Abstract

Activation of osteoclasts and their acidification-dependent resorption of bone is thought to maintain proper serum calcium levels. Here we show that osteoclast dysfunction alone does not generally affect calcium homeostasis. Indeed, mice deficient in Src, encoding a tyrosine kinase critical for osteoclast activity, show signs of osteopetrosis, but without hypocalcemia or defects in bone mineralization. Mice deficient in Cckbr, encoding a gastrin receptor that affects acid secretion by parietal cells, have the expected defects in gastric acidification but also secondary hyperparathyroidism and osteoporosis and modest hypocalcemia. These results suggest that alterations in calcium homeostasis can be driven by defects in gastric acidification, especially given that calcium gluconate supplementation fully rescues the phenotype of the Cckbr-mutant mice. Finally, mice deficient in Tcirg1, encoding a subunit of the vacuolar proton pump specifically expressed in both osteoclasts and parietal cells, show hypocalcemia and osteopetrorickets. Although neither Src- nor Cckbr-deficient mice have this latter phenotype, the combined deficiency of both genes results in osteopetrorickets. Thus, we find that osteopetrosis and osteopetrorickets are distinct phenotypes, depending on the site or sites of defective acidification (pages 610–612).

Introduction

Bone is constantly remodeled through the balanced activities of osteoblasts and osteoclasts. Whereas genetic defects of bone-forming osteoblasts typically result in delayed bone development and osteoporosis, mutational inactivation of bone-resorbing osteoclasts leads to osteopetrosis, a condition of high bone mass accompanied by immunologic defects due to bone marrow displacement^{1, 2}. Through their ability to release calcium from the mineralized bone matrix, osteoclasts are also involved in calcium homeostasis, which is mediated mostly by the action of parathyroid hormone (PTH) triggering increased bone resorption³. Likewise, hypocalcemia and hyperparathyroidism have been observed in individuals with infantile malignant osteopetrosis, and there are a few case reports describing rachitic widening of the growth plate, a phenotype that has been termed osteopetrorickets^{4, 5, 6}.

Because osteoclasts resorb bone through extracellular acidification, it is not surprising that most of the known human osteopetrosis mutations affect genes involved in acid production and secretion⁷. These include CA2, encoding the proton-generating cytoplasmic enzyme carbonic anhydrase 2; CLCN7, encoding the transmembrane chloride transporter CLC-7; and OSTM1, encoding a presumptive -subunit of CLC-7 (refs. 8,9,10). Moreover, TCIRG1, the gene most commonly mutated in individuals with osteopetrosis, encodes a subunit of the vacuolar ATPase, a proton pump mediating acidification of intracellular organelles¹¹. The physiological importance of the Tcirg1 protein (also termed ATP6i, ATP6v0a3 or OC116) for bone resorption was first demonstrated through the generation of a strain of mouse deficient in this gene that then developed osteopetrosis due to a cell-autonomous defect of osteoclast-mediated extracellular acidification¹². A deletion within the Tcirg1 gene was subsequently found in oc/oc mice, a spontaneously occurring osteopetrotic mouse model^{13, 14}. Moreover, genetic screening of humans with infantile malignant osteopetrosis revealed that inactivating TCIRG1 mutations account for approximately 50% of cases^{15, 16, 17, 18}.

Notably, the initial characterization of Tcirg1-deficient oc/oc mice revealed an osteopetrorickets phenotype—again, high bone mass accompanied by rachitic widening of the growth plates and defects of skeletal mineralization¹⁹. However, the importance of Tcirg1 for cartilage and bone mineralization was not fully investigated thereafter, as histology was performed only after decalcification of the specimens. Moreover, to our knowledge there have been no systematic studies on the importance of functional osteoclasts for calcium homeostasis and skeletal mineralization, and the reported osteoclast-specific expression of Tcirg1 did not provide a sufficient explanation for the osteopetrorickets phenotype of the oc/oc mice^{12, 20}.

Here we have analyzed non-decalcified bone biopsy specimens from individuals with osteopetrosis and found defects of skeletal mineralization in fewer than 50% of them, suggesting that osteopetrosis and osteopetrorickets are distinct phenotypes. This was confirmed by comparison of two mouse models, Src^-/- and oc/oc, which show osteopetrosis and osteopetrorickets phenotypes, respectively^{19, 21}. The defects of skeletal mineralization in oc/oc mice result from severe hypocalcemia and can be prevented by feeding the mice a high-calcium diet. To assess how Tcirg1 deficiency triggers hypocalcemia, we next analyzed the expression pattern of Tcirg1 and found specific expression not only in osteoclasts but also in parietal cells. Our subsequent finding that oc/oc mice display low gastric acid levels (hypochlorhydria) demonstrated that Tcirg1 is physiologically involved in gastric acidification. Finally, to determine whether this previously unknown function of Tcirg1 could explain the osteopetrorickets phenotype of oc/oc mice, we studied Cckbr^-/- mice, which lack the cholecystokinin B–gastrin receptor²². We found that these mice, which are known to have a hypochlorhydria phenotype, also have mild hypocalcemia, secondary hyperparathyroidism and increased bone resorption; in addition, mice with deficiencies of both Cckbr and Src have osteopetrorickets. Moreover, Cckbr deficiency by itself triggers an osteoporotic phenotype, which is fully rescued by calcium gluconate supplementation. Together, these results suggest that gastric acid levels, rather than solely affecting osteoclast activity, help determine proper serum calcium levels, and that osteopetrorickets and osteopetrosis are distinct phenotypes.

Top of page

---

Results

Osteopetrosis and osteopetrorickets are distinct phenotypes

To analyze whether osteoclast dysfunction is generally associated with defects of skeletal mineralization, we first performed histomorphometry on non-decalcified transiliac bone biopsy specimens from 21 individuals with diagnosed osteopetrosis who had not been previously genotyped. As expected, we observed pathologically high trabecular bone volume, as compared to control biopsy specimens, in all 21 cases (Fig. 1a). In contrast, we found pathologic enrichment of non-mineralized bone matrix (osteoid), indicative of osteopetrorickets, in only ten cases. That this mineralization defect was not caused by excessive bone formation was demonstrated by our observation of diffuse fluorescence, with only occasional discrete fluorochrome labels, following tetracycline administration in the biopsy specimens from these individuals (Fig. 1b). To underscore the differences between the two groups, we carried out histomorphometry, which suggested that osteopetrosis and osteopetrorickets are distinct phenotypes that can be distinguished by quantifying osteoid thickness and osteoid volume (Fig. 1c).

Figure 1: Osteopetrosis (OPT) and osteopetrorickets (OPR) are distinct phenotypes.

(a) Von Kossa/van Gieson staining of non-decalcified bone biopsy specimens. Normally mineralized bone matrix is stained black and non-mineralized osteoid is stained red. The Histomorphometric quantification of the trabecular bone volume per tissue volume (BV/TV) is shown below (here and throughout, values are given as mean s.d.). ^**P < 0.005 versus the control group (n = 12). Scale bars, 400 m. (b) Fluorescence micrographs from the same specimens showing tetracycline-labeled bone surfaces. Scale bars, 200 m. (c) Histomorphometric quantification of osteoid thickness (O.Th.) and osteoid volume per bone volume (OV/BV). ^**P < 0.005 versus the control group. Here and throughout, box plots denote the mean and the 25th–75th percentile range. (d) Contact radiographs of whole skeletons (top) and vertebral bodies (middle), as well as von Kossa/van Gieson staining of non-decalcified vertebral sections (bottom), from wild-type (WT), Src^-/- and oc/oc mice at 2 weeks of age. Scale bars, 2 cm, 2 mm and 0.5 mm. (e) Contact radiographs (top) and non-decalcified histology of tibias (middle and bottom). Scale bars, 2 mm, 1 mm and 200 m. (f) Contact radiographs (top) and non-decalcified histology of calvariae (bottom). Scale bars, 1 cm and 100 m. (g) Histomorphometric quantification of femoral length, growth-plate thickness (GpTh), BV/TV and OV/BV. ^**P < 0.005 versus WT (n = 5 mice per group).

Full size image (97 KB)

We confirmed this phenotypic difference between these two conditions through analysis of two osteopetrotic mouse models, Src-deficient mice and Tcirg1-deficient oc/oc mice^{19, 21}. Contact radiographs performed at 2 weeks of age revealed an osteopetrorickets phenotype specifically in oc/oc mice, whose high bone mass was accompanied by rachitic widening of the growth plates and by severe growth retardation—two characteristics not seen in Src^-/- mice (Fig. 1d,e). Non-decalcified histology confirmed that only oc/oc mice had a mineralization defect of hypertrophic cartilage and bone matrix, whereas in Src^-/- mice the increased bone matrix was normally mineralized (Fig. 1e). We also observed enrichment of osteoid in the absence of Tcirg1 in craniofacial bones, which is important because these bones develop without the formation of a cartilage intermediate (Fig. 1f). Finally, we quantified our observations by histomorphometry, which confirmed that Src^-/- mice displayed osteopetrosis, whereas oc/oc mice displayed osteopetrorickets, demonstrating that osteoclast dysfunction does not necessarily cause a defect in skeletal mineralization (Fig. 1g).

Osteopetrorickets in oc/oc mice is caused by hypocalcemia

To assess whether the absence of Tcirg1 in oc/oc mice results in a cell-autonomous defect of skeletal mineralization, we first analyzed Tcirg1 expression in ex vivo cultures of bone cells. As expected, we found Tcirg1 expressed by osteoclasts but not by chondrocytes or osteoblasts. We observed the same pattern in vivo when we performed in situ hybridization on tibia sections from wild-type mice (Fig. 2a,b). Likewise, we did not observe any difference in extracellular matrix mineralization, alkaline phosphatase activity or osteocalcin production when primary osteoblasts from wild-type and oc/oc mice were differentiated ex vivo (Fig. 2c,d). An intrinsic osteoblast defect was further ruled out by the findings that bone mineralization was not affected in oc/oc mice at embryonic day 17.5 (E17.5) and that the osteopetrorickets phenotype developed postnatally (Fig. 2e,f). These observations can be explained by the fact that embryonic development of oc/oc mice occurs in oc/+ mothers, which were normocalcemic (Fig. 2g); after birth, however, oc/oc mice developed hypocalcemia and were not viable beyond the age of 3 weeks.

Figure 2: Osteoid enrichment in oc/oc mice is caused by hypocalcemia.

(a) Top, northern blot for Tcirg1 expression in cultured mouse osteoclasts (Ocl; differentiated for 0, 5 or 10 d as indicated), growth-plate chondrocytes (C; differentiated for 20 d) and osteoblasts (Obl; differentiated for 0, 5, 15 or 25 d) derived from wild-type (WT) mice. Bottom, ethidium bromide staining of RNA. (b) In situ hybridization for Tcirg1 expression in tibias from 3-day-old WT mice at low (left) and at high magnification (right). GP, growth plate; PS, primary spongiosa; Ocl, osteoclasts; Obl, osteoblasts; Ocy, osteocytes. Scale bars, 200 m (left) and 50 m (right). Unless otherwise indicated, arrows indicate osteoclasts. (c) Von Kossa staining for mineralized matrix in primary osteoblasts from wild-type and oc/oc mice after differentiation for 10 or 20 d. Scale bars, 1 cm. (d) Alkaline phosphatase activity (AP) and osteocalcin level (OC) after 20 d of differentiation. (e) Von Kossa/van Gieson staining of non-decalcified vertebral sections from WT and oc/oc mice before (E17.5, left) and after birth (postnatal day 2 (P2), right). Scale bars, 100 m. (f) Quantification of BV/TV and OV/BV. ^*P < 0.05 versus WT (n = 4 mice per group). (g) Serum calcium and PTH in 2-week-old WT, Src^-/-, oc/+ and oc/oc mice. ^**P < 0.005 versus WT (n = 5 mice per group). (h) Contact radiographs (scale bars, 2 mm) from vertebral bodies (top left) and tibias (bottom left), and von Kossa/van Gieson staining of non-decalcified sections from vertebral bodies (middle left, scale bars, 500 m), tibias (top right, scale bars, 200 m) and calvariae (bottom right, scale bars, 100 m) from 6-week-old WT and oc/oc mice kept on a high-calcium liquid diet. Quantification of OV/BV is also shown.

Full size image (97 KB)

We next analyzed whether calcium homeostasis would be affected by inactivation of Tcirg1. We found that oc/oc mice were characterized by severe hypocalcemia and hyperparathyroidism, conditions not seen in oc/+ mice or in Src^-/- mice (Fig. 2g). To demonstrate that the osteopetrorickets phenotype of oc/oc mice is caused by hypocalcemia, we fed the mice a high-calcium liquid diet, which resulted in normalization of serum calcium levels (10.4 0.2 mg dl^-1 in wild-type mice given the same diet versus 9.5 + 3.1 mg dl^-1 in oc/oc mice). As the majority of the oc/oc mice receiving this diet survived beyond 3 weeks of age, we performed our histologic analysis at 6 weeks and found that treated oc/oc mice had elevated trabecular bone volume (89.0 6.9% versus 21.6 2.2% in wild-type mice, P < 0.005) but no enrichment of osteoid (Fig. 2h). Taken together, these results demonstrated that the rachitic aspects of osteopetrorickets are caused by hypocalcemia and that the disturbance of calcium homeostasis in the absence of Tcirg1 is not solely explained by osteoclast dysfunction.

Tcirg1 is involved in gastric acidification

To assess how Tcirg1 deficiency triggers hypocalcemia, we first determined urinary calcium levels and found that they were significantly lower in oc/oc mice than in their wild-type littermates (Fig. 3a). Because this finding ruled out a defect of renal calcium reabsorption, it seemed possible that intestinal calcium uptake might be affected in oc/oc mice, and we therefore analyzed Tcirg1 expression in the gastrointestinal tract by RT-PCR. We detected specific expression of Tcirg1 in the fundus, a region of the stomach involved in gastric acid production (Fig. 3b). This finding led us to determine the gastric pH in wild-type, oc/oc and Cckbr^-/- mice—the latter serving as a control, as they are known to display hypochlorhydria due to their lower numbers of acid-producing parietal cells²². Whereas all the wild-type mice had gastric pH below 2.5, we observed higher gastric pH not only in Cckbr^-/- mice but also in oc/oc mice, thereby demonstrating, for what we believe is the first time, that hypochlorhydria is associated with Tcirg1 deficiency (Fig. 3c).

Figure 3: Tcirg1 is expressed by parietal cells and involved in gastric acidification.

(a) Urinary calcium concentrations in wild-type (WT), oc/+ and oc/oc mice. ^*P < 0.05 versus WT (n = 4 mice per group). (b) RT-PCR analysis for Tcirg1 expression in the gastrointestinal tract. Expression of the gastric-specific Atp4b gene and the housekeeping gene Gapdh were monitored as controls. (c) Gastric pH in WT, oc/oc and Cckbr^-/- mice. ^*P < 0.05 versus WT (n = 5 mice per group). (d) Immunohistochemistry on a human tissue microarray at low (left) and high magnification (right) using a TCIRG1-specific antibody. Scale bars, 2 mm and 100 m. (e) Non-decalcified histology of a transiliac bone biopsy specimen from an 18-month-old child carrying a homozygous point mutation (R56W) in TCIRG1 and from an age-matched control subject (scale bars, 100 m). Quantification of BV/TV and OV/BV is shown at right. (f) Retrospective analysis of serum parameters from 12 individuals with TCIRG1-dependent autosomal recessive osteopetrosis (OPT). Shown are values for calcium, phosphorus, PTH, 1,25-(OH)₂-vitamin D3 and gastrin. ^*P < 0.05 versus the control group (n = 12).

Full size image (93 KB)

To evaluate whether this deduced function of Tcirg1 in gastric acidification is also relevant in humans, we used a monoclonal antibody directed against the human TCIRG1 protein to perform immunohistochemistry on tissue microarrays. TCIRG1 was detectable only in the parietal cells of the stomach but not in sections from other organs (Fig. 3d). To determine whether TCIRG1 deficiency in humans results in the same phenotypic abnormalities as are seen in oc/oc mice, we next analyzed bone mineralization and gastric acidification in an individual with TCIRG1-dependent osteopetrosis (Fig. 3e). Although this individual's serum calcium was in the low normal range (2.3 mM), non-decalcified histology revealed high bone mass and a pathologic enrichment of osteoid, indicative of osteopetrorickets. Moreover, the individual's gastric pH was pathologically elevated, remaining between 2.7 and 4.1 and never reaching a normal value (below 2.2) in the course of 24 h of monitoring. We further performed a retrospective analysis of serum parameters from 12 individuals with TCIRG1-dependent osteopetrosis, in which we observed significantly lower serum calcium levels compared to age-matched controls. We also found elevated serum concentrations of PTH, 1,25-vitamin D3 and gastrin, indicative of disturbed calcium homeostasis and hypochlorhydria (Fig. 3f).

Osteopetrorickets due to impaired gastric acidification

By analyzing the bone phenotype of Cckbr^-/- mice, we next addressed the question of whether hypochlorhydria results in impaired calcium homeostasis and bone mineralization. In Cckbr^-/- mice at 2 weeks of age, although osteoid volume was no larger, we did observe greater growth-plate thickness and significantly lower serum calcium levels compared to those of wild-type littermates (Fig. 4a). Moreover, the Cckbr^-/- mice had higher PTH levels and osteoclast numbers, suggesting that a more severe hypocalcemia is prevented by calcium mobilization from the bone matrix. This observation raised the possibility that the combination of hypochlorhydria and osteoclast dysfunction could explain the osteopetrorickets phenotype of oc/oc mice. To confirm this hypothesis, we generated mice deficient in both Cckbr and Src, thereby combining hypochlorhydria and osteoclast dysfunction caused by two independent genetic defects. As expected, when we analyzed the Cckbr^-/-Src^-/- mice at 2 weeks of age, we found a rachitic widening of the growth plates, a pathologic increase of the osteoid volume and early lethality, three phenotypes that are observed neither in Cckbr^-/- nor in Src^-/- mice (Fig. 4b). The lethality caused by the combined deficiency of Cckbr and Src is comparable to that of the oc/oc mice, and both the hypocalcemia (7.2 0.7 mg calcium per dl serum) and the hyperparathyroidism (214 82 mg PTH per ml serum) were more pronounced than in mice lacking Cckbr alone.

Figure 4: Osteopetrorickets caused by a combined defect of bone resorption and gastric acidification.

(a) Top, toluidine blue–stained sections from tibias of 2-week-old WT and Cckbr^-/- mice. The white bar in each panel indicates the growth plate (scale bar, 100 m). Bottom, TRAP activity staining of osteoclasts (stained in red; scale bar, 50 m). Quantification of GpTh, serum calcium, PTH and osteoclast number per bone surface (OcN/BS) is given on the right. ^*P < 0.05 versus WT (n = 5 mice per group). (b) Von Kossa/van Gieson staining of non-decalcified sections from tibias of 2-week-old Cckbr^-/- and Cckbr^-/-Src^-/- mice. Scale bars, 500 m and 200 m. Quantification of BV/TV, GpTh, OV/BV and survival curves of three mutant strains are given at right. ^*P < 0.05 versus Cckbr^-/- littermates (n = 5 mice per group). (c) Immunohistochemistry on decalcified human bone biopsy specimens demonstrating osteoclast-specific presence of TCIRG1, CLC-7 and SLC4a2, but no expression of ATP4b. Arrows indicate multinucleated osteoclasts. Scale bars, 100 m. (d) Contact radiographs (scale bar, 2 mm) and von Kossa/van Gieson staining of non-decalcified sections from vertebral bodies (lower left) and tibias (middle top) from 2-week-old WT and Slc4a2^-/- mice. Scale bars, 500 m (vertebrae) and 200 m (tibias). Quantification of serum calcium, PTH, BV/TV and OV/BV is also shown. ^**P < 0.005 versus WT (n = 6 mice per group).

Full size image (80 KB)

We next investigated whether mice lacking the anion exchanger Slc4a2 would have a similar phenotype, given that Slc4a2^-/- mice display not only achlorhydria but also failure of tooth eruption, a typical feature of osteopetrosis²³. To address a possible function of Slc4a2 in bone resorption, we first performed immunohistochemistry on sections from decalcified human bone biopsy specimens and found that Slc4a2 was present on osteoclasts (Fig. 4c). Consistent with this finding, Slc4a2^-/- mice, which do not survive beyond the age of 3 weeks, had high bone mass, impaired calcium homeostasis and osteoid enrichment—all hallmarks of osteopetrorickets (Fig. 4d). Taken together, these data demonstrate that impaired gastric acidification negatively affects calcium homeostasis and that the combination of this condition with an osteoclast defect causes more severe hypocalcemia and osteopetrorickets.

Cckbr^-/- mice develop an osteoporotic phenotype

Because the analysis of 2-week-old Cckbr^-/- mice revealed secondary hyperparathyroidism, we next addressed the question of whether hypochlorhydria, in the absence of an osteoclast defect, would trigger osteoporosis. We therefore analyzed the skeletal phenotype of wild-type and Cckbr^-/- mice at 12 and 52 weeks of age. Using histomorphometry following non-decalcified sectioning of vertebral bodies, we found that trabecular bone volume was significantly lower in the Cckbr^-/- mice at both ages (Fig. 5a,b). We found the same in sections from the tibia, where we also observed severe cortical porosity in 52-week-old Cckbr^-/- mice, which we did not see in wild-type controls. To understand the underlying cause of the observed phenotype, we first performed dynamic histomorphometry following dual injection of calcein. Although we did not observe any defect of bone matrix mineralization in the Cckbr^-/- mice, their rates of bone formation were significantly lower than those of wild-type controls (Fig. 5c). Yet neither osteocalcin level nor alkaline phosphatase activity were significantly lower in the serum of Cckbr^-/- mice versus wild-type controls, suggesting that their osteoporotic phenotype is caused primarily by higher levels of bone resorption (Fig. 5d).

Figure 5: Osteoporotic phenotype of Cckbr^-/- mice.

(a) Non-decalcified histology of vertebral bodies (top) and tibias (bottom) from 12- and 52-week-old WT and Cckbr^-/- mice. Arrow indicates cortical porosity in the 52-week-old Cckbr^-/- mice. Scale bars, 1 mm. (b) Histomorphometric quantification of BV/TV (top, vertebral bodies; bottom, tibia) and trabecular thickness (TbTh). ^*P < 0.05 versus WT (n = 6 mice per group). (c) Fluorescence micrographs showing calcein-labeled bone surfaces (left) and the deduced bone formation rate per bone surface (BFR/BS) (right) in 52-week-old WT and Cckbr^-/- mice. Scale bars, 20 m. ^*P < 0.05 versus WT (n = 6 mice per group). (d) Serum osteocalcin levels (OC) and alkaline phosphatase activity (AP) of 52-week-old WT and Cckbr^-/- mice. (e) Contact radiographs (top; scale bars, 5 mm and 500 m), non-decalcified histology (middle; scale bars, 500 m) and TRAP activity staining (bottom; scale bars, 500 m). Arrows indicate the severe cortical porosity caused by increased bone resorption in the Cckbr^-/- mice. (f) Histomorphometric quantification of fibrous tissue per bone surface (FT/BS) and osteoclast surface per bone surface (OcS/BS). ^**P < 0.005 versus WT (n = 6 mice per group). (g) Urinary abundance of Dpd cross-links in WT and Cckbr^-/- mice. Values were normalized to creatinine to control for differences in urine water content. ^*P < 0.05 versus WT (n = 6 mice per group). (h) Three-point bending assays of the femurs of WT and Cckbr^-/- mice. F_max, required force in newtons (N) until bone failure. ^*P < 0.05 versus WT (n = 6 mice per group).

Full size image (97 KB)

In fact, although contact radiography and histologic analysis confirmed the presence of cortical lesions in the femurs of Cckbr^-/- mice as well, activity staining for the osteoclast marker TRAP (tartrate-resistant acid phosphatase) revealed that this was due to higher levels of bone resorption (Fig. 5e). To quantify these observations, we performed histomorphometry, which showed that fibrotic tissue and osteoclast number, both indicative of secondary hyperparathyroidism, were pathologically elevated in the cortical bone of the Cckbr^-/- mice (Fig. 5f). Higher levels of systemic bone resorption in the Cckbr^-/- mice was further confirmed by the elevated concentrations of bone-specific collagen degradation products (Dpd cross-links) in the urine (Fig. 5g). Finally, to demonstrate that these abnormalities negatively affected biomechanical bone stability, we performed three-point bending assays of the femurs and observed that the required force until bone fracture was significantly lower in the absence of Cckbr (Fig. 5h).

The Cckbr^-/- phenotype is prevented by calcium supplementation

We next assessed whether normalization of calcium homeostasis through dietary manipulation could prevent the osteoporotic phenotype of Cckbr^-/- mice. Our reference diet included 0.8% calcium carbonate and 500 IU kg^-1 vitamin D3, and we also used the same diet with 2% calcium carbonate and 2,000 IU kg^-1 vitamin D3. In addition, because calcium carbonate is known to be poorly absorbed at elevated gastric pH, we also tested the effect of replacing calcium carbonate with calcium gluconate, which has greater solubility at neutral pH. To compare the efficiency of calcium carbonate and calcium gluconate in preventing hypochlorhydria-induced bone loss, we fed Cckbr^-/- mice one of the four diets (0.8% calcium carbonate or calcium gluconate, 2% calcium carbonate or calcium gluconate) and studied their skeletal phenotypes at the age of 52 weeks.

We first analyzed vertebral bodies by contact radiography, which revealed higher bone density in the groups receiving either calcium gluconate or 2% calcium carbonate, results subsequently confirmed by non-decalcified histology and histomorphometry (Fig. 6a,b). Quantification of serum PTH and osteoclast numbers on the trabecular bone surfaces provided an explanation for this observation. Although Cckbr^-/- mice receiving the control diet (0.8% calcium carbonate) had secondary hyperparathyroidism and pathologically elevated numbers of osteoclasts, both parameters were significantly improved by the other diets, but only the 2% calcium gluconate diet resulted in complete normalization (Fig. 6c). We observed the same pattern when we determined the biomechanical stability of the vertebral bodies by microcompression testing (Fig. 6d).

Figure 6: Hypochlorhydria-induced bone loss is prevented by calcium gluconate supplementation.

(a) Contact radiographs (top) and von Kossa/van Gieson staining (bottom) of non-decalcified sections from vertebral bodies of Cckbr^-/- mice receiving diets containing 0.8% or 2% of either calcium carbonate or calcium gluconate. Scale bars, 2 mm and 1 mm. (b–d) Quantification of BV/TV (b), serum PTH and OcN/BS (c) and microcompression testing of the vertebrae (d). ^*P < 0.05 for calcium gluconate versus calcium carbonate (n = 6 mice per group). (e) Cross-sectional CT scanning (top) and von Kossa/van Gieson staining (bottom) of non-decalcified sections from femurs. Scale bars, 500 m and 100 m. (f) Quantification of the cortical porosity and data from three-point bending assays of the femora. ^*P < 0.05 for calcium gluconate versus calcium carbonate (n = 6 mice per group). (g) Proposed model for the impact of hypochlorhydria on calcium homeostasis and bone mass. In healthy individuals (top), intact gastric acidification is a prerequisite for efficient intestinal uptake of calcium (blue circles). In hypochlorhydria (middle), intestinal calcium uptake (blue arrow) is lower, thereby leading to secondary hyperparathyroidism, osteoclast activation and osteoporosis. In the case of a combined acidification defect of gastric parietal cells and osteoclasts (bottom), PTH-dependent activation of bone resorption is blocked, resulting in severe hypocalcemia and osteopetrorickets.

Full size image (97 KB)

We next analyzed the femurs of hypochlorhydric Cckbr^-/- mice, quantifying their cortical porosity by cross-sectional micro-computed tomography (CT) scanning and by histology and their biomechanical stability through three-point bending assays. Although mice receiving either 2% calcium carbonate or 0.8% calcium gluconate had reduced cortical porosity, only the diet containing 2% calcium gluconate fully prevented the osteoporotic phenotype (Fig. 6e,f). Likewise, at both concentrations (0.8% and 2%), calcium gluconate was more efficient than calcium carbonate both in reducing cortical porosity and in increasing biomechanical stability of the mouse femurs. Taken together, these results not only demonstrate that hypochlorhydria-induced bone loss is a consequence of secondary hyperparathyroidism but also reveal that calcium gluconate is the most effective calcium formulation for preventing this phenotype, at least in mice.

Acknowledgments

We thank O. Winter, M. Dietzmann, C. Erdmann, T.O. Klatte, S. Kessler, G. Arndt and S. Conrad for technical assistance. Moreover, we are grateful to A.S. Kopin (Tufts Medical Center) and G.E. Shull (University of Cincinnati) for providing the Cckbr^-/- and Slc4a2^+/- mice, respectively. This work was supported by grants from the Deutsche Forschungsgemeinschaft to M.A. (AM103/14-1) and M.B. (BL423/4-3), from the Deutsches Zentrum für Luft- und Raumfahrt within the framework of the E-Rare JTC 2007 to M.A., A.S., U.K., A.T. and A.V., from Telethon to A.T. (GGP06119) and by the NOBEL program from Fondazione Cariplo to A.V.

Received 9 December 2008; Accepted 6 April 2009; Published online 17 May 2009.

References

1. Harada, S. & Rodan, G.A. Control of osteoblast function and regulation of bone mass. Nature 423, 349–355 (2003). | Article | PubMed | ISI | ChemPort |
2. Teitelbaum, S.L. & Ross, F.P. Genetic regulation of osteoclast development and function. Nat. Rev. Genet. 4, 638–649 (2003). | Article | PubMed | ISI | ChemPort |
3. Karaplis, A.C. & Goltzman, D. PTH and PTHrP effects on the skeleton. Rev. Endocr. Metab. Disord. 1, 331–341 (2000). | Article | PubMed | ChemPort |
4. Kaplan, F.S., August, C.S., Fallon, M.D., Gannon, F. & Haddad, J.G. Osteopetrorickets. The paradox of plenty. Pathophysiology and treatment. Clin. Orthop. Relat. Res. 294, 64–78 (1993). | PubMed |
5. Taranta, A. et al. Genotype-phenotype relationship in human ATP6i-dependent autosomal recessive osteopetrosis. Am. J. Pathol. 162, 57–68 (2003). | PubMed | ChemPort |
6. Kirubakaran, C., Ranjini, K., Scott, J.X., Basker, M. & Sridhar, G. Osteopetrorickets. J. Trop. Pediatr. 50, 185–186 (2004). | Article | PubMed |
7. Del Fattore, A., Cappariello, A. & Teti, A. Genetics, pathogenesis and complications of osteopetrosis. Bone 42, 19–29 (2008). | Article | PubMed | ChemPort |
8. Sly, W.S. et al. Carbonic anhydrase II deficiency in 12 families with the autosomal recessive syndrome of osteopetrosis with renal tubular acidosis and cerebral calcification. N. Engl. J. Med. 313, 139–145 (1985). | PubMed | ISI | ChemPort |
9. Kornak, U. et al. Loss of the ClC-7 chloride channel leads to osteopetrosis in mice and man. Cell 104, 205–215 (2001). | Article | PubMed | ISI | ChemPort |
10. Lange, P.F., Wartosch, L., Jentsch, T.J. & Fuhrmann, J.C. ClC-7 requires Ostm1 as a beta-subunit to support bone resorption and lysosomal function. Nature 440, 220–223 (2006). | Article | PubMed | ChemPort |
11. Nishi, T. & Forgac, M. The vacuolar H⁺-ATPases—nature's most versatile proton pumps. Nat. Rev. Mol. Cell Biol. 3, 94–103 (2002). | Article | PubMed | ISI | ChemPort |
12. Li, Y.P., Chen, W., Liang, Y., Li, E. & Stashenko, P. Atp6i-deficient mice exhibit severe osteopetrosis due to loss of osteoclast-mediated extracellular acidification. Nat. Genet. 23, 447–451 (1999). | Article | PubMed | ISI | ChemPort |
13. Scimeca, J.C. et al. The gene encoding the mouse homologue of the human osteoclast-specific 116-kDa V-ATPase subunit bears a deletion in osteosclerotic (oc/oc) mutants. Bone 26, 207–213 (2000). | Article | PubMed | ISI | ChemPort |
14. Marks, S.C. Jr., Seifert, M.F. & Lane, P.W. Osteosclerosis, a recessive skeletal mutation on chromosome 19 in the mouse. J. Hered. 76, 171–176 (1985). | PubMed | ISI |
15. Frattini, A. et al. Defects in TCIRG1 subunit of the vacuolar proton pump are responsible for a subset of human autosomal recessive osteopetrosis. Nat. Genet. 25, 343–346 (2000). | Article | PubMed | ISI | ChemPort |
16. Kornak, U. et al. Mutations in the a3 subunit of the vacuolar H⁺-ATPase cause infantile malignant osteopetrosis. Hum. Mol. Genet. 9, 2059–2063 (2000). | Article | PubMed | ISI | ChemPort |
17. Sobacchi, C. et al. The mutational spectrum of human malignant autosomal recessive osteopetrosis. Hum. Mol. Genet. 10, 1767–1773 (2001). | Article | PubMed | ISI | ChemPort |
18. Del Fattore, A. et al. Clinical, genetic, and cellular analysis of 49 osteopetrotic patients: implications for diagnosis and treatment. J. Med. Genet. 43, 315–325 (2006). | Article | PubMed | ChemPort |
19. Banco, R., Seifert, M.F., Marks, S.C. Jr. & McGuire, J.L. Rickets and osteopetrosis: the osteosclerotic (oc) mouse. Clin. Orthop. Relat. Res. 201, 238–246 (1985). | PubMed |
20. Li, Y.P., Chen, W. & Stashenko, P. Molecular cloning and characterization of a putative novel human osteoclast-specific 116-kDa vacuolar proton pump subunit. Biochem. Biophys. Res. Commun. 218, 813–821 (1996). | Article | PubMed | ISI | ChemPort |
21. Soriano, P., Montgomery, C., Geske, R. & Bradley, A. Targeted disruption of the c-src proto-oncogene leads to osteopetrosis in mice. Cell 64, 693–702 (1991). | Article | PubMed | ISI | ChemPort |
22. Langhans, N. et al. Abnormal gastric histology and decreased acid production in cholecystokinin-B/gastrin receptor-deficient mice. Gastroenterology 112, 280–286 (1997). | Article | PubMed | ISI | ChemPort |
23. Gawenis, L.R. et al. Mice with a targeted disruption of the AE2 Cl^-/HCO₃^- exchanger are achlorhydric. J. Biol. Chem. 279, 30531–30539 (2004). | Article | PubMed | ChemPort |
24. Wilson, C.J. & Velodi, A. Autosomal recessive osteopetrosis: diagnosis, management, and outcome. Arch. Dis. Child. 83, 449–452 (2000). | Article | PubMed | ISI | ChemPort |
25. Driessen, G.J. et al. Long-term outcome of haematopoietic stem cell transplantation in autosomal recessive osteopetrosis: an EBMT report. Bone Marrow Transplant. 32, 657–663 (2003). | Article | PubMed | ISI | ChemPort |
26. Frattini, A. et al. Chloride channel ClCN7 mutations are responsible for severe recessive, dominant, and intermediate osteopetrosis. J. Bone Miner. Res. 18, 1740–1747 (2003). | Article | PubMed | ISI | ChemPort |
27. Kasper, D. et al. Loss of the chloride channel ClC-7 leads to lysosomal storage disease and neurodegeneration. EMBO J. 24, 1079–1091 (2005). | Article | PubMed | ISI | ChemPort |
28. Pangrazio, A. et al. Mutations in OSTM1 (grey lethal) define a particularly severe form of autosomal recessive osteopetrosis with neural involvement. J. Bone Miner. Res. 21, 1098–1105 (2006). | Article | PubMed | ISI | ChemPort |
29. Chalhoub, N. et al. Grey-lethal mutation induces severe malignant autosomal recessive osteopetrosis in mouse and human. Nat. Med. 9, 399–406 (2003). | Article | PubMed | ISI | ChemPort |
30. Kassarjian, Z. & Russell, R.M. Hypochlorhydria: a factor in nutrition. Annu. Rev. Nutr. 9, 271–285 (1989). | Article | PubMed | ChemPort |
31. Aoki, K. et al. Comparison of prevalence of chronic atrophic gastritis in Japan, China, Tanzania, and the Dominican Republic. Ann. Epidemiol. 15, 598–606 (2005). | Article | PubMed |
32. Jacobson, B.C. et al. Who is using chronic acid suppression therapy and why? Am. J. Gastroenterol. 98, 51–58 (2003). | Article | PubMed |
33. Recker, R.R. Calcium absorption and achlorhydria. N. Engl. J. Med. 313, 70–73 (1985). | PubMed | ISI | ChemPort |
34. O'Connell, M.B., Madden, D.M., Murray, A.M., Heaney, R.P. & Kerzner, L.J. Effects of proton pump inhibitors on calcium carbonate absorption in women: a randomized crossover trial. Am. J. Med. 118, 778–781 (2005). | Article | PubMed | ISI | ChemPort |
35. Yang, Y.-X., Lewis, J.D., Epstein, S. & Metz, D.C. Long-term proton pump inhibitor therapy and risk of hip fracture. J. Am. Med. Assoc. 296, 2947–2953 (2006). | Article | ChemPort |
36. Cummings, S.R. & Melton, L.J. Epidemiology and outcomes of osteoporotic fractures. Lancet 359, 1761–1767 (2002). | Article | PubMed | ISI |
37. Straub, D.A. Calcium supplementation in clinical practice: a review of forms, doses, and indications. Nutr. Clin. Pract. 22, 286–296 (2007). | Article | PubMed |
38. Seitz, S. et al. Paget's disease of bone—histologic analysis of 754 patients. J. Bone Miner. Res. 24, 62–69 (2009). | Article | PubMed |
39. Parfitt, A.M. et al. Bone histomorphometry: standardization of nomenclature, symbols, and units. Report of the ASBMR Histomorphometry Nomenclature Committee. J. Bone Miner. Res. 2, 595–610 (1987). | PubMed | ISI | ChemPort |
40. Huebner, A.K. et al. Calcitonin deficiency in mice progressively results in high bone turnover. J. Bone Miner. Res. 21, 1924–1934 (2006). | Article | PubMed | ChemPort |
41. Hoff, A.O. et al. Increased bone mass is an unexpected phenotype associated with deletion of the calcitonin gene. J. Clin. Invest. 110, 1849–1857 (2002). | Article | PubMed | ChemPort |
42. Schmidt, K. et al. The high mobility group transcription factor Sox8 is a negative regulator of osteoblast differentiation. J. Cell Biol. 168, 899–910 (2005). | Article | PubMed | ChemPort |
43. Amling, M. et al. Bcl-2 lies downstream of parathyroid hormone-related peptide in a signaling pathway that regulates chondrocyte maturation during skeletal development. J. Cell Biol. 136, 205–213 (1997). | Article | PubMed | ISI | ChemPort |
44. Simon, R., Mirlacher, M. & Sauter, G. Tissue microarrays. Biotechniques 36, 98–105 (2004). | PubMed | ISI | ChemPort |
45. Takagi, H., Jhappan, C., Sharp, R. & Merlino, G. Hypertrophic gastropathy resembling Ménétrier's disease in transgenic mice overexpressing transforming growth factor alpha in the stomach. J. Clin. Invest. 90, 1161–1167 (1992). | Article | PubMed | ISI | ChemPort |

Online methods

Human bone biopsy specimens. A collection of transiliac bone biopsy specimens from non-genotyped individuals with various disorders was established at the Institute of Pathology at Hamburg University Medical Center over the last 36 years, including 21 specimens from individuals with osteopetrosis³⁸. All of the latter individuals received tetracycline twice orally before biopsy. Biopsy specimens were embedded non-decalcified into methylmethacrylate and stored light protected thereafter. Sections were subjected to von Kossa/van Gieson and toluidine blue staining, and histomorphometric analysis was carried out using the OsteoMeasure system (Osteometrics) following the guidelines of the American Society for Bone and Mineral Research³⁹. A case of TCIRG1-dependent osteopetrosis in an 18-month-old child was confirmed by automated sequencing of all TCIRG1-encoding exons and also by conventional X-ray analysis demonstrating growth-plate widening. A bone biopsy was taken during bone marrow transplantation and subjected to non-decalcified histology as described above. Age- and sex-matched control biopsy specimens for all subjects were taken from skeletal intact donors. Experiments were approved by the ethics committee of the Aerztekammer Hamburg (OB-020/07), and informed consent was obtained from human subjects.

Mice. Oc/+ and Src^+/- mice were obtained from the Jackson Laboratory. Cckbr^-/- mice and Slc4a2^+/- mice were kindly provided by A.S. Kopin and by G.E. Shull, respectively. Unless stated otherwise, all mice were fed a standard rodent diet containing 1% calcium. For the determination of life span, Src^-/-, Cckbr^-/-Src^-/- and oc/oc mice were fed a liquid diet containing 0.8% calcium carbonate (Altromin no. C0199) starting at 2 weeks of age. The same diet containing 2% calcium carbonate was given to oc/oc mice to normalize their serum calcium. For the treatment of hypochlorhydria-induced bone loss, Cckbr^-/- mice were fed a diet (Altromin no. C1032) containing either 0.8% calcium carbonate or 0.8% calcium gluconate. Another group of mice was given the same diet supplemented with higher concentrations of vitamin D3 (2000 IU/kg) and either 2% calcium carbonate or calcium gluconate. Experiments were approved by the animal facility of the University Medical Center Hamburg-Eppendorf and by the Amt für Gesundheit und Verbraucherschutz (Org139).

Skeletal analysis. The skeletal analysis, including contact X-ray, CT scanning, non-decalcified bone histology and TRAP activity staining, has been described previously⁴⁰. Femoral length and growth-plate thickness were determined with a microscopic ruler. To determine the bone formation rate, Cckbr^-/- mice were injected twice with calcein before being killed. Biomechanical stability of the femora was determined by three-point bending assays as described, and maximum strength and overall stiffness of the vertebrae was determined by microcompression testing⁴¹.

Primary cell culture. Cultures of primary osteoclasts, chondrocytes or osteoblasts were generated as described previously^{40, 42, 43}. For the assessment of ex vivo mineralization, primary osteoblasts were isolated from single calvariae of 3-day-old wild-type and oc/oc littermates and stained by the method of von Kossa at two stages of differentiation induced by ascorbic acid and -glycerophosphate. Alkaline phosphatase activity was determined using p-nitrophenylphosphate as a substrate (Ponte Scientific, no. A7516-150), and osteocalcin levels were measured by radioimmunoassay (Immutopics, no. 50-1300).

Expression analysis. RNA was isolated using Trizol reagent (Invitrogen). A full-length cDNA encoding Tcirg1 was obtained from RZPD (no. 2416867) and used for northern blotting following standard protocols. For in situ hybridization, the Tcirg1 cDNA was subcloned into pBluescript (Stratagene) to generate [³⁵S]UTP-labeled sense and antisense probes using the riboprobe combination system (Promega). For RT-PCR, RNA was reverse transcribed using the Cloned AMV First-Strand cDNA synthesis Kit (Invitrogen). The PCR-amplified cDNA fragments were verified by automatic sequencing. Immunohistochemistry was performed according to standard protocols using antibodies against human TCIRG1 (Abnova, no. H00010312-M01), CLC-7 (Santa Cruz, no. sc-28755), ATP4b (Abcam, no. ab2866) and SLC4a2 (Santa Cruz, no. sc-46710), respectively, all diluted 100-fold. The tissue microarray experiment was performed as described previously⁴⁴.

Analysis of calcium homeostasis. Calcium concentrations were measured using a colorimetric assay (BioAssay Systems, no. DICA-500). Serum PTH was quantified by ELISA (Immutopics, no. 60-2300). Urinary deoxypyridinoline (Dpd) cross-links were measured using Pyrilinks-D ELISA (Metra Biosystems, no. 8007) and normalized to creatinine level as determined by an alkaline picrate assay (Metra Biosystems, no. 8009). For human subjects, we applied the routine measurements that were performed at the Institute of Clinical Chemistry of the University Medical Center Hamburg-Eppendorf for all parameters shown in Figure 3f.

Determination of gastric pH. The gastric pH of 2-week-old mice was measured using a modification of a published method^{22, 45}. In brief, mice were fasted for 2 h and then their esophagus and pylorus were ligated under anesthesia. After injection of 0.5 ml saline, the fluid within the gastric lumen was collected for pH measurement. The gastric pH of an individual with TCIRG1-dependent osteopetrosis and control individuals was monitored for 24 h at the Department of Gastroenterology of the University Medical Center Ulm.

Statistical analysis. Results are either presented as box plots, including data from the 25th to 75th percentile, or as bar graphs, indicating mean s.d. Statistical analysis was performed using an unpaired, two-tailed Student's t-test, and P values <0.05 were considered statistically significant.

MORE ARTICLES LIKE THIS

These links to content published by NPG are automatically generated.

NPG Journals

by Subject Area

• Chemistry

  • Chemistry
  • Drug discovery
  • Biotechnology
  • Materials
  • Methods & Protocols

• Clinical Practice & Research

  • Cancer
  • Cardiovascular medicine
  • Dentistry
  • Endocrinology
  • Gastroenterology & Hepatology
  • Methods & Protocols
  • Pathology & Pathobiology
  • Urology

• Earth & Environment

  • Earth sciences
  • Evolution & Ecology

• Life sciences

  • Biotechnology
  • Cancer
  • Development
  • Drug discovery
  • Evolution & Ecology
  • Genetics
  • Immunology
  • Medical research
  • Methods & Protocols
  • Microbiology
  • Molecular cell biology
  • Neuroscience
  • Pharmacology
  • Systems biology

• Physical sciences

  • Physics
  • Materials

by A - Z Index