Bone morphogenetic protein-7 (OP1) and transforming growth factor-beta1 modulate 1,25(OH)2-vitamin D3-induced differentiation of human osteoblasts

Bone morphogenetic proteins (BMPs) and transforming growth factor-beta (TGFbeta) are potent regulators of osteoblast differentiation and proliferation, processes that are crucial in bone remodeling. BMPs and TGFbeta act in concert with other local factors and hormones, among them 1,25(OH)2-vitamin D3 and insulin. Here we show that BMP7 inhibits 1,25(OH)2-vitamin D3-induced differentiation of human osteoblasts, whereas TGFbeta1 stimulates it, as assessed by assays for alkaline phosphatase (ALP) induction, matrix mineralization, and morphology changes. BMP7 or TGFbeta1 alone affects the differentiation of human osteoblasts. Similar results were obtained in assays for ALP induction using conditionally immortalized human osteoblasts (hFOB) and primary osteoblasts obtained from trabecular bone of the femoral head after hip replacement surgery. BMP7 stimulation led to a decrease of 1,25(OH)2-vitamin D3-induced binding of nuclear proteins to a vitamin D response element, as shown by electrophoretic mobility shift assay. Our results suggest that 1,25(OH)2-vitamin D3 modulates in opposite ways the effects of BMP7 and TGFbeta1 on osteoblast differentiation.     

Transforming growth factor beta inhibits plasminogen activator (PA) activity and stimulates production of urokinase-type PA, PA inhibitor-1 mRNA, and protein in rat osteoblast-like cells.

Transforming growth factor beta (TGF beta) treatment of rat osteoblast-rich calvarial cells or of the clonal osteogenic sarcoma cells, UMR 106-01, resulted in dose-dependent inhibition of plasminogen activator (PA) activity, and increased production of 3.2 kb mRNA and protein for PA inhibitor -1 (PAI-1). Although tissue-type PA (tPA) protein was not measured, TGF beta did not influence production of mRNA for tPA. Production of 2.3 kb mRNA for urokinase-type PA (uPA) was also increased by TGF beta in a dose-dependent manner. The effects of TGF beta on synthesis of mRNA for PAI-1 and uPA were maintained when protein synthesis was inhibited, and were abolished by inhibition of RNA synthesis. Although uPA had not been detected previously as a product of rat osteoblasts, treatment of lysates of osteoblast-like cells with plasmin yielded a band of PA activity on reverse fibrin autography, corresponding to a low Mr form of uPA. Untreated conditioned media from normal osteoblasts or UMR 106-01 cells contained no significant TGF beta activity, but activity could be detected in acidified medium. Treatment of conditioned media with plasmin resulted in activation of approximately 50% of the TGF beta detectable in acidified media. The results identify several effects of TGF beta on the PA-PA inhibitor system in osteoblasts. Net regulation of tPA activity through the stimulatory actions of several calciotropic hormones and the promotion of PAI-1 formation by TGF beta could determine the amount of osteoblast-derived TGF beta activated locally in bone. Stimulation of osteoblast production of mRNA for uPA could reflect effects on the synthesis of sc-uPA, a precursor for the active form of the enzyme.     

Possible involvement of the p57(Kip2) gene in bone metabolism

We previously uncovered that growth stimulation of rat primary osteoblasts by transforming growth factor-beta1 (TGF-beta1) resulted in a dramatic decrease in p57(Kip2), a member of cyclin-dependent kinase (CDK) inhibitors, through the proteasomal degradation pathway (Urano et al., J. Biol. Chem. 274, 12197-12200, 1999). Here we demonstrated that the amount of p57 protein increases markedly, when rat calvarial primary osteoblasts treated with 1,25-dihydroxyvitamin D3 transit from proliferation toward differentiation. Next, we have analyzed the association of four amino acids deletion polymorphism of p57 and bone mineral density (BMD). The p57 genotype was determined in 154 postmenopausal Japanese women. When we separated the subjects into two groups, one having one or two copies of deletion polymorphism and the other without the deletion, the former subjects had higher BMD (Z score of total body, 0.67 +/- 0.93 vs 0. 23 +/- 0.90, mean +/- standard deviation; P = 0.021). Taken together, these findings suggest that the p57 regulated in the osteoblast proliferation and differentiation may play a role in determination of bone mineral density and pathogenesis of osteoporosis.     

Species difference exists in the effects of 1alpha,25(OH)(2)D(3) and its analogue 2-methylene-19-nor-(20S)-1,25-dihydroxyvitamin D(3) (2MD) on osteoblastic cells

The direct effect of 1alpha,25(OH)(2)D(3) on osteoblasts remains unclear. In this study, we evaluated the in vitro effects of 1alpha,25(OH)(2)D(3) and its analogue, 2-methylene-19-nor-(20S)-1,25-dihydroxyvitamin D(3) (2MD), on osteoblasts from three different species, i.e. bone marrow stromal cells from the Sprague-Dawley (SD) rat, from the C57BL/6 mouse, as well as human osteoblast NHOst cells and human osteosarcoma derived MG-63 cells. We found that in rat cells, both compounds increased cell proliferation, inhibited cell apoptosis and increased alkaline phosphatase (ALP) activity. In mouse cells, however, both compounds initiated cell apoptosis and inhibited ALP activity. In human cells, although cell proliferation was inhibited by both compounds, cell apoptosis was inhibited and ALP activity was enhanced. In each species, 2MD was much more potent than 1alpha,25(OH)(2)D(3). To summarize, species differences should be taken into account in studies of vitamin D effects. However, in all tested species - rat, mouse and human - 2MD is considerably more potent in its effects on osteoblastic cells in vitro than 1alpha,25(OH)(2)D(3).     

Evidence of vitamin D and interferon-β cross-talk in human osteoblasts with 1α,25-dihydroxyvitamin D3 being dominant over interferon-β in stimulating mineralization

It is well established that 1α-25-dihydroxyvitamin D3 (1,25D3) regulates osteoblast function and stimulates mineralization by human osteoblasts. The aim of this study was to identify processes underlying the 1,25D3 effects on mineralization. We started with gene expression profiling analyses of differentiating human pre-osteoblast treated with 1,25D3. Bioinformatic analyses showed interferon-related and -regulated genes (ISG) to be overrepresented in the set of 1,25D3-regulated genes. 1,25D3 down-regulated ISGs predominantly during the pre-mineralization period. This pointed to an interaction between the vitamin D and IFN signaling cascades in the regulation of osteoblast function. Separately, 1,25D3 enhances while IFNβ inhibits mineralization. Treatment of human osteoblasts with 1,25D3 and IFNβ showed that 1,25D3 completely overrules the IFNβ inhibition of mineralization. This was supported by analyses of extracellular matrix gene expression, showing a dominant effect of 1,25D3 over the inhibitory effect of IFNβ. We identified processes shared by IFNβ- and 1,25D3-mediated signaling by performing gene expression profiling during early osteoblast differentiation. Bioinformatic analyses revealed that genes being correlated or anti-correlated with interferon-induced protein with tetratricopeptide repeats 1 (IFIT1) were associated with osteoblast proliferation. In conclusion, the current study demonstrates a cross talk between 1,25D3 and IFNβ in osteoblast differentiation and bone formation/mineralization. The interaction is complex and depends on the process but importantly, 1,25D3 stimulation of mineralization is dominant over the inhibitory effect of IFNβ. These observations are of potential clinical relevance considering the impact of the immune system on bone metabolism in conditions such as rheumatoid arthritis.     

Hormonal regulation of [Ca(2+)](i) in periosteal-derived osteoblasts: effects of parathyroid hormone, 1,25(OH)(2)D(3) and prostaglandin E(2)

The effects of hormonal modulators of osteoblast function, parathyroid hormone, 1,25(OH)(2)D(3) and prostaglandins on [Ca(2+)](i) in periosteal-derived osteoblasts from rat femurs have been investigated. Our results show that application of parathyroid hormone PTH (10(-5) M) and prostaglandin E(2) (PGE(2)) (4 microM) result in a rapid heterogeneous elevation in [Ca(2+)](i) that, in the case of PTH, is dependent on both extracellular and intracellular sources of calcium. Variable responses to treatments have been found within populations of cells. The PGE(2) response is dose dependent. Treatment with 1,25(OH)(2)D(3) (10(-8) M) induces a brief (60-90 sec) elevation in [Ca(2+)](i) that is almost totally abolished in EGTA-buffered Ca(2+)-free medium. Interactive effects of multiple hormone treatments have been observed. Pretreatment with 1,25(OH)(2)D(3) results in near-total inhibition of the PTH and PGE(2) responses. In conclusion, modulation of [Ca(2+)](i) appears to play a role not only in the direct effects of osteotropic hormones on osteoblasts but also in the synergistic and antagonistic effects between circulating hormones.     

Effects of combining transforming growth factor beta and 1,25-dihydroxyvitamin D3 on differentiation of a human osteosarcoma (MG-63).

Transforming growth factor beta (TGF beta) and 1,25-dihydroxyvitamin D3 (1,25D3), when added simultaneously to a human osteosarcoma cell line, MG-63, induce alkaline phosphatase activity 40-70-fold over basal levels, 6-7-fold over 1,25D3 treatment alone, and 15-20-fold over TGF beta treatment alone. TGF beta and 1,25D3 synergistically increased alkaline phosphatase specific activity in both matrix vesicles and plasma membrane isolated from the cultures, but the specific activity was greater in and targeted to the matrix vesicle fraction. Inhibitor and cleavage studies proved that the enzymatic activity was liver/bone/kidney alkaline phosphatase. Preincubation of MG-63 cells with TGF beta for 30 min before addition of 1,25D3 was sufficient for maximal induction of enzyme activity. Messenger RNA for liver/bone/kidney alkaline phosphatase was increased 2.1-fold with TGF beta, 1.7-fold with 1,25D3, and 4.8-fold with the combination at 72 h. Human alkaline phosphatase protein as detected by radioimmunoassay was stimulated only 6.3-fold over control levels with the combination. This combination of factors was tested for their effect on production of three other osteoblast cell proteins: collagen type I, osteocalcin, and fibronectin. TGF beta inhibited 1,25D3-induced osteocalcin production, whereas both factors were additive for fibronectin and collagen type I production. TGF beta appears to modulate the differentiation effects of 1,25D3 on this human osteoblast-like cell and thereby retain the cell in a non-fully differentiated state.     

Effect of ageing on the expression of protein kinase C and its activation by 1,25(OH)2-vitamin D3 in rat skeletal muscle

To characterize age-induced effects on muscle protein kinase C (PKC) and its regulation by the steroid hormone 1,25(OH)2-vitamin D3 [1,25(OH)2D3], changes in PKC activity and the expression and translocation of the specific PKC conventional isoforms alpha and beta, novel isoforms delta, epsilon, and theta and atypical isoform zeta were studied in homogenates and subcellular fractions from skeletal muscle of young (3 months) and aged (24 months) rats treated in vitro with 1,25(OH)2D3. The hormone (10(-9) M) increased total and membrane PKC activity, within 1 min, and these effects were completely blunted in muscle from aged rats. The presence of PKC isoenzymes was shown by Western blot analysis with the use of specific antibodies. The expression of PKC alpha, beta and delta was greatly diminished in old rats, whereas age-related changes were less pronounced in the isoforms epsilon, theta and zeta. After a short exposure (1 min) of muscle to 1,25(OH)2D3, increased amounts of PKC alpha and beta in muscle membranes and reverse translocation (from membrane to cytosol) of PKC epsilon were observed only in young animals. The data indicate that, in rat muscle, ageing impairs calcium-dependent PKC (alpha and beta) and calcium-independent PKC (delta, epsilon, theta and zeta) signal transduction pathways under selective regulation by 1,25(OH)2D3.     

Estrogens (E2) regulate expression and response of 1,25-dihydroxyvitamin D3 receptors in bone cells: changes with aging and hormone deprivation

Studies on the effect of estrogens (E(2)) on the expression of vitamin D receptor (VDR) and its bioresponse in bone have demonstrated that E(2) modulate activity and increase the number of VDRs in vitro; however, no in vivo studies have been pursued to assess this interaction. Our study identifies the changes in the number of VDR-expressing cells in bone of C57BL/6J young and old oophorectomized mice (4 and 24 months) with and without 17beta estradiol (E(2)) replacement. A total of 36 mice were sacrificed; both tibiae and femora were isolated and VDR expression was quantified by Northern blot, immunohistochemistry, immunofluorescence, and flow cytometry. Among the intact mice there was a significant difference in the number of VDR-expressing osteoblasts between young (68%) and old (56%) (p<0.04). In young oophorectomized mice the number of VDR-expressing osteoblasts decreased from 68% to 46% after oophorectomy and recovered to 72% after E(2) administration (p<0.02), while in the group of old mice, the number of VDR-expressing osteoblasts decreased from 56% to 48% after oophorectomy (p<0.01) and recovered to 85% after E(2) administration (p<0.001). Our results show that VDR expression in bone decreases with aging and estrogen deprivation but recovers after E(2) supplementation in both young and old mice with a more significant level of response in older bone. To evaluate the level of VDR bioresponse to E(2) we assessed the effect of E(2) supplementation to human osteoblasts (N-976) in vitro. Northern blot showed a significant up-regulation of VDR expression in E(2) treated cells as compared to non-treated cells (p<0.05). We also assessed the previously known anti-apoptotic effect of vitamin D in osteoblasts in vitro after serum deprivation by using either E(2), E(2)+1,25(OH)(2)D(3), or 1,25(OH)(2)D(3) alone. We found a lower number of apoptotic cells and longer cell survival after 48 h of treatment with 1,25(OH)(2)D(3)+E(2) as compared to 1,25(OH)(2)D(3) or E(2) alone (p<0.002). In summary, our results demonstrate that E(2) increases VDR expression in bone in vivo and potentiate the bioresponse of VDR in osteoblasts in vitro.     

Parathyroid hormone (1-34) regulates integrin expression in vivo in rat osteoblasts.

Intermittent administration of parathyroid hormone (PTH) activates new sites of bone formation by stimulating osteoblast differentiation and function resulting in an increase in bone mass. Because integrins have been shown to play a crucial role in osteoblast differentiation and bone formation, in the present study, we evaluated whether human PTH (1-34) upon administration to rats, influenced integrin expression in osteoblastic cells isolated from the metaphysis and the diaphysis of rat long bones. Initial immunohistochemical evaluation of bone sections demonstrated that the osteoblasts expressed at least alphav, alpha2, alpha3, and alpha5beta1 integrins. Immunocolocalization studies for integrins and vinculin established that alphav, alpha2, and alpha5beta1, but not alpha3 integrins were present in the focal adhesion sites of osteoblasts attached to FN coated surfaces. Osteoprogenitor cells isolated from metaphyseal (but not diaphyseal) marrow of rats injected with intermittent PTH (1-34) exhibited greater alphav and reduced alpha2 levels, with no apparent changes in alpha3, and alpha5beta1 integrin levels, as assessed by immunohistochemistry, Northern, and Western blot analyses. However, these changes were not observed on the same cells treated with PTH in vitro. These observations suggest that integrin modulation by PTH is likely to be indirect and that selective phenotypic expression of integrin subtypes is part of the cascade of events that lead to PTH (1-34) mediated osteoblast differentiation.     

Effect of 17 beta-estradiol on the proliferation of osteoblastic MC3T3-E1 cells via human monocytes

Effects of human monocyte-conditioned medium on the proliferation of osteoblastic MC3T3-E1 cells were investigated in serum-free cultured condition. Monocyte-conditioned medium significantly stimulated osteoblast proliferation at the concentration between 10 and 30%, compared to that in the absence of monocytes. 17 beta-estradiol directly stimulated osteoblast proliferation at the concentrations of 10(-8) and 10(-10)M. On the contrary, the conditioned medium prepared by monocytes cultured in the presence of 17 beta-estradiol at the concentrations of 10(-8) and 10(-10)M significantly inhibited osteoblast proliferation. Present data indicate that in addition to direct effect on osteoblasts, 17 beta-estradiol affected osteoblast proliferation presumably through modulating the release of several local regulators of bone turnover from monocytes. The effect on osteoblastic activity via monocytes might be linked to the coupling of osteoclast and osteoblast actions.     

Prolactin decreases the expression ratio of receptor activator of nuclear factor kappaB ligand/osteoprotegerin in human fetal osteoblast cells

Prolactin (PRL) enhanced bone remodeling leading to net bone loss in adult and net bone gain in young animals. Studies in PRL-exposed osteoblasts derived from adult humans revealed an increase in the expression ratio of receptor activator of nuclear factor kappaB ligand (RANKL) and osteoprotegerin (OPG), thus supporting the previous finding of PRL-induced bone loss in adults. This study thus investigated the effects of PRL on the osteoblast functions and the RANKL/OPG ratio in human fetal osteoblast (hFOB) cells which strongly expressed PRL receptors. After 48h incubation, PRL increased osteocalcin expression, but had no effect on cell proliferation. However, the alkaline phosphatase activity was decreased in a dose-response manner within 24h. The effect of PRL on alkaline phosphatase was abolished by LY294002, a phosphoinositide 3-kinase (PI3K) inhibitor. PRL also decreased the RANKL/OPG ratio by downregulating RANKL and upregulating OPG expression, implicating a reduction in the osteoblast signal for osteoclastic bone resorption. It could be concluded that, unlike the osteoblasts derived from adult humans, PRL-exposed hFOB cells exhibited indices suggestive of bone gain, which could explain the in vivo findings in young rats. The signal transduction of PRL in osteoblasts involved the PI3K pathway.     

1,25(OH)2D3 and calcipotriol (MC903) have similar effects on the induction of osteoclast-like cell formation in human bone marrow cultures

MC903 is a novel analogue of 1,25(OH)2D3 which exhibits similar inhibitory effects on cell proliferation and like, 1,25(OH)2D3, stimulates synthesis of osteoblast specific proteins by osteoblast-like cells in vitro. It is less active than 1,25(OH)2D3 in causing hypercalcemia in vivo. Since 1,25(OH)2D3 is known to stimulate bone resorption and increase the number of osteoclasts in several systems (in vivo and in vitro) we examined the effects of MC903 on the formation of osteoclast-like cells in vitro. As reported previously 1,25(OH)2D3 promoted the formation of multinucleated cells with phenotypic and functional characteristics of osteoclasts from adult human bone-marrow cultures at concentrations between 10(-8)M to 10(-12)M. Higher doses consistently suppressed multinucleated cell formation to values seen in the absence of 1,25(OH)2D3. Cells cultured in the presence of MC903 or for three weeks consistently induced the formation of multinucleated cells at concentrations 10(-8)M to 10(-12)M. As seen with 1,25(OH)2D3, MC903 also inhibited multinucleated cell formation at very high concentrations (10(-6)M). In two separate experiments MC903 appeared to be more potent than 1,25(OH)2D3 at lower concentrations (10(-10)M - 10(-12)M). From this study we conclude that MC903 is at least as potent as 1,25(OH)2D3 in inducing the formation human osteoclast-like cells in vitro. The decreased ability of MC903 to induce hypercalcemia in vivo is not therefore a result of a less marked effect than 1,25(OH)2D3 on the regulation of osteoclast formation.     

Type beta transforming growth factor (TGF beta) regulation of alkaline phosphatase expression and other phenotype-related mRNAs in osteoblastic rat osteosarcoma cells

TGF beta 1 from porcine platelets increased alkaline phosphatase (AP) activity in the rat osteoblastic cell line ROS 17/2.8 about three-fold. This effect was dose-dependent with an ED50 of about approximately 0.2 ng/ml and was larger during logarithmic growth than at confluence. TGF beta 1 inhibited cell growth by about 30% with similar dose dependence. Thirty min exposure to TGF beta 1 was sufficient to increase AP activity 3 days later by about two-fold but did not affect cell growth, suggesting dissociation between effects on proliferation and differentiation. The rise in AP activity started 6 h after TGF beta 1 addition and was blocked by cycloheximide and actinomycin D. TGF beta 1 also increased AP mRNA by two- to three-fold and this effect was not blocked by cycloheximide. The half-life of AP mRNA, estimated following the addition of 5,6-dichloro-1-beta-D-ribofuranosylbenzimidazole was about ten h in both control and TGF beta 1-treated cells. The mRNAs for type I procollagen and osteonectin were also increased by TGF beta 1 but fibronectin mRNA was decreased. TGF beta 2 effects on AP and cell growth were similar to those of TGF beta 1, except for lack of activity following transient exposure. At saturating concentrations, TGF beta 2 (2 ng/ml) or dexamethasone (10(-7) M), which has similar effects on these cells, did not further augment the effects of TGF beta 1 (at 2 ng/ml). Above findings suggest that TGF beta promotes osteoblastic differentiation in rat osteosarcoma cells at least in part by acting at the pretranslational level.     

Nitric oxide mediates 17beta-estradiol-stimulated human and rodent osteoblast proliferation and differentiation

Oestradiol can stimulate osteoblast activity. Osteoblast function is thought to be regulated by nitric oxide (NO). We hypothesised that the effect of 17beta-oestradiol (17beta-E(2)) on osteoblast activity is mediated by NO. This hypothesis was tested using osteoblasts isolated from human trabecular bone, calvariae of rats, endothelial NO synthase (eNOS) gene-deficient mice, and their wild-type counterparts. Our results show that 17beta-E(2) dose-dependently stimulated proliferation and differentiation of primary human, rat and wild-typeosteoblasts. The presence of N(G)-monomethyl-l-arginine (10(-3) M), an inhibitor of NOS activity, blocked the 17beta-E(2)-(10(-7) M)-induced increases in thymidine incorporation (P < 0.01), alkaline phosphatase activity (P < 0.01) and bone nodule formation (P < 0.01) of wild-type, human and rat osteoblasts, respectively. Moreover, 17beta-E(2) did not induce a response in eNOS gene-deficient osteoblasts. 17beta-E(2) also increased total eNOS enzyme expression in rat osteoblasts. These findings indicate 17beta-E(2) modulates osteoblast function by NO-dependent mechanisms mediated via the eNOS isoform.     

Osteoblast migration on poly(alpha-hydroxy esters)

We investigated the migration of rat calvaria osteoblast populations on poly(alpha-hydroxy ester) films for up to 14 days to determine effects of substrate composition and culture conditions on the migratory characteristics of osteoblasts. Initial osteoblast culture conditions included cell colonies formed by seeding a high (84,000 cells/cm(2)) or low (42,000 cells/cm(2)) density of isolated osteoblasts on the polymer films, and bone tissue cultures formed by plating bone chips directly on the substrates. High density osteoblast colonies cultured and allowed to migrate and proliferate radially on 85:15 poly(DL-lactic-co-glycolic acid) (PLGA) films, 75:25 PLGA films, and tissue culture polystyrene controls demonstrated that the copolymer ratio in the polymer films did not affect the rate of increase in substrate surface area (or culture area) covered by the growing cell colony. However, the rate of increase in culture area was dependent on the initial osteoblast seeding density. Initial cell colonies formed with a lower osteoblast seeding density on 75:25 PLGA resulted in a lower rate of increase in culture area, specifically 4.9 +/- 0.3 mm(2)/day, versus 14.1 +/- 0.7 mm(2)/day for colonies seeded with a higher density of cells on the same polymer films. The proliferation rate for osteoblasts in the high and low density seeded osteoblast colonies did not differ, whereas the proliferation rate for the osteoblasts arising from the bone chips was lower than either of these isolated cell colonies. Confocal and light microscopy revealed that the osteoblast migration occurred as a monolayer of individual osteoblasts and not a calcified tissue front. These results demonstrated that cell seeding conditions strongly affect the rates of osteoblast migration and proliferation on biodegradable poly(alpha-hydroxy esters). (c) 1996 John Wiley & Sons, Inc.