Dramatic Decrease of Innervation Density in Bone after Ovariectomy
http://www.100md.com
内分泌学杂志 2005年第1期
Institut National de la Santé et de la Recherche Medicale Unit 403 (B.B.-P., F.D., C.I., P.D.D., C.C.), H?pital E. HERRIOT, Pavillon F, Lyon, 69437 Cedex 03, France; Faculté de Médecine (M.H.L.-P., N.L., L.V.), Laboratoire de biologie du tissu osseux, E Institut National de la Santé et de la Recherche Medicale 366, Saint-Etienne, 42023 Cedex 2, France; and Royal Veterinary College (C.C.), London NW1 OTU, United Kingdom
Address all correspondence and requests for reprints to: Dr. C. Chenu, Royal Veterinary College, Royal College street, London NW1 OTU, United Kingdom. E-mail: cchenu@rvc.ac.uk.
Abstract
Recent studies have demonstrated that bone is highly innervated and contains neuromediators that have functional receptors on bone cells. However, no data exist concerning the quantitative changes of innervation during bone loss associated with estrogen withdrawal. To study the involvement of nerve fibers in the regulation of bone remodeling, we have evaluated the modifications of innervation in a classical in vivo model of osteopenia in rats, ovariectomy (OVX). Skeletal innervation was studied by immunocytochemistry using antibodies directed against specific neuronal markers, neurofilament 200 and synaptophysin, and the neuromediator glutamate. Sciatic neurectomy, another model of bone loss due to limb denervation and paralysis, was used to validate our quantitative image analysis technique of immunostaining for nerve markers. Female Wistar rats at 12 wk of age were sham-operated (SHAM) or ovariectomized (OVX). Bone mineral density measurement and bone histomorphometry analysis of tibiae 14 d after surgery demonstrated a significant bone loss in OVX compared with SHAM. We observed an important reduction of nerve profile density in tibiae of OVX animals compared with SHAM animals, whereas innervation density in skin and muscles was similar for OVX and control rats. Quantitative image analysis of immunostainings demonstrated a significant decrease of the percentage of immunolabeling per total bone volume of neurofilament 200, synaptophysin, and glutamate in both the primary and secondary spongiosa of OVX rats compared with SHAM. These data indicate for the first time that OVX-induced bone loss in rat tibiae is associated with a reduction in nerve profile density, suggesting a functional link between the nervous system and the bone loss after ovariectomy.
Introduction
IT IS WELL DOCUMENTED that both sympathetic and sensory nerve fibers are present in periosteum and bone (1, 2, 3, 4) and form dense parallel networks around blood vessels adjacent to bone trabeculae, in close contact to bone cells (5, 6, 7). Experimental and clinical studies have shown increasing evidence for a neural control of bone development, growth, turnover, and repair (8, 9, 10, 11, 12, 13). Recently, a direct control of bone formation through a hypothalamic relay was demonstrated using leptin-deficient ob/ob mice (14). Ob/ob mice have an increased bone formation leading to high bone mass, and leptin binding to its hypothalamic receptor is sufficient to induce bone loss by decreasing osteoblastic function, revealing the central nature of bone remodeling regulation. Interestingly, these central effects of leptin on bone were recently shown to be mediated by the sympathetic nervous system (15). Similarly, neuropeptide Y2 receptors-deficient mice have a 2-fold increase in trabecular bone volume compared with control mice, and intracerebroventricular administration of neuropeptide Y also causes bone loss (16). Those animal studies suggest that bone formation is under ?-adrenergic control. A recent human population-based study also demonstrates that ?-adrenergic blockers are associated with a reduction in fracture risk and higher bone mineral density, consistent with the hypothesis that ?-adrenergic system may play a role in bone cell regulation (17). Recent work also indicated that the ?-adrenergic pathway of the sympathetic nervous system is a mediator of mechanical loading in bone (18).
Immunohistochemical studies have shown the presence in bone of a number of neuromediators conveyed by sympathetic and sensory nerve fibers. Neuropeptides including vasoactive-intestinal peptide, pituitary adenylate cyclase activating peptides, neuropeptide Y, substance P, and calcitonin gene-related peptide (CGRP), were identified in skeletal nerve axons, and their detection was confirmed by biochemical measurements after bone extraction (3, 4). Functional receptors for these neuropeptides are expressed by bone cells, and many in vitro studies have shown that neuropeptides affect their biological functions (3, 4, 19). Classical neuromediators and their functional receptors were also identified in bone, such as noradrenaline and adrenergic receptors and, recently, serotonin receptors and transporters (15, 20, 21, 22). We and others (23, 24, 25, 26, 27, 28, 29) demonstrated that bone cells express receptors for glutamate, a major neuromediator of both the central and peripheral nervous system, and that these receptors are involved in bone cell function. The presence in the vicinity of osteoblasts of axons containing glutamate and catecholamines has also been established in vivo (7, 15).
Very few in vivo studies have investigated the alterations of innervation associated with bone growth or changes in bone modeling or remodeling. An intense nerve regeneration occurs at an early stage of fracture healing, indicating that nerve supply is essential for normal fracture repair (12, 30). In addition, the alterations of bone remodeling observed after sympathectomy and sensory denervation are associated with changes in skeletal nerve fibers immunoreactive for neuropeptides (3, 8, 10). No data related to the potential relationships between bone innervation and estrogen deficiency-induced increase in bone remodeling are available. However, there is accumulated evidence that estrogen exhibits pleiotropic effects on neurons (31) and has a neuroprotective role in both central and peripheral nervous system through various signal transduction pathways. For instance, it was shown that estrogens are able to modulate neuronal apoptosis under various conditions (32, 33). In this context, our working hypothesis was that estrogen deficiency- induced bone loss might involve bone innervation. Therefore, we evaluated the changes of innervation after ovariectomy (OVX) in rats. To validate the quantitative image analysis technique of nerve marker immunostaining that we developed, we first studied bone innervation after sciatic neurectomy (SN), another bone loss condition. Expectedly, we found that limb denervation induced a significant decrease in nerve profile density in the tibiae of the paralyzed limbs. Moreover, we showed for the first time that OVX-induced bone loss in rat tibiae was associated with a specific reduction in skeletal innervation density, suggesting that neural regulation may play a role in bone loss after OVX.
Materials and Methods
Materials
The monoclonal antibodies directed against neurofilament 200 (NF200) and synaptophysin (SY) were obtained from Sigma (St. Louis, MO). The antiglutamate polyclonal antibody was purchased from Chemicon (Temecula, CA).
Animals
Male and female Wistar rats were purchased from IFFA-CREDO (L’arbresle, France). Rats were maintained at the animal research facility at the university La?nnec (Lyon, France).
The experimental protocols were carried out according to the guidelines defined by the animal welfare and ethical review committees of the Institut National de Santé et de Recherche Médicale.
Experimental designs
Ovariectomy.
Forty-seven female Wistar rats at 12 wk of age were randomly divided into two groups, one sham-operated (SHAM) group (n = 23) and one OVX group (n = 24). Ten animals in each group were designed for histomorphometry analysis and injected with calcein 3 and 9 d before killing the rats. The rats were killed 14 d after surgery by pentobarbital injection (Centravet, Dinan, France). The success of OVX was confirmed by uterus weight measurement. Left tibiae of rats dedicated to histomorphometry were dissected, measured for bone mineral density, and processed for bone histomorphometry analysis. Rats dedicated to immunocytochemistry were infused with 1% glutaraldehyde in PBS by intracardiac injection. Right tibiae, skin of the hind paws, and hindlimb muscles were immediately collected and processed for immunocytochemistry.
Sciatic neurectomy.
Twenty male Wistar rats at 4 wk of age were randomly divided into three groups, one baseline control group (n = 6) to be killed on d 0, one SHAM-operated group on the right hindlimb (n = 7), and one sciatic and crural neurectomized group (SN) on the right hindlimb (n = 7). The rats were killed 14 d after surgery by pentobarbital injection and infused with 1% glutaraldehyde in PBS by intracardiac injection. Right and left (used as contralateral controls) tibiae were dissected and cut longitudinally; one half was processed for bone histomorphometry analysis, whereas the other half was used for immunocytochemistry analysis.
Bone mineral density (BMD) measurement.
BMD of OVX and SHAM left tibiae were measured at d 0 and 14 by dual energy x-ray absorptiometry on Hologic 1000 plus (Bedford, MA), using a software modified for small animals (Regional High Resolution version 4.76; Hologic). The coefficient of variation for BMD measurements was 1.1%.
Histomorphometric analysis.
For bone histomorphometric analysis, tibiae were fixed in ethanol 70%, dehydrated in graded alcohols, and embedded undecalcified in methyl methacrylate at low temperature as previously described (34). Seven-micrometer-thick-sections were stained with Goldner’s trichrome. Measurements of the trabecular bone volume (BV/TV) expressed in percentage of the total volume, trabecular bone thickness (TbTh), trabecular bone number (TbN), and trabecular separation (TbSp) of secondary spongiosa were all carried out on Goldner’s-stained sections with Osteolab Image Analyser (Biocom, Les Ulis, France). Bone mass and microarchitectural parameters were measured on both primary and secondary spongiosae. Parameters of bone remodeling were measured on secondary spongiosae only using a semiautomatic system, as previously described (35). Osteoclast surfaces (OcS/BS) were measured on four sections stained for tartrate-resistant acid phosphatase activity. Histodynamic measurements (mineral apposition rate expressed in μm/d), mineralizing surfaces (%), and bone formation rates surface referent (expressed in μm3/μm2/d) were measured on unstained sections under UV light.
Immunocytochemistry.
Tibiae were fixed in 1% glutaraldehyde, decalcified in 4% EDTA/1% glutaraldehyde, and embedded in Epon. Semithin sections from decalcified samples were laid on sylanized slides (Dako, Carpinteria, CA) and dried overnight at 37 C. Epon was removed with 13.3% (wt/vol) potassium hydroxide in a methanol/propylene oxide (2/1) mixture and rehydrated. Sections were treated for 20 min with 100 mM glycine and 50 mM ammonium chloride in Tris buffer (pH 7.6) to saturate free aldehydic groups. Endogenase peroxidase activity was inhibited by incubation for 15 min with 1% sodium azide and 1.5% H2O2 in 50% methanol, and nonspecific immunoreactions were blocked for 30 min in 10% normal goat serum in Tris-buffered saline (TBS) containing 0.01% BSA. Sections were incubated overnight with specific antibodies directed against nerve markers diluted in TBS containing 1% normal goat serum: NF200 (30 μg/ml), SY (6 μg/ml), and glutamate (Glu; 1/1000). Control sections were incubated with the monoclonal antibody MOPC21, which has no known hapten or antigen binding activity. Antigen-antibody complexes were detected with the EnVision/HRP System (Dako) and were revealed with 3-3'-diaminobenzidine tetrahydrochloride (Dako) in Tris buffer containing 0.01% H2O2. Sections were counterstained with Meyer’s hemalum, dehydrated, and mounted in Xam (Gurr-BDH Laboratory, Poole, UK).
Quantitative image analysis of immunostaining for these markers was performed in primary and secondary spongiosa of rat tibiae using Osteolab (Biocom). After image acquisition of immunostaining, an immunolabeling memorized color threshold was applied together with a constant smooth filter to remove immunostaining noise background. Measurement of detected area was performed after calibration. These different steps are illustrated in Fig. 1. Results are expressed as percentages of immunolabelings per total bone volume. The coefficients of variation for nerve marker measurements were 8.1, 11.0, and 7.2% for NF200, SY, and Glu, respectively. Because Glu is expressed both by nerve fibers in bone and by osteoblastic cells (7, 36), the analysis of Glu-immunoreactive nerve fibers was performed after the exclusion in each field of the immunostaining for Glu in osteoblastic cells along bone trabeculae.
FIG. 1. Quantification of immunostaining for nerve markers. A, Image acquisition of NF200 immunostaining in primary spongiosa of control tibiae. Immunoreactive nerve fibers were predominantly nonvascular, identified in the bone marrow closed to bone trabeculae. B, An immunolabeling memorized color threshold was applied to the image, together with a constant smooth filter to remove immunostaining noise background. C, After calibration, detected areas corresponding exactly to the immunostaining observed for NF200 (A) were measured.
Skin and muscle were fixed with 4% paraformaldehyde for 48 h and embedded in paraffin. Paraffin sections (7 μm) were laid on sylanized slides and dried overnight at 37 C before dewaxing with methylcyclohexane followed by rehydration. Sections were then treated as described for tibiae.
Statistical analyses
BMD measurement analyses were carried out using SPSS 10 software (SPSS Inc., Chicago, IL). Other analyses were performed using Instat software (GaphPad, San Diego, CA). The significance of differences in BMD, percentages of immunostainings, and quantitative histomorphometric parameters between control and treatment groups was determined based on nonparametric Wilcoxon’s or Mann-Whitney U tests.
Results
Effect of sciatectomy on tibiae innervation
To confirm that denervation was able to induce changes in skeletal nerve fibers that could be quantified by immunocytochemistry using antibodies directed against nerve fibers, we studied innervation in tibial semi-thin sections of SN limb and a contralateral intact section immunostained for NF200 and SY, two specific markers of nerve fibers. NF200 is specific of neurofilaments present in nerve cells, whereas SY is a nerve terminal marker. Two cancellous bone regions of the metaphysis were studied: the primary spongiosa, which is the trabecular bone formed close to the growth plate following the resorption of calcified endochondral cartilage; and the secondary spongiosa, which arises from remodeled primary spongiosa. We observed a dramatic reduction of immunoreactivity for both markers in the primary spongiosa of SN tibiae compared with the contralateral ones (Fig. 2). A less striking decrease of nerve profile density was observed in secondary spongiosa of SN tibiae compared with contralateral controls (data not shown). Quantitative image analysis of immunostaining for NF200 and SY demonstrated a significant decrease in the percentage of immunolabeling per total bone volume for both NF200 (P = 0.01) and SY (P = 0.01) in the primary spongiosa of SN tibiae compared with contralateral controls (Fig. 3A). Trend to decrease for these markers was also observed in secondary spongiosa of SN tibiae, but it did not reach statistical significance (Fig. 3B). This decrease of nerve fiber density in tibiae after SN indicates that changes in innervation can be quantified in tibiae with this technique, thus allowing studies on modifications of innervation when bone remodeling is affected.
FIG. 2. Immunolabellings for nerve markers on semithin sections from neurectomized and contralateral intact tibiae. Primary spongiosa of tibiae from neurectomized (SN) and contralateral intact limbs, immunostained for NF200 and SY. No labeling was observed in control sections incubated with the monoclonal antibody MOPC21. Original magnification, x400.
FIG. 3. Quantitative analysis of immunostaining for nerve markers in tibiae from neurectomized and contralateral intact limbs. Quantitative image analysis of immunostaining for NF200 and SY in primary (A) and secondary (B) spongiosa of tibiae from neurectomized (SN) or contralateral intact limbs was performed using Osteolab software. Immunostainings were quantified in 10 consecutive fields for both tibial primary and secondary spongiosa, and results are expressed in percentage of immunolabelings per total bone volume. Data are means ± SD (n = 6), *, Significantly different from contralateral intact tibiae, P < 0.05, Wilcoxon’s test).
Effects of ovariectomy on BMD
SHAM and OVX groups of female rats had similar weights before surgery. Two weeks after surgery, the uterine weight was strikingly lower in every OVX animal compared with SHAM, confirming their estrogen deficiency. BMD was significantly lower (–4%) in tibiae of OVX rats (0.287 g/cm2) compared with SHAM (0.296 g/cm2) 14 d after ovariectomy (P < 0.005, Mann-Whitney U test).
Bone histomorphometric analysis of OVX rat tibiae
Histomorphometric analysis of rat tibiae showed a significant reduction of BV/TV (–29%) in secondary spongiosa of OVX rats compared with SHAM (P = 0.0012, Mann-Whitney U test) 14 d after ovariectomy (Table 1). This was associated with a significant decrease in trabecular number, whereas no change in trabecular thickness was observed yet. As expected, OVX stimulated bone resorption activity, as assessed by active resorption surfaces using osteoclast tartrate-resistant acid phosphatase staining (Table 1). Dynamic parameters measurements showed the expected significant increase in both mineral apposition rate (+33%) and bone formation rates surface referent (+200%) after OVX, reflecting the stimulation of bone formation activity induced by OVX.
TABLE 1. Bone histomorphometry parameters in the secondary spongiosa of tibial metaphysis after 14 d of OVX
Effect of OVX on tibiae innervation
To analyze the involvement of bone innervation in a model of bone loss associated with increased bone remodeling due to estrogen deficiency, we studied innervation in rat tibiae after OVX. We observed a substantial reduction of nerve profiles immunolabeled for NF200, SY, and Glu, a major neuromediator involved in bone metabolism (23), in the tibial primary and secondary spongiosae of OVX rats compared with SHAM (Fig. 4). Quantitative image analysis of immunostaining for nerve markers demonstrated a significant decrease in the percentage of immunolabeling of NF200, SY, and Glu in both the primary and secondary spongiosae of OVX rats compared with SHAM (Fig. 5, A and B). Similar results were observed when the percentages of immunolabeling for these markers were expressed per total bone volume or per bone marrow volume (data not shown), indicating that the dramatic decrease of innervation observed after OVX was not dependent on a reduction of the number of bone trabeculae.
FIG. 4. Immunolabellings for nerve markers on tibial semithin sections from OVX and SHAM rats. Primary spongiosa of tibiae from ovariectomized (OVX) and SHAM rats, immunostained for NF200, SY, and Glu. Original magnification, x400.
FIG. 5. Quantitative analysis of immunostaining for nerve markers in tibiae from OVX and SHAM rats. Quantitative image analysis of immunostaining for NF200, SY, and Glu in primary (A) and secondary (B) spongiosa of tibiae from OVX or SHAM rats was performed using Osteolab software. Immunostainings were quantified in 10 consecutive fields for both tibial primary and secondary spongiosae, and results are expressed in percentage of immunolabelings per total bone volume. Data are means ± SD of 13 SHAM and 14 OVX rats. Significantly different from the SHAM group (*, P < 0.05; **, P < 0.01; ***, P < 0.001, Mann-Whitney U test).
Effect of OVX on muscle and skin innervation
To investigate whether the decrease in innervation observed after OVX is specific for bone and not a ubiquitous consequence of estrogen depletion, we studied innervation in the skin of rat hind paws and in the hindlimb muscles. We did not observe any difference in nerve profiles immunolabeled for NF200 in the skin (data not shown) or muscle of OVX rats compared with SHAM (Fig. 6). Quantitative image analysis of NF200 immunostaining demonstrated no significant difference in the percentage of immunolabeling for NF200 in muscle of OVX rats compared with SHAM (not shown).
FIG. 6. Immunolabelling for NF200 of muscle sections from OVX and SHAM rats. Muscle sections from OVX and SHAM rats immunostained for the nerve marker NF200 (A and C). No labeling was observed in control sections incubated with the monoclonal antibody MOPC21 (B and D). Original magnification, x250 (A and B); x320 (C and D).
Discussion
This study clearly shows evidence that skeletal innervation is decreased after OVX, suggesting a link between innervation and the bone loss induced by OVX. Previous work by our group (7) has shown that bone is rich in nerve fibers localized in the vicinity of bone cells, indicating a possible role of nerve fibers in bone cell metabolism. Although numerous in vitro studies have shown roles for neuromediators in bone cell functions (3, 4, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29), the relationship between the presence of these nerve fibers and bone cell metabolism has been poorly investigated in vivo. We first used a model of denervation-induced bone loss, sciatic neurectomy, to investigate whether changes in skeletal nerve fibers could be quantified in bone. Then, we analyzed the alterations of bone innervation after ovariectomy to check whether estrogen deficiency, known to induce increase in bone remodeling and to modulate neuronal functions (37), could affect nerve density in bone.
The tibia can be denervated by surgical sciatic neurectomy, a largely used model of bone loss under less mechanical stress. Bone deficit in this model is caused by a significant early increase in bone resorption and by depressed bone formation as a result of impaired osteoblast activity (38). In our study, we chose to investigate the effects of denervation in young 4-wk-old rats, because we had previously demonstrated that innervation is abundant in bone of young growing rats (7), and our first aim was to quantify potential changes in bone innervation occurring after sciatectomy. We used immunocytochemistry with antibodies directed against two specific neuronal markers, NF200 and SY. NF200 is specific of neurofilaments that are structural proteins found in all nerve processes. There are the major proteins found in myelinated sensory neurons but are also expressed in sympathetic neurons. SY is one of the most abundant membrane proteins of small synaptic vesicles, so it is expressed in all nerve terminals. Therefore, NF200 and SY are general nerve markers and can’t discriminate between sympathetic and sensory nerve fibers. As previously illustrated, immunoreactivity for NF200 and SY was found in thin cell processes in the proximity of blood vessels and bone cells, and both markers showed a similar distribution in bone (7). We observed a marked decrease in innervation in the tibial primary spongiosa of neurectomized limbs compared with the contralateral ones, whereas only a mild decrease in innervation was observed in secondary spongiosa, indicating that sciatectomy in young growing rats primarily affects the tibial modeling zone. These results might explain the fact that neurectomy mainly alters the regions of bone modeling, as previously reported in rapidly growing young rats (39) as well as in slowly growing mature rats (40). Complete denervation of the long bones was not achieved after sciatectomy, as, in addition to sciatic nerve, other nerves and their branches supply nerve fibers to the hindlimbs (11). After quantification of SY and NF200 immunolabelings per total bone volume, we demonstrated a 50% reduction of bone innervation in the primary spongiosa of sciatectomized limbs, whereas the decrease in the secondary spongiosa did not reach statistical significance. Our results demonstrate that innervation is decreased in tibiae after sciatectomy and confirm that our quantitative image analysis technique of immunostaining for nerve markers is adequate for measuring changes in skeletal innervation.
Estrogen deficiency in the model of OVX rats induces an increase in bone turnover with a rate of bone resorption exceeding that of bone formation, resulting in the reduction of bone mass (41). OVX resulted in significant rapid decrease in BMD within 14 days. This bone loss was illustrated at the trabecular level by a significant decrease in BV/TV and number of trabeculae in secondary spongiosae. As expected, the OVX-induced increase in bone cellular activities associated both increased active osteoclast surfaces and bone formation parameters.
So far, no study has yet investigated bone innervation after OVX. This model of bone loss was chosen for two reasons. First, it is the most widely used model in the bone field when studying high remodeling-induced bone loss; second, we wanted to study innervation because estrogen has many effects on brain neurons and on the autonomic nervous system (42, 43). Surprisingly, our results demonstrate a dramatic decrease of innervation in both primary and secondary spongiosae of OVX rat tibiae compared with SHAM. This decrease was not related to the important reduction of trabecular bone volume after OVX, as we observed a similar decrease of innervation when normalized by bone marrow or bone volume. Whether the decrease in innervation is the cause or the consequence of the increase in bone remodeling is not yet known, and kinetic studies of innervation after OVX should be performed to answer this question. The fact that no decrease of innervation was observed in muscle and skin after OVX indicates that decrease of innervation after OVX is specific to bone. Interestingly, several authors reported site-specific effects of estrogen deficiency on various neuronal functions. Excitatory or inhibitory responses of neurons in the gracile nucleus vary after OVX as a function of the tissular origin of the stimulus (44). In a different field, nociception seems more evident in the orofacial area than in many other parts of the body in post menopausal women. Pajot et al. (45) showed that gonadectomy induces site-specific differences in nociception in rats, with higher sensitivity of the upper lip region compared with hind paws. Finally, there is now evidence that estrogen specifically modulates sympathetic innervation in the uterus, with a decrease in nerve density induced by estrogen deficiency and restoration of innervation under estrogen supplementation in OVX rats and mice (46). After uterus and breast, bone is the third major target of estrogens; however, our study is the first to demonstrate that estrogen depletion also affects innervation in the skeleton. Whether estrogen regulates the number of skeletal nerves through direct action on sympathetic and sensory ganglions or indirectly is unknown. In uterus, estrogen modulates sympathetic neurite outgrowth by regulating neurotrophic factors synthesis such as nerve growth factor and brain-derived neurotrophic factor (47), presumably through interactions with estrogen response elements that are present in the promoter regions of these genes (48, 49, 50). Interestingly, these neurotropins are expressed by osteoblasts and we could hypothesize that a similar regulatory mechanism exists in bone (51). Expression of estrogen receptor (ER) or ER? may also explain the specific effects of estrogen deficiency on bone innervation. Bone, skin, and muscles exhibit specific ER and ER? expression pattern and regulation. Although both receptors play important roles in bone metabolism in females, ER? is the main mediator of estrogen action in skin, and ER is highly expressed in muscles (52, 53, 54). Interestingly, ER deficiency sensibly affects uterus innervation in mice (46).
The demonstration that OVX induces a specific bone loss associated with a marked reduction of nerve profile density in both modeling and remodeling zones of tibiae suggests an association between bone remodeling and innervation. Several studies have previously demonstrated that innervation may control local bone turnover. Selective denervations of either sensory or sympathetic nerves were shown to affect bone remodeling (8, 10, 55). Increased osteoclast formation was observed in long-term bone marrow cultures from patients with a spinal cord injury, which might be attributed to a deficit in CGRP (56, 57). Contacts between osteoclasts and CGRP-positive nerve fibers were observed at the epiphyseal trabeculae facing the growth plate (6), and in vitro experiments have demonstrated that several neuropeptides regulate osteoclast and osteoblast activities (4). Recently, the concept that osteoporosis might in part originate from the hypothalamus has been proposed following the demonstration of a central control of bone formation via the peripheral sympathetic nervous system (14, 15, 16). Furthermore, it was recently suggested that the ?-adrenergic pathway of the sympathetic nervous system is a potential mediator of mechanical loading in bone (18). Our findings that demonstrate a decrease in bone innervation after OVX suggest that neural regulation could also modulate bone remodeling and play a role in postmenopausal osteoporosis. All these observations were, however, made in trabecular bone, and it is possible that the differences of innervation in the various bones of the skeleton and in the different envelopes within a bone may result in diverse responses to estrogen deficiency. Recent work (58) indeed indicates that leptin deficiency produces contrasting phenotypes in bones of the limb and spine.
Among the broad range of neurotransmitters present in bone, Glu has profound effects on the skeleton and may be involved in the neural control of bone remodeling. We showed that Glu contributes to the local regulation of bone resorption via NMDA Glu receptors (23, 24, 27). Because accumulating evidence suggests that estrogen may exert its effect in the brain through the regulation of these receptors (59), we therefore studied glutamatergic innervation after OVX. Because Glu itself is expressed in osteoblasts and their precursors as well as in nerve processes in the vicinity of bone cells (7, 36), we excluded immunolabeling for Glu in osteoblastic cells along bone trabeculae to quantify glutamatergic innervation. We found that the percentage of immunolabeling per total bone volume for Glu was similar to those determined for NF200 and SY, suggesting that the majority of nerve fibers in bone express Glu. Nevertheless, although Glu-containing osteoblasts were excluded, Glu is also expressed by some bone marrow cells, suggesting a possible contribution of these cells to Glu immunolabeling quantification. We observed a dramatic decrease of the percentage of immunolabeling for Glu in both the primary and secondary spongiosa of OVX rats compared with SHAM, suggesting that glutamatergic innervation may play a role in the bone loss induced after OVX. As demonstrated for neurons that innervate genital organs (42), it is possible that the glutamate-containing neurons, which innervate bone, express estrogen receptors. Whereas NF200 and SY are general nerve markers, Glu is the major neuromediator released by sensory nerves. The demonstration that the percentage of immunolabeling for Glu decreases after OVX suggests that sensory denervation may be primarily affected by OVX. However, we cannot exclude that sympathetic innervation is also decreased after OVX, as both sympathetic and sensory innervation are regulated by estrogens (45, 46). Additional studies, using specific markers of the sympathetic and the sensory nervous system, are necessary to investigate which nerve cell populations are most affected by OVX.
Our findings, which demonstrate a dramatic decrease of skeletal innervation after OVX, suggest that neural regulation may play a role in the bone loss during osteoporosis. These data support the increased evidence for a major role of innervation in the local control of skeletal metabolism. Clearly, additional studies are required to better understand the mechanisms linking bone remodeling to innervation.
Acknowledgments
We thank Annie Desenfants for animal care and Fran?oise Munoz and Pavel Szulc for helpful comments on statistical analyses.
References
Hill EL, Elde R 1991 Distribution of CGRP-, VIP-, D?H-, SP-, and NPY-immunoreactive nerves in the periosteum of the rat. Cell Tissue Res 264: 469–480
Hohmann EL, Elde R, Rysavy J, Einzig S, Gebhard R 1986 Innervation of periosteum and bone by sympathetic vasoactive intestinal peptide-containing nerve fibers. Science 232:869–887
Konttinen Y, Imai S, Suda A 1996 Neuropeptides and the puzzle of bone remodeling. Acta Orthop Scand 67:632–639
Lerner UH, Lundberg P 2002 Kinins and neuro-osteogenic factors. In: Bilezikian JP, Raisz LG, and Rodan GA, eds. Principles of bone biology. 2nd ed. San Diego: Academic Press; 773–799
Goto T, Yamaza T, Kido MA, Tanaka T 1998 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
Hara-Hirie F, Amizuka N, Ozawa H 1996 Immunohistochemical and ultrastructural localization of CGRP-positive nerve fibers at the epiphyseal trabecules facing the growth plate of rat femurs. Bone 18:29–39
Serre CM, Farlay D, Delmas PD, Chenu C 1999 Evidence for a dense and intimate innervation of the bone tissue, including glutamate-containing fibers. Bone 25:623–629
Cherruau M, Facchinetti P, Baroukh B, Saffar JL 1999 Chemical sympathectomy impairs bone resorption in rats: a role for the sympathetic system on bone metabolism. Bone 25:545–551
Edoff K, Hellman J, Persliden J, Hildebrand C 1997 The developmental skeletal growth in the rat foot is reduced after denervation. Anat Embryol 195:531–538
Hill EL, Turner R, Elde R 1991 Effects of neonatal sympathectomy and capsaicin treatment on bone remodeling in rats. Neuroscience 44:747–755
Hukkanen M, Konttinen YT, Santavirta S, Nordsletten L, Madsen JE, Almaas R, Oestreicher AB, Rootwelt T, Polak JM 1995 Effect of sciatic nerve section on neural ingrowth into the rat tibial fracture callus. Clin Orthop 311:247–257
Li J, Ahmad T, Spetea M, Ahmed M, Kreicbergs A 2001 Bone reinnervation after fracture: a study in the rat. J Bone Miner Res 16:1505–1510
Ramnemark A, Nyberg L, Lorentzon R, Englund U, Gustafson Y 1999 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
Ducy P, Amling M, Takeda S, Priemel M, Schilling AF, Beil FT, Shen J, Vinson C, Rueger JM, Karsenty G 2000 Leptin inhibits bone formation through a hypothalamic relay: a central control of bone mass. Cell 100:197–207
Takeda S, Elefteriou F, Levasseur R, Liu X, Zhao L, Parker KL, Amstrong D, Ducy P, Karsenty G 2002 Leptin regulates bone formation via the sympathetic nervous system. Cell 111:305–317
Baldock PA, Sainsbury A, Couzens M, Enriquez RF, Thomas GP, Gardiner EM, Herzog H 2002 Hypothalamic Y2 receptors regulate bone formation. J Clin Invest 109:915–921
Pasco JA, Henry MJ, Sanders KM, Kotowicz MA, Seeman E, Nicholson GC 2004 ?-adrenergic blockers reduce the risk of fractures partly by increasing bone mineral density: Geelong osteoporosis study. J Bone Miner Res 19:19–24
Levasseur R, Sabatier JP, Potrel-Burgot C, Lecoq B, Creveuil C, Marcelli C 2003 Sympathetic nervous system as transmitter of mechanical loading in bone. Joint Bone Spine 70:515–519
Lundberg P, Lie A, Bjurholm A, Lehenkari PP, Horton MA, Lerner UH, Ransj? M 2000 Vasoactive intestinal peptide regulates osteoclast activity via specific binding sites on both osteoclasts and osteoblasts. Bone 27:803–810
Bliziotes MM, Eshleman AJ, Zhang XW, Wiren KM 2001 Neurotransmitter action in osteoblasts: expression of a functional system for serotonin receptor activation and reuptake. Bone 29:477–486
Susiki A, Palmer G, Bonjour JP, Caverzasio J 1998 Catecholamines stimulate the proliferation and alkaline phosphatase activity of MC3T3–E1 osteoblast-like cells. Bone 23:197–203
Wesbroek I, Van der Plas A, de Rooij KE, Klein-Nulends J, Nijweide PJ 2001 Expression of serotonin receptors in bone. J Biol Chem 276:28961–28968
Chenu C, Serre CM, Raynal C, Burt-Pichat B, Delmas PD 1998 Glutamate receptors are expressed by bone cells and are involved in bone resorption. Bone 22:295–299
Espinosa L, Itzstein C, Cheynel H, Delmas PD, Chenu C 1999 Active NMDA glutamate receptors are expressed by mammalian osteoclasts. J Physiol 518:47–53
Gu Y, Publicover S 2000 Expression of functional metabotropic glutamate receptors in primary cultured rat osteoblasts. J Biol Chem 275:34252–34259
Itzstein C, Cheynel H, Burt-Pichat B, Merle B, Espinosa L, Delmas PD, Chenu C 2001 Molecular identification of NMDA glutamate receptors expressed in bone cells. J Cell Biochem 82:134–144
Itzstein C, Espinosa L, Delmas PD, Chenu C 2000 Specific antagonists of NMDA receptors prevent osteoclast sealing zone formation required for bone resorption. Biochem Biophys Res Commun 268:201–209
Patton AJ, Genever PG, Birch MA, Suva LJ, Skerry TM 1998 Expression of N-methyl-D-aspartate-type receptor by human and rat osteoblasts and osteoclasts suggests a novel glutamate signaling pathway in bone. Bone 22:645–649
Peet NM, Grabowski PS, Laketic-Ljubojevic I, Skerry TM 1999 The glutamate receptor antagonist MK801 modulates bone resorption in vitro by a mechanism predominantly involving osteoclast differentiation. FASEB J 13:2179–2185
Hukkanen M, Konttinen YT, Santavirta S, Paavolainen P, Gu XH, Terenghi G, Polak JM 1993 Rapid proliferation of calcitonin gene-related peptide-immunoreactive nerves during healing of rat tibial fracture suggests neural involvement in bone growth and remodelling. Neuroscience 54:969–979
Haynes LE, Lendon CL, Barber DJ, Mitchell IJ 2003 17?-Oestradiol attenuates dexamethasone-induced lethal and sublethal neuronal damage in the striatum and hippocampus. Neuroscience 120:799–806
Koski CL, Hila S, Hoffman GE 2004 Regulation of cytokine-induced neuron death by ovarian hormones: involvement of antiapoptotic protein expression and c-JUN N-terminal kinase-mediated proapoptotic signaling. Endocrinology 145:95–103
Yu X, Rajala RV, McGinnis JF, Li F, Anderson RE, Yan X, Li S, Elias RV, Knapp RR, Zhou X, Cao W 2004 Involvement of insulin/phosphoinositide 3-kinase/Akt signal pathway in 17?-estradiol-mediated neuroprotection. J Biol Chem 279:13086–13094
Chappard D, Palle S, Alexandre C, Vico L, Riffat G 1987 Bone embedding in pure methyl methacrylate at low temperature preserves enzyme activities. Acta Histochem 81:175–183
Chappard D, Alexandre C, Riffat G 1983 Histochemical identification of osteoclasts. Review of current method and reappraisal of simple procedure for routine diagnosis on undecalcified human iliac bone biopsies. Basic Appl Histochem 27:75–85
Bhangu PS, Genever PG, Spencer GJ, Grewal TS, Skerry TM 2001 Evidence for targeted vesicular glutamate exocytosis in osteoblasts. Bone 29:16–23
Maggi A, Ciana P, Belcredito S, Vegeto E 2004 Estrogens in the nervous system: mechanisms and nonreproductive functions. Annu Rev Physiol 66:291–313
Weinreb M, Rodan GA, Thompson DD 1991 Depression of osteoblastic activity in immobilized limbs of suckling rats. J Bone Miner Res 6:725–731
Iwamoto J, Takeda T, Ichumira S, Sato Y, Yeh JK 2003 Comparative effects of orchidectomy and sciatic neurectomy on cortical and cancellous bone in young growing rats. J Bone Miner Metab 21:211–216
Shen V, Liang XG, Birchman R, Wu DD, Healy D, Lindsay R, Dempster DW 1997 Short-term immobilisation-induced cancellous bone loss is limited to regions undergoing high turnover and/or modeling in mature rats. Bone 21:71–78
Kalu DN 1991 The ovariectomized rat model of postmenopausal bone loss. Bone Miner 15:175–191
Papka RE, Storey-Workley M, Shughrue PJ, Merchenthaler I, Collins JJ, Usip S, Saunders PT, Shupnik M 2001 Estrogen receptor- and ?-immunoreactivity and mRNA in neurons of sensory and autonomic ganglia and spinal cord. Cell Tissue Res 304:193–214
Wickelgren I 1997 Estrogen stakes claim to cognition. Science 276:675–678
Bradshaw HB, Berkley KJ 2003 The influence of ovariectomy with or without estrogen replacement on responses of rat gracile nucleus neurons to stimulation of hindquarter skin and pelvic viscera. Brain Res 986:82–90
Pajot J, Ressot C, Ngom I, Woda A 2003 Gonadectomy induces site-specific differences in nociception in rats. Pain 104:367–373
Zoubina EV, Smith PG 2001 Sympathetic hyperinnervation of the uterus in the estrogen receptor knock-out mouse. Neuroscience 103:237–244
Krizsan-Agbas D, Smith PG 2002 Estrogen modulates myometrium-induced sympathetic neurite formation through actions on target and ganglion. Neuroscience 114:339–347
Sohrabji F, Miranda RC, Toran-Allerand CD 1995 Identification of a putative estrogen response element in the gene encoding brain-derived neurotrophic factor. Proc Natl Acad Sci USA 92:11110–11114
Gibbs RB 1998 Levels of trkA and BDNF mRNA, but not NGF mRNA, fluctuate across the estrous cycle and increase in response to acute hormone replacement. Brain Res 787:259–268
Siminoski K, Murphy RA, Rennert P, Heinrich G 1987 Cortisone, testosterone, and aldosterone reduce levels of nerve growth factor messenger ribonucleic acid in L-929 fibroblasts. Endocrinology 121:1432–1437
Togari A 2002 Adrenergic regulation of bone metabolism: possible involvement of sympathetic innervation of osteoblastic and osteoclastic cells. Microsc Res Tech 58:77–84
Sims NA, Dupont S, Krust A, Clement-Lacroix P, Minet D, Resche-Rigon M, Gaillard-Kelly M, Baron R 2002 Deletion of estrogen receptors reveals a regulatory role for estrogen receptors-? in bone remodeling in females but not in males. Bone 30:18–25
Thornton MJ, Taylor AH, Mulligan K, Al-Azzawi F, Lyon CC, O’Driscoll J, Messenger AG 2003 The distribution of estrogen receptor ? is distinct to that of estrogen receptor and the androgen receptor in human skin and the pilosebaceus unit. J Investig Dermatol Symp Proc 8:100–103
Pfaffl MW, Lange IG, Daxenberger A, Meyer HHD 2001 Tissue-specific expression pattern of estrogen receptors (ER): quantification of ER and ER? mRNA with real-time RT-PCR. APMIS 109:345–355
Sherman BE, Chole RA 2000 Sympathectomy which induces membranous bone remodeling, has no effect on endochondral long bone remodeling in vivo. J Bone Miner Res 15:1354–1360
Akopian A, Demulder A, Ouriaghli F, Corazza F, Fondu P, Bergmann P 2000 Effects of CGRP on human osteoclast-like cell formation: a possible connection with the bone loss in neurological disorders? Peptides 21:559–564
Demulder A, Guns M, Ismail A, Wilmet E, Fondu P, Bergmann P 1998 Increased osteoclast-like cells formation in long-term bone marrow cultures from patients with a spinal cord injury. Calcif Tissue Int 63:396–400
Hamrick MW, Pennington C, Newton D, Xie D, Isales C 2004 Leptin deficiency produces contrasting phenotypes in bones of the limb and spine. Bone 34:376–383
Watanabe T, Inoue S, Hiroi H, Orimo A, Muramatsu M 1999 NMDA receptor type 2D gene as a target for estrogen receptor in the brain. Mol Brain Res 63:375–379(B. Burt-Pichat, M. H. Laf)
Address all correspondence and requests for reprints to: Dr. C. Chenu, Royal Veterinary College, Royal College street, London NW1 OTU, United Kingdom. E-mail: cchenu@rvc.ac.uk.
Abstract
Recent studies have demonstrated that bone is highly innervated and contains neuromediators that have functional receptors on bone cells. However, no data exist concerning the quantitative changes of innervation during bone loss associated with estrogen withdrawal. To study the involvement of nerve fibers in the regulation of bone remodeling, we have evaluated the modifications of innervation in a classical in vivo model of osteopenia in rats, ovariectomy (OVX). Skeletal innervation was studied by immunocytochemistry using antibodies directed against specific neuronal markers, neurofilament 200 and synaptophysin, and the neuromediator glutamate. Sciatic neurectomy, another model of bone loss due to limb denervation and paralysis, was used to validate our quantitative image analysis technique of immunostaining for nerve markers. Female Wistar rats at 12 wk of age were sham-operated (SHAM) or ovariectomized (OVX). Bone mineral density measurement and bone histomorphometry analysis of tibiae 14 d after surgery demonstrated a significant bone loss in OVX compared with SHAM. We observed an important reduction of nerve profile density in tibiae of OVX animals compared with SHAM animals, whereas innervation density in skin and muscles was similar for OVX and control rats. Quantitative image analysis of immunostainings demonstrated a significant decrease of the percentage of immunolabeling per total bone volume of neurofilament 200, synaptophysin, and glutamate in both the primary and secondary spongiosa of OVX rats compared with SHAM. These data indicate for the first time that OVX-induced bone loss in rat tibiae is associated with a reduction in nerve profile density, suggesting a functional link between the nervous system and the bone loss after ovariectomy.
Introduction
IT IS WELL DOCUMENTED that both sympathetic and sensory nerve fibers are present in periosteum and bone (1, 2, 3, 4) and form dense parallel networks around blood vessels adjacent to bone trabeculae, in close contact to bone cells (5, 6, 7). Experimental and clinical studies have shown increasing evidence for a neural control of bone development, growth, turnover, and repair (8, 9, 10, 11, 12, 13). Recently, a direct control of bone formation through a hypothalamic relay was demonstrated using leptin-deficient ob/ob mice (14). Ob/ob mice have an increased bone formation leading to high bone mass, and leptin binding to its hypothalamic receptor is sufficient to induce bone loss by decreasing osteoblastic function, revealing the central nature of bone remodeling regulation. Interestingly, these central effects of leptin on bone were recently shown to be mediated by the sympathetic nervous system (15). Similarly, neuropeptide Y2 receptors-deficient mice have a 2-fold increase in trabecular bone volume compared with control mice, and intracerebroventricular administration of neuropeptide Y also causes bone loss (16). Those animal studies suggest that bone formation is under ?-adrenergic control. A recent human population-based study also demonstrates that ?-adrenergic blockers are associated with a reduction in fracture risk and higher bone mineral density, consistent with the hypothesis that ?-adrenergic system may play a role in bone cell regulation (17). Recent work also indicated that the ?-adrenergic pathway of the sympathetic nervous system is a mediator of mechanical loading in bone (18).
Immunohistochemical studies have shown the presence in bone of a number of neuromediators conveyed by sympathetic and sensory nerve fibers. Neuropeptides including vasoactive-intestinal peptide, pituitary adenylate cyclase activating peptides, neuropeptide Y, substance P, and calcitonin gene-related peptide (CGRP), were identified in skeletal nerve axons, and their detection was confirmed by biochemical measurements after bone extraction (3, 4). Functional receptors for these neuropeptides are expressed by bone cells, and many in vitro studies have shown that neuropeptides affect their biological functions (3, 4, 19). Classical neuromediators and their functional receptors were also identified in bone, such as noradrenaline and adrenergic receptors and, recently, serotonin receptors and transporters (15, 20, 21, 22). We and others (23, 24, 25, 26, 27, 28, 29) demonstrated that bone cells express receptors for glutamate, a major neuromediator of both the central and peripheral nervous system, and that these receptors are involved in bone cell function. The presence in the vicinity of osteoblasts of axons containing glutamate and catecholamines has also been established in vivo (7, 15).
Very few in vivo studies have investigated the alterations of innervation associated with bone growth or changes in bone modeling or remodeling. An intense nerve regeneration occurs at an early stage of fracture healing, indicating that nerve supply is essential for normal fracture repair (12, 30). In addition, the alterations of bone remodeling observed after sympathectomy and sensory denervation are associated with changes in skeletal nerve fibers immunoreactive for neuropeptides (3, 8, 10). No data related to the potential relationships between bone innervation and estrogen deficiency-induced increase in bone remodeling are available. However, there is accumulated evidence that estrogen exhibits pleiotropic effects on neurons (31) and has a neuroprotective role in both central and peripheral nervous system through various signal transduction pathways. For instance, it was shown that estrogens are able to modulate neuronal apoptosis under various conditions (32, 33). In this context, our working hypothesis was that estrogen deficiency- induced bone loss might involve bone innervation. Therefore, we evaluated the changes of innervation after ovariectomy (OVX) in rats. To validate the quantitative image analysis technique of nerve marker immunostaining that we developed, we first studied bone innervation after sciatic neurectomy (SN), another bone loss condition. Expectedly, we found that limb denervation induced a significant decrease in nerve profile density in the tibiae of the paralyzed limbs. Moreover, we showed for the first time that OVX-induced bone loss in rat tibiae was associated with a specific reduction in skeletal innervation density, suggesting that neural regulation may play a role in bone loss after OVX.
Materials and Methods
Materials
The monoclonal antibodies directed against neurofilament 200 (NF200) and synaptophysin (SY) were obtained from Sigma (St. Louis, MO). The antiglutamate polyclonal antibody was purchased from Chemicon (Temecula, CA).
Animals
Male and female Wistar rats were purchased from IFFA-CREDO (L’arbresle, France). Rats were maintained at the animal research facility at the university La?nnec (Lyon, France).
The experimental protocols were carried out according to the guidelines defined by the animal welfare and ethical review committees of the Institut National de Santé et de Recherche Médicale.
Experimental designs
Ovariectomy.
Forty-seven female Wistar rats at 12 wk of age were randomly divided into two groups, one sham-operated (SHAM) group (n = 23) and one OVX group (n = 24). Ten animals in each group were designed for histomorphometry analysis and injected with calcein 3 and 9 d before killing the rats. The rats were killed 14 d after surgery by pentobarbital injection (Centravet, Dinan, France). The success of OVX was confirmed by uterus weight measurement. Left tibiae of rats dedicated to histomorphometry were dissected, measured for bone mineral density, and processed for bone histomorphometry analysis. Rats dedicated to immunocytochemistry were infused with 1% glutaraldehyde in PBS by intracardiac injection. Right tibiae, skin of the hind paws, and hindlimb muscles were immediately collected and processed for immunocytochemistry.
Sciatic neurectomy.
Twenty male Wistar rats at 4 wk of age were randomly divided into three groups, one baseline control group (n = 6) to be killed on d 0, one SHAM-operated group on the right hindlimb (n = 7), and one sciatic and crural neurectomized group (SN) on the right hindlimb (n = 7). The rats were killed 14 d after surgery by pentobarbital injection and infused with 1% glutaraldehyde in PBS by intracardiac injection. Right and left (used as contralateral controls) tibiae were dissected and cut longitudinally; one half was processed for bone histomorphometry analysis, whereas the other half was used for immunocytochemistry analysis.
Bone mineral density (BMD) measurement.
BMD of OVX and SHAM left tibiae were measured at d 0 and 14 by dual energy x-ray absorptiometry on Hologic 1000 plus (Bedford, MA), using a software modified for small animals (Regional High Resolution version 4.76; Hologic). The coefficient of variation for BMD measurements was 1.1%.
Histomorphometric analysis.
For bone histomorphometric analysis, tibiae were fixed in ethanol 70%, dehydrated in graded alcohols, and embedded undecalcified in methyl methacrylate at low temperature as previously described (34). Seven-micrometer-thick-sections were stained with Goldner’s trichrome. Measurements of the trabecular bone volume (BV/TV) expressed in percentage of the total volume, trabecular bone thickness (TbTh), trabecular bone number (TbN), and trabecular separation (TbSp) of secondary spongiosa were all carried out on Goldner’s-stained sections with Osteolab Image Analyser (Biocom, Les Ulis, France). Bone mass and microarchitectural parameters were measured on both primary and secondary spongiosae. Parameters of bone remodeling were measured on secondary spongiosae only using a semiautomatic system, as previously described (35). Osteoclast surfaces (OcS/BS) were measured on four sections stained for tartrate-resistant acid phosphatase activity. Histodynamic measurements (mineral apposition rate expressed in μm/d), mineralizing surfaces (%), and bone formation rates surface referent (expressed in μm3/μm2/d) were measured on unstained sections under UV light.
Immunocytochemistry.
Tibiae were fixed in 1% glutaraldehyde, decalcified in 4% EDTA/1% glutaraldehyde, and embedded in Epon. Semithin sections from decalcified samples were laid on sylanized slides (Dako, Carpinteria, CA) and dried overnight at 37 C. Epon was removed with 13.3% (wt/vol) potassium hydroxide in a methanol/propylene oxide (2/1) mixture and rehydrated. Sections were treated for 20 min with 100 mM glycine and 50 mM ammonium chloride in Tris buffer (pH 7.6) to saturate free aldehydic groups. Endogenase peroxidase activity was inhibited by incubation for 15 min with 1% sodium azide and 1.5% H2O2 in 50% methanol, and nonspecific immunoreactions were blocked for 30 min in 10% normal goat serum in Tris-buffered saline (TBS) containing 0.01% BSA. Sections were incubated overnight with specific antibodies directed against nerve markers diluted in TBS containing 1% normal goat serum: NF200 (30 μg/ml), SY (6 μg/ml), and glutamate (Glu; 1/1000). Control sections were incubated with the monoclonal antibody MOPC21, which has no known hapten or antigen binding activity. Antigen-antibody complexes were detected with the EnVision/HRP System (Dako) and were revealed with 3-3'-diaminobenzidine tetrahydrochloride (Dako) in Tris buffer containing 0.01% H2O2. Sections were counterstained with Meyer’s hemalum, dehydrated, and mounted in Xam (Gurr-BDH Laboratory, Poole, UK).
Quantitative image analysis of immunostaining for these markers was performed in primary and secondary spongiosa of rat tibiae using Osteolab (Biocom). After image acquisition of immunostaining, an immunolabeling memorized color threshold was applied together with a constant smooth filter to remove immunostaining noise background. Measurement of detected area was performed after calibration. These different steps are illustrated in Fig. 1. Results are expressed as percentages of immunolabelings per total bone volume. The coefficients of variation for nerve marker measurements were 8.1, 11.0, and 7.2% for NF200, SY, and Glu, respectively. Because Glu is expressed both by nerve fibers in bone and by osteoblastic cells (7, 36), the analysis of Glu-immunoreactive nerve fibers was performed after the exclusion in each field of the immunostaining for Glu in osteoblastic cells along bone trabeculae.
FIG. 1. Quantification of immunostaining for nerve markers. A, Image acquisition of NF200 immunostaining in primary spongiosa of control tibiae. Immunoreactive nerve fibers were predominantly nonvascular, identified in the bone marrow closed to bone trabeculae. B, An immunolabeling memorized color threshold was applied to the image, together with a constant smooth filter to remove immunostaining noise background. C, After calibration, detected areas corresponding exactly to the immunostaining observed for NF200 (A) were measured.
Skin and muscle were fixed with 4% paraformaldehyde for 48 h and embedded in paraffin. Paraffin sections (7 μm) were laid on sylanized slides and dried overnight at 37 C before dewaxing with methylcyclohexane followed by rehydration. Sections were then treated as described for tibiae.
Statistical analyses
BMD measurement analyses were carried out using SPSS 10 software (SPSS Inc., Chicago, IL). Other analyses were performed using Instat software (GaphPad, San Diego, CA). The significance of differences in BMD, percentages of immunostainings, and quantitative histomorphometric parameters between control and treatment groups was determined based on nonparametric Wilcoxon’s or Mann-Whitney U tests.
Results
Effect of sciatectomy on tibiae innervation
To confirm that denervation was able to induce changes in skeletal nerve fibers that could be quantified by immunocytochemistry using antibodies directed against nerve fibers, we studied innervation in tibial semi-thin sections of SN limb and a contralateral intact section immunostained for NF200 and SY, two specific markers of nerve fibers. NF200 is specific of neurofilaments present in nerve cells, whereas SY is a nerve terminal marker. Two cancellous bone regions of the metaphysis were studied: the primary spongiosa, which is the trabecular bone formed close to the growth plate following the resorption of calcified endochondral cartilage; and the secondary spongiosa, which arises from remodeled primary spongiosa. We observed a dramatic reduction of immunoreactivity for both markers in the primary spongiosa of SN tibiae compared with the contralateral ones (Fig. 2). A less striking decrease of nerve profile density was observed in secondary spongiosa of SN tibiae compared with contralateral controls (data not shown). Quantitative image analysis of immunostaining for NF200 and SY demonstrated a significant decrease in the percentage of immunolabeling per total bone volume for both NF200 (P = 0.01) and SY (P = 0.01) in the primary spongiosa of SN tibiae compared with contralateral controls (Fig. 3A). Trend to decrease for these markers was also observed in secondary spongiosa of SN tibiae, but it did not reach statistical significance (Fig. 3B). This decrease of nerve fiber density in tibiae after SN indicates that changes in innervation can be quantified in tibiae with this technique, thus allowing studies on modifications of innervation when bone remodeling is affected.
FIG. 2. Immunolabellings for nerve markers on semithin sections from neurectomized and contralateral intact tibiae. Primary spongiosa of tibiae from neurectomized (SN) and contralateral intact limbs, immunostained for NF200 and SY. No labeling was observed in control sections incubated with the monoclonal antibody MOPC21. Original magnification, x400.
FIG. 3. Quantitative analysis of immunostaining for nerve markers in tibiae from neurectomized and contralateral intact limbs. Quantitative image analysis of immunostaining for NF200 and SY in primary (A) and secondary (B) spongiosa of tibiae from neurectomized (SN) or contralateral intact limbs was performed using Osteolab software. Immunostainings were quantified in 10 consecutive fields for both tibial primary and secondary spongiosa, and results are expressed in percentage of immunolabelings per total bone volume. Data are means ± SD (n = 6), *, Significantly different from contralateral intact tibiae, P < 0.05, Wilcoxon’s test).
Effects of ovariectomy on BMD
SHAM and OVX groups of female rats had similar weights before surgery. Two weeks after surgery, the uterine weight was strikingly lower in every OVX animal compared with SHAM, confirming their estrogen deficiency. BMD was significantly lower (–4%) in tibiae of OVX rats (0.287 g/cm2) compared with SHAM (0.296 g/cm2) 14 d after ovariectomy (P < 0.005, Mann-Whitney U test).
Bone histomorphometric analysis of OVX rat tibiae
Histomorphometric analysis of rat tibiae showed a significant reduction of BV/TV (–29%) in secondary spongiosa of OVX rats compared with SHAM (P = 0.0012, Mann-Whitney U test) 14 d after ovariectomy (Table 1). This was associated with a significant decrease in trabecular number, whereas no change in trabecular thickness was observed yet. As expected, OVX stimulated bone resorption activity, as assessed by active resorption surfaces using osteoclast tartrate-resistant acid phosphatase staining (Table 1). Dynamic parameters measurements showed the expected significant increase in both mineral apposition rate (+33%) and bone formation rates surface referent (+200%) after OVX, reflecting the stimulation of bone formation activity induced by OVX.
TABLE 1. Bone histomorphometry parameters in the secondary spongiosa of tibial metaphysis after 14 d of OVX
Effect of OVX on tibiae innervation
To analyze the involvement of bone innervation in a model of bone loss associated with increased bone remodeling due to estrogen deficiency, we studied innervation in rat tibiae after OVX. We observed a substantial reduction of nerve profiles immunolabeled for NF200, SY, and Glu, a major neuromediator involved in bone metabolism (23), in the tibial primary and secondary spongiosae of OVX rats compared with SHAM (Fig. 4). Quantitative image analysis of immunostaining for nerve markers demonstrated a significant decrease in the percentage of immunolabeling of NF200, SY, and Glu in both the primary and secondary spongiosae of OVX rats compared with SHAM (Fig. 5, A and B). Similar results were observed when the percentages of immunolabeling for these markers were expressed per total bone volume or per bone marrow volume (data not shown), indicating that the dramatic decrease of innervation observed after OVX was not dependent on a reduction of the number of bone trabeculae.
FIG. 4. Immunolabellings for nerve markers on tibial semithin sections from OVX and SHAM rats. Primary spongiosa of tibiae from ovariectomized (OVX) and SHAM rats, immunostained for NF200, SY, and Glu. Original magnification, x400.
FIG. 5. Quantitative analysis of immunostaining for nerve markers in tibiae from OVX and SHAM rats. Quantitative image analysis of immunostaining for NF200, SY, and Glu in primary (A) and secondary (B) spongiosa of tibiae from OVX or SHAM rats was performed using Osteolab software. Immunostainings were quantified in 10 consecutive fields for both tibial primary and secondary spongiosae, and results are expressed in percentage of immunolabelings per total bone volume. Data are means ± SD of 13 SHAM and 14 OVX rats. Significantly different from the SHAM group (*, P < 0.05; **, P < 0.01; ***, P < 0.001, Mann-Whitney U test).
Effect of OVX on muscle and skin innervation
To investigate whether the decrease in innervation observed after OVX is specific for bone and not a ubiquitous consequence of estrogen depletion, we studied innervation in the skin of rat hind paws and in the hindlimb muscles. We did not observe any difference in nerve profiles immunolabeled for NF200 in the skin (data not shown) or muscle of OVX rats compared with SHAM (Fig. 6). Quantitative image analysis of NF200 immunostaining demonstrated no significant difference in the percentage of immunolabeling for NF200 in muscle of OVX rats compared with SHAM (not shown).
FIG. 6. Immunolabelling for NF200 of muscle sections from OVX and SHAM rats. Muscle sections from OVX and SHAM rats immunostained for the nerve marker NF200 (A and C). No labeling was observed in control sections incubated with the monoclonal antibody MOPC21 (B and D). Original magnification, x250 (A and B); x320 (C and D).
Discussion
This study clearly shows evidence that skeletal innervation is decreased after OVX, suggesting a link between innervation and the bone loss induced by OVX. Previous work by our group (7) has shown that bone is rich in nerve fibers localized in the vicinity of bone cells, indicating a possible role of nerve fibers in bone cell metabolism. Although numerous in vitro studies have shown roles for neuromediators in bone cell functions (3, 4, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29), the relationship between the presence of these nerve fibers and bone cell metabolism has been poorly investigated in vivo. We first used a model of denervation-induced bone loss, sciatic neurectomy, to investigate whether changes in skeletal nerve fibers could be quantified in bone. Then, we analyzed the alterations of bone innervation after ovariectomy to check whether estrogen deficiency, known to induce increase in bone remodeling and to modulate neuronal functions (37), could affect nerve density in bone.
The tibia can be denervated by surgical sciatic neurectomy, a largely used model of bone loss under less mechanical stress. Bone deficit in this model is caused by a significant early increase in bone resorption and by depressed bone formation as a result of impaired osteoblast activity (38). In our study, we chose to investigate the effects of denervation in young 4-wk-old rats, because we had previously demonstrated that innervation is abundant in bone of young growing rats (7), and our first aim was to quantify potential changes in bone innervation occurring after sciatectomy. We used immunocytochemistry with antibodies directed against two specific neuronal markers, NF200 and SY. NF200 is specific of neurofilaments that are structural proteins found in all nerve processes. There are the major proteins found in myelinated sensory neurons but are also expressed in sympathetic neurons. SY is one of the most abundant membrane proteins of small synaptic vesicles, so it is expressed in all nerve terminals. Therefore, NF200 and SY are general nerve markers and can’t discriminate between sympathetic and sensory nerve fibers. As previously illustrated, immunoreactivity for NF200 and SY was found in thin cell processes in the proximity of blood vessels and bone cells, and both markers showed a similar distribution in bone (7). We observed a marked decrease in innervation in the tibial primary spongiosa of neurectomized limbs compared with the contralateral ones, whereas only a mild decrease in innervation was observed in secondary spongiosa, indicating that sciatectomy in young growing rats primarily affects the tibial modeling zone. These results might explain the fact that neurectomy mainly alters the regions of bone modeling, as previously reported in rapidly growing young rats (39) as well as in slowly growing mature rats (40). Complete denervation of the long bones was not achieved after sciatectomy, as, in addition to sciatic nerve, other nerves and their branches supply nerve fibers to the hindlimbs (11). After quantification of SY and NF200 immunolabelings per total bone volume, we demonstrated a 50% reduction of bone innervation in the primary spongiosa of sciatectomized limbs, whereas the decrease in the secondary spongiosa did not reach statistical significance. Our results demonstrate that innervation is decreased in tibiae after sciatectomy and confirm that our quantitative image analysis technique of immunostaining for nerve markers is adequate for measuring changes in skeletal innervation.
Estrogen deficiency in the model of OVX rats induces an increase in bone turnover with a rate of bone resorption exceeding that of bone formation, resulting in the reduction of bone mass (41). OVX resulted in significant rapid decrease in BMD within 14 days. This bone loss was illustrated at the trabecular level by a significant decrease in BV/TV and number of trabeculae in secondary spongiosae. As expected, the OVX-induced increase in bone cellular activities associated both increased active osteoclast surfaces and bone formation parameters.
So far, no study has yet investigated bone innervation after OVX. This model of bone loss was chosen for two reasons. First, it is the most widely used model in the bone field when studying high remodeling-induced bone loss; second, we wanted to study innervation because estrogen has many effects on brain neurons and on the autonomic nervous system (42, 43). Surprisingly, our results demonstrate a dramatic decrease of innervation in both primary and secondary spongiosae of OVX rat tibiae compared with SHAM. This decrease was not related to the important reduction of trabecular bone volume after OVX, as we observed a similar decrease of innervation when normalized by bone marrow or bone volume. Whether the decrease in innervation is the cause or the consequence of the increase in bone remodeling is not yet known, and kinetic studies of innervation after OVX should be performed to answer this question. The fact that no decrease of innervation was observed in muscle and skin after OVX indicates that decrease of innervation after OVX is specific to bone. Interestingly, several authors reported site-specific effects of estrogen deficiency on various neuronal functions. Excitatory or inhibitory responses of neurons in the gracile nucleus vary after OVX as a function of the tissular origin of the stimulus (44). In a different field, nociception seems more evident in the orofacial area than in many other parts of the body in post menopausal women. Pajot et al. (45) showed that gonadectomy induces site-specific differences in nociception in rats, with higher sensitivity of the upper lip region compared with hind paws. Finally, there is now evidence that estrogen specifically modulates sympathetic innervation in the uterus, with a decrease in nerve density induced by estrogen deficiency and restoration of innervation under estrogen supplementation in OVX rats and mice (46). After uterus and breast, bone is the third major target of estrogens; however, our study is the first to demonstrate that estrogen depletion also affects innervation in the skeleton. Whether estrogen regulates the number of skeletal nerves through direct action on sympathetic and sensory ganglions or indirectly is unknown. In uterus, estrogen modulates sympathetic neurite outgrowth by regulating neurotrophic factors synthesis such as nerve growth factor and brain-derived neurotrophic factor (47), presumably through interactions with estrogen response elements that are present in the promoter regions of these genes (48, 49, 50). Interestingly, these neurotropins are expressed by osteoblasts and we could hypothesize that a similar regulatory mechanism exists in bone (51). Expression of estrogen receptor (ER) or ER? may also explain the specific effects of estrogen deficiency on bone innervation. Bone, skin, and muscles exhibit specific ER and ER? expression pattern and regulation. Although both receptors play important roles in bone metabolism in females, ER? is the main mediator of estrogen action in skin, and ER is highly expressed in muscles (52, 53, 54). Interestingly, ER deficiency sensibly affects uterus innervation in mice (46).
The demonstration that OVX induces a specific bone loss associated with a marked reduction of nerve profile density in both modeling and remodeling zones of tibiae suggests an association between bone remodeling and innervation. Several studies have previously demonstrated that innervation may control local bone turnover. Selective denervations of either sensory or sympathetic nerves were shown to affect bone remodeling (8, 10, 55). Increased osteoclast formation was observed in long-term bone marrow cultures from patients with a spinal cord injury, which might be attributed to a deficit in CGRP (56, 57). Contacts between osteoclasts and CGRP-positive nerve fibers were observed at the epiphyseal trabeculae facing the growth plate (6), and in vitro experiments have demonstrated that several neuropeptides regulate osteoclast and osteoblast activities (4). Recently, the concept that osteoporosis might in part originate from the hypothalamus has been proposed following the demonstration of a central control of bone formation via the peripheral sympathetic nervous system (14, 15, 16). Furthermore, it was recently suggested that the ?-adrenergic pathway of the sympathetic nervous system is a potential mediator of mechanical loading in bone (18). Our findings that demonstrate a decrease in bone innervation after OVX suggest that neural regulation could also modulate bone remodeling and play a role in postmenopausal osteoporosis. All these observations were, however, made in trabecular bone, and it is possible that the differences of innervation in the various bones of the skeleton and in the different envelopes within a bone may result in diverse responses to estrogen deficiency. Recent work (58) indeed indicates that leptin deficiency produces contrasting phenotypes in bones of the limb and spine.
Among the broad range of neurotransmitters present in bone, Glu has profound effects on the skeleton and may be involved in the neural control of bone remodeling. We showed that Glu contributes to the local regulation of bone resorption via NMDA Glu receptors (23, 24, 27). Because accumulating evidence suggests that estrogen may exert its effect in the brain through the regulation of these receptors (59), we therefore studied glutamatergic innervation after OVX. Because Glu itself is expressed in osteoblasts and their precursors as well as in nerve processes in the vicinity of bone cells (7, 36), we excluded immunolabeling for Glu in osteoblastic cells along bone trabeculae to quantify glutamatergic innervation. We found that the percentage of immunolabeling per total bone volume for Glu was similar to those determined for NF200 and SY, suggesting that the majority of nerve fibers in bone express Glu. Nevertheless, although Glu-containing osteoblasts were excluded, Glu is also expressed by some bone marrow cells, suggesting a possible contribution of these cells to Glu immunolabeling quantification. We observed a dramatic decrease of the percentage of immunolabeling for Glu in both the primary and secondary spongiosa of OVX rats compared with SHAM, suggesting that glutamatergic innervation may play a role in the bone loss induced after OVX. As demonstrated for neurons that innervate genital organs (42), it is possible that the glutamate-containing neurons, which innervate bone, express estrogen receptors. Whereas NF200 and SY are general nerve markers, Glu is the major neuromediator released by sensory nerves. The demonstration that the percentage of immunolabeling for Glu decreases after OVX suggests that sensory denervation may be primarily affected by OVX. However, we cannot exclude that sympathetic innervation is also decreased after OVX, as both sympathetic and sensory innervation are regulated by estrogens (45, 46). Additional studies, using specific markers of the sympathetic and the sensory nervous system, are necessary to investigate which nerve cell populations are most affected by OVX.
Our findings, which demonstrate a dramatic decrease of skeletal innervation after OVX, suggest that neural regulation may play a role in the bone loss during osteoporosis. These data support the increased evidence for a major role of innervation in the local control of skeletal metabolism. Clearly, additional studies are required to better understand the mechanisms linking bone remodeling to innervation.
Acknowledgments
We thank Annie Desenfants for animal care and Fran?oise Munoz and Pavel Szulc for helpful comments on statistical analyses.
References
Hill EL, Elde R 1991 Distribution of CGRP-, VIP-, D?H-, SP-, and NPY-immunoreactive nerves in the periosteum of the rat. Cell Tissue Res 264: 469–480
Hohmann EL, Elde R, Rysavy J, Einzig S, Gebhard R 1986 Innervation of periosteum and bone by sympathetic vasoactive intestinal peptide-containing nerve fibers. Science 232:869–887
Konttinen Y, Imai S, Suda A 1996 Neuropeptides and the puzzle of bone remodeling. Acta Orthop Scand 67:632–639
Lerner UH, Lundberg P 2002 Kinins and neuro-osteogenic factors. In: Bilezikian JP, Raisz LG, and Rodan GA, eds. Principles of bone biology. 2nd ed. San Diego: Academic Press; 773–799
Goto T, Yamaza T, Kido MA, Tanaka T 1998 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
Hara-Hirie F, Amizuka N, Ozawa H 1996 Immunohistochemical and ultrastructural localization of CGRP-positive nerve fibers at the epiphyseal trabecules facing the growth plate of rat femurs. Bone 18:29–39
Serre CM, Farlay D, Delmas PD, Chenu C 1999 Evidence for a dense and intimate innervation of the bone tissue, including glutamate-containing fibers. Bone 25:623–629
Cherruau M, Facchinetti P, Baroukh B, Saffar JL 1999 Chemical sympathectomy impairs bone resorption in rats: a role for the sympathetic system on bone metabolism. Bone 25:545–551
Edoff K, Hellman J, Persliden J, Hildebrand C 1997 The developmental skeletal growth in the rat foot is reduced after denervation. Anat Embryol 195:531–538
Hill EL, Turner R, Elde R 1991 Effects of neonatal sympathectomy and capsaicin treatment on bone remodeling in rats. Neuroscience 44:747–755
Hukkanen M, Konttinen YT, Santavirta S, Nordsletten L, Madsen JE, Almaas R, Oestreicher AB, Rootwelt T, Polak JM 1995 Effect of sciatic nerve section on neural ingrowth into the rat tibial fracture callus. Clin Orthop 311:247–257
Li J, Ahmad T, Spetea M, Ahmed M, Kreicbergs A 2001 Bone reinnervation after fracture: a study in the rat. J Bone Miner Res 16:1505–1510
Ramnemark A, Nyberg L, Lorentzon R, Englund U, Gustafson Y 1999 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
Ducy P, Amling M, Takeda S, Priemel M, Schilling AF, Beil FT, Shen J, Vinson C, Rueger JM, Karsenty G 2000 Leptin inhibits bone formation through a hypothalamic relay: a central control of bone mass. Cell 100:197–207
Takeda S, Elefteriou F, Levasseur R, Liu X, Zhao L, Parker KL, Amstrong D, Ducy P, Karsenty G 2002 Leptin regulates bone formation via the sympathetic nervous system. Cell 111:305–317
Baldock PA, Sainsbury A, Couzens M, Enriquez RF, Thomas GP, Gardiner EM, Herzog H 2002 Hypothalamic Y2 receptors regulate bone formation. J Clin Invest 109:915–921
Pasco JA, Henry MJ, Sanders KM, Kotowicz MA, Seeman E, Nicholson GC 2004 ?-adrenergic blockers reduce the risk of fractures partly by increasing bone mineral density: Geelong osteoporosis study. J Bone Miner Res 19:19–24
Levasseur R, Sabatier JP, Potrel-Burgot C, Lecoq B, Creveuil C, Marcelli C 2003 Sympathetic nervous system as transmitter of mechanical loading in bone. Joint Bone Spine 70:515–519
Lundberg P, Lie A, Bjurholm A, Lehenkari PP, Horton MA, Lerner UH, Ransj? M 2000 Vasoactive intestinal peptide regulates osteoclast activity via specific binding sites on both osteoclasts and osteoblasts. Bone 27:803–810
Bliziotes MM, Eshleman AJ, Zhang XW, Wiren KM 2001 Neurotransmitter action in osteoblasts: expression of a functional system for serotonin receptor activation and reuptake. Bone 29:477–486
Susiki A, Palmer G, Bonjour JP, Caverzasio J 1998 Catecholamines stimulate the proliferation and alkaline phosphatase activity of MC3T3–E1 osteoblast-like cells. Bone 23:197–203
Wesbroek I, Van der Plas A, de Rooij KE, Klein-Nulends J, Nijweide PJ 2001 Expression of serotonin receptors in bone. J Biol Chem 276:28961–28968
Chenu C, Serre CM, Raynal C, Burt-Pichat B, Delmas PD 1998 Glutamate receptors are expressed by bone cells and are involved in bone resorption. Bone 22:295–299
Espinosa L, Itzstein C, Cheynel H, Delmas PD, Chenu C 1999 Active NMDA glutamate receptors are expressed by mammalian osteoclasts. J Physiol 518:47–53
Gu Y, Publicover S 2000 Expression of functional metabotropic glutamate receptors in primary cultured rat osteoblasts. J Biol Chem 275:34252–34259
Itzstein C, Cheynel H, Burt-Pichat B, Merle B, Espinosa L, Delmas PD, Chenu C 2001 Molecular identification of NMDA glutamate receptors expressed in bone cells. J Cell Biochem 82:134–144
Itzstein C, Espinosa L, Delmas PD, Chenu C 2000 Specific antagonists of NMDA receptors prevent osteoclast sealing zone formation required for bone resorption. Biochem Biophys Res Commun 268:201–209
Patton AJ, Genever PG, Birch MA, Suva LJ, Skerry TM 1998 Expression of N-methyl-D-aspartate-type receptor by human and rat osteoblasts and osteoclasts suggests a novel glutamate signaling pathway in bone. Bone 22:645–649
Peet NM, Grabowski PS, Laketic-Ljubojevic I, Skerry TM 1999 The glutamate receptor antagonist MK801 modulates bone resorption in vitro by a mechanism predominantly involving osteoclast differentiation. FASEB J 13:2179–2185
Hukkanen M, Konttinen YT, Santavirta S, Paavolainen P, Gu XH, Terenghi G, Polak JM 1993 Rapid proliferation of calcitonin gene-related peptide-immunoreactive nerves during healing of rat tibial fracture suggests neural involvement in bone growth and remodelling. Neuroscience 54:969–979
Haynes LE, Lendon CL, Barber DJ, Mitchell IJ 2003 17?-Oestradiol attenuates dexamethasone-induced lethal and sublethal neuronal damage in the striatum and hippocampus. Neuroscience 120:799–806
Koski CL, Hila S, Hoffman GE 2004 Regulation of cytokine-induced neuron death by ovarian hormones: involvement of antiapoptotic protein expression and c-JUN N-terminal kinase-mediated proapoptotic signaling. Endocrinology 145:95–103
Yu X, Rajala RV, McGinnis JF, Li F, Anderson RE, Yan X, Li S, Elias RV, Knapp RR, Zhou X, Cao W 2004 Involvement of insulin/phosphoinositide 3-kinase/Akt signal pathway in 17?-estradiol-mediated neuroprotection. J Biol Chem 279:13086–13094
Chappard D, Palle S, Alexandre C, Vico L, Riffat G 1987 Bone embedding in pure methyl methacrylate at low temperature preserves enzyme activities. Acta Histochem 81:175–183
Chappard D, Alexandre C, Riffat G 1983 Histochemical identification of osteoclasts. Review of current method and reappraisal of simple procedure for routine diagnosis on undecalcified human iliac bone biopsies. Basic Appl Histochem 27:75–85
Bhangu PS, Genever PG, Spencer GJ, Grewal TS, Skerry TM 2001 Evidence for targeted vesicular glutamate exocytosis in osteoblasts. Bone 29:16–23
Maggi A, Ciana P, Belcredito S, Vegeto E 2004 Estrogens in the nervous system: mechanisms and nonreproductive functions. Annu Rev Physiol 66:291–313
Weinreb M, Rodan GA, Thompson DD 1991 Depression of osteoblastic activity in immobilized limbs of suckling rats. J Bone Miner Res 6:725–731
Iwamoto J, Takeda T, Ichumira S, Sato Y, Yeh JK 2003 Comparative effects of orchidectomy and sciatic neurectomy on cortical and cancellous bone in young growing rats. J Bone Miner Metab 21:211–216
Shen V, Liang XG, Birchman R, Wu DD, Healy D, Lindsay R, Dempster DW 1997 Short-term immobilisation-induced cancellous bone loss is limited to regions undergoing high turnover and/or modeling in mature rats. Bone 21:71–78
Kalu DN 1991 The ovariectomized rat model of postmenopausal bone loss. Bone Miner 15:175–191
Papka RE, Storey-Workley M, Shughrue PJ, Merchenthaler I, Collins JJ, Usip S, Saunders PT, Shupnik M 2001 Estrogen receptor- and ?-immunoreactivity and mRNA in neurons of sensory and autonomic ganglia and spinal cord. Cell Tissue Res 304:193–214
Wickelgren I 1997 Estrogen stakes claim to cognition. Science 276:675–678
Bradshaw HB, Berkley KJ 2003 The influence of ovariectomy with or without estrogen replacement on responses of rat gracile nucleus neurons to stimulation of hindquarter skin and pelvic viscera. Brain Res 986:82–90
Pajot J, Ressot C, Ngom I, Woda A 2003 Gonadectomy induces site-specific differences in nociception in rats. Pain 104:367–373
Zoubina EV, Smith PG 2001 Sympathetic hyperinnervation of the uterus in the estrogen receptor knock-out mouse. Neuroscience 103:237–244
Krizsan-Agbas D, Smith PG 2002 Estrogen modulates myometrium-induced sympathetic neurite formation through actions on target and ganglion. Neuroscience 114:339–347
Sohrabji F, Miranda RC, Toran-Allerand CD 1995 Identification of a putative estrogen response element in the gene encoding brain-derived neurotrophic factor. Proc Natl Acad Sci USA 92:11110–11114
Gibbs RB 1998 Levels of trkA and BDNF mRNA, but not NGF mRNA, fluctuate across the estrous cycle and increase in response to acute hormone replacement. Brain Res 787:259–268
Siminoski K, Murphy RA, Rennert P, Heinrich G 1987 Cortisone, testosterone, and aldosterone reduce levels of nerve growth factor messenger ribonucleic acid in L-929 fibroblasts. Endocrinology 121:1432–1437
Togari A 2002 Adrenergic regulation of bone metabolism: possible involvement of sympathetic innervation of osteoblastic and osteoclastic cells. Microsc Res Tech 58:77–84
Sims NA, Dupont S, Krust A, Clement-Lacroix P, Minet D, Resche-Rigon M, Gaillard-Kelly M, Baron R 2002 Deletion of estrogen receptors reveals a regulatory role for estrogen receptors-? in bone remodeling in females but not in males. Bone 30:18–25
Thornton MJ, Taylor AH, Mulligan K, Al-Azzawi F, Lyon CC, O’Driscoll J, Messenger AG 2003 The distribution of estrogen receptor ? is distinct to that of estrogen receptor and the androgen receptor in human skin and the pilosebaceus unit. J Investig Dermatol Symp Proc 8:100–103
Pfaffl MW, Lange IG, Daxenberger A, Meyer HHD 2001 Tissue-specific expression pattern of estrogen receptors (ER): quantification of ER and ER? mRNA with real-time RT-PCR. APMIS 109:345–355
Sherman BE, Chole RA 2000 Sympathectomy which induces membranous bone remodeling, has no effect on endochondral long bone remodeling in vivo. J Bone Miner Res 15:1354–1360
Akopian A, Demulder A, Ouriaghli F, Corazza F, Fondu P, Bergmann P 2000 Effects of CGRP on human osteoclast-like cell formation: a possible connection with the bone loss in neurological disorders? Peptides 21:559–564
Demulder A, Guns M, Ismail A, Wilmet E, Fondu P, Bergmann P 1998 Increased osteoclast-like cells formation in long-term bone marrow cultures from patients with a spinal cord injury. Calcif Tissue Int 63:396–400
Hamrick MW, Pennington C, Newton D, Xie D, Isales C 2004 Leptin deficiency produces contrasting phenotypes in bones of the limb and spine. Bone 34:376–383
Watanabe T, Inoue S, Hiroi H, Orimo A, Muramatsu M 1999 NMDA receptor type 2D gene as a target for estrogen receptor in the brain. Mol Brain Res 63:375–379(B. Burt-Pichat, M. H. Laf)