-AR, a1D-AR, a2C-AR, b1-AR, b3-AR, SP-R, VIP-2R, and pituitary adenylate cyclase-activating polypeptide (PACAP)-R, were expressed in human periosteum-derived osteoblastic cells (SaM-1) and human osteosarcomaderived cells (SaOS-2, HOS, MG-63) by the use of the reverse transcription-polymerase chain reaction (RT-PCR) [8]. The expression of these receptors seems to be a common feature of osteoblastic cells, but the magnitude of expression was not dependent upon the relative state of commitment of the osteoblastic cells to the osteoblast lineage (Table 1). In the osteoclastic cells, a1B-AR, a2BAR, b2-AR, CGRP-R, SP-R and VIP-1R were expressed (Table 1). SaM-1 and human osteoclastic cells, generated from bone marrow, expressed several phenotypes typical of mature cells. Recently, the expression of ARs was also detected by immunofluorescence microscopy and Western blotting in human osteoblasts [18]. From these results, it was revealed that human osteoblastic as well as ost eoclastic cells are equipped with ARs and neuropeptide receptors.

2.2. Diffusible axon guidance molecules During the development of the nervous system, neuronal growth cones traverse appropriate pathways to find their targets. Extending axons in the developing nervous system are guided to their targets through the coordinated actions of attractive and repulsive guidance cues. These mechanisms are mediated by many different families of guidance molecules secreted by target cells such as neurotrophins (chemoattractants), semaphorin-III (a chemorepellent), and netrins (chemoattractants and chemorepellents) [19]. In order to determine the ability of osteoblastic and osteoclastic cells to produce these diffusible axon guidance molecules, the steady-state expression of neurotrophins (NGF, nerve growth factor; BDNF, brain derived neurotrophic factor; NT-3, neurotrophin-3), semaphorin-III (Sema-III), netrins (NTN1, netrin-1; NTN2L, netrin-2-like protein) has been analyzed in SaM-1, SaOS-2, HOS, MG-63, and human osteoclastic cells by RTPCR, ELISA, and Western blot analysis [9]. SaM-1 cells expressed NGF, BDNF, NT-3, Sema-III, NTN1, and NTN2L after reaching confluence (Table 1). Their expression was also detected in osteosarcoma-derived cells, though the magnitude of expression was different. Human osteoclastic cells expressed NGF, BDNF, Sema-III, and NTN1, but not NTN2L. Thus, both osteoblastic and osteoclastic cells also

The neuro-osteogenic network

63

Table 1 Expression of mRNAs for adrenergic receptors, neuropeptide receptors, and axon guidance molecules in human osteoblastic and osteoclastic cells.

Adrenergic receptors

Neuropeptide receptors

Axon guidance molecules

a1A-R a1B-R a1D-R a2A-R a2B-R a2C-R b1-R b2-R b3-R

(60) (30) (60) (60) (60) (60) (30) (30) (30)

CGRP-R NPY-R SP-R VIP-1R VIP-2R PACAP-R NGF BDNF NT-3 Sema-III NTN1 NTN2L

SaM-1

SaOS-2

HOS

MG-63

Osteoclast

+

+++

+++

++

++

+ +

+ +

++

++

++

++

++

++

(45) (60) (60) (35) (35) (35)

+ ++

+ +

+ +

+++

++

+

+++

++

+

++ ++

(30) (30) (30) (35) (32) (32)

++ ++ ++ + + +

+++ ++ +++ ++ ++ +++

+++ ++ +++ +++

+ + ++ ++ ++ ++

+ + +++ +

( ) no, (+) faint, (++) medium, (+++) strong expression. Number: cycles for PCR amplification.

constitutively express diffusible axon guidance molecules known to function as chemoattractants and/or chemorepellents for growing nerve fibers. These findings may suggest that the extension of axons of sympathetic and peripheral sensory neurons to osteoblastic and osteoclastic cells is required for the dynamic neural regulation of local bone metabolism. Therefore, it has been proposed that signaling molecules in the nervous system may participate in the control of local bone metabolism and that, consequently, a neuro-osteogenic network may exist, one similar to previously proposed neuroimmune and neuro-immune-endocrine interacting systems [20—22].

2.3. Nerve-bone cell communication Takeda et al. [2] demonstrated electron microscopically the presence of peripheral nerve axons coursing through the marrow adjacent to osteoblasts in bone tissue, in which case actual membrane-membrane contacts were formed between nerve and osteoblastic cells. However, whether the activation of both osteoblastic and osteoclastic cells occurs as a direct response to neuronal activation or requires an intermediary cell is unclear. In 2007, direct nerve-osteoblastic cell communication was elucidated using an in vitro co-culture model comprising mouse osteoblastic cells, MC3T3-E1 cells, and neurite-spouting mouse superior cervical ganglia (Fig. 1) [10]. Following loading with the calcium fluorophore Fluo-3, neurite-osteoblastic cell units were examined by confocal laser scanning microscope. Addition of scorpion venom (SV) elicited neurite activation (i.e., Ca2+ mobilization) and, after a lag period, osteoblastic Ca2+ mobilization. SV had no direct effect on the MC3T3-E1 cells

in the absence of neurites. Addition of an a1-AR antagonist, prazosin, concentration-dependently prevented the osteoblastic activation that resulted as a consequence of neural activation by SV. Thus, our recent findings demonstrate that MC3T3-E1 cell activation, as judged by Ca2+ mobilization, can be a direct consequence of contact with a specific activated nerve fiber. This evidence obtained in vitro shows that nerve-osteoblastic cell cross-talk can occur in the absence of an intermediary transducing cell, and that NA is an important mediator of this communication. A similar technique also demonstrated nerve-osteoclastic cell crosstalk [11]. Although, in the co-culture experiment, both osteoblastic and osteoclastic activation was observed via a1-AR as a direct response to neuronal activation, there are very few reports of a physiological role for a-AR. In MC3T3-E1 cells, a1-AR stimulation increased cell proliferation, alkaline phosphatase activity, and type III Pi transporter activity [23,24], and increased RANKL expression via protein kinase C and extracellular signal-regulated kinase pathways [25]. Recently, using whole-cell patch clamp recordings, we also found that a1B-adrenergic stimulation suppressed Cs-sensitive and tetraethylammonium-insensitive potassium channels in SaM-1 cells [26]. Since potassium channel activity is known to regulate membrane potential and cell proliferative capacity in various cells [27—30], a1-AR stimulation may facilitate cell proliferation via a1B-AR through the regulation of potassium channels in human osteoblasts. Anyway, several in vivo and in vitro studies have demonstrated sympathomimetic effects on bone formation and resorption via osteoblastic and osteoclastic cells equipped with a- and b-ARs.

64

A. Togari et al.

Figure 1 Nerve-induced osteoblastic or osteoclastic cell activation. (A) Representative differential interference contrast (DIC) images of the SCG-osteoblast (a) and the SCG-osteoclast (b) co-cultures. (B) Pharmacological analysis of the functional interaction between neurons and osteoblastic or osteoclastic cells in a co-culture experiment. Scorpion venom, a neurotoxin, opens neural Na+ channels to cause depolarization and increase [Ca2+]i. Then, it stimulates neurons to release neurotransmitters from the synaptic vesicles of nerve endings. If there is functional cellular interaction, the neurotransmitters released from neurons act on the receptors of osteoblastic or osteoclastic cells to increase [Ca2+]i within them as well.

3. Sympathetic effect on bone resorption 3.1. Effect of adrenergic agonists on osteoclastogenesis Bone marrow culture techniques have been successfully employed to study the development of osteoclasts from their precursor cells. Such cultures provide an appropriate system to investigate osteotrophic hormones, cytokines, and other bone-active factors that may be involved in the generation of osteoclasts. In this culture system, receptor activator of NFkB ligand (RANKL) and osteoprotegerin (OPG) were reported to play an essential role in osteoclastic differentiation. The expression of both proteins was reported to be regulated by several osteotrophic factors including 1a,25-dihydroxyvitamin D3 (1a,25(OH)2D3), interleukin (IL)-1a, IL-11, prostaglandin (PG)E2, transforming growth factor-b1, and parathyroid hormone [31—35]. In 2001, adrenaline and isoprenaline, b-AR agonists, have been also demonstrated to modulate osteoclastogenesis. The involvement of RANKL and/or OPG in adrenaline-induced bone resorption was shown by determining the effect of adrenaline on the mRNA expression of RANKL and OPG in MC3T3-E1 cells and the formation of tartrateresistant acid phosphatase (TRAP)-positive multinuclear cells (MNCs) in mouse bone marrow cultures, thus providing a better understanding of the bone resorption induced by the sympathetic system [36]. Use of the RT-PCR procedure revealed that the expression of RANKL and OPG mRNAs in osteoblastic cells was regulated by adrenergic stimulation [36]. The expression of RANKL and OPG elicited by adrenaline appeared to be mediated by b-adrenergic and a-adrenergic stimulation, respectively. Treatment of mouse bone marrow

cells with adrenaline or isoprenaline generated TRAP-positive MNCs capable of excavating resorptive pits on dentine slices, and caused an increase in RANKL and a decrease in OPG production by the marrow cells [5,15,36]. The osteoclast formation was significantly inhibited by OPG, suggesting the involvement of the RANKL-RANK system. Since the osteoclastogenesis in mouse bone marrow cells was not stimulated by an a-AR agonist, it may be regulated by the balance between RANKL and OPG production in osteoblasts/stromal cells. Fig. 2 schematically shows a possible mechanism for the adrenergic stimulation of osteoclastogenesis.

3.2. Effect of adrenergic agonists on osteoclastic activity In neonatal mouse calvariae, AR agonists stimulated cyclic AMP (cAMP) synthesis and bone resorption in the presence of a phosphodiesterase inhibitor and an antioxidant [37]. The stimulation of cAMP synthesis by b-AR agonists was inhibited by propranolol in a bone organ culture [38]. The b-adrenergic stimulation of bone resorption might be mediated by directly activated osteoclasts and osteoclastogenesis enhanced by osteotrophic factors released from osteoblasts. In human osteoclastic cells constitutively expressing a1B-, a2B-, and b2-ARs, b-AR agonists upregulated the expression of characteristic markers of the mature osteoclast, such as integrin, carbonic anhydrase II, and cathepsin K; increased osteoclastic bone-resorbing activity; and clearly caused actin ring formation [39]. These findings were not obtained on treatment with a-AR agonists, and suggest that b-AR agonists directly stimulate bone-resorbing activity in mature osteoclasts. In a clonal cell line of human osteoclast precursors

The neuro-osteogenic network

65

Figure 2 Schematic diagram of adrenergic action on osteoclastogenesis and osteoclastic avtivity. Both osteoblasts and osteoclasts constitutively express a1B-, a2B- and b2-ARs. Stimulation of b2-ARs increases osteoclast formation and the expression of RANKL, IL-6, IL11, and PGE2 in bone marrow or clonal osteoblastic cells. The increase in osteoclasts is inhibited by OPG treatment, suggesting the involvement of the RANKL-RANK system in osteoclastogenesis caused by stimulating b2-ARs. The expression of RANKL and OPG in MC3T3E1 cells is increased by stimulating b2-ARs and a-ARs, respectively. In human osteoclasts, stimulation of b-ARs activates osteoclastic activity. Thus, increased sympathetic activity stimulates osteoclast formation as well as osteoclastic activity through b2-ARs.

(FLG 29.1 cells), catecholamines were also demonstrated to act as inducers of osteoclastic maturation in vitro and as stimulators of osteoclastic activity via binding to b2-ARs [40]. As osteoclastogenesis-enhancing osteotrophic factors produced by b-adrenergic stimulation, IL-6, IL-11, and PGE2 were detected in human and mouse osteoblastic cells [36,41]. The coexpression of IL-6 and IL-11 induced by the activation of b-ARs, which appears to be a common feature of osteoblastic cells, has been shown to be mediated via a common signaling pathway involving the protein kinase A and p38 mitogen-activated protein kinase systems, leading to the transcriptional activation of activator protein-1 in human osteoblastic cells. Thus, AR agonists cause the catabolic effect on bone metabolism via the b-adrenergic system. Nevertheless, Elefteriou et al. reported that isoprenaline did not stimulate cAMP production in bone marrow macrophages treated with RANKL and macrophage colony-stimulating factor [5], indicating the b-agonist to have an indirect rather than direct effect on mature osteoclasts. Further research may be necessary to gain a better understanding of how b-agonists act on mature osteoclasts.

3.3. Sympathetic regulation of bone resorption in vivo There are several lines of evidence that the sympathetic nervous system modulates bone resorption in vivo. Surgical removal of the superior cervical ganglion increased bone resorption [42], as did chemical treatment with guanethidine in newborn rats [43]. However, in adult rats treated with guanethidine, bone resorption was reduced [44]. The reduction in bone resorption was assumed to reflect the acute

effects of the sympathectomy. By using a compartmentalization procedure, the resorption surface in the osteogenic compartment was found to be significantly reduced in the guanethidine-treated rats, accompanied by a fall in the number of osteoclasts and impaired access to the bone surface. The effect on resorption that the sympathectomy had by inhibiting preosteoclastic differentiation and disturbing osteoclastic activation suggests that depletion of sympathetic mediators could disturb osteogenic cell-mediated osteoclastic differentiation. In addition, sympathectomyinduced depletion of NA may be another possible mechanism for the reduction in bone resorption. Such a notion is supported by several significant findings including the stimulation of bone resorption in a tissue culture system [37], an increase in preosteoclastic cell activity [40], and stimulation of the synthesis of osteoclast-like cell formation-stimulating factors in osteoblastic cells by b-adrenergic stimulation [36,41]. The intracerebroventricular (i.c.v.) injection of lipopolysaccharide (LPS), an inflammatory stimulus in the brain, was demonstrated to increase output from the peripheral sympathetic nervous system. To prove the physiological role of the sympathetic nervous system in bone metabolism in vivo, RT-PCR was performed to examine the effect of an i.c.v. injection of LPS on cyclooxygenase-2 and IL-6 mRNA expression in mouse calvaria [45,46]. The expression of both mRNA was increased by the injection. The increases were inhibited by treatment with the neurotoxin 6-hydroxydopamine (6OHDA) or b-AR antagonists. Similarly, restraint stress induced the expression of IL-6 mRNA in mouse calvaria [46]. This induction was not influenced by 6-OHDA, but was inhibited by propranolol. In addition, the treatment of calvaria with isoprenaline or NA increased PGE2 and IL-6 synthesis in the

66

A. Togari et al.

Figure 3 Schematic diagram of bone metabolism modulated by the sympathetic nervous system. Central activation of the peripheral sympathetic nervous system caused by the intracerebroventricular (i.c.v.) injection of lipopolysaccharide (LPS) or leptin and by restraint stress alters bone formation and bone resorption. Centrally mediated activation of the nervous system is associated with increased catecholamine levels. Leptin-induced sympathetic activation has been reported to decrease bone formation via osteoblastic activity. Conversely, both LPS- and stress-induced activation increase the synthesis of osteoclastogenetic factors such as IL-6, and PGE2, which leads to bone resorption. Chronic stimulation of b-AR also induces bone loss due to increased osteoclastic activity rather than inhibition of bone formation.

organ culture system. These findings show that the increase in gene expression caused by a restraint stress or i.c.v. injection of LPS was mediated by the activation of sympathetic nerve fibers and b-ARs in mouse calvaria and suggest that in vivo activation of the sympathetic nervous system modulates bone resorption. Such results are strongly supported by the findings of Karsenty’s group [5], who reported that b2-AR-deficient (Adrb2 / ) mice had a more severe high bone mass phenotype than wild-type mice receiving b-AR antagonists and that longterm leptin i.c.v. infusion did not reduce bone mass in Adrb2 / mice. Furthermore, they showed that the sympathetic nervous system favors bone resorption by increasing the expression of RANKL and that isoprenaline enhanced the generation of osteoclasts when wild-type, but not Adrb2 / , osteoblasts were co-cultured with wild-type bone marrow macrophages. Moreover, using osmotic minipumps implanted into the subcutaneous tissue in the back, we recently demonstrated that chronic stimulation of b-AR with low-dose isoprenaline treatment induces bone loss due to increased osteoclastic activity rather than inhibition of bone formation [47]. Thus, these in vivo experiments modulating peripheral sympathetic nervous activity suggest that increased sympathetic nervous activity leads to increase bone resorption through b2-ARs. To integrate these recent findings, we have presented a possible mechanism for the regulation of bone metabolism via the sympathetic nervous system in Fig. 3.

3.4. Experimental and clinical evaluation of the effects of b-AR antagonists (b-blockers) on bone fracture risk Both in vitro and in vivo experimental studies indicate bblockers to be effective against osteoporosis attributed to

increased sympathetic nervous activity. The use of b-blockers to inhibit bone resorption and/or to stimulate bone formation could, therefore, be an important new approach to treating osteoporosis. In population-based, case-control studies involving adult women [48], adult men and young women [49], the use of b-blockers, taken alone as well as in combination with thiazide diuretics, was demonstrated to be associated with a reduced risk of fractures. Thus, b-blockers generally do cause a reduction in bone fracture risk and higher bone mineral density. Another prospective study, however, found no association between b-blocker use and fracture risk in perimenopausal and older women [50—52]. Therefore, there is currently no convincing evidence supporting the hypothesis that pharmacological blockade of the badrenergic system is beneficial to the human skeleton after menopause. Although b-adrenergic stimulation can be proposed as one of the causes of osteoporosis in experimental studies, the clinical usefulness of b-blockers for fracture risk must be analyzed in several patients with increased sympathetic nervous activity. Specifically, it is important to find a difference between users and nonusers with increased sympathetic tone. To evaluate the effectiveness of b-blockers for experimental osteoporosis with hyperactivity of the peripheral sympathetic nervous system, bone mass was analyzed in the spontaneously hypertensive rat (SHR), a hypertensive model with enhanced sympathetic nervous activity. The SHR exhibited significantly decreased cancellous bone density as well as markedly increased blood pressure [53]. Specifically, bone density and strength in the lumbar spine decreased. Histochemistry showed decreased bone formation, increased numbers of osteoclasts, decreased serum levels of osteocalcin, a bone formation marker, and increased TRAP 5b activity, a systemic bone resorption marker. These

The neuro-osteogenic network

67

4. Physiological modification of sympathetic effect on bone resorption 4.1. Neuropeptides modifying adrenergic osteoclastogenesis in vitro

Figure 4 Schematic diagram of the b-adrenergic regulation of bone metabolism involved in bone fracture risk. Several experimental studies suggested that b-AR antagonists decreased the risk of bone fractures. Then, clinical studies showed the pharmacological effectiveness of b-blockers in reducing bone-fracture risk. Recently, in SHR showing osteoporosis as well as hypertension, the effectiveness of b-blockers against bone loss and bone fragility was confirmed. The reduced bone mass in SHR with enhanced sympathetic nerve activity is caused by increased bone resorption and decreased bone formation, as compared with control rats (WKY). The reduced bone mass is accompanied by bone fragility. The bone mass in SHR recovered to WKY’s level when b2-AR was suppressed.

changes were improved with a low dose of propranolol, which has no effect on blood pressure [53]. The b-blocker timolol and selective b2-blocker butoxamine, which have no membrane-stabilizing effect, also improved the bone density. Thus, in SHR with enhanced sympathetic nerve activity, bone loss was improved by blocking b2-ARs with a low dose of b-blockers (Fig. 4). These results are similar to the effects of propranolol, as observed by Bonnet et al. [54] in ovariectomized (OVX) rats. The results may be associated with enhanced sympathetic activity in OVX rats.

In mammalian species, nerve fibers containing several neuropeptides, such as NPY, CGRP, VIP, and SP, as well as NA, a classical neurotransmitter, have been identified in the vicinity of bone tissue [55—58]. In accordance with the neuroosteogenic hypothesis, these neuropeptides can be released from nerve endings and transmit physiological signals to osteoblastic and osteoclastic cells present close by. Although these neuropeptides produce significant osteotropic effects on bone metabolism [12,13], NPY and CGRP have been demonstrated to modulate osteoclastogenesis elicited by adrenergic stimulation [14,15]. NPY is co-localized with NA in sympathetic nerve terminals [59—62] and recognized as a co-transmitter with NA in peripheral sympathetic nerve fibers. Studies have revealed that NPY inhibits cAMP production in the target cell [63,64]. Indeed, NPY inhibited the stimulatory effect of NA on cAMP in UMR-106-01 cells and isolated bone cells [65]. In addition, NPY has been demonstrated to inhibit b-adrenergic- or VIPergic-stimulated accumulation of cAMP in the pineal gland, which is mediated through a pertussis toxin-sensitive G protein. Recently, the effect of NPY on osteoclastogenesis has been demonstrated in mouse bone marrow cell cultures treated with isoprenaline [14]. The mouse bone marrow cells constitutively expressed mRNAs for the NPY-Y1 receptor and b2-AR. NPY inhibited the formation of osteoclast-like cells induced by isoprenaline but not by 1a,25(OH)2D3 or soluble RANKL; and suppressed the production of RANKL and cAMP increased by isoprenaline but not by 1a,25(OH)2D3. NPY also inhibited osteoclastogenesis induced by forskolin, an activator of adenylate cyclase; but not that induced by dibutyryl cAMP, a cell-permeable cAMP analog that activates cAMP-

Figure 5 Schematic diagram of a working hypothesis for the possible interaction between sympathetic and sensory neurons. Stimulation of sympathetic neurons as well as denervation of sensory neurons causes bone loss via an increase in bone resorption and a decrease in bone formation. The strong resemblance between sensory denervation and sympathetic hyperfunction, suggests that sensory activity functionally interacts with the sympathetic activity in bone metabolism. Black arrow: effects on osteoblastic bone formation; white arrow: effects on osteoclastic bone resorption.

68 dependent protein kinases. These results demonstrate that NPY inhibited isoprenaline-induced osteoclastogenesis by blocking agonist-elicited increases in the production of cAMP and RANKL in mouse bone marrow cells, suggesting an interaction between NPY and b-AR agonists in bone resorption. The present results indicate NPY to be a specific inhibitor of b2-AR agonist-induced osteoclastic formation. The physiological relevance of this observation is that NPY substantially co-existed with NA in the sympathetic nerve terminals to control the b-adrenergic stimulation of NA released simultaneously; because similar effects of NPY had been observed in the cells and tissues controlled by b-adrenergic regulation. CGRP is a neuropeptide abundant in sensory neurons innervating the skeleton and regulates bone integrity. CGRP has been demonstrated to inhibit the bone-resorbing activity of isolated osteoclasts [66,67], and calcium release in bone tissue cultures [68,69]. It also significantly inhibited 1a,25(OH)2D3-induced osteoclastogenesis in macrophages [70]. In bone marrow cell cultures, CGRP inhibited the formation of osteoclasts caused by isoprenaline or soluble RANKL but had no influence on RANKL or OPG production by the bone marrow cells treated with isoprenaline, suggesting that CGRP inhibited the osteoclastogenesis by interfering with the action of RANKL produced by the isoprenalinetreated bone marrow cells without affecting RANKL or OPG production [15]. Thus, the b-AR agonist generated osteoclasts from mouse bone marrow cells via the RANKL-RANK system. Also, CGRP significantly inhibited this RANKLmediated osteoclastogenesis by interfering with the action of RANKL in bone marrow macrophages [71]. Although these in vitro data suggest the physiological interaction of sympathetic and sensory nerves in osteoclastogenesis in vivo, the physiological significance remains to be established.

4.2. Relationship between the sympathetic and sensory nervous systems Substantial evidence has accumulated that sympathetic nervous activity causes bone loss via an increase in bone resorption and a decrease in bone formation. Sympathetic denervation is known to be associated with a significant increase in the number of CGRP-immunoreactive sensory nerve fibers [72]. b-AR blockers not only improve bone loss in SHR with hyperactivity of the peripheral sympathetic tone [53] but also increase the release of CGRP from sensory nerve fibers in SHR [73]. Furthermore, like b-blocker, CGRP partially inhibits bone loss in OVX rats [74]. These studies may suggest physiological interaction between sympathetic and sensory nerves in bone metabolism. The effects of sensory denervation on bone metabolism are usually examined in animals treated with capsaicin, which destroys unmyelinated and small-diameter myelinated sensory neurons via the activation of transient receptor potential vanilloid 1 expressed by sensory neurons. There are less SP-containing nerve fibers than CGRP-containing nerve fibers [75], so the sensory nerve fibers ablated by capsaicin are mainly CGRP-containing nerve fibers. Capsaicin treatment caused the destruction of the skeletal structure in adult rats [76,77]. In our experiment [77], it reduced trabecular bone volume due to increased trabecular separation in the proximal tibia and the modification of mechanical

A. Togari et al. properties such as strength, ductility, and toughness toward increasing bone fragility in the trunk of the sixth lumber vertebrae. Bone histomorphometry showed an increase in osteoclast numbers and surface area. Capsaicin significantly increased the level of TRAP 5b in plasma but had no influence on the plasma osteocalcin concentration suggesting that capsaicin-induced sensory nerve denervation increased bone resoption but had no influence on bone formation. These results show that sensory nerve innervation contributes to the maintenance of trabecular bone mass and its mechanical properties by inhibiting bone resorption. Thus, there is a strong resemblance between sensory denervation and sympathetic activation in bone metabolism, suggesting that a sensory activity functionally interacts with the sympathetic activity in osteoclastic formation. Fig. 5 schematically represents a working hypothesis for the possible interaction of sympathetic and sensory neurons.

5. Concluding remarks An in vitro co-culture experiment demonstrated that the responses to osteoblastic and osteoclastic cells produced by neural stimulation were inhibited by AR antagonists, suggesting that synaptic transmission occurs from nerve terminals to these cells. Specifically, the peripheral nerve may be functionally and directly connected to the osteoblastic and osteoclastic cells for regulating bone metabolism in vivo. If osteoblastic activation is judged by cAMP, instead of the Ca2+ used in our study, it would be possible to analyze the molecular mechanism behind the communication mediated by NA acting through b-ARs. There is currently great interest in comparing the cellular interaction with other cellular interactions, and in elucidating the physiological regulation of NA secretion in neuro-osteogenic synapses. Thus, the in vitro coculture model is useful for studying the molecular mechanism responsible for the neuro-osteogenic cross-talk. In SHR with hyperactivity of the sympathetic nervous system, bone mass and biomechanical fragility were markedly reduced by increased bone resorption and decreased bone formation, and improved by the b-blocker propranolol at lower doses than those required to improve hypertension. Hypertension is often accompanied by enhanced sympathetic nerve activity and reduced bone mass. Comprehensive investigation may be necessary to understand the mutual relationship among calcium metabolism, bone metabolism, hypertension, and sympathetic nerve activity. Notably, sympathetic regulation of bone metabolism may be modified by sensory nervous activity, as described in this article. Although there is currently great interest in physiological modifications to signal transduction and the synaptic integrity of sympathetic and sensory neurons, further studies should clarify the mechanisms responsible for these modifications in peripheral nervous systems and give further insight into the neural regulation in bone metabolism.

Acknowledgements We are grateful to Shoko Imamura for excellent technical assistance. Our studies mentioned in this review were partly supported by a grant-in-aid for AGU High-Tech Research Center Project for Private Universities: matching fund subsidy from the Ministry of Education, Culture, Sports, Science and Technology

The neuro-osteogenic network (MEXT: 2003—2007), by a grant-in-aid from Strategic Research AGU-Plantform Formation (2008—2012), and by grants-in-aid from MEXT (14571782 and 17591956to AT).

References [1] Togari A. Adrenergic regulation of bone metabolism: possible involvement of sympathetic innervation of osteoblastic and osteoclastic cells. Microsc Res Technol 2002;58:77—84. [2] Takeda S, Elefteriou F, Levasseur R, Liu X, Zhao L, Parker KL, et al. Leptin regulates bone formation via the sympathetic nervous system. Cell 2002;111:305—17. [3] Flier JS. Physiology: is brain sympathetic to bone? Nature 2002;420(619):621—2. [4] Togari A, Arai M, Kondo A. The role of the sympathetic nervous system in controlling bone metabolism. Expert Opin Ther Targets 2005;9:931—40. [5] Elefteriou F, Ahn JD, Takeda S, Starbuck M, Yang X, Liu X, et al. Leptin regulation of bone resorption by the sympathetic nervous system and CART. Nature 2005;434:514—20. [6] Togari A, Arai M. Pharmacological topics of bone metabolism: the physiological function of the sympathetic nervous system in modulating bone resorption. J Pharmacol Sci 2008;106: 542—6. [7] Takeda S. Osteoporosis: a neuroskeletal disease? Int J Biochem Cell Biol 2009;41:455—9. [8] Togari A, Arai M, Mizutani S, Mizutani S, KoshiharaY, Nagatsu T. Expression of mRNAs for neuropeptide receptors and b-adrenergic receptors in human osteoblasts and human osteogenic sarcoma cells. Neurosci Lett 1997;233:125—8. [9] Togari A, Mogi M, Arai M, Yamamoto S, Koshihara Y. Expression of mRNA for axon guidance molecules, such as semaphorin-III, netrins and neurotrophins, in human osteoblasts and osteoclasts. Brain Res 2000;878:204—9. [10] Obata K, Furuno T, Nakanishi M, Togari A. Direct neurite-osteoblastic cell communication, as demonstrated by use of an in vitro co-culture system. FEBS Lett 2007;581:5917—22. [11] Suga S, Goto S, Togari A. Demonstration of direct neuriteosteoclastic cell communication in vitro via the adrenergic receptor. J Pharmacol Sci 2010;112:184—91. [12] Lerner UH, Persson E. Osteotropic effects by the neuropeptides calcitonin gene-related peptide, substance P and vasoactive intestinal peptide. J Musculoskelet Neuronal Interact 2008;8:154—65. [13] Franquinho F, Liz MA, Nunes AF, Neto E, Lamghari M, Sousa MM. Neuropeptide Y and osteoblast differentiation — the balance between the neuro-osteogenic network and local control. FEBS J 2010;277:3664—74. [14] Amano S, Arai M, Goto S, Togari A. Inhibitory effect of NPY on isoprenaline-induced osteoclastogenesis in mouse bone marrow cells. Biochim Biophys Acta 2007;1770:966—73. [15] Ishizuka K, Hirukawa K, Nakamura H, Togari A. Inhibitory effect of CGRP on osteoclast formation by mouse bone marrow cells treated with isoproterenol. Neurosci Lett 2005;379:47—51. [16] Bjurholm A. Neuroendocrine peptides in bone. Int Orthop 1991;15:325—9. [17] Hohmann EL, Elde RP, Rysavy JA, Einzig S, Gebhard RL. Innervation of periosteum and bone by sympathetic vasoactive intestinal peptide-containing nerve fibers. Science 1986;232:868—71. [18] Huang HH, Brennan TC, Muir MM, Mason RS. Functional a1- and b2-adrenergic receptors in human osteoblasts. J Cell Physiol 2009;220:267—75. [19] Goodman CS. Mechanisms and molecules that control growth cone guidance. Annu Rev Neurosci 1996;19:341—77. [20] Opp MR, Imeri L. Sleep as a behavioral model of neuro-immune interactions. Acta Neurobiol Exp 1999;59:45—53. [21] Aller MA, Arias JL, Lorente L, Nava MP, Duran HJ, Arias J. Neuroimmune-endocrine functional system and vascular pathology. Med Hypotheses 2001;57:561—9.

69 [22] Chesnokova V, Melmed S, Minireview:. Neuro-immuno-endocrine modulation of the hypothalamic—pituitary—adrenal (HPA) axis by gp130 signaling molecules. Endocrinology 2002;143:1571—4. [23] Suzuki A, Palmer G, Bonjour JP, Caverzasio J. Stimulation of sodium-dependent inorganic phosphate transport by activation of Gi/o-protein-coupled receptors by epinephrine in MC3T3-E1 osteoblast-like cells. Bone 2001;28:589—94. [24] Suzuki A, Guicheux J, Palmer G, Miura Y, Oiso Y, Bonjour JP, et al. Evidence for a role of p38 MAP kinase in expression of alkaline phosphatase during osteoblastic cell differentiation. Bone 2002;30:91—8. [25] Nishiura T, Abe K. a1-adrenergic receptor stimulation induces the expression of receptor activator of nuclear factor kappaB ligand gene via protein kinase C and extracellular signal-regulated kinase pathways in MC3T3-E1 osteoblast-like cells. Arch Oral Biol 2007;52:778—85. [26] Kodama D, Togari A. Modulation of potassium channels via the a1B-adrenergic receptor in human osteoblasts. Neurosci Lett 2010;485:102—6. [27] Pardo LA, Contreras-Jurado C, Zientkowska M, Alves F, Stu ¨hmer W. Role of voltage-gated potassium channels in cancer. J Membr Biol 2005;205:115—24. [28] Liebau S, Vaida B, Proepper C, Grissmer S, Storch A, Boeckers TM, et al. Formation of cellular projections in neural progenitor cells depends on SK3 channel activity. J Neurochem 2007;101:1338—50. [29] Henney NC, Li B, Elford C, Reviriego P, Campbell AK, Wann KT, et al. A large-conductance (BK) potassium channel subtype affects both growth and mineralization of human osteoblasts. Am J Physiol Cell Physiol 2009;297:C1397—408. [30] Jang SS, Park J, Hur SW, Hong YH, Hur J, Chae JH, et al. Endothelial progenitor cells functionally express inward rectifier potassium channels. Am J Physiol Cell Physiol 2011;301:C150—61. [31] Yasuda H, Shima N, Nakagawa N, Mochizuki SI, Yano K, Fujise N, et al. Identity of osteoclastogenesis inhibitory factor (OCIF) and osteoprotegerin (OPG): a mechanism by which OPG/OCIF inhibits osteoclastogenesis in vitro. Endocrinology 1998;139:1329—37. [32] Gao YH, Shinki T, Yuasa T, Kataoka-Enomoto H, Komori T, Suda T, et al. Potential role of cbfa1, an essential transcriptional factor for osteoblast differentiation, in osteoclastogenesis: regulation of mRNA expression of osteoclast differentiation factor (ODF). Biochem Biophys Res Commun 1998;252:697—702. [33] Hofbauer LC, Dunstan CR, Spelsberg TC, Riggs BL, Khosla S. Osteoprotegerin production by human osteoblast lineage cells is stimulated by vitamin D, bone morphogenetic protein-2, and cytokines. Biochem Biophys Res Commun 1998;250:776—81. [34] Horwood NJ, Elliott J, Martin TJ, Gillespie MT. Osteotropic agents regulate the expression of osteoclast differentiation factor and osteoprotegerin in osteoblastic stromal cells. Endocrinology 1998;139:4743—6. [35] Murakami T, Yamamoto M, Ono K, Nishikawa M, Nagata N, Motoyoshi K, et al. Transforming growth factor-b1 increases mRNA levels of osteoclastogenesis inhibitory factor in osteoblastic/stromal cells and inhibits the survival of murine osteoclast-like cells. Biochem Biophys Res Commun 1998;252: 747—52. [36] Takeuchi T, Tsuboi T, Arai M, Togari A. Adrenergic stimulation of osteoclastogenesis mediated by expression of osteoclast differentiation factor in MC3T3-E1 osteoblast-like cells. Biochem Pharmacol 2001;61:579—86. [37] Moore RE, Smith II CK, Bailey CS, Voelkel EF, Tashjian Jr AH. Characterization of beta-adrenergic receptors on rat and human osteoblast-like cells and demonstration that b-receptor agonists can stimulate bone resorption in organ culture. Bone Miner 1993;23:301—15. [38] Majeska RJ, Minkowitz B, Bastian W, Einhorn TA. Effects of badrenergic blockade in an osteoblast-like cell line. J Orthop Res 1992;10:379—84.

70 [39] Arai M, Nagasawa T, Koshihara Y, Yamamoto S, Togari A. Effects of b-adrenergic agonists on bone-resorbing activity in human osteoclast-like cells. Biochim Biophys Acta 2003;1640:137—42. [40] Frediani U, Becherini L, Lasagni L, Tanini A, Brandi ML. Catecholamines modulate growth and differentiation of human preosteoclastic cells. Osteoporos Int 1996;6:14—21. [41] Kondo A, Mogi M, Koshihara Y, Togari A. Signal transduction system for interleukin-6 and interleukin-11 synthesis stimulated by epinephrine in human osteoblasts and human osteogenic sarcoma cells. Biochem Pharmacol 2001;61:319—26. [42] Sandhu HS, Herskovits MS, Singh IJ. Effect of surgical sympathectomy on bone remodeling at rat incisor and molar root sockets. Anat Rec 1987;219:32—8. [43] Hill EL, Turner R, Elde R. Effects of neonatal sympathectomy and capsaicin treatment on bone remodeling in rats. Neuroscience 1991;44:747—55. [44] Cherruau M, Facchinetti P, Baroukh B, Saffar JL. Chemical sympathectomy impairs bone resorption in rats: a role for the sympathetic system on bone metabolism. Bone 1999;25:545—51. [45] Inoue H, Kondo A, Togari A. Activation of the peripheral sympathetic nervous system increased the expression of cyclooxygenase-2 (COX-2) mRNA in mouse calvaria. Neurosci Lett 2003;338:37—40. [46] Kondo A, Togari A. In vivo stimulation of sympathetic nervous system modulates osteoblastic activity in mouse calvaria. Am J Physiol Endocrinol Metab 2003;285:E661—7. [47] Kondo H, Togari A. Continuous treatment with a low-dose bagonist reduces bone mass by increasing bone resorption without suppressing bone formation. Calcif Tissue Int 2011;88:23—32. [48] Pasco JA, Henry MJ, Sanders KM, Kotowicz MA, Seeman E, Nicholson GC. b-Adrenergic blockers reduce the risk of fracture partly by increasing bone mineral density: Geelong Osteoporosis Study. J Bone Miner Res 2004;19:19—24. [49] Schlienger RG, Kraenzlin ME, Jick SS, Meier CR. Use of bblockers and risk of fractures. J Am Med Assoc 2004;292: 1326—32. [50] Rejnmark L, Vestergaard P, Kassem M, Christoffersen BR, Kolthoff N, Brixen K, et al. Fracture risk in perimenopausal women treated with beta-blockers. Calcif Tissue Int 2004;75:365—72. [51] Levasseur R, Dargent-Molina P, Sabatier JP, Marcelli C, Breart G. Beta-blocker use, bone mineral density, and fracture risk in older women: results from the Epidemiologie de l’Osteoporose prospective study. J Am Geriatr Soc 2005;53:550—2. [52] Reid IR, Gamble GD, Grey AB, Black DM, Ensrud KE, Browner WS, et al. b-Blocker use, BMD, and fractures in the study of osteoporotic fractures. J Bone Miner Res 2005;20:613—8. [53] Sato T, Arai M, Goto S, Togari A. Effects of propranolol on bone metabolism in spontaneously hypertensive rats. J Pharmacol Exp Ther 2010;334:99—105. [54] Bonnet N, Benhamou CL, Malaval L, Goncalves C, Vico L, Eder V, et al. Low dose b-blocker prevents ovariectomy-induced bone loss in rats without affecting heart functions. J Cell Physiol 2008;217:819—27. [55] Sisask G, Bjurholm A, Ahmed M, Kreicbergs A. The development of autonomic innervation in bone and joints of the rat. J Auton Nerv Syst 1996;59:27—33. [56] Hill EL, Elde R. Distribution of CGRP-, VIP-, DbH-, SP-, and NPYimmuno-reactive nerves in the periosteum of the rat. Cell Tissue Res 1991;264:469—80. [57] Bjurholm A, Kreicbergs A, Terenius L, Goldstein M, Schultzberg M. Neuropeptide Y-, tyrosine hydroxylase- and vasoactive intestinal polypeptide-immunoreactive nerves in bone and surrounding tissues. J Auton Nerv Syst 1988;25:119—25. [58] Chenu C. Role of innervation in the control of bone remodeling. J Musculoskelet Neuronal Interact 2004;4:132—4.

A. Togari et al. [59] Straub RH, Schaller T, Miller LE, von Horsten S, Jessop DS, Falk W, et al. cotransmission with norepinephrine in the sympathetic nerve-macrophage interplay. J Neurochem 2000;75:2464—71. [60] Lundberg JM, Anggard A, Pernow J, Hokfelt T. Neuropeptide Y-, substance P- and VIP-immunoreactive nerves in cat spleen in relation to autonomic vascular and volume control. Cell Tissue Res 1985;239:9—18. [61] Weihe E, Nohr D, Michel S, Muller S, Zentel HJ, Fink T, et al. Molecular anatomy of the neuro-immune connection. Int J Neurosci 1991;59:1—23. [62] Jessop D, Biswas S, D’Souza L, Chowdrey H, Lightman S. Neuropeptide Y immunoreactivity in the spleen and thymus of normal rats and following adjuvant-induced arthritis. Neuropeptides 1992;23:203—7. [63] Harfstrand A, Fredholm B, Fuxe K. Inhibitory effects of neuropeptide Y on cyclic AMP accumulation in slices of the nucleus tractus solitarius region of the rat. Neurosci Lett 1987;76:185—90. [64] van Valen F, Keck E, Jurgens H. Neuropeptide Y inhibits vasoactive intestinal peptide- and dopamine-induced cyclic AMP formation in human Ewing’s sarcoma WE-68 cells. FEBS Lett 1989;249:271—4. [65] Bjurholm A, Kreicbergs A, Schultzberg M, Lerner UH. Neuroendocrine regulation of cyclic AMP formation in osteoblastic cell lines (UMR-106-01, ROS 17/2.8, MC3T3-E1, and Saos-2) and primary bone cells. J Bone Miner Res 1992;7:1011—9. [66] Alam AS, Moonga BS, Bevis PJ, Huang CL, Zaidi M. Amylin inhibits bone resorption by a direct effect on the motility of rat osteoclasts. Exp Physiol 1993;78:183—96. [67] Maclntyre I. Amylinamide, bone conservation, and pancreatic beta cells. Lancet 1989;2:1026—7. [68] Reid IR, Cornish J. Amylin and CGRP. In: Bilezikian JP, Raisz LG, Rodan GA, editors. Principles of bone biology. San Diego: Academic Press; 1996. p. 495—505. [69] Tamura T, Miyaura C, Owan I, Suda T. Mechanism of action of amylin in bone. J Cell Physiol 1992;153:6—14. [70] Cornish J, Callon KE, Bava U, Kamona SA, Cooper GJ, Reid IR. Effects of calcitonin, amylin, and calcitonin gene-related peptide on osteoclast development. Bone 2001;29:162—8. [71] Wang L, Shi X, Zhao R, Halloran BP, Clark DJ, Jacobs CR, et al. Calcitonin-gene-related peptide stimulates stromal cell osteogenic differentiation and inhibits RANKL induced NF-kB activation, osteoclastogenesis and bone resorption. Bone 2010;46: 1369—79. [72] Cherruau M, Morvan FO, Schirar A, Saffar JL. Chemical sympathectomy-induced changes in TH-, VIP-, and CGRP-immunoreactive fibers in the rat mandible periosteum: influence on bone resorption. J Cell Physiol 2003;194:341—8. [73] Okajima K, Harada N, Uchiba M, Isobe H. Activation of capsaicin-sensitive sensory neurons by carvedilol, a nonselective bblocker, in spontaneous hypertensive rats. J Pharmacol Exp Ther 2004;309:684—91. [74] Valentijn K, Gutow AP, Troiano N, Gundberg C, Gilligan JP, Vignery A. Effects of calcitonin gene-related peptide on bone turnover in ovariectomized rats. Bone 1997;21:269—74. [75] Imai S, Matsusue Y. Neuronal regulation of bone metabolism and anabolism: calcitonin gene-related peptide-, substance P-, and tyrosine hydroxylase-containing nerves and the bone. Microsc Res Technol 2002;58:61—9. [76] Offley SC, Guo TZ, Wei T, Clark JD, Vogel H, Lindsey DP, et al. Capsaicin-sensitive sensory neurons contribute to the maintenance of trabecular bone integrity. J Bone Miner Res 2005;20:257—67. [77] Ding Y, Arai M, Kondo H, Togari A. Effects of capsaicin-induced sensory denervation on bone metabolism in adult rats. Bone 2010;46:1591—6.