Increased formation and decreased resorption of ... - The FASEB Journal

Increased formation and decreased resorption of bone in mice with elevated vitamin D receptor in mature cells of the osteoblastic lineage EDITH M. GARDINER,*,1 PAUL A. BALDOCK,* GETHIN P. THOMAS,* NATALIE A. SIMS,*,2 N. KATHRYN HENDERSON,* BRUCE HOLLIS,† CHRISTOPHER P. WHITE,*,3 KATHRYN L. SUNN,* NIGEL A. MORRISON,*,4 WILLIAM R. WALSH,‡ AND JOHN A. EISMAN* *Bone and Mineral Research Program, Garvan Institute of Medical Research, Darlinghurst, Sydney, New South Wales, Australia; †Department of Pediatrics, Medical University of South Carolina, Charleston, South Carolina, USA; and ‡Department of Orthopaedics, Prince of Wales Hospital, Sydney, New South Wales ABSTRACT The microarchitecture of bone is regulated by complex interactions between the boneforming and resorbing cells, and several compounds regulate both actions. For example, vitamin D, which is required for bone mineralization, also stimulates bone resorption. Transgenic mice overexpressing the vitamin D receptor solely in mature cells of the osteoblastic bone-forming lineage were generated to test the potential therapeutic value of shifting the balance of vitamin D activity in favor of bone formation. Cortical bone was 5% wider and 15% stronger in these mice due to a doubling of periosteal mineral apposition rate without altered body weight or calcium homeostatic hormone levels. A 20% increase in trabecular bone volume in transgenic vertebrae was also observed, unexpectedly associated with a 30% reduction in resorption surface rather than greater bone formation. These findings indicate anabolic vitamin D activity in bone and identify a previously unknown pathway from mature osteoblastic cells to inhibit osteoclastic bone resorption, counterbalancing the known stimulatory action through immature osteoblastic cells. A therapeutic approach that both stimulates cortical anabolic and inhibits trabecular resorptive pathways would be ideal for treatment of osteoporosis and other osteopenic disorders.— Gardiner, E. M., Baldock, P. A., Thomas, G. P., Sims, N. A., Henderson, N. K., Hollis, B., White, C. P., Sunn, K. L., Morrison, N. A., Walsh, W. R., Eisman, J. A. Increased formation and decreased resorption of bone in mice with elevated vitamin D receptor in mature cells of the osteoblastic lineage. FASEB J. 14, 1908 –1916 (2000)

Key Words: osteoclast 䡠 osteocyte 䡠 periosteum 䡠 uncoupling 䡠 turnover

Osteoporosis is a condition in which fractures occur with minimal trauma due to underlying bone 1908

fragility. Peak bone mass and optimum bone structure achieved during early adulthood and the subsequent age and menopause-related bone loss are key determinants of this condition. Bone loss at menopause relates to increased osteoclastic resorption, predominantly in trabecular bone, whereas the more gradual but continuous loss that occurs with aging is thought to be the result of decreased osteoblastic bone formation (1) and primarily affects cortical bone (2). In both situations, there is gradual deterioration of the bone microarchitecture as well as the overall loss of bone mass. Effective osteoporosis therapies minimize bone loss, essentially by reducing resorption (3, 4). Therapeutic use of the active 1␣-hydroxylated forms of vitamin D (i.e., 1,25-dihydroxyvitamin D2 and D3, referred to collectively as 1,25-dihydroxyvitamin D) has yielded somewhat disparate clinical results (5– 8), presumably because of diverse actions on the osteoblastic and osteoclastic cell lineages as well as on other tissues such as intestine, parathyroid, and kidney. 1,25-Dihydroxyvitamin D directly inhibits osteoblastic differentiation (9, 10), but stimulates mineralized matrix formation by mature osteoblasts (11); however, it also indirectly stimulates osteoclastic recruitment and differentiation, acting through immature cells of the osteoblastic lineage via the vitamin D receptor (VDR) (12). Regulation of osteoclastic 1

Correspondence: Garvan Institute of Medical Research, 384 Victoria St., Darlinghurst, Sydney, New South Wales 2010, Australia. E-mail [email protected] 2 Current address: Department of Orthopedics, Yale University School of Medicine, 333 Cedar St., New Haven, CT 06520, USA. 3 Current address: Department of Endocrinology, Prince of Wales Hospital, Sydney, New South Wales 2031 Australia. 4 Current address: Genomics Research Centre, Griffith University Gold Coast Campus, Queensland 9726, Australia. 0892-6638/00/0014-1908/$02.25 © FASEB

bone resorption by cells of the bone-forming lineage provides a mechanism by which the opposing actions of these two cell lineages are coordinated in healthy bone, maintaining overall bone mass and calcium homeostasis (13). This model, however, is inherently incomplete as it does not explain how bone resorption is restrained once homeostatic requirements have been met. An inhibitory pathway that locally limits bone resorption could complete the model of coupled bone turnover, ensuring the maintenance of bone microarchitectural integrity. The present study tested the hypothesis that if mature osteoblastic cells were made more responsive to 1,25-dihydroxyvitamin D relative to immature osteoblasts as well as to the osteoclastic lineage and other responsive tissues, a net positive bone effect would ensue. This strategy has been shown to be effective in a transgenic model system, with an increase in cortical bone formation. However, presence of the transgene was also associated with a decrease in trabecular bone resorption, indicating for the first time that a counter-regulatory pathway acts via mature osteoblastic cells to restrain osteoclastic activity.

MATERIALS AND METHODS Transgenic mice pOSVDR was generated by inserting a human VDR cDNA 2.1 kb EcoRI fragment from phVDR1/3 (14) into the human osteocalcin-based pGOSCAS vector (15), followed by SV40 small t antigen splice and polyadenylation signals (Promega Corporation, Madison, Wis.) immediately downstream. The human osteocalcin gene is expressed solely in mature cells of the osteoblastic lineage, and this cell type specificity was duplicated by a pGOSCAS reporter transgene in previous work (15). Transgenic mice were generated by pronuclear injection of FVB/N embryos. The OSV9 and OSV3 lines were independently derived, each carrying a single insertion of 5 to 10 copies of the transgene. Hemizygous experimental animals, bred by mating homozygous males to FVB/N females, were studied. Age-matched female nontransgenic and transgenic mice were mixed and group housed at weaning. This study was conducted twice. The first study included FVB/N and OSV9; a repetition including these lines plus OSV3 confirmed initial observations. Data shown are from the second study. A third line, OSV8 line, showed inconsistent growth characteristics unrelated to transgene status and was excluded from final analyses as the level of VDR protein in its bones was not significantly different from the FVB/N level (P⫽0.35). Four weeks prior to collection, the mice were randomly assigned to two groups and changed from standard laboratory chow (0.9% calcium) to semi-synthetic diets (16) with moderate (0.5%) or low (0.1%) calcium content. Vitamin D was supplied (1000 IU/kg) in all diets. Mice were injected with the fluorescent tetracycline compounds calcein and demeclocycline (Sigma Chemical Company, St. Louis, Mo.), each at 15 mg/kg, 10 and 3 days prior to collection. Tibiae were collected and stored for mechanical testing at ⫺20°C in phosphate-buffered saline. Femora and vertebrae were col-

lected, fixed in 4% paraformaldehyde and prepared for histomorphometry. Calvaria, radius, kidney, liver, brain, muscle, heart, lung, and spleen were collected for molecular analyses. Analysis of transgene expression Total RNA was prepared from tissues of 8 wk old mice and analyzed by Northern blot as described previously (15). Filters were probed with a random primed ␣-32P-dCTP- (Amersham, Buckinghamshire, England) labeled mouse VDR cDNA fragment that was cloned after reverse transcription-polymerase chain reaction (RT-PCR) from nontransgenic mouse kidney RNA, using primers derived from the human cDNA sequence (forward primer 5⬘-CGGAATTCTCATTCTGACAGATGAGGAAGTGC-3⬘ and reverse primer 5⬘-AACTGCAGTCCTGGTATCATCTTAGCAAAGCC-3⬘). The filter was stripped and reprobed for osteocalcin using a radiolabeled rat osteocalcin cDNA insert from pOC918 (a kind gift from Dr. S. E. Harris) and for GAPDH using a radiolabeled PCR product. Relative signals were quantitated by PhosphorImager (Molecular Dynamics 445SI, Sunnyvale, Calif.). Total VDR protein was measured by ELISA (see method below) from long bones of six to eight 9 month old mice for each line, with equal numbers of mice from the low and moderate calcium diet groups. Biochemistry Serum 1,25-dihydroxyvitamin D was measured by radioimmunoassay (RIA) (17). PTH was also measured by RIA (Immutopics, San Clemente, Calif.), as was serum osteocalcin, using the method of Gundberg (18) except 50 ␮l sample sizes were assayed. Primary antibody and osteocalcin standards were generously provided by Dr. C. Gundberg. Iodinated osteocalcin was purchased from Biomedical Technologies, Inc. (Stoughton, Mass.) and donkey anti-goat IgG secondary antibody from Sigma. Total VDR protein (sample collection described above) was measured by ELISA (19) using antibodies generously provided by Dr. H. DeLuca, commercially supplied biotinconjugated alkaline phosphatase (Bio-Rad, Hercules, Calif.), and purified VDR protein standards (Pan Vera, Madison, Wis.). Nuclear protein extracts for VDR assay were prepared using a protocol adapted from Pierce (Rockford, Ill.) (20). Whole bones were homogenized initially using a Polytron homogenizer and subsequently by Dounce homogenizer. Total protein levels were determined by Bradford colorimetric assay (Bio-Rad). Values are means ⫾ se. Histology For histology, 8 wk old animals were treated with a single intraperitoneal injection of 1,25-dihydroxyvitamin D3 (Tetrionics Inc., Madison, Wis.) at a dose of 2 ␮g per kg body weight. Femora were collected 6 h later, distal segments were paraffin-embedded, and sagittal sections were analyzed as previously (15). In situ hybridization used antisense and sense human VDR cDNA riboprobes generated from linearized pGhVcEBx, which contains 400 bp of 5⬘ sequence from the human VDR cDNA. The antisense riboprobe detects both mouse and transgenic human transcripts. Hybridized probe was detected using the alkaline phosphatase-coupled antidigoxigenin antibody method (Boehringer Mannheim, Mannheim, Germany) with the addition of 1.2 mg/ml levamisole (Sigma) in the final staining solution. Immunohistochemistry on sections of the same specimens used a humanspecific anti-VDR antibody (21), kindly provided by Dr. P.

LOWER BONE RESORPTION WITH HIGH VDR IN MATURE OSTEOBLASTIC CELLS

1909

Tuohimaa. Biotinylated goat anti-rabbit IgG was used as secondary antibody, and avidin-biotinylated peroxidase complex and diamino-benzidine staining (Vector Laboratories, Burlingame, Calif.) were used for detection. Sections were not counterstained. Histomorphometry The fourth caudal vertebra and the distal half of the right femur from 4- and 9-month-old animals were fixed and embedded undecalcified in K-Plast resin (Medim-Medizinische Diagnostik, Giessen, Germany) and 5 ␮m sagittal sections were analyzed (Bioquant, R&M Biometrics Inc., Nashville, Tenn.). Femoral width was measured using bright field microscopy, and periosteal mineral appositional rate and vertebral bone formation rate (BFR⫽double-labeled surface ⫻ MAR) by fluorescence microscopy (Leica, Heerbrugg, Switzerland). Sections were stained for mineralized bone (22); trabecular bone volume (BV/TV), thickness (Tb.Th), and number (Tb.N) were quantitated (23). For measurements of osteoclast surface (Oc.S) and number (Oc.N), sections were stained for tartrate-resistant acid phosphatase activity as described previously (15). Mechanical testing and morphometry The biomechanical and physiological consequences of osteoblastic and osteocytic VDR elevation were evaluated in 4 and 9 month old female mice. Tibiae were dissected free of remaining soft tissue prior to three-point bending tests using an MTS 858 Bionix Testing Machine (MTS Systems Corporation, Minneapolis, Minn.) at 2 mm/min until failure. Samples were tested immersed in normal saline at room temperature with a support span of 10 mm. After mechanical testing, tibiae were imaged at the fracture site at 40⫻ magnification using a Leica stereo microscope. Measurements and analyses were completed using the Bioquant System (R&M Biometrics). Cortical height (diameter) and width were measured in the sagittal plane and cortical moments of inertia were calculated (24). Statistics Statistical analyses were performed by one-way analysis of variance (ANOVA) within age groups with linear contrasts selected a priori to compare results from each of the transgenic lines with those from the FVB/N control line, *P ⬍ 0.05 (SPSS for Macintosh v. 4.02; SPSS Inc., Chicago, Ill.). The effects of transgenic line and dietary calcium group were examined by two-way ANOVA; n⫽17–20 mice per line per age group unless otherwise indicated.

RESULTS Transgenic VDR expression was detected in the bones of adult mice from two OSV transgenic lines (OSV3 and OSV9), whereas VDR transcripts were not detected in the bones of normal FVB/N mice (Fig. 1A). VDR expression was undetectable in transgenic or nontransgenic bones by immunohistochemistry or in situ hybridization. Injection of mice with 1,25-dihydroxyvitamin D3, however, elevated VDR expression in cuboidal and flattened osteoblasts (Fig. 1C–D) and in hypertrophic chondrocytes of 1910

Vol. 14

October 2000

Figure 1. OSVDR transgene expression increases osteoblastic VDR levels. A) The 3.3 kb OSVDR transgene transcript, indicated by dots, was detectable by Northern blot in transgenic bone (Bo) but not in kidney (Ki), liver (Li), or brain (Br). Endogenous VDR transcripts (4.3 and 2.5 kb) were undetectable in bone regardless of transgene status, but readily detectable in kidney in all lines. No VDR signal was detected in muscle, heart, lung, or spleen samples (not shown). B) Total VDR protein was elevated in OSV9 and OSV3 bones. Values are means ⫾ se. Significant differences from FVB/N are indicated by asterisks above individual lines (P⬍0.05). After pretreatment with 1, 25-dihydroxyvitamin D3, OSVDR transgene expression was detected in mature osteoblasts and osteocytes of endocortical bone (C–F ). Trabecular osteoblasts and osteocytes were similarly stained (not shown). VDR RNA was detected by in situ hybridization in osteoblasts and osteocytes of 1,25-dihydroxyvitamin D3-treated transgenic OSV9 bones (D) but was not present in treated FVB/N bones (C). Similarly, transgenic VDR protein expression was evident in osteoblasts and osteocytes of treated OSV9 bones (F) but not treated FVB/N bones (E) by immunohistochemistry using an antibody specific for human VDR. Ob, osteoblast, Oy, osteocyte.

transgenic but not nontransgenic bones, as expected based on previous studies of the transgenic human osteocalcin promoter (15). Signal was also detected in a minority of osteocytes (10 –20%), primarily those immediately adjacent to positively stained osteoblastic surfaces. Immunohistochemistry with an antibody specific for human but not mouse VDR detected VDR protein in osteoblasts and osteocytes of bones from treated transgenic but not nontransgenic mice (Fig. 1E–F). Elevation of VDR protein in untreated transgenic mice was also apparent by

The FASEB Journal

GARDINER ET AL.

TABLE 1. Mouse line age (months)

Body weight (g) Tibial length (mm) Serum Ca (mg/dl) Serum OC (ng/ml) Serum 1,25 D (pg/ml) Serum PTH (ng/ml)

FVB/N

OSV9 4

OSV3

FVB/N

OSV9 9

OSV3

34.0 ⫾ 0.9 18.2 ⫾ 0.1 9.5 ⫾ 0.2 168 ⫾ 8 64 ⫾ 6 61 ⫾ 5

34.3 ⫾ 1.0 18.3 ⫾ 0.1 9.1 ⫾ 0.1 148 ⫾ 7 58 ⫾ 5 77 ⫾ 6

36.7 ⫾ 1.2 18.2 ⫾ 0.1 9.1 ⫾ 0.2 160 ⫾ 9 58 ⫾ 8 82 ⫾ 6*

35.0 ⫾ 1.3 18.0 ⫾ 0.1 11.4 ⫾ 0.2 140 ⫾ 6 61 ⫾ 2 109 ⫾ 8

35.9 ⫾ 1.5 18.1 ⫾ 0.1 11.0 ⫾ 0.3 117 ⫾ 7 58 ⫾ 3 100 ⫾ 8

38.3 ⫾ 1.2 18.4 ⫾ 0.1 11.8 ⫾ 0 110 ⫾ 5* 66 ⫾ 2 106 ⫾ 6

ELISA measurement (Fig. 1B), with a threefold greater level in OSV3 bones relative to FVB/N (7.0⫾0.6 vs. 1.8⫾0.5 fmol/mg protein) and intermediate elevation in OSV9 bones (5.7⫾0.7). Transgenic animals were healthy and phenotypically normal. As there were no interactions between transgenic line and dietary calcium content, diet groups were combined for all analyses. There was no consistent transgene-associated effect on body weight, bone length (tibia), or serum levels of calcium or calcium homeostatic hormones (Table 1). The transient elevation of serum PTH at 4 months was not present in older animals. Similarly, serum osteocalcin was lower in transgenic mice at 9 months but not at 4 months (Table 1) or in older animals (data not shown). Northern blot analysis revealed a reduction in osteocalcin gene expression in 8 wk old transgenic animals (Fig. 1A). Cortical bone Tibiae from transgenic females were significantly stronger in a three-point bending test than nontransgenic bones (mean peak load for OSV9 and OSV3, 16% higher than FVB/N at 4 months and up to 25% higher at 9 months; Fig. 2A). Transgenic bones were also stiffer than their wild-type counterparts (up to 24%, Fig. 2B). Similar differences in peak load and stiffness were also observed for males at the same ages, with mean peak loads up to 27% higher than FVB/N males (16.3⫾0.3, OSV3 vs. 12.8⫾0.3, FVB/N at 9 months, P⫽0.001) and stiffness values elevated by up to 30% (78.0⫾2.2, OSV3 vs. 62.7⫾1.7, FVB/N at 9 months, P⫽0.001). Cortical area moments of inertia (a measure of bone geometry and determinant of bone strength) (24) of the female tibiae were greater in OSV9 and OSV3 than in FVB/N mice by 8 and 11%, respectively, at 4 months, and by 23 and 34% at 9 months (Fig. 2C). There were similar differences in tibial cortical areas at both ages, with OSV3 up to 18% larger than FVB/N (Fig. 2D). The greater crosssectional areas of the transgenic tibiae were positively correlated with their greater strengths in the three-point bending test (Fig. 2E). The increases in moments of inertia and cross-sectional areas were

associated with increased cortical diameter (Fig. 2F), consistent with an increase in periosteal bone formation. This parameter was therefore investigated in femora from these mice. By histomorphometry, the OSV3 femora were also wider than those of FVB/N femora, with diameter 7% greater at 4 and 9 months (Fig. 3A). Femoral periosteal mineral appositional rate (MAR), indicated by greater separation of tetracycline labels (pictured at 4 months, Fig. 3B) was elevated in both transgenic lines at 4 months (130% increase in OSV3, 66% in OSV9), and showed a similar pattern at 9 months (P⫽0.07; Fig. 3C). These MAR increases are consistent with the greater cortical dimensions of the OSV3 bones. Endocortical mineral apposition rate was not altered in the transgenic femora (data not shown). Trabecular bone Trabecular bone volume measured in the fourth caudal vertebral body (Fig. 4A) was at least 17% greater in OSV3 and OSV9 mice than in FVB/N at 4 months and 25% greater at 9 months (Fig. 4B). This difference was associated with thicker trabeculae (OSV3, 17% greater than FVB/N at both ages) without a change in trabecular number (Fig. 4C–D). In contrast to the pattern observed in cortical bone, however, this difference in trabecular thickness was not attributable to greater bone formation in the transgenic mice, as neither trabecular mineral apposition rate nor double-labeled surface (data not shown), nor bone formation rate (Fig. 4E) was affected by transgene status. The observed increase in trabecular bone volume was instead associated with a reduction in bone resorption, with osteoclast surface on vertebral trabeculae of both transgenic lines reduced by 32% at 4 months and by greater than 40% at 9 months (Fig. 4F). An apparent transgene-related reduction in osteoclast number was not significant (Fig. 4G). A similar trend to increased trabecular bone volume and reduction in osteoclast surface was also apparent at 4 months in the femoral metaphysis but not the epiphysis (Fig. 4H–J).


1911

Figure 2. Tibiae from two lines of transgenic mice were stronger than nontransgenic tibiae. A) Tibial peak load, a measure of the maximum bending force withstood by bone prior to fracture, was greater in OSV9 and OSV3 mice, as was tibial stiffness (B). Cortical area moment of inertia (C), a measure of the distribution of bone mass around the central axis and determinant of bone strength, was significantly greater in OSV9 and OSV3 mice as was tibial cortical area (D). Tibial cortical area and peak load were correlated (E). Dashed lines indicate mean values for FVB/N, OSV9, and OSV3 in order from left to right. Peak load ⫽ 1.6 ⫻ tibial cortical area ⫻10⫺5 ⫹ 1. 7, R2 ⫽ 0.41, P⬍0.0001. FVB/N (open circles), OSV9 (open triangles), and OSV3 (filled boxes). Tibial diameter was significantly greater in OSV9 and OSV3 mice (F ). Values are means ⫾ se. Significant differences from FVB/N are indicated by asterisks above individual lines (P⬍0.05). In Figs. 2 to 4, x axis categories are nontransgenic (FVB, white bars) and transgenic lines OSV9 and 3 (OS9 and OS3, striped and black bars, respectively). Sets of three mouse lines are bracketed in age groups of 4 or 9 months.

DISCUSSION Elevation of vitamin D receptor in cuboidal osteoblasts, osteocytes, and lining cells in transgenic mice improved bone characteristics, with the phenotypic effects of transgene expression differing between long bones and caudal vertebrae. In long bones the transgene effect was most apparent in cortical bone 1912

Vol. 14

October 2000

and was evidenced as elevated cortical area with greater periosteal mineral apposition rates. In caudal vertebrae the transgene effect was manifest in thicker trabeculae and elevated trabecular bone volume. At this site, however, the presence of the transgene was associated with decreased bone resorption rather than increased formation. The lack of a concomitant change in formation is suggestive of uncoupled bone turnover in the transgenic mice. This is the first demonstration in vivo that 1,25dihydroxyvitamin D directly stimulates bone-forming activity by mature osteoblasts and is the first detection of a specific osteoclastic inhibitory pathway from cuboidal osteoblasts, lining cells, or osteocytes. Increased mineral apposition rate reflects an increase in the anabolic ability of individual osteoblasts rather than a change in osteoblastic proliferation or survival (1). The increased osteoblastic bone formation on the OSV periosteal surface may most simply be explained by a transgene-enhanced ability of 1,25-dihydroxyvitamin D to regulate expression of bone structural or regulatory genes (25–28). Although the effect of 1,25-dihydroxyvitamin D on expression of individual genes may be either positive or negative, the net consequence of VDR elevation in mature osteoblasts in these mice is anabolic. As body weight and tibial length were not altered by the transgene, it appears that periosteal osteoblasts in the transgenic mice have a greater osteogenic response for a given range of body weights. This is of particular interest in light of the suggestion that age-related deficiencies in the anabolic potential of osteoblasts may be a major contributor to age-related bone loss in humans (29). The change in osteoblast activity was associated with wider tibiae and femora in OSVDR mice. As a consequence, transgenic tibiae were stronger and stiffer than nontransgenic bones. The modest increases in periosteal bone deposition result in substantial gains in long bone strength because area moment of inertia increases with the fourth power of the radius of the bone (24). The transgene effect on bone formation was envelope specific, being evident only on the periosteal surface of the long bones but not on endosteal or trabecular surfaces, suggesting that the transgene affects osteoblasts of periosteal but not bone marrow origin. In contrast, the transgene effect on bone resorption presumably was mediated by osteoblastic cells originating from the bone marrow, as trabecular osteoclast surface was reduced in the fourth caudal vertebrae and the femoral metaphyseal region. No transgene effect on resorption or formation was detected in the femoral epiphysis, indicating site specificity. These variations in cellular responses to transgene expression suggest that the osteoblastic and osteoclastic phenotypes may result from trans-

The FASEB Journal

GARDINER ET AL.

Figure 3. Cellular responses to the OSVDR transgene in femoral cortical and vertebral cancellous bone. Femoral width (A) was greater than FVB/N in OSV3 (7%) but not in OSV9 mice. Periosteal mineral appositional rates as measured by double tetracycline labeling (B) were elevated (66 –130%) in both transgenic lines (C). Values are means ⫾ se. significant differences from FVB/N are indicated by asterisks above individual lines (P⬍0.05). Scale bar in panel B represents 10 ␮m.

gene interaction with local factors in the bone microenvironment, as discussed below. In caudal vertebrae, thickening of trabeculae and reduction of bone resorption occurred despite the maintenance of normal or slightly elevated levels of serum 1,25-dihydroxyvitamin D and PTH in the transgenic mice. This surprising result was not predicted by the large number of previous studies showing that 1,25-dihydroxyvitamin D acts through immature osteoblasts and stromal cells to stimulate osteoclastic recruitment and activity by direct and indirect mechanisms (12). Rather, the inhibition of

bone resorption evidently results from the elevated sensitivity of mature osteoblasts to normal endogenous levels of 1,25-dihydroxyvitamin D. Such a coupling of pathways may provide a system for local control of bone turnover and maintenance of microarchitectural integrity. The present approach allowed the specific responses of mature osteoblasts to be assessed without altering the sensitivity of immature osteoblastic and stromal cells, in contrast to earlier studies in which intact animals, mixed cell populations in culture, or tissues were treated with active vitamin D compounds. Given the strength of

Figure 4. The OSVDR transgene increases trabecular bone volume by reducing resorption. Photographs of mid-sagittal sections of fourth caudal vertebrae from 4 month old mice (A) show more abundant mineralized tissue in OSV3 bones. Vertebral trabecular bone volume (B) was 17–20% greater in both transgenic lines with similar increases (14 –17%) in vertebral trabecular thickness (C) but not trabecular number (D). Bone formation rate (E) was not affected by the transgene. Osteoclast surface (F ) in vertebral bone was reduced (⬎ 30%) in both transgenic lines. Trends to reduced osteoclast number (G) of 15% for OSV3 and 9% for OSV9 were not significant. In the femoral metaphyseal region, trabecular bone showed a trend for greater bone volume (P⫽0.18) (H) and reduced osteoclast surface (P⫽0.053) (I), with no trend in the epiphysis (J). Values are means ⫾ se. Significant differences from FVB/N are indicated by asterisks above individual lines (P⬍0.05). For Oc.S and Oc.N measurements of 9 month age group, n⫽9 to 13; for all other histomorphometry, n⫽17 to 20. LOWER BONE RESORPTION WITH HIGH VDR IN MATURE OSTEOBLASTIC CELLS

1913

the proresorptive response of immature osteoblastic and stromal cells to 1,25-dihydroxyvitamin D, it is unlikely that this counter-regulatory pathway would be detected in vivo without an experimental enhancement of the mature osteoblast response. A previous study that used a similar mouse osteocalcin gene-based transgene expression system to transiently or chronically ablate differentiated but still proliferating osteoblastic cells indicated that there is no obligatory cross-regulation between bone formation and resorption (30). Although our data do not address the issue of necessity, they do indicate that differentiated osteoblastic cells are able to decrease resorption. There was no indication in the previous study that ablation of differentiated osteoblastic cells caused an increase in osteoclastic resorption, as would be expected from the OSVDR findings. This may be because only proliferating cells were ablated, leaving lining cells and osteocytes intact and functional. If so, the implication is that the negative regulation of osteoclastic resorption may be mediated by these post-proliferative cells. Possible mechanisms Levels of the circulatory factors 1,25-dihydroxyvitamin D and PTH were not consistently changed and certainly were not reduced in the OSVDR mice, indicating that cortical and trabecular transgene effects are paracrine rather than endocrine. Paracrine pathways may act via soluble mediators such as growth factors or cytokines or via cell– cell or cell– matrix interactions. The mechanisms underlying the transgene effects on bone formation and resorption may involve multiple vitamin D-responsive pathways, some of which may affect both processes. One system that could mediate the resorption effect is the network of tumor necrosis factor family members, which has recently been shown to regulate osteoclastogenesis. This family includes the NF␬-B receptor activator RANK on osteoclast precursors, its ligand RANKL on immature osteoblastic cells, and the soluble decoy receptor osteoprotegerin (OPG) (31). As RANKL and OPG are normally expressed by osteoblastic cells and regulated by 1,25-dihydroxyvitamin D (32), a transgene-associated decrease in the local RANKL/OPG ratio could reduce osteoclastic recruitment/activation. To date, however, there is limited information about the regulation of these genes in mature osteoblastic cells. Another cytokine that may participate in similar regulatory systems is OCIL, a recently described cytokine that inhibits osteoclastogenesis, is expressed by mature osteoblastic cells and is up regulated by 1,25-dihydroxyvitamin D (33–35). Identification of the factor(s) involved in the transgenic reduction in bone resorption is the subject of ongoing investigation. 1914

Vol. 14

October 2000

In addition to the structural changes already described, material properties of the transgenic bones may also contribute to increased OSVDR bone strength and/or stiffness. Moreover, extracellular matrix composition can alter bone cell biology and gene expression (36) and thus may also contribute to the OSVDR bone phenotype, as 1,25-dihydroxyvitamin D is a common regulator of bone matrix protein genes. For example, a reduction in osteocalcin protein, as suggested by Northern blots, could contribute to the increase in periosteal mineral apposition rate, as it has been suggested that osteocalcin is an inhibitor of bone formation (37, 38). Studies of the osteocalcin-deficient mouse indicate, however, that a significant decrease in osteocalcin protein would cause an increase in osteoclast surface on trabecular bone rather than the decrease observed in the OSVDR mice (38). Given that the mechano-transducing osteocytes (39) and osteoblastic lining cells are sites of elevated VDR in the transgenic mice, response to weight bearing may be altered in the OSVDR bones. As noted above, however, periosteal bone formation was enhanced but endosteal and trabecular formation were unaffected by the transgene, indicating that any mechano-sensory contributions to the transgenic osteoblastic phenotype may be envelope specific. The transgene effect on trabecular osteoclast surface is also site specific, as it was evident in the fourth caudal vertebrae and the femoral metaphyseal region but not apparent in the more heavily loaded (40) femoral epiphysis. These local differences in cortical and trabecular bone responses to OSVDR could relate to differential expression of regulatory genes that are expressed by osteocytes, regulated by 1,25-dihydroxyvitamin D, and change levels in response to mechanical loading such as c-fos or osteopontin (28, 41– 43). This project was undertaken to model a potential therapeutic approach based on the hypothesis that net bone formation would result if mature osteoblastic cells could be made more responsive to vitamin D. As the VDR is normally expressed in this cell type, this genetic alteration represents a subtle enhancement in the context of otherwise normal calcium homeostasis and bone physiology. The reduced trabecular bone resorption in the OSVDR mice highlights a specific osteoclastic inhibitory mechanism. It acts via mature osteoblasts, where the transgene is expressed, and counterbalances osteoclastogenic signals from immature osteoblasts and osteoblastic stromal cells. This novel negative regulatory activity may also be controlled by other calcium homeostatic regulators such as PTH, interleukins, and prostanoids. A therapeutic approach that specifically enhances mature osteoblastic responses to endogenous or exogenous agents could constitute an ideal strat-

The FASEB Journal

GARDINER ET AL.

egy for osteoporosis treatment, decreasing trabecular bone resorption and increasing cortical bone formation, thereby reversing the typical patterns of osteoporotic and age-related bone loss. The authors thank Dr. H. DeLuca (University of Wisconsin, Madison), Dr. C. Gundberg (Yale University School of Medicine, New Haven), Dr. S. E. Harris (University of Texas Health Sciences Center, San Antonio), and Dr. P. Tuohimaa (Tampere University Medical School, Finland) for providing antisera and assay reagents. Thanks also to A. Bourne, R. Enriquez, R. McCabe, S. Gillies, and M. Svehla for technical assistance and to J. Ferguson, T. Chaplin, and P. Gregory for animal care. The work described here was performed in compliance with animal care and use guidelines from the Australian NH & MRC and the New South Wales State Government. The work was supported by a Center Grant and postgraduate scholarship support of C.P.W. from the NH & MRC, by NIH Grant 1RO1-AR43421 (U.S.) and by funding from Aza Research Pty. Ltd.

REFERENCES 1.

2. 3. 4. 5.

6.

7. 8. 9.

10.

11. 12. 13. 14.

Parfitt, A. M., Villanueva, A. R., Foldes, J., and Rao, D. S. (1995) Relations between histologic indices of bone formation: implications for the pathogenesis of spinal osteoporosis. J. Bone Miner. Res. 10, 466 – 473 Rico, H. (1997) The therapy of osteoporosis and the importance of cortical bone [editorial]. Calcif. Tissue Int. 61, 431– 432 McClung, M. R. (1996) Current bone mineral density data on bisphosphonates in postmenopausal osteoporosis. Bone 19, 195S–198S. Vedi, S., and Compston, J. E. (1996) The effects of long-term hormone replacement therapy on bone remodeling in postmenopausal women. Bone 19, 535–539 Gallagher, J. C., and Goldgar, D. (1990) Treatment of postmenopausal osteoporosis with high doses of synthetic calcitriol. A randomized controlled study [see comments]. Ann. Int. Med. 113, 649 – 655 Lips, P., Graafmans, W. C., Ooms, M. E., Bezemer, P. D., and Bouter, L. M. (1996) Vitamin D supplementation and fracture incidence in elderly persons. A randomized, placebo-controlled clinical trial [see comments]. Ann. Int. Med. 124, 400 – 406 Ott, S. M., and Chesnut, C. H., 3rd (1989) Calcitriol treatment is not effective in postmenopausal osteoporosis [see comments]. Ann. Int. Med. 110, 267–274 Tilyard, M. W., Spears, G. F., Thomson, J., and Dovey, S. (1992) Treatment of postmenopausal osteoporosis with calcitriol or calcium [see comments]. N. Engl. J. Med. 326, 357–362 Owen, T. A., Aronow, M. S., Barone, L. M., Bettencourt, B., Stein, G. S., and Lian, J. B. (1991) Pleiotropic effects of vitamin D on osteoblast gene expression are related to the proliferative and differentiated state of the bone cell phenotype: dependency upon basal levels of gene expression, duration of exposure, and bone matrix competency in normal rat osteoblast cultures. Endocrinology 128, 1496 –1504 Ishida, H., Bellows, C. G., Aubin, J. E., and Heersche, J. N. (1993) Characterization of the 1,25-(OH)2D3-induced inhibition of bone nodule formation in long-term cultures of fetal rat calvaria cells. Endocrinology 132, 61– 66 Marie, P. J., Hott, M., and Garba, M. T. (1985) Contrasting effects of 1,25-dihydroxyvitamin D3 on bone matrix and mineral appositional rates in the mouse. Metabolism 34, 777–783 Martin, T. J., and Ng, K. W. (1994) Mechanisms by which cells of the osteoblast lineage control osteoclast formation and activity. J. Cell. Biochem. 56, 357–366 Rodan, G. A., and Martin, T. J. (1982) Role of osteoblasts in hormonal control of bone resorption—a hypothesis [letter]. Calcif. Tissue Int. 34, 311 McDonnell, D. P., Scott, R. A., Kerner, S. A., O’Malley, B. W., and Pike, J. W. (1989) Functional domains of the human

15.

16.

17.

18.

19. 20. 21.

22.

23.

24. 25.

26.

27. 28.

29. 30.

31.

32.

vitamin D3 receptor regulate osteocalcin gene expression. Mol. Endocrinol. 3, 635– 644 Sims, N. A., White, C. P., Sunn, K. L., Thomas, G. P., Drummond, M. L., Morrison, N. A., Eisman, J. A., and Gardiner, E. M. (1997) Human and murine osteocalcin gene expression: conserved tissue restricted expression and divergent responses to 1,25-dihydroxyvitamin D3 in vivo. Mol. Endocrinol. 11, 1695–1708 Reeves, P. G., Nielsen, F. H., and Fahey, G. C., Jr. (1993) AIN-93 purified diets for laboratory rodents: final report of the American Institute of Nutrition ad hoc writing committee on the reformulation of the AIN-76 rodent diet. J. Nutr. 123, 1939 – 1951 Hollis, B. W., Kamerud, J. Q., Kurkowski, A., Beaulieu, J., and Napoli, J. L. (1996) Quantification of circulating 1,25-dihydroxyvitamin D by radioimmunoassay with 125I-labeled tracer. Clin. Chem. 42, 586 –592 Gundberg, C. M., Clough, M. E., and Carpenter, T. O. (1992) Development and validation of a radioimmunoassay for mouse osteocalcin: paradoxical response in the Hyp mouse. Endocrinology 130, 1909 –1915 Uhland-Smith, A., Prahl, J. M., and DeLuca, H. F. (1996) An enzyme-linked immunoassay for the 1,25-dihydroxyvitamin D3 receptor protein. J. Bone Miner. Res. 11, 1921–1925 Pierce, E. A., Dame, M. C., and DeLuca, H. F. (1987) Size and charge of the functional 1,25-dihydroxyvitamin D receptor in porcine intestine. J. Biol. Chem. 262, 17092–17099 Tuohimaa, P., Blaüer, M., Jaä¨skelaïnen, T., Itkonen, A., Lindfors, M., Mahonen, A., Palvimo, J., Vilja, P., and Maënpaä¨, P. H. (1992) Characterization of human 1,25-dihydroxyvitamin D3 receptor anti-peptide antibodies. J. Steroid Biochem. Mol. Biol. 43, 649 – 657 Page, K. (1977) Bone and preparation of bone sections. In Theory and Practice of Histological Techniques (Bancroft, J. D., and Stevens, A., eds) pp. 223–248, Churchill Livingston, London, England Parfitt, A. M., Mathews, C. H., Villanueva, A. R., Kleerekoper, M., Frame, B., and Rao, D. S. (1983) Relationships between surface, volume, and thickness of iliac trabecular bone in aging and in osteoporosis. Implications for the microanatomic and cellular mechanisms of bone loss. J. Clin. Invest. 72, 1396 –1409 Bak, B., and Jensen, K. S. (1992) Standardization of tibial fractures in the rat. Bone 13, 289 –295 Murakami, T., Yamamoto, M., Ono, K., Nishikawa, M., Nagata, N., Motoyoshi, K., and Akatsu, T. (1998) Transforming growth factor-beta1 increases mRNA levels of osteoclastogenesis inhibitory factor in osteoblastic/stromal cells and inhibits the survival of murine osteoclast-like cells. Biochem. Biophys. Res. Commun. 252, 747–752 Yasuda, H., Shima, N., Nakagawa, N., Yamaguchi, K., Kinosaki, M., Goto, M., Mochizuki, S. I., Tsuda, E., Morinaga, T., Udagawa, N., Takahashi, N., Suda, T., and Higashio, K. (1999) A novel molecular mechanism modulating osteoclast differentiation and function. Bone 25, 109 –113 Zhang, R., Ducy, P., and Karsenty, G. (1997) 1,25-Dihydroxyvitamin D3 inhibits osteocalcin expression in mouse through an indirect mechanism. J. Biol. Chem. 272, 110 –116 Lian, J. B., Shalhoub, V., Aslam, F., Frenkel, B., Green, J., Hamrah, M., Stein, G. S., and Stein, J. L. (1997) Species-specific glucocorticoid and 1,25-dihydroxyvitamin D responsiveness in mouse MC3T3–E1 osteoblasts: dexamethasone inhibits osteoblast differentiation and vitamin D down-regulates osteocalcin gene expression. Endocrinology 138, 2117–2127 Thomsen, J. S., Mosekilde, L., and Mosekilde, E. (1996) Quantification of remodeling parameter sensitivity—assessed by a computer simulation model. Bone 19, 505–511 Corral, D. A., Amling, M., Priemel, M., Loyer, E., Fuchs, S., Ducy, P., Baron, R., and Karsenty, G. (1998) Dissociation between bone resorption and bone formation in osteopenic transgenic mice. Proc. Natl. Acad., Sci. USA 95, 13835–13840 Suda, T., Takahashi, N., Udagawa, N., Jimi, E., Gillespie, M. T., and Martin, T. J. (1999) Modulation of osteoclast differentiation and function by the new members of the tumor necrosis factor receptor and ligand families. Endocr. Rev. 20, 345–357 Horwood, N. J., Elliott, J., Martine, T. J., Gillespie, M. T. (1998) Osteotropic agents regulate the expression of osteoclast differentiation factor and osteoprotegerin in osteoblastic stromal cells. Endocrinology 139, 4743– 4746


1915

33.

34.

35.

36.

37.

38.

39.

1916

Zhou, H., Hu, Y. S., Kartsogiannis, V., Elliott, J., Quinn, J., Gillespie, M. T., Martin, T. J., Ng, K. W. (1999) A novel osteoblast-derived molecule that inhibits the early phase of osteoclast formation. J. Bone Miner. Res. 14, S198 Kartsogiannis, V., Zhou, H., Hu, Y. S., Gillespie, M. T., Martin, T. J., Ng, K. W. (1999) Tissue localization of a novel protein. OCIL, that inhibits the early phase of osteoclast formation. J. Bone Miner Res 14, S361 Zhou, H., Y. Hu, S. Wong, M. Gillespie, T. Martin, and K. Ng. (1999) Regulatory studies of a novel protein, OCIL, that inhibits the early phase of osteoclast formation. J. Bone Miner. Res. 14, S361 Gerstenfeld, L. C. (1999) Osteopontin in skeletal tissue homeostasis: an emerging picture of the autocrine/paracrine functions of the extracellular matrix [comment]. J. Bone Miner. Res. 14, 850 – 855 Price, P. A., Williamson, M. K., Haba, T., Dell, R. B., and Jee, W. S. (1982) Excessive mineralization with growth plate closure in rats on chronic warfarin treatment. Proc. Natl. Acad. Sci. USA 79, 7734 –7738 Ducy, P., Desbois, C., Boyce, B., Pinero, G., Story, B., Dunstan, C., Smith, E., Bonadio, J., Goldstein, S., Gundberg, C., Bradley, A., and Karsenty, G. (1996) Increased bone formation in osteocalcin-deficient mice. Nature (London) 382, 448 – 452 Nijweide, P. J., Burger, E. H., Klein Nulend, J., Van der Plas, A. (1996) The osteocyte. In Principles of Bone Biology (Belizekian,

Vol. 14

October 2000

40.

41.

42.

43.

J. P., Raisz, L. G., and Rodan, G. A., eds) pp. 115–126, Academic Press, San Diego Westerlind, K. C., Wronski, T. J., Ritman, E. L., Luo, Z. P., An, K. N., Bell, N. H., and Turner, R. T. (1997) Estrogen regulates the rate of bone turnover but bone balance in ovariectomized rats is modulated by prevailing mechanical strain. Proc. Natl. Acad. Sci. USA 94, 4199 – 4204 Candeliere, G. A., Jurutka, P. W., Haussler, M. R., and StArnaud, R. (1996) A composite element binding the vitamin D receptor, retinoid X receptor alpha, and a member of the CTF/NF-1 family of transcription factors mediates the vitamin D responsiveness of the c-fos promoter. Mol. Cell. Biol. 16, 584 –592 Inaoka, T., Lean, J. M., Bessho, T., Chow, J. W., Mackay, A., Kokubo, T., and Chambers, T. J. (1995) Sequential analysis of gene expression after an osteogenic stimulus: c-fos expression is induced in osteocytes. Biochem. Biophys. Res. Commun. 217, 264 –270 Terai, K., Takano-Yamamoto, T., Ohba, Y., Hiura, K., Sugimoto, M., Sato, M., Kawahata, H., Inaguma, N., Kitamura, Y., and Nomura, S. (1999) Role of osteopontin in bone remodeling caused by mechanical stress [see comments]. J. Bone Miner. Res. 6, 839 – 849

The FASEB Journal

Received for publication December 24, 1999. Revised for publication March 29, 2000.

GARDINER ET AL.