Calcitonin Deficiency in Mice Progressively Results in High Bone Turnover

Authors

  • Antje K Huebner,

    1. Center for Biomechanics and Skeletal Biology, University Medical Center Hamburg Eppendorf, Hamburg, Germany
    2. Department of Trauma, Hand, and Reconstructive Surgery, University Medical Center Hamburg Eppendorf, Hamburg, Germany
    Search for more papers by this author
    • These authors contributed equally to this study.

  • Thorsten Schinke,

    1. Center for Biomechanics and Skeletal Biology, University Medical Center Hamburg Eppendorf, Hamburg, Germany
    2. Department of Trauma, Hand, and Reconstructive Surgery, University Medical Center Hamburg Eppendorf, Hamburg, Germany
    Search for more papers by this author
    • These authors contributed equally to this study.

  • Matthias Priemel,

    1. Center for Biomechanics and Skeletal Biology, University Medical Center Hamburg Eppendorf, Hamburg, Germany
    2. Department of Trauma, Hand, and Reconstructive Surgery, University Medical Center Hamburg Eppendorf, Hamburg, Germany
    Search for more papers by this author
  • Sarah Schilling,

    1. Center for Biomechanics and Skeletal Biology, University Medical Center Hamburg Eppendorf, Hamburg, Germany
    2. Department of Trauma, Hand, and Reconstructive Surgery, University Medical Center Hamburg Eppendorf, Hamburg, Germany
    Search for more papers by this author
  • Arndt F Schilling,

    1. Center for Biomechanics and Skeletal Biology, University Medical Center Hamburg Eppendorf, Hamburg, Germany
    2. Department of Trauma, Hand, and Reconstructive Surgery, University Medical Center Hamburg Eppendorf, Hamburg, Germany
    Search for more papers by this author
  • Ronald B Emeson,

    1. Departments of Pharmacology, Molecular Physiology, and Biophysics, Vanderbilt University School of Medicine, Nashville, Tennessee, USA
    Search for more papers by this author
  • Johannes M Rueger,

    1. Center for Biomechanics and Skeletal Biology, University Medical Center Hamburg Eppendorf, Hamburg, Germany
    2. Department of Trauma, Hand, and Reconstructive Surgery, University Medical Center Hamburg Eppendorf, Hamburg, Germany
    Search for more papers by this author
  • Michael Amling MD

    Corresponding author
    1. Center for Biomechanics and Skeletal Biology, University Medical Center Hamburg Eppendorf, Hamburg, Germany
    2. Department of Trauma, Hand, and Reconstructive Surgery, University Medical Center Hamburg Eppendorf, Hamburg, Germany
    • Center for Biomechanics and Skeletal Biology, Department of Trauma, Hand, and Reconstructive Surgery, University Medical Center Hamburg Eppendorf, Martinistrasse 52, 20246 Hamburg, Germany
    Search for more papers by this author

  • The authors state that they have no conflicts of interest.

Abstract

Although the pharmacological action of calcitonin (CT) as an inhibitor of bone resorption is well established, there is still some controversy regarding its physiological function. Unexpectedly, Calca-deficient mice lacking CT and α-calcitonin gene-related peptide (αCGRP) were described to have a high bone mass phenotype caused by increased bone formation with normal bone resorption. Here we show that these mice develop a phenotype of high bone turnover with age, suggesting that CT is a physiological inhibitor of bone remodeling.

Introduction: The absence of significant changes in bone mineral density caused by decline or overproduction of CT in humans has raised the question, whether the pharmacological action of CT as an inhibitor of bone resorption is also of physiological relevance. To study the physiological role of mammalian CT, we have analyzed the age-dependent bone phenotype of two mouse models, one lacking CT and αCGRP (Calca−/−), the other one lacking only αCGRP (αCGRP−/−).

Materials and Methods: Bones from wildtype, Calca−/−-mice and αCGRP−/−-mice were analyzed at the ages of 6, 12 and 18 months using undecalcified histology. Differences of bone remodeling were quantified by static and dynamic histomorphometry as well as by measuring the urinary collagen degradation products. To rule out secondary mechanisms underlying the observed phenotype, we determined serum concentrations of relevant hormones using commercially available antibody-based detection kits.

Results: Whereas αCGRP−/−-mice display an osteopenia at all ages analyzed, the Calca−/−-mice develop a phenotype of high bone turnover with age. Histomorphometric analysis performed at the age of 12 months revealed significant increases of bone formation and bone resorption specifically in the Calca−/−-mice. This severe phenotype that can result in hyperostotic lesions, can not be explained by obvious endocrine abnormalities other than the absence of CT.

Conclusions: In addition to the previously described increase of bone formation in the Calca-deficient mice, we have observed that there is also an increase of bone resorption with age. This suggests that CT has a dual action as an inhibitor of bone remodeling, which may explain why alterations of CT serum levels in humans do not result in major changes of bone mineral density.

INTRODUCTION

Calcitonin (CT) is a polypeptide of 32 amino acids produced by thyroidal C cells.(1,2 When administered at high pharmacological doses, it triggers a hypocalcemic response that is partially mediated through an inhibition of bone resorption.(3,4 This effect is well explained by the binding of CT to its receptor present on osteoclasts, although comparative studies have shown that mammalian CT is less potent than salmon CT as an inhibitor of their resorptive activity.(5–7 This has led to the therapeutic use of salmon CT in conditions associated with high bone resorption such as Paget's disease or osteoporosis.(8,9 It has also led to the assumption that the physiological role of mammalian CT is to participate in calcium hemostasis through an inhibitory effect on osteoclasts. This concept has, however, been challenged by two clinical observations. In fact, it was always puzzling that thyroidectomy does not result in osteoporosis and that high circulating levels of CT in patients with medullary thyroid carcinoma do not cause the expected osteopetrosis.(10,11

Whereas this absence of evidence was not necessarily in contradiction to a physiological role of mammalian CT as an inhibitor of bone resorption, the analysis of a CT-deficient mouse model was. These mice, which are lacking exons 2–5 of the Calca gene, display an unexpected high bone mass phenotype caused by an increased bone formation at the age of 3 months.(12 Even more surprising was the fact that there was no significant change of bone resorption and basal calcium hemostasis associated with the absence of CT at this age. These results suggested that mammalian CT is a physiological inhibitor of bone formation with no apparent influence on bone resorption. However, because the deletion of exons 2–5 from the Calca gene also results in the lack of α-calcitonin gene-related peptide (αCGRP), it was not clear at that point, whether the unexpected phenotype of the Calca−/− mice was indeed caused by the absence of CT.(12

Therefore, we took advantage of another mouse model, where a translational termination codon was introduced into exon 5 of the Calca gene, thereby selectively preventing the production of αCGRP without affecting the expression of CT.(13,14 These αCGRP−/− mice did not display the high bone mass phenotype that was observed in the Calca−/− mice. In contrast, they even displayed a mild osteopenia caused by decreased bone formation.(14 These results did not only establish a physiological role of αCGRP as an activator of bone formation, but they also suggested that the additional absence of CT in the Calca-deficient mice was counteracting the absence of αCGRP and causing their high bone mass phenotype.

In this manuscript we have continued our study and analyzed the progressive development of the bone phenotypes of both mouse models with age. Whereas the sole absence of αCGRP leads to osteopenia at 6, 12, and 18 months of age, the deficiency of CT and αCGRP in the Calca−/− mice results in high bone mass. More importantly, we observed major structural changes of trabecular bone as well as an increased cortical porosity in the Calca−/− mice at the age of 12 months or older. Histomorphometric analysis revealed that 12-month-old Calca−/−-mice display a phenotype of high bone turnover with increased bone formation, but also bone resorption. This high bone turnover resulted in hyperostotic lesions in 20% of all Calca−/− mice analyzed, but it could not be explained by alterations of the serum levels of several hormones with known effects on bone remodeling. The deduced dual action of CT as an inhibitor of both bone formation and bone resorption may explain why there are no major changes of BMD in human patients with altered CT serum concentrations.

MATERIALS AND METHODS

Mice

The colonies of Calca−/− and αCGRP−/− mice used for this study have been described previously.(12,13 Because we did not observe significant differences in the two corresponding wildtype control groups, their data were combined in this manuscript. All animal experiments were approved by the Animal Care Facility of the Hamburg University. Mice were fed a standard rodent diet and housed in a regular light/dark cycle. The corresponding bone phenotypes were analyzed at the ages of 6, 12, and 18 months. Because we did not find significant sex differences at these ages, only the data from female mice are presented in this manuscript. To assess dynamic histomorphometric indices mice were given two injections of calcein 9 and 2 days before death. At least six mice per group were subjected to histomorphometry and serum analysis to obtain statistically significant results. For the quantification of hyperostotic lesion development a total of 30 female Calca-deficient mice or wildtype controls were screened by radiography, before these lesions were confirmed histologically.

Histomorphometry

Skeletons were fixed in 3.7% PBS-buffered formaldehyde for 18 h at 4°C. After a 24-h incubation in 70% ethanol the lumbar vertebral bodies (L3—L5) and one tibia of each mouse were dehydrated in ascending alcohol concentrations and embedded in methylmethacrylate as described previously.(15 Sections of 5 μm were cut in the sagittal plane on a Microtec rotation microtome (Techno-Med, Munich, Germany). These sections were stained by toluidine blue and by the van Gieson/von Kossa procedure as described.(15 Nonstained sections of 12 μm were used to determine the bone formation rate.

Parameters of static and dynamic histomorphometry were quantified on toluidine blue—stained undecalcified proximal tibia and lumbar vertebral sections of 5 μm. Analysis of bone volume, trabecular number, trabecular spacing, trabecular thickness, and the determination of osteoblast and osteoclast numbers and surface were carried out according to standardized protocols using the OsteoMeasure histomorphometry system (Osteometrics, Atlanta, GA, USA).(16 Fluorochrome measurements for the determination of the bone formation rate were performed on two nonconsecutive 12-μm sections for each animal. Statistical differences between the groups (n =6) were assessed by the Student t-test.

Cell culture

Bone marrow cells were isolated from the femora by flushing with α-MEM containing 10% FBS and seeded into 6-well plates. To induce osteoblast differentiation the medium was supplemented with 50 μg/ml ascorbic acid and 10 mM β-glycerophosphate. Mineralized nodule formation was determined after 20 days by von Kossa staining. Alkaline phosphatase activity was measured after standard protocols with p-nitrophenylphosphate as a substrate, whereas protein concentrations were determined using the BioRad Protein Assay. For osteoclastogenesis, the marrow cultures were incubated with 10 nM 1,25(OH)2 vitamin D3 for 10 days. Formation of multinuclear cells was assessed by TRACP activity staining as described below. To determine their resorption activity, cells were additionally differentiated on dentin chips that were subsequently stained by toluidine blue.

Radiographic and μCT analysis

After death and removal of internal organs, the whole skeletons of all mice were analyzed by contact radiography using a Faxitron X-ray cabinet (Faxitron X-ray Corp., Wheeling, IL, USA). For 3D visualization the lumbar vertebra L6 was scanned (40 kV/114 μA) in a μCT 40 (Scanco Medical, Bassersdorf, Switzerland) at a resolution of 12 μm. For the assessment of the cortical porosity, femora were scanned at the midshaft at a resolution of 10 μm. The raw data were manually segmented and analyzed with the μCT Evaluation Program V4.4A (Scanco Medical). For visualization, the segmented data were imported and displayed in μCT Ray V3.0 (Scanco Medical).

Biochemical assays

To visualize functional osteoclasts on the bone surface, TRACP activity assays were performed on decalcified bone sections. Sections were preincubated for 1 h in 10 mM sodium tartrate dissolved in 40 mM acetate buffer (pH 5). The activity staining was performed in the same buffer including 0.1 mg/ml naphtol AS-MX phosphate (N-5000; Sigma Biochemicals) and 0.6 mg/ml Fast Red Violet LB salt (F-3881; Sigma Biochemicals). Serum TRACP5b activities were determined using the mouse TRACP assay (SB-TR-103; IDS). To quantify osteoclastic bone resorption, we measured the urinary excretion of deoxypyridinoline (Dpd) cross-links with the Pyrilinks-D ELISA (8007; Metra Biosystems). Values are expressed relative to creatinine concentrations as determined by a standardized colorimetric assay using alkaline picrate (8009; Metra Biosystems).

Serum concentrations of total calcium and inorganic phosphorus were determined using colorimetric assays (587-A and 360-3; Sigma Biochemicals). Serum concentrations of hormones were quantified using antibody-based detection kits (PTH and osteocalcin, 60-2300 and 50-1300; Immutopics; Leptin, 90030; Crystal Chem; Opg and Rankl, MOP00 and MTR00; R&D Systems). Estradiol was measured in the Department of Clinical Chemistry of the University Medical Center Hamburg Eppendorf according to standard procedures.

RESULTS

Age-dependent bone phenotypes of Calca- and αCGRP-deficient mice

In continuation of our previous study,(14 we analyzed the bone phenotypes of Calca-deficient mice (lacking CT and αCGRP) and αCGRP-deficient mice (lacking only αCGRP) at the ages of 6, 12, and 18 months. Von Kossa staining of undecalcified vertebral sections revealed that the trabecular bone volume of the αCGRP−/− mice was not only decreased at the age of 6 months,(14 but even more in the older mice, thereby showing that the sole absence of αCGRP in mice leads to progressive osteopenia (Fig. 1A, bottom). In contrast, the additional absence of CT in the Calca−/− mice leads to an increased trabecular bone volume, not only at 6 months of age,(14 but also thereafter (Fig. 1A, middle). There were, however, distinct structural changes of the trabecular bone in the Calca−/− mice that became apparent at the age of 12 months (see below).

Figure Figure 1.

Progressive phenotype development of Calca- and αCGRP-deficient mice. (A) Von Kossa staining of undecalcified sections from vertebral bodies of wildtype and Calca- and αCGRP-deficient mice at 6, 12, and 18 months of age. The trabecular bone volume is increased in the Calca−/− mice but decreased in the αCGRP−/−-mice. (B) Von Kossa staining of undecalcified tibia sections from the same groups of mice. Only the Calca−/− mice display severe cortical porosity at the ages of 12 and 18 months.

When we looked at the tibia sections from the same groups of mice we observed a striking phenotype of the Calca−/− mice that has not been described before, because it only appears at the age of 12 months or older. In fact, at these ages the Calca−/− mice, but not the αCGRP−/− mice displayed severe cortical porosity suggesting that bone resorption is now affected as well (Fig. 1B). To quantify these observations we next performed a full histomorphometric characterization of both mouse models at the age of 12 months, where the phenotype of the Calca−/− mice was most pronounced.

Histomorphometric analysis of 12 months old Calca- and αCGRP-deficient mice

To assess the structural parameters of trabecular bone remodeling, we applied static histomorphometry on toluidine blue—stained undecalcified vertebral sections. As we have described for the age of 6 months,(14 the αCGRP−/− mice have a decreased trabecular bone volume with increased trabecular spacing also at 12 months of age (Fig. 2A, dotted bars). The opposite is the case in the Calca−/− mice, indicating that the additional absence of CT in these mice reverses the osteopenia caused by the sole absence of αCGRP. In fact, as it has been described for the ages of 3 and 6 months,(12,14 the Calca−/− mice still have an increased trabecular bone volume and reduced trabecular spacing (Fig. 2A, striped bars). These changes of trabecular bone volumes could be explained by significant alterations of the trabecular number in both mouse models (Fig. 2B). In contrast, there was also a reduction of trabecular thickness specifically in the Calca−/− mice that was not observed at younger age.(12

Figure Figure 2.

Histomorphometric analysis of trabecular bone architecture. (A) Bone volume per tissue volume (BV/TV) is increased in 12-month-old Calca−/− mice (striped bars), but decreased in age-matched αCGRP−/− mice (dotted bars) compared with the corresponding wildtype controls (white bars). The opposite is the case for the trabecular spacing (TbSp). (B) Trabecular number (TbN) is increased in the Calca−/− mice, but decreased in the αCGRP−/− mice, whereas trabecular thickness (TbTh) is only decreased in the Calca−/− mice. Bars represent mean ± SD (n =6). Asterisks represent statistically significant differences (p < 0.05) compared with wildtype controls as determined by Student t-test.

We next looked at bone formation by cellular and dynamic histomorphometry. Whereas the numbers of osteoblasts and the surface covered by them were not significantly altered compared with wildtype controls (Fig. 3A), we observed significant changes of the bone formation rates in both mouse models. As it has been observed in the younger animals,(14 the sole deficiency of αCGRP leads to a reduction of bone formation, thereby explaining the osteopenic phenotype of the αCGRP-deficient mice (Fig. 3B). In contrast, the additional absence of CT reverses this state of low bone formation, and the Calca−/− mice, because it is the case in the younger animals,(12 still have an increased bone formation rate at the age of 12 months, which explains their high bone mass phenotype (Fig. 3B).

Figure Figure 3.

Analysis of bone formation. (A) The histomorphometric quantification of osteoblast number per bone perimeter (NOb/BPm) and osteoblast surface per bone surface (ObS/BS) in 12-month-old wildtype and Calca- and αCGRP-deficient mice revealed no significant differences between the groups. (B) Bone formation rates were determined after dual calcein labeling. Representative fluorescent micrographs show that the distance between the two labeled mineralization fronts is increased in Calca−/− mice but decreased in αCGRP−/− mice. The quantification of the bone formation rate per bone surface (BFR/BS) is given below. Values represent mean ± SD (n =6). Asterisks represent statistically significant differences (p < 0.05) compared with wildtype controls as determined by Student t-test.

When we measured the parameters of bone resorption, we found a striking phenotype specifically in the Calca−/− mice that was not observed at younger ages.(12 In fact, osteoclast numbers and surfaces covered by them were elevated 4-fold in the absence of CT and αCGRP, whereas they were normal in the sole absence of αCGRP (Fig. 4A). The increased cortical porosity in the Calca−/− mice is also explained by an elevated number of osteoclasts that were visualized by TRACP activity staining of tibia sections (Fig. 4B). Moreover, measuring the TRACP5b activities in the serum as well as the collagen degradation products in the urine further showed that bone resorption is strongly increased in the Calca−/− mice but not in the αCGRP−/− mice (Fig. 4C).

Figure Figure 4.

Analysis of bone resorption. (A) Histomorphometric analysis of osteoclast number per bone perimeter (NOc/BPm) and osteoclast surface per bone surface (OcS/BS) in 12-month-old wildtype and Calca- and αCGRP-deficient mice revealed a strong increase of both parameters specifically in the Calca−/− mice. (B) Toluidine blue staining of undecalcified tibia sections showed an increased cortical porosity with multinucleated osteoclasts only in the Calca−/− mice (top). The identity of these cells was further confirmed by TRACP activity staining (bottom). (C) Serum TRACP5b activities and urinary deoxypyridinoline (Dpd) cross-links were >2-fold elevated in the Calca−/− mice but not in the αCGRP−/− mice. Values represent mean ± SD (n =6). Asterisks represent statistically significant differences (*p < 0.05; **p < 0.005) compared with wildtype controls as determined by Student t-test.

To study whether the observed differences are caused by cell-autonomous mechanisms, we next isolated bone marrow stromal cells and analyzed their differentiation potential ex vivo. When osteoblast differentiation was induced by adding ascorbic acid and β-glycerophosphate, we observed no difference in nodule formation and mineralization between the different mouse models, thus ruling out intrinsic defects of osteoblast function (Fig. 5A). In contrast, alkaline phosphatase activities were slightly, but significantly, altered in the absence of CT and/or αCGRP, thereby providing an unexpected observation that warrants further study. When the cells were cultured in the presence of vitamin D3 to induce osteoclast formation, we observed no significant differences between the three genotypes, neither in the number of TRACP+ multinuclear cells, nor concerning the resorption of dentine chips (Fig. 5B). Taken together, these results show that the absence of CT in mice results in a phenotype of high bone turnover, which can be explained by an endocrine mechanism.

Figure Figure 5.

Analysis of bone marrow cultures from wildtype and Calca- and αCGRP-deficient mice. (A) Bone marrow cells were cultured in the presence of ascorbic acid and β-glycerophosphate for 20 days to allow osteoblast differentiation. Von Kossa staining for mineralized bone nodules did not reveal statistically significant differences between the three genotypes after quantification of the mineralized area. In contrast, the alkaline phosphatase activity was slightly increased in Calca-deficient cultures and decreased in αCGRP-deficient cultures. (B) Bone marrow cells from the same mice were cultured for 10 days in the presence of vitamin D3 to allow osteoclast differentiation. No statistically significant differences were observed between the three genotypes concerning the number of TRACP+ multinuclear cells (MNC/visual field) and their ability to form resorption pits on dentin chips. Values represent mean ± SD (n =6). Asterisks represent statistically significant differences (*p < 0.05) compared with wildtype controls as determined by Student t-test.

High turnover bone remodeling in 12 months old Calca-deficient mice

To confirm the phenotype of high bone turnover caused by the absence of CT, we next performed 3D μCT scans from vertebral bodies and cross-sectional scans from femora. Specifically in the Calca−/− mice, we observed a strong increase of trabecular structures that virtually leads to a trabecularization of cortical bone at the age of 12 months (Fig. 6A). Another consequence of this high turnover state is the development of hyperostotic lesions that were found in 20% of all Calca-deficient mice analyzed (6 of 30 mice at the age of 12 months), but never in wildtype control animals or in αCGRP−/− mice at the same age. These lesions, which were confirmed by von Kossa staining of undecalcified sections, were characterized by a strong local increase of bone formation but also bone resorption (Fig. 6B).

Figure Figure 6.

High bone turnover and hyperostotic lesions in Calca-deficient mice. (A) 3D μCT scans from vertebral bodies (top) and cross-sectional μCT-scans from femora (bottom) of 12-month-old wildtype and Calca- and αCGRP-deficient mice. Note the high degree of trabecularization in the Calca−/−-mice showing a phenotype of high bone turnover. (B) Von Kossa staining (top) of three representative undecalcified sections from vertebral bodies or a tibia of 12-month-old Calca−/− mice with hyperostotic lesions. The bottom panels show a strong increase of local bone formation as indicated by calcein labeling (left), but also an increase of bone resorption as determined by TRACP activity staining (middle). Such lesions were found in 20% of 12-month-old Calca−/− mice (right) but never in wildtype or αCGRP−/− mice (n =30).

We next addressed the question of whether this severe bone remodeling phenotype in the absence of CT could be explained by secondary mechanisms involving other hormones. As was the case in the younger animals,(12 we did not find any changes in the serum levels of calcium, phosphorus, and PTH in the Calca−/− mice at 12 months of age (Fig. 7A). Whereas the serum concentration of leptin was also not affected in the absence of CT, we did observe a slight increase of estradiol levels, but because the Calca-deficient mice were not hypogonadic, this change is unlikely to explain their high bone turnover phenotype (Fig. 7B). As expected, we also observed a 2-fold increase in serum osteocalcin levels reflecting the increased bone formation in the Calca−/− mice. In contrast, serum levels of Rankl were not increased, but even slightly decreased, whereas Opg levels were in the normal range (Fig. 7C). Taken together, these results show that the Calca-deficient mice progressively develop a phenotype of high bone turnover that can not be explained by major endocrine abnormalities other than the absence of CT.

Figure Figure 7.

Analysis of serum parameters in Calca-deficient mice. (A) Twelve-month-old Calca−/− mice display normal serum concentrations of calcium, phosphorus, and intact PTH. (B) The serum concentration of leptin is not significantly altered in Calca−/− mice compared with wildtype controls. Estradiol levels are slightly increased in the Calca−/− mice, but gonad size was found to be normal. (C) Circulating osteocalcin levels are strongly elevated in 12-month-old Calca−/− mice, reflecting their increased bone formation. In contrast, serum levels of Rankl are slightly decreased, whereas Opg concentrations are normal. Bars represent mean ± SD (n =6). Asterisks represent statistically significant differences (*p < 0.05, **p < 0.005) compared with wildtype controls as determined by Student t-test.

DISCUSSION

Distinct functions of CT and αCGRP in bone remodeling

The Calca gene encodes two polypeptides, CT and αCGRP, that are generated by alternative splicing.(17–19 Whereas CT is produced by thyroidal C cells and is thought to act as a hypocalcemic hormone inhibiting bone resorption,(2,3 αCGRP is expressed in neuronal cells of the central and peripheral nervous system and has mostly been implicated as a regulator of vascular tone.(19–21 Whether these actions are of physiological importance is still not fully clarified, especially because mouse deficiency models have been described that did not display the expected phenotypes.(12,13 In mice that are specifically lacking αCGRP, the absence of a vascular phenotype is possibly explained by the fact that βCGRP, a polypeptide closely related to αCGRP but encoded by a different gene, is still expressed in these mice.(13,22 In a Calca-deficient mouse model, however, that lacked expression of both CT and αCGRP, the absence of a bone resorption phenotype until the age of 6 months was indeed surprising, especially because these mice displayed a high bone mass phenotype caused by increased bone formation.(12,14

The comparison of the bone remodeling phenotypes of both mouse models described in this manuscript is therefore important for a better understanding of the physiological functions of the two peptides derived from the Calca gene. What can be concluded from the histomorphometric analysis of Calca- and αCGRP-deficient mice at various ages is that both peptides play specific, but distinct, roles in bone remodeling. Taken together, this analysis has shown that αCGRP−/− mice are characterized by an osteopenic phenotype caused by decreased bone formation, whereas the Calca−/− mice progressively develop a phenotype of high bone turnover. This suggests that the two polypeptides derived from the Calca gene have antagonistic functions on bone formation, whereas CT has a specific additional action as an inhibitor of bone resorption. Because both mouse models do not display any other obvious abnormalities, it seems that these functions are indeed physiologically relevant, at least in mice.

CT as an inhibitor of bone resorption

Since its discovery >40 years ago, it is well established that CT has a pharmacological effect on bone resorption.(3–5,23,24 After cloning of the CT receptor, it became evident that the binding of CT to this receptor that is present on mature osteoclasts triggers an intracellular signaling cascade, resulting in an inhibition of their resorptive activity.(3–6,25,26 Although these effects have been confirmed in vivo using high pharmacological doses of CT, there was thus far no evidence that the action of CT as an inhibitor of osteoclast function was also of physiological relevance. Whereas the therapeutic use of CT in bone remodeling disorders such as osteoporosis and Paget's disease was mostly involving salmon CT, whose antiresorptive effect is at least 50 times stronger compared with human CT,(7,27 it has always been surprising that there were no changes of BMD in human patients after thyroidectomy or with medullary thyroid carcinoma, although the serum levels of endogenous CT in these two conditions are indeed significantly altered.(10,11 These observations have even led some investigators to speculate that CT is not relevant in human physiology.(28,29

Because there is no human CT deficiency model established thus far to fully address this issue, we took advantage of mouse genetics as an experimental tool and have analyzed a Calca-deficient mouse model lacking CT, but also αCGRP.(12 At 3 months of age, these mice display an unexpected high bone mass phenotype caused by increased bone formation that also protects against ovariectomy-induced bone loss.(12 In contrast, osteoclast number, bone resorption, and serum calcium were all not affected in the Calca−/− mice, thereby challenging the classical concept of CT biology. Therefore, the results from the 12-month-old Calca−/− mice presented in this manuscript are indeed important, because they provide the first evidence that the deficiency of mammalian CT does also result in increased bone resorption. In fact, they show for the first time that CT is a physiological inhibitor of osteoclast function, at least in mice.

Although the increased bone resorption is only detectable in Calca−/− mice at the age of 12 months or older, it is completely in line with the classical action of CT. Because the difference in osteoclastogenesis was not apparent ex vivo, it further seems to be mediated by the well-established endocrine mechanism involving binding of the thyroid-derived circulating CT to its receptor present on osteoclasts. Moreover, the normal histologic appearance of several organs (data not shown) and the absence of hyperparathyroidism and hypogonadism in the Calca−/− mice rule out common secondary mechanisms that can result in elevated bone resorption.(30,31 The slight increase in estradiol levels, as well as the decrease of Rankl concentrations in the serum of Calca−/− mice, can also not explain their increased bone resorption and seem to be rather the consequence of a counter-regulatory mechanism. Finally, because osteoclast differentiation and function were found to be normal in the αCGRP−/− mice, we can rule out the possibility that the increased bone resorption in 12-month-old Calca−/− mice is caused by their deficiency in αCGRP.

CT as an inhibitor of bone formation

Regardless of the role of CT in bone resorption, it seems that another major physiological function of CT lies in the inhibition of bone formation. The importance of this is underscored by several arguments. First, the increased bone formation of the Calca−/− mice precedes the increase in bone resorption and is readily detectable at the age of 3 months, where bone resorption is still not affected.(12 Second, the absence of CT in the Calca−/− mice overcomes the absence of αCGRP, and the Calca−/− mice have increased bone formation despite the fact that the sole absence of αCGRP has the opposite effect. Third, even in the light of 4-fold elevated numbers of active osteoclasts at the age of 12 months, the Calca−/− mice still have a high bone mass phenotype. This indicates that the increased bone formation caused by the absence of CT outweighs not only the absence of αCGRP, but also the high level of bone resorption that should by itself result in an osteoporotic phenotype.

Although it is clear from our analysis that the absence of CT in mice leads to a strong increase in bone formation, this aspect of the Calca-deficient phenotype is not as easy to explain as their increased resorption. In fact, like others, we were unable to detect expression of CT and the CT receptor in bone-forming osteoblasts (data not shown), thus suggesting an indirect mechanism. Likewise, we did not observe cell-autonomous defects of osteoblast mineralization in the absence of CT, albeit the activity of alkaline phosphatase was slightly increased in Calca-deficient bone marrow stromal cells. Thus, we believe that there are basically three possibilities to explain the increased bone formation in the Calca−/− mice. First, it is possible that another yet unidentified gene is differentially expressed in the Calca-deficient mice that is causing their high bone mass phenotype. Second, we can not completely rule out that CT or differentially processed peptides derived from the Calca gene bind to a not yet identified receptor that is expressed in osteoblasts. Third, it is possible that the inhibitory action of CT on bone formation is not caused by the direct interaction of CT with osteoblasts, but involving other organs; for example, the hypothalamus where CT receptors are expressed and where certain nuclei have been shown to play important roles in bone remodeling.(32–35

One possibility to address these issues would be the analysis of mouse models with cell-specific deletions of the CT receptor. This is especially needed, because the complete deficiency of the CT receptor in mice causes embryonic lethality.(36 Interestingly, mice lacking only one allele of the CT receptor are viable and display a high bone mass phenotype caused by an increased bone formation.(36 Unfortunately these mice were only analyzed at younger age thus far, and it would be interesting to know, whether they also display a phenotype of high bone turnover, similar to the one observed in the Calca-deficient mice, at the age of 12 months or older.

Regardless of these open questions, however, we believe that our analysis of the Calca-deficient mice already provides one potential explanation for the absence of major changes in BMD in patients with altered levels of serum CT. In fact, if the dual function of CT as an inhibitor of bone formation and bone resorption is also true for human physiology, one would not expect that decreased CT levels after thyroidectomy would result in osteoporosis. Likewise, the patients with medullary thyroid carcinoma should rather develop a state of low bone turnover, but not an osteopetrosis.

Acknowledgements

The authors thank Dr Robert F. Gagel for providing the Calca-deficient mice, for helpful discussion and advice, as well as for the critical reading of the manuscript. Sarah Schilling is the recipient of a research fellowship of the Werner Otto Foundation. This work was supported by DFG Grant AM 103/9-2 (MA).

Ancillary