Endocrinology and Metabolism

1,25-Dihydroxyvitamin D3 contributes to regulating mammary calcium transport and modulates neonatal skeletal growth and turnover cooperatively with calcium

Ji Ji, Ruinan Lu, Xiaojie Zhou, Yingben Xue, Chunmin Shi, David Goltzman, Dengshun Miao

Abstract

To assess the interaction of 1,25(OH)2D3 and dietary calcium on mammary calcium transport in lactating dams and skeletal growth and turnover in the neonate, female lactating 1α(OH)ase+/− or 1α(OH)ase−/− mice were fed either a high-calcium diet containing 1.5% calcium in the drinking water or a “rescue diet.” Dietary effects on the expression of molecules mediating mammary calcium transport were determined in the dams, and the effects of milk calcium content were assessed on skeletal growth and turnover in 2-wk-old 1,25(OH)2D3-deficient pups. Results showed that the reduction of milk calcium levels in the 1α(OH)ase−/− dams and the elevation of milk calcium levels in dams fed the rescue diet were associated with the down- or upregulation of calbindin D9k and plasma membrane Ca2+ ATPase isoform 2b expression, respectively, in mammary epithelial cells. The action of ambient calcium in stimulating skeletal growth in the neonates appeared to supercede the direct action of 1,25(OH)2D3, and the response of chondrocytes in the neonates to elevated calcium was more sensitive in hypocalcemic animals. Osteopenia was more apparent in pups nursed by dams with lower milk calcium than in 1,25(OH)2D3-deficient pups nursed by dams with higher milk calcium. Bone formation parameters were increased significantly in all pups fed by dams on the rescue diet but were still lower in 1α(OH)ase−/− pups than in 1α(OH)ase+/− pups. Consequently, there is an important contributory role of calcium in conjunction with 1,25(OH)2D3 to mammary calcium transport in lactating dams and skeletal growth and turnover in the neonate.

  • vitamin D
  • dietary calcium

vitamin d is essential for the maintenance of a mineralized skeleton. We (29) and others (9) have previously described a mouse model with targeted ablation of the gene encoding the enzyme 25-hydroxyvitamin D 1αhydroxylase [1α(OH)ase] that results in 1α,25(OH)2D deficiency. After weaning, mice fed regular mouse chow developed secondary hyperparathyroidism, retarded growth, and skeletal abnormalities characteristic of rickets and osteomalacia. These abnormalities mimic those described in the human genetic disorder VDDR-I (11, 13). When the phenotype of 1α(OH)ase−/− mice was analyzed, we found that the skeletal phenotype was different before and after weaning. At 2 wk of age, the trabecular volume and osteoblast numbers in the 1α(OH)ase−/− mice were decreased, and the osteoid volume was not increased significantly (45). In contrast, at 4 mo of age, the trabecular volume, osteoblast number, and osteoid volume were all increased significantly in the 1α(OH)ase−/− mice, even on a high-calcium diet of 1.5% calcium in the drinking water (28). These differences were thought to result from the elevations in circulating parathyroid hormone (PTH) along with hypocalcemia and hypophosphatemia. The serum PTH was increased 0.5-fold at 2 wk of age (45), but 30-fold at 4 mo of age (28), in the 1α(OH)ase−/− mice compared with their wild-type counterparts. This possibility was confirmed in the 1α(OH)ase−/− mice fed a “rescue diet” containing 2% calcium, 1.25% phosphorus, and 20% lactose. After serum PTH, calcium, and phosphate levels were normalized, skeletal mineralization was normalized, and the trabecular volume and osteoblast number were reduced in the 1α(OH)ase−/− mice (28). These findings suggest that the secondary hyperparathyroidism occurring in the 1α(OH)ase−/− mice on the normal diet was responsible for the increased bone formation. Moreover, the bone volume of the normocalcemic 1α(OH)ase−/− mice on the rescue diet was below that observed in the wild-type mice, suggesting that 1,25-dihydroxyvitamin D [1,25(OH)2D] was necessary for baseline bone formation. However, it is unclear whether the effect of enhancing dietary calcium absorption that we observed in adult life also occurs in neonates.

Considerable evidence demonstrates that sufficient calcium intake augments bone gain during growth and may retard age-related bone loss and reduce osteoporotic fracture risk (16, 21, 31, 36), but it is unknown what the mechanisms of the calcium intake are that result in a positive bone mass balance. Calcium intake is especially crucial during pregnancy and lactation because of the potential adverse effect on maternal bone health if maternal calcium stores are depleted (39). There is often a transient lowered bone mineral density and increased rate of bone resorption, with the greatest effect taking place during the third trimester and throughout lactation. Some studies indicate that calcium consumption should be encouraged, especially during pregnancy and lactation, to replace maternal skeletal calcium stores that are depleted during these periods, although it has also been reported that calcium supplementation does not prevent bone loss in the mother during lactation and only slightly enhances the gain in bone density after weaning (20). Because the fetus in utero and the neonate through breastfeeding are dependent on maternal sources for acquired calcium, adequate maternal calcium intake also can affect fetal bone health positively. Proper calcium consumption can ensure maternal and fetal bone health without the danger of adverse effects on the neonate.

During lactation, mammary epithelial cells extract large quantities of ionized calcium from plasma and produce a calcium-rich secretion. In doing so the mammary gland generates a large transepithelial calcium gradient that favors increased calcium in milk (15). The process of transepithelial calcium transport consists of at least three important steps (37, 42). The first is the transport of free ionized calcium across the blood surface (i.e., basolateral membrane) of the mammary epithelium. The second step is the transfer of calcium across the epithelial cells. The third step involves the movement of calcium across the apical membrane of the secretory epithelial cell and into milk. Of the three steps in transcellular calcium transport, the least well characterized in the mammary gland is calcium entry. Recently, two “apical” epithelial calcium entry channels, Trpv5 and Trpv6, that are thought to be involved in calcium absorption/reabsorption in the small intestine and kidney were identified (18, 30). However, in the mouse mammary gland, Trpv5 expression was reported to be lower during pregnancy and lactation compared with the nulliparous gland, and Trpv6 expression was undetectable during lactation, although it was expressed in the virgin mammary gland and, in decreasing amounts, throughout pregnancy (42). Therefore, the calcium entry mechanism is still unknown in the lactating mammary gland. Calcium-binding proteins could be involved in the process of shuttling calcium across the epithelial cell, although at present there is no obvious candidate for this role (41, 42). By avidly binding free calcium within the cell, calbindins protect against calcium toxicity and allow calcium signaling to occur at the same time as calcium transport (3, 17). Calbindins are of major importance in calcium absorption and reabsorption in the small intestine and kidney, respectively (17). However, a previous study did not find significantly increased expression of calbindins in the mouse mammary gland during lactation compared with other developmental stages (41). Previous studies have suggested that plasma membrane Ca2+ ATPase isoform 2b (PMCA2b) is responsible for the majority of calcium transported from the cytoplasm across the apical membrane of the mammary epithelial cell into milk. PMCA2 has also been localized to the apical membrane by immunofluorescence and immunoelectron microscopy (41). Furthermore, a null mutation in the gene encoding PMCA2 markedly impairs the transport of calcium into mouse milk (33). Indeed, milk from PMCA2-null mice had 60% less calcium compared with the milk of wild-type mice. Therefore, Reinhardt et al. (33) concluded that calcium transport across the apical membrane via PMCA2 was a fundamentally important step in the process of milk calcium secretion. However, it is unknown whether PMCA2 is involved in the regulation of mammary calcium transport by 1,25-dihydroxyvitamin D3 [1,25(OH)2D3] and calcium.

In the present study, we examined 1) whether 1,25(OH)2D3 and dietary calcium can increase milk calcium concentration by upregulating expression of mammary calcium transport molecules and 2) whether the skeletal anabolic action of 1,25(OH)2D3 in neonates results not only from its direct action via the vitamin D receptor (VDR) to stimulate endochondral bone formation and osteoblastic bone formation but also from an indirect action mediated through increasing extracellular calcium concentrations. To test this hypothesis, 1α(OH)ase+/− and 1α(OH)ase−/− pups were generated, and nursing 1α(OH)ase+/− and 1α(OH)ase−/− dams were fed either a high-calcium diet containing 1.5% calcium in the drinking water or a rescue diet containing 2% calcium, 1.25% phosphorus, and 20% lactose. The effects of the genetic 1,25(OH)2D3 deficiency and of the dietary manipulations were then determined on the content of minerals in the milk and the expression of calcium-transporting machinery in lactating dams and on skeletal metabolism in 2-wk-old pups.

MATERIALS AND METHODS

Animals and in vivo experiments.

The derivation of the parental strain of 1α(OH)ase−/− mice by homologous recombination in embryonic stem cells and the genotyping of the mice were described previously by Panda et al. (29). To determine whether 1,25(OH)2D3 and calcium interact in neonatal bone formation, male 1α(OH)ase−/− and female 1α(OH)ase+/− or male 1α(OH)ase+/− and female 1α(OH)ase−/− mice were mated to generate 1α(OH)ase+/− and 1α(OH)ase−/− pups. The 1α(OH)ase+/− and 1α(OH)ase−/− dams were fed either a high-calcium diet containing 1.5% calcium in the drinking water or a rescue diet containing 2% calcium, 1.25% phosphorus, and 20% lactose (28). Lactose in the rescue diet is known to facilitate intestinal calcium absorption in the absence of 1,25(OH)2D. The litter size was equalized to five to six pups per dam to equalize suckling intensity. On day 14 postpartum, dams were injected intraperitoneally with 0.1 mIU of oxytocin (Sigma-Aldrich) and milked manually, as described previously (40). Blood was collected from 2-wk-old 1α(OH)ase+/− and 1α(OH)ase−/− pups, and serum was isolated for biochemical analysis. Then femurs and tibiae were removed for subsequent analysis. All animal experiments were carried out in compliance with and approval by the Institutional Animal Care and Use Committee at Nanjing Medical University.

Biochemistry and hormone analyses.

The milk was diluted 1:100 in distilled water, and the calcium concentration was measured with an atomic absorptiometer (45). Serum calcium and phosphorus were determined by autoanalyzer (Beckman Synchron 67; Beckman Instruments). Milk and serum 1,25(OH)2D3 were measured by radioimmunoassay (ImmunoDiagnostic Systems, Bolden, UK), and PTH-related protein (PTHrP) was measured by a two-site immunoradiometric assay (Immutopics, San Clemente, CA). Serum intact PTH was measured using an ELISA (Immutopics, San Clemente, CA).

Quantitative real-time RT-PCR.

RNA was isolated from mammary glands, using Trizol reagent (Invitrogen) according to the manufacturer's protocol. Reverse transcription reactions were performed using the SuperScript First-Strand Synthesis System (Invitrogen), as described previously (45). To determine the number of cDNA molecules in the reverse-transcribed samples, real-time PCR analyses were performed using the LightCycler system (Roche, Indianapolis, IN). PCR was performed using 2 μl of LightCycler DNA Master SYBR Green I (Roche), 0.25 μM of each 5′ and 3′ primer, and 2-μl samples or H2O to a final volume of 20 μl. The MgCl2 concentration was adjusted to 3 mM. Samples were denatured at 95°C for 10 s, with a temperature transition rate of 20°C/s. Amplification and fluorescence determination were carried out in four steps: denaturation at 95°C for 0 s, with a temperature transition rate of 20°C/s; annealing for 5 s, with a temperature transition rate of 8°C/s; extension at 72°C for 20 s, with a temperature transition rate of 4°C/s; and detection of SYBR Green fluorescence, which reflects the amount of double-stranded DNA, at 86°C for 3 s. The amplification cycle number was 35. To discriminate specific from nonspecific cDNA products, a melting curve was obtained at the end of each run. Products were denatured at 95°C for 3 s, and the temperature was then decreased to 58°C for 15 s and raised slowly from 58 to 95°C using a temperature transition rate of 0.1°C/s. To determine the number of copies of the targeted DNA in the samples, purified PCR fragments of known concentrations were serially diluted and served as external standards that were measured in each experiment. Data were normalized with GAPDH levels in the samples. The primer sequences of calbindin D9k and PMCA2b used for the real-time PCR were the same as described previously (32, 45).

Western blot analysis.

Proteins were extracted from mammary glands and quantitated using a commercial kit (Bio-Rad, Mississauga, ON, Canada). Thirty-microgram protein samples were fractionated by SDS-PAGE and transferred to nitrocellulose membranes. Immunoblotting was carried out as described (45) using antibodies against PMCA2b (Swant, Bellinzona, Switzerland) and β-tubulin (Santa Cruz Biotechnology, Santa Cruz, CA). Bands were visualized using enhanced chemiluminescence (Amersham) and quantitated by Scion Image Beta 4.02.

Radiography.

For radiography, femurs were removed and dissected free of soft tissue, and X-ray images were taken with a Faxitron (model 805; Faxitron X-ray) under constant conditions (22 kV, 4-min exposure) using Kodak X-Omat TL film (Eastman Kodak).

Microcomputed tomography.

Femurs obtained from 2-wk-old mice were dissected free of soft tissue, fixed overnight in 70% ethanol, and analyzed by microcomputed tomography (μCT) with a SkyScan 1072 scanner and associated analysis software (SkyScan, Antwerp, Belgium), as described (44). Briefly, image acquisition was performed at 100 kV and 98 μA with a 0.9° rotation between frames. During scanning, the samples were enclosed in tightly fitting plastic wrap to prevent movement and dehydration. Thresholding was applied to the images to segment the bone from the background. Two-dimensional images were used to generate three-dimensional renderings using the 3D Creator software supplied with the instrument. The resolution of the μCT images is 11.26 μm.

Histology.

Tibiae were removed and fixed in PLP fixative (2% paraformaldehyde containing 0.075 M lysine and 0.01 M sodium periodate) overnight at 4°C and processed histologically, as described (24). Proximal ends of tibiae were decalcified in EDTA glycerol solution for 5–7 days at 4°C. Decalcified right tibiae were dehydrated and embedded in paraffin, after which 5-μm sections were cut on a rotary microtome. The sections were stained with hematoxylin and eosin or histochemically for tartrate-resistant acid phosphatase (TRAP) activity or immunohistochemically, as described below. Alternatively, undecalcified left tibiae were embedded in LR White acrylic resin (London Resin, London, UK), and 1-μm sections were cut on an ultramicrotome. These sections were stained for mineral with the von Kossa staining procedure and counterstained with toluidine blue.

Histochemical staining for alkaline phosphatase and TRAP.

Enzyme histochemistry for alkaline phosphatase (ALP) activity was performed as described previously (25). Briefly, following preincubation overnight in 1% magnesium chloride in 100 mM Tris-maleate buffer (pH 9.2), dewaxed sections were incubated for 2 h at room temperature in a 100-mM Tris-maleate buffer containing naphthol AS-MX phosphate (0.2 mg/ml; Sigma) dissolved in ethylene glycol monomethyl ether (Sigma) as substrate and fast red TR (0.4 mg/ml; Sigma) as a stain for the reaction product. After washing with distilled water, the sections were counterstained with Vector methyl green nuclear counterstain (Vector Laboratories) and mounted with Kaiser's glycerol jelly.

Enzyme histochemistry for TRAP was performed as described previously (26). Dewaxed sections were preincubated for 20 min in buffer containing 50 mM sodium acetate and 40 mM sodium tartrate at pH 5.0. Sections were then incubated for 15 min at room temperature in the same buffer containing 2.5 mg/ml naphthol AS-MX phosphate (Sigma) in dimethylformamide as substrate and 0.5 mg/ml fast garnet GBC (Sigma) as a color indicator for the reaction product. After washing with distilled water, the sections were counterstained with methyl green and mounted in Kaiser's glycerol jelly.

Immunohistochemical staining.

Immunohistochemical staining for calbindin D9k, proliferating cell nuclear antigen (PCNA), and type I collagens were performed using the avidin-biotin-peroxidase complex technique with a polyclonal rabbit antibody specific to CaBP-9k (Swant, Bellinzona, Switzerland), mouse monoclonal antibody against PCNA (Medicorp, Montreal, QC, Canada), and affinity-purified goat anti-human type I collagen antibody (Southern Biotechnology Associates, Birmingham, AL). Briefly, dewaxed and rehydrated paraffin-embedded sections were incubated with methanol-hydrogen peroxide (1:10) to block endogenous peroxidase activity and then washed in Tris-buffered saline (pH 7.6). The slides were then incubated with the primary antibodies overnight at room temperature. After being rinsed with Tris-buffered saline for 15 min, tissues were incubated with biotinylated secondary antibody (Sigma). Sections were then washed and incubated with the Vectastain Elite ABC reagent (Vector Laboratories) for 45 min. After washing, brown pigmentation was produced using 3,3-diaminobenzidine (2.5 mg/ml). After being washed with distilled water, the sections were counterstained with Mayer's hematoxylin, dehydrated in graded ethanol and xylene, and mounted with Biomount medium.

Computer-assisted image analysis.

After hematoxylin and eosin staining or histochemical or immunohistochemical staining of sections from five mice of each genotype, images of fields were photographed with a Sony digital camera. Images of micrographs from single sections were digitally recorded using a rectangular template, and recordings were processed and analyzed using Northern Eclipse image analysis software, as described previously (24).

Statistical analysis.

Data from image analyses are presented as means ± SE. Statistical comparisons were made using a one-way ANOVA, with P < 0.05 being considered significant.

RESULTS

Effects of 1,25(OH)2D3 deficiency and dietary calcium on serum calcium, phosphorus, and PTH in virgin and lactating mice.

To determine whether 1,25(OH)2D3 deficiency and dietary calcium alter ion homeostasis and calcium-regulating hormone levels, serum calcium, phosphorus, and PTH were examined in virgin and lactating 1α(OH)ase+/− or 1α(OH)ase−/− mice fed either a high-calcium diet or a rescue diet. Serum calcium and phosphorus levels were reduced, but serum PTH levels were increased in both virgin and lactating 1α(OH)ase−/− mice compared with virgin and lactating 1α(OH)ase+/− mice when they were fed the high-calcium diet (Fig. 1, AC). When the rescue diet was administered, serum calcium and phosphorus levels in both virgin and lactating 1α(OH)ase−/− mice rose to levels observed in virgin and lactating 1α(OH)ase+/− mice, whereas serum PTH levels in both virgin and lactating 1α(OH)ase−/− mice were reduced to levels seen in virgin and lactating 1α(OH)ase+/− mice (Fig. 1, AC). Serum calcium levels were increased in both lactating 1α(OH)ase+/− or lactating 1α(OH)ase−/− mice compared with virgin genotype-matched mice on the same diet; however, serum phosphorus and PTH levels were not altered significantly between virgin and lactating mice (Fig. 1, AC).

Fig. 1.

Effects of 1,25-dihydroxyvitamin D3 [1,25(OH)2D3] deficiency and high dietary calcium on serum calcium, phosphorus and parathyroid hormone (PTH), milk calcium contents, PTH-related protein (PTHrP), and 1,25(OH)2D3 levels and mammary epithelial calcium transport in dams. Serum calcium (A), phosphorus (B), and PTH levels (C) were determined in samples from virgin and lactating 25-hydroxyvitamin D 1αhydroxylase [1α(OH)ase]+/− and 1α(OH)ase−/− mice on the high-calcium diet (HCa) or on the rescue diet (RD), as described in materials and methods. Milk calcium (D), PTHrP (E), 1,25(OH)2D3 (F), and protein levels (G) were determined in samples from 1α(OH)ase+/− and 1α(OH)ase−/− dams on day 14 of lactation on the high-calcium diet (high Ca) or on the rescue diet (rescue), as described in materials and methods. Comparison of calbindin D9k (H) and plasma membrane Ca2+ ATPase isoform 2b (PMCA2b; I) gene expression levels in mammary glands from dams on day 14 of lactation on the high Ca or on the rescue diet. Specific calbindin D9k and PMCA2b products were quantified from tissue RNAs by real-time RT-PCR, as described in materials and methods. Messenger RNA relative levels were calculated as a ratio to the GAPDH mRNA level and expressed relative to levels of 1α(OH)ase+/− dams on the high Ca. J: representative micrographs of sections of mammary glands stained immunohistochemically for calbindin D9k. Scale bars represent 25 μm. K: quantitative ratio of calbindin D9k-positive area relative to total tissue area (%). L: Western blots of mammary gland extracts were carried out for expression of PMCA2b, and β-tubulin was used as a loading control. M: PMCA2b protein levels relative to β-tubulin protein levels were assessed by densitometric analysis and are expressed relative to levels present in 1α(OH)ase+/− dams on high Ca. Each value is the mean ± SE of determinations in 5 mice of each group. ***P < 0.001 compared with 1α(OH)ase+/− dams; ##P < 0.01; ###P < 0.001 compared with genotype-matched dams on high Ca; ΔΔP < 0.01 compared with genotype-matched virgin mice.

Effects of dietary calcium on milk calcium content and calcium-regulating hormones in milk.

To determine whether 1,25(OH)2D3 deficiency and dietary calcium alter milk calcium content and calcium-regulating hormones, milk calcium, 1,25(OH)2D3, and PTHrP were examined in the lactating 1α(OH)ase+/− or 1α(OH)ase−/− dams fed either a high-calcium or rescue diet. Milk calcium levels were reduced, but milk PTHrP levels were increased in the 1α(OH)ase−/− dams compared with the 1α(OH)ase+/− dams when they were fed the high-calcium diet (Fig. 1, D and E). Milk 1,25(OH)2D3 levels were significantly lower in 1α(OH)ase+/− dams on the rescue diet compared with 1α(OH)ase+/− dams on the high-calcium diet and were undetectable in 1α(OH)ase−/− dams on any diet (Fig. 1F). Milk calcium levels were increased, but milk PTHrP levels were decreased in both 1α(OH)ase+/− and 1α(OH)ase−/− dams on the rescue diet compared with the genotype-matched dams on the high-calcium diet (Fig. 1, D and E). Milk protein concentrations were not significantly different in the two genotypes on either diet (Fig. 1G).

Effects of 1,25(OH)2D3 deficiency and dietary calcium on mammary calcium transport.

To determine whether the alteration of milk calcium content caused by 1,25(OH)2D3 deficiency and dietary calcium was associated with regulation of mammary transcellular calcium transport, the expression of calbindin D9k and PMCA2b in mammary glands was examined at the mRNA level by real-time RT-PCR and at the protein level by immunohistochemistry or Western blots. Our results showed that mRNA levels of calbindin D9k and PMCA2b were reduced in the 1α(OH)ase−/− dams compared with the 1α(OH)ase+/− dams on both diets and were upregulated significantly in both 1α(OH)ase+/− and 1α(OH)ase−/− dams on the rescue diet compared with the genotype-matched dams on the high-calcium diet. However, these parameters were still lower in 1α(OH)ase−/− dams than in 1α(OH)ase+/− dams (Figs. 1, H and I). Calbindin D9k was localized within the cytoplasm in the mammary epithelial cells (Fig. 1J). The alterations of calbindin D9k-positive area (Fig. 1, J and K) and PMCA2b protein levels (Fig. 1, L and M) were consistent with the changes in the expression levels of these genes.

Effects of 1,25(OH)2D3 deficiency and maternal milk contents on serum calcium, phosphorus, PTH, and 1,25(OH)2D3 in suckling pups.

To determine whether maternal 1,25(OH)2D3 deficiency and maternal dietary calcium affected mineral homeostasis in the suckling pups, serum calcium, phosphorus, PTH, and 1,25(OH)2D3 were measured in 2-wk-old pups nursed by 1α(OH)ase+/− or 1α(OH)ase−/− dams on either the high-calcium diet or the rescue diet. Serum calcium and phosphorus levels were decreased in the 1α(OH)ase−/− pups nursed by the 1α(OH)ase+/− dams on the high-calcium diet and were decreased further in the 1α(OH)ase−/− pups fed by the 1α(OH)ase−/− dams on the high-calcium diet (Fig. 2, A and B). Serum calcium and phosphorus levels were normalized in the 1α(OH)ase−/− pups fed by either 1α(OH)ase+/− or 1α(OH)ase−/− dams on the rescue diet (Fig. 2, A and B). Serum PTH levels were increased in 1α(OH)ase−/− pups fed by either 1α(OH)ase+/− or 1α(OH)ase−/− dams on the high-calcium diet. However, serum PTH levels were decreased significantly in both 1α(OH)ase+/− and 1α(OH)ase−/− pups fed by dams on the rescue diet compared with genotype-matched pups fed by dams on the high-calcium diet (Fig. 2C). Serum 1,25(OH)2D3 levels were decreased significantly in the 1α(OH)ase+/− pups fed by 1α(OH)ase−/− dams compared with those fed by 1α(OH)ase+/− dams on either diet and were decreased significantly in 1α(OH)ase+/− pups fed by dams on the rescue diet compared with genotype-matched pups fed by dams on the high-calcium diet (Fig. 2D). Serum 1,25(OH)2D3 was undetectable in 1α(OH)ase−/− pups (Fig. 2D).

Fig. 2.

Effects of 1,25(OH)2D3 deficiency and high dietary calcium on serum calcium, phosphorus, PTH, and 1,25(OH)2D3 and on skeletal growth in pups. Serum calcium (A), phosphorus (B), PTH (C), and 1,25(OH)2D3 levels (D) were determined in samples from 2-wk-old 1α(OH)ase+/− [p-1α(OH)ase+/−] and 1α(OH)ase−/− [p-1α(OH)ase−/−] pups fed by 1α(OH)ase+/− [m-1α(OH)ase+/−] and 1α(OH)ase−/− [m-1α(OH)ase−/−] dams on high Ca or on the rescue diet. X-rays of femurs (E) and the length of femurs (F) were determined in 2-wk-old pups. Each value is the mean ± SE of determinations in 5 mice of each group. *P < 0.05, **P < 0.01, and ***P < 0.001 compared with 1α(OH)ase+/− littermates. #P < 0.05, ##P < 0.01, and ###P < 0.001 compared with pups fed by 1α(OH)ase+/− dams. ΔΔΔP < 0.001 compared with genotype-matched pups fed by dams on high Ca.

Effects of 1,25(OH)2D3 deficiency and maternal milk contents on skeletal growth and endochondral bone formation in the suckling pups.

We then assessed whether 1,25(OH)2D3 deficiency and the alterations of maternal milk mineral content affected skeletal growth in the suckling pups. The length of femurs were measured in 2-wk-old pups fed by 1α(OH)ase+/− or 1α(OH)ase−/− dams on either the high-calcium diet or the rescue diet. The length of femurs was significantly shorter in the 1α(OH)ase−/− pups compared with the 1α(OH)ase+/− pups fed by either 1α(OH)ase+/− or 1α(OH)ase−/− dams on the high-calcium diet. The femur length was also shorter in both 1α(OH)ase+/− or 1α(OH)ase−/− pups fed by 1α(OH)ase−/− dams compared with genotype-matched pups fed by 1α(OH)ase+/− dams on the high-calcium diet (Fig. 2, E and F). In contrast, the length of femurs was rescued in 1α(OH)ase−/− pups fed by 1α(OH)ase+/− dams. The length of femurs was also increased in 1α(OH)ase+/− and 1α(OH)ase−/− pups fed by 1α(OH)ase−/− dams on the rescue diet compared with that of 1α(OH)ase+/− pups fed by 1α(OH)ase+/− dams on the high-calcium diet (Fig. 2, E and F).

To determine whether the alterations of skeletal growth were associated with alterations of endochondral bone formation, the proliferation of chondrocytes was examined by immunostaining for PCNA, and the width of the hypertrophic zone in tibial cartilaginous growth plates was measured. The percentage of PCNA-positive chondrocytes was decreased, but not significantly, in the 1α(OH)ase−/− pups compared with the 1α(OH)ase+/− pups fed by either 1α(OH)ase+/− or 1α(OH)ase−/− dams on the high-calcium diet. However, it was decreased significantly in both 1α(OH)ase+/− and 1α(OH)ase−/− pups fed by 1α(OH)ase−/− dams compared with pups fed by 1α(OH)ase+/− dams on the high-calcium diet (Fig. 3, A and C). In contrast, the percentage of PCNA-positive chondrocytes was increased significantly in both 1α(OH)ase+/− and 1α(OH)ase−/− pups fed by either 1α(OH)ase+/− or 1α(OH)ase−/− dams on the rescue diet compared with genotype-matched pups fed by dams on the high-calcium diet. This increase was more apparent in 1α(OH)ase−/− pups than 1α(OH)ase+/− pups fed by either 1α(OH)ase+/− or 1α(OH)ase−/− dams on the rescue diet. The increase in PCNA-positive cells was also more apparent in both 1α(OH)ase+/− and 1α(OH)ase−/− pups fed by 1α(OH)ase−/− dams on the rescue diet than genotype-matched pups fed by 1α(OH)ase+/− dams on the same diet (Fig. 3, A and C). The width of the hypertrophic zone was increased in both 1α(OH)ase−/− pups fed by either 1α(OH)ase+/− or 1α(OH)ase−/− dams on the high-calcium diet. The width of this zone was normalized in 1α(OH)ase−/− pups fed by dams on the rescue diet compared with genotype-matched pups fed by dams on the high-calcium diet (Fig. 3, B and D).

Fig. 3.

Effects of 1,25(OH)2D3 deficiency and high dietary calcium on endochodral bone formation in pups. Representative micrographs of sections of growth plates of tibiae stained immunohistochemically for PCNA (A) and with hematoxylin and eosin (HE; B) in p-1α(OH)ase+/− and p-1α(OH)ase−/− pups fed by m-1α(OH)ase+/− and m-1α(OH)ase−/− dams on high Ca or on the rescue diet. Scale bars in A and B represent 25 and 50 μm, respectively. C and D: %PCNA-positive chondrocytes of total chondrocytes (C) and width of hypertrophic zone (D). Each value is the mean ± SE of determinations in 5 mice in each group. *P < 0.05, **P < 0.01, and ***P < 0.001 compared with 1α(OH)ase+/− littermates. #P < 0.05 and ###P < 0.001 compared with pups fed by 1α(OH)ase+/− dams. ΔP < 0.05 and ΔΔΔP < 0.001 compared with genotype-matched pups fed by dams on high Ca.

Effects of 1,25(OH)2D3 deficiency and maternal milk content on bone volume and collagen deposition in bone matrix in the suckling pups.

To determine whether 1,25(OH)2D3 deficiency and altering maternal milk mineral content affected bone volume and the collagen deposition in bone matrix in the suckling pups, the trabecular and cortical bone volume were assessed by μCT (Fig. 4, A and B) and in undecalcified plastic sections of the proximal ends of tibiae stained by the von Kossa technique (Fig. 4C), and the collagen deposition in bone matrix was examined by immunostaining for type I collagen (Fig. 4D). The trabecular and cortical bone volume and type I collagen positive areas were reduced significantly in the 1α(OH)ase−/− pups compared with the 1α(OH)ase+/− pups (Fig. 4) and were also reduced in both 1α(OH)ase+/− and 1α(OH)ase−/− pups fed by 1α(OH)ase−/− dams compared with genotype-matched pups fed by 1α(OH)ase+/− dams on any diet (Fig. 4). In contrast, these two parameters were increased significantly in both 1α(OH)ase+/− and 1α(OH)ase−/− pups fed by dams on the rescue diet compared with genotype-matched pups fed by dams on the high-calcium diet (Fig. 4).

Fig. 4.

Effects of 1,25(OH)2D3 deficiency and maternal milk contents on bone volume and type I collagen deposition in the bone matrix in pups. Representative longitudinal sections (A) and cross-sections (B) of the distal end of femurs by microcomputed tomography (μCT) scan and 3D reconstruction and sections of the proximal end of tibiae stained with Von Kossa procedure (C) and immunohistochemically (D) for type I collagen (Col I) in p-1α(OH)ase+/− and p-1α(OH)ase−/− pups fed by m-1α(OH)ase+/− and m-1α(OH)ase−/− dams on high Ca or on the rescue diet. Trabecular bone volume relative to tissue area (BV/TV; E), cortical bone volume (F), and type I collagen positive area (G) as %tissue area were measured as described in materials and methods. Each value is the mean ± SE of determinations in 5 mice in each group. ***P < 0.001 compared with 1α(OH)ase+/− littermates. ###P < 0.001 compared with pups fed by 1α(OH)ase+/− dams. ΔΔΔP < 0.001 compared with genotype-matched pups fed by dams on high Ca.

Effects of 1,25(OH)2D3 deficiency and maternal milk content on osteoblastic bone formation in the suckling pups.

To determine whether the alterations in bone volume were associated with alterations in osteoblastic bone formation in the suckling pups, the paraffin-embedded longitudinal tibial sections were stained with hematoxylin and eosin (Fig. 5A) and histochemically for ALP (Fig. 5B). The osteoblast number and ALP-positive area were determined by computer-assisted image analysis (Fig. 5, C and D). Osteoblast number and ALP-positive area were reduced significantly in the 1α(OH)ase−/− pups compared with the 1α(OH)ase+/− pups and in both 1α(OH)ase+/− and 1α(OH)ase−/− pups fed by 1α(OH)ase−/− dams compared with pups fed by 1α(OH)ase+/− dams on the high-calcium diet (Fig. 5). In contrast, the osteoblast number and ALP-positive area were increased significantly in both 1α(OH)ase+/− and 1α(OH)ase−/− pups fed by dams on the rescue diet compared with genotype-matched pups fed by dams on the high-calcium diet. However, these two parameters were still lower in 1α(OH)ase−/− pups compared with 1α(OH)ase+/− pups when they were fed by dams on the rescue diet (Fig. 5).

Fig. 5.

Effects of 1,25(OH)2D3 deficiency and maternal milk contents on bone formation in 2-wk-old pups. Representative micrographs of decalcified, paraffin-embedded sections stained with HE (A) and histochemically (B) for alkine phosphatase (ALP) in p-1α(OH)ase+/− and p-1α(OH)ase−/− pups fed by m-1α(OH)ase+/− and m-1α(OH)ase−/− dams on high Ca or on the rescue diet. Scale bars in A and B represent 50 and 100 μm, respectively. C: nos. of osteoblasts per mm2 were counted in the metaphyseal region of HE-stained tibiae of the mice. D: ALP-positive area as %tissue area was determined in the metaphyseal regions for each group. Each value is the mean ± SE of determinations in 5 mice in each group. *P < 0.05, **P < 0.01, and ***P < 0.001 compared with 1α(OH)ase+/− littermates. ###P < 0.001 compared with pups fed by 1α(OH)ase+/− dams. ΔP < 0.05 and ΔΔΔP < 0.001 compared with genotype-matched pups fed by dams on high Ca.

Effects of 1,25(OH)2D3 deficiency and maternal milk content on bone resorption in suckling pups.

To determine whether the alterations of bone volume were associated with alterations of osteoclastic bone resorption in the suckling pups, the paraffin-embedded longitudinal tibial sections were stained histochemically for TRAP (Fig. 6A), and the TRAP-positive osteoclast number and surface were determined by computer-assisted image analysis (Figs. 6, B and C). Osteoclast number and surface were reduced significantly in the 1α(OH)ase−/− pups compared with the 1α(OH)ase+/− pups fed by either 1α(OH)ase+/− or 1α(OH)ase−/− dams on any diets. These parameters were reduced, albeit not significantly, in both 1α(OH)ase+/− and 1α(OH)ase−/− pups fed by 1α(OH)ase−/− dams compared with genotype-matched pups fed by 1α(OH)ase+/− dams on any diets. These parameters were also reduced significantly in both 1α(OH)ase+/− and 1α(OH)ase−/− pups fed by dams on the rescue diet compared with genotype-matched pups fed by dams on the high-calcium diet (Fig. 6).

Fig. 6.

Effects of 1,25(OH)2D3 deficiency and maternal milk content on bone resorption in pups A: representative micrographs of sections of the tibial metaphysis stained histochemically for tartrate-resistant acid phosphatase activity (TRAP) in p-1α(OH)ase+/− and p-1α(OH)ase−/− pups fed by m-1α(OH)ase+/− and m-1α(OH)ase−/− dams on high Ca or on the rescue diet. Scale bars represent 50 μm. B and C: no. of TRAP-positive osteoclasts related to tissue area [N.Oc/T.Ar (no./mm2); B] and osteoclastic surface relative to bone surface (Oc.S/B.S, %; C) were counted in the metaphyseal regions for each group. Each value is the mean ± SE of determinations in 5 mice in each group. **P < 0.01 and ***P < 0.001 compared with 1α(OH)ase+/− littermates. ΔΔP < 0.01 and ΔΔΔP < 0.001 compared with genotype-matched pups fed by dams on high Ca.

DISCUSSION

In the present study, we assessed the interaction of 1,25(OH)2D3 and calcium on neonatal skeletal growth and turnover by a combination of genetic and environmental manipulations. We used a genetic approach to modulate circulating 1,25(OH)2D3 levels in the neonates and dams and compared mice that were homozygous for targeted deleletion of the Cyp27b1 gene and that have undetectable circulating 1,25(OH)2D3 levels with mice that were heterozygotes and have normal circulating 1,25(OH)2D3 levels. Nevertheless, although heterozygous adult mice have normal 1,25(OH)2D3 levels, there may be differences in calcium and/or milk calcium content in wild-type compared with heterozygous dams during a time of calcemic stress, such as lactation, and further studies will be required to address this issue. We used environmental manipulation to modulate serum calcium in the neonates. In view of the fact that suckling neonates receive their calcium intake via maternal milk, we altered the dietary content of the dams in an attempt to modulate neonatal calcium intake. In view of the fact that neonates were derived from heterozygous and homozygous dams, we were able to observe the effect of the presence or absence of circulating 1,25(OH)2D3 in the dams on the levels of calcium in maternal milk when the dietary calcium intake was altered.

Our results show that calcium in milk is altered by altering the calcium in the diet of the lactating mothers and that, for each level of dietary calcium presented to the dams, milk calcium was higher in those with circulating 1,25(OH)2D3 than in those without circulating 1,25(OH)2D3. In view of the fact that the dietary levels of protein were not different and the protein levels in milk were not significantly different, dietary calcium appeared to influence milk calcium independent of changes in protein. Several studies have now implicated the calcium-sensing receptor (CaR) in the transport of calcium from the systemic circulation of lactating animals to milk (1, 40, 41). One previous study has shown that Ca2+ accumulation by cultured mouse mammary explants is stimulated by 1,25(OH)2D3 in a dose-dependent fashion (23). However, studies in humans have suggested that milk calcium levels are independent of the vitamin D status (2) and the calcium status (19) of the lactating mother. These inconsistencies may have resulted from a lack of a direct correlation between 1,25(OH)2D3 levels and milk calcium levels or from the assessment of varying degrees of vitamin D insufficiency rather than from assessment of the complete absence of 1,25(OH)2D3 as in our study.

To clarify the mechanism whereby 1,25(OH)2D3 and dietary calcium increase milk calcium concentration, we assessed the effects of 1,25(OH)2D3 deficiency and dietary calcium on mammary calcium transport. Our previous studies found that Trpv5 expression at both the mRNA and protein levels was downregulated in kidneys of 1α(OH)ase−/− mice (45) and upregulated by the administration of exogenous 1,25(OH)2D3 (44). However, we were unable to detect Trpv5 and Trpv6 gene expression in in mammary glands from dams on day 14 of lactation (results not shown). Therefore, the entry mechanism for mammary gland transepithelial calcium transport is still unknown.

However, we did find that calbindin D9k mRNA was expressed in the mouse mammary gland and that calbindin D9k protein was localized within the cytoplasm in mammary epithelial cells during lactation. We also found that the expression of calbindin D9k at both the gene and protein levels was downregulated in 1,25(OH)2D3-deficient dams and upregulated in both wild-type and 1,25(OH)2D3-deficient dams by dietary calcium supplementation. Our previous studies demonstrated that calbindin D9k expression at both mRNA and protein levels was downregulated in kidneys of 1α(OH)ase−/− mice (45) and upregulated by the administration of exogenous 1,25(OH)2D3 (44). Our current results suggest that, in mammary gland, calbindin D9k may function to buffer calcium and shuttle it through the cytoplasm as in other calcium-transporting epithelia and that both 1,25(OH)2D3 and dietary calcium stimulate mammary calcium transport by upregulation of calbindin D9k expression in mammary epithelial cells during lactation.

Previous studies have suggested that calcium transport across the apical membrane via PMCA2 is a fundamentally important step in the process of milk calcium secretion (33). Not only did we confirm that PMCA2b mRNA and protein were expressed in the mouse mammary gland during lactation, but we also found that the expression of PMCA2b at both mRNA and protein levels was downregulated in 1,25(OH)2D3-deficient dams and upregulated in both wild-type and 1,25(OH)2D3-deficient dams by dietary calcium supplementation. A previous study found that activation of the CaR increased calcium-ATPase activity in cultured mammary epithelial cells (41). This effect was lost when CaR expression or PMCA2 expression was reduced with siRNAs, demonstrating that the effect of CaR was mediated by PMCA2 (41). Our study demonstrated that both 1,25(OH)2D3 and dietary calcium stimulate mammary calcium transport by upregulation of PMCA2b expression in mammary epithelial cells during lactation.

Serum calcium in the neonates was lowest in the 1α(OH)ase−/− mice that received milk calcium from the 1α(OH)ase−/− dams that had the lowest milk calcium. Serum calcium was higher in the 1α(OH)ase+/− neonates even when suckling from the same 1α(OH)ase−/− dams with the same low milk calcium. Consequently, 1,25(OH)2D3 per se appeared to contribute to elevating the serum calcium in the neonate. When 1α(OH)ase−/− neonates were exposed to the higher calcium in the milk of the 1α(OH)ase+/− and 1α(OH)ase−/− dams on the rescue diet, serum calcium was normalized. Our previous study also found that serum calcium in adult mice was normalized on the lactose-containing rescue diet in the absence of 1,25(OH)2D, the VDR, or both (28). Consequently high milk calcium intake contributes to elevating the serum calcium in the neonate in a 1,25(OH)2D3-independent manner, possibly facilitated by lactose in maternal milk. In the hypocalcemic neonates on the high-calcium intake, secondary hyperparathyroidism appeared less severe in the 1α(OH)ase−/− pups than in adult 1α(OH)ase−/− mice (28). Nevertheless, when serum calcium was raised in the 1α(OH)ase−/− pups fed by dams on the rescue diet, normalization of the elevated serum PTH concentrations occurred in mutant neonates. Consequently, the ambient calcium concentration suppressed PTH secretion independently of 1,25(OH)2D3.

Our previous study demonstrated clear abnormalities in skeletal growth in 2-wk-old 1α(OH)ase−/− mice prior to weaning, which included shorter long bones (45). In the current study, we found that the abnormalities of skeletal growth correlated with the levels of milk calcium to which the genetically modified pups were exposed. Surprisingly, the abnormalities of skeletal growth were rescued even in 1,25(OH)2D3-deficient pups fed by 1,25(OH)2D3-deficient dams on the rescue diet, and the increase in chondrocyte proliferation was most pronounced in these pups. Consequently, the action of ambient calcium on stimulating skeletal growth appears to be more important than the direct action of 1,25(OH)2D3, and the response of chondrocytes to elevated calcium is more sensitive under hypocalcemic conditions. Studies with mouse chondrocytes support roles for high extracellular calcium and the CaR in promoting differentiation (34, 35). Deletion of the Casr gene in chondrocytes has been reported to delay growth plate development and produce a rickets-like phenotype with expansion and reduced mineralization of the hypertrophic zone and decreased abundance of markers of mature, terminally differentiated chondrocytes (5). Other studies have shown that high extracellular calcium promotes chondrocyte differentiation and that suppressing Casr expression with antisense or dominant-negative cDNA constructs blocks this effect (4, 6, 7). Taken together, these data support the concept that increased extracellular calcium by 1,25(OH)2D3 action and dietary supplementation can stimulate endochondral bone formation mediated by the CaR.

We then demonstrated that the degree of osteopenia in the 2-wk-old pups correlated with the degree of reduction in milk calcium (45). However, the effect of altering milk content on osteoblastic bone formation was different from its effect on the skeletal growth. Thus, although the trabecular bone volume and osteoblast activity also increased in parallel with the increase in milk calcium, in contrast with the parameters in the cartilaginous growth plate it was also dependent on the presence or absence of 1,25(OH)2D3 in the pups. This is consistent with our previous findings in 4-mo-old 1α(OH)ase−/−, VDR−/−, and 1α(OH)ase−/−VDR−/− mutant mice (28) and with our previous studies examining the effects of exogenous 1,25(OH)2D3 trabecular bone volume and osteoblast activity (44). Overall, these data suggest that the 1,25(OH)2D-VDR system may exert a skeletal “anabolic” effect, which is necessary to sustain basal bone-forming activity.

The CaR has been localized on osteoblastic cells (4). In tissue culture models, elevations in extracellular free ionized calcium concentrations increase osteoblast chemotaxis and proliferation (8, 14, 46) and alter the levels of expression of some differentiation markers (10, 12, 27). More recent studies have reported that deletion of the Casr gene in osteoblasts profoundly blocked postnatal growth and skeletal development, a finding that was evident by 3 days of age (5). Our recent study has demonstrated that CaR deficiency abolishes skeletal responses to dietary calcium supplementation in suckling neonates (38) and suggests that raising serum calcium concentration by dietary calcium supplementation may activate CaR and subsequently produce a skeletal anabolic action, at least in the neonates. Taken together, these data suggest that 1,25(OH)2D3 and calcium can each exert cooperative roles on osteoblastic bone formation in the neonate.

Our results examining TRAP-positive osteoclast numbers and surface in the 2-wk-old pups with varying levels of calcium and 1,25(OH)2D3 suggest that the absence of the bone-resorbing activity of 1,25(OH)2D3 may supersede any effect of calcium on bone resorption, and in fact calcium supplementation in the 2-wk-old mice may inhibit osteoclastic bone resorption, as we reported previously (38), either directly (22), indirectly, or both. Thus, calcitonin release in rodents has been implicated in bone sparing in lactating females (43) and could help explain the reduced osteoclasts in the pups with elevations in calcium.

In summary, both 1,25(OH)2D3 and dietary calcium can stimulate mammary calcium transport to elevate milk calcium concentration in lactating females. Defective skeletal growth appeared to be rescued completely by normalizing serum calcium in 2-wk-old mice deficient in 1,25(OH)2D3. However, normalization of serum calcium and normal 1,25(OH)2D3 levels were needed for optimizing osteoblastic bone formation. Consequently, 1,25(OH)2D3 and calcium exert cooperative roles in regulating skeletal function in the neonate.

GRANTS

This work was supported by an operating grant (no. H200627) from the Health Department of Jiangsu Province, China, a grant from the Program for Changjiang Scholars and Innovative Research Team in University (to D. Miao), and a grant from the Canadian Institutes of Health Research (to D. Goltzman).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

REFERENCES

  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
  29. 29.
  30. 30.
  31. 31.
  32. 32.
  33. 33.
  34. 34.
  35. 35.
  36. 36.
  37. 37.
  38. 38.
  39. 39.
  40. 40.
  41. 41.
  42. 42.
  43. 43.
  44. 44.
  45. 45.
  46. 46.
View Abstract