Hormonal Regulation of the Osteoblastic Phenotype Expression in Neonatal Murine Calvarial Cells

Osteoblastic cell cultures from fetal rat calvariae have provided a popular model for studying the effects of dexamethasone (DEX) and 1,25 dihydroxyvitamin D₃ [1,25(OH)₂D₃] on gene expression but data from murine calvarial cells are scarce. Species-specific responses of rat and mouse osteoblastic cells to these hormones have been reported previously. In the present study, we investigated the effects of DEX and 1,25(OH)₂D₃ on expression of the osteoblastic phenotype by mouse calvarial cells. These murine osteoblast-like (MOB) cells expressed alkaline phosphatase (ALP) activity and osteocalcin and formed calcified nodules. Unlike the rat calvarial cells, ALP activities and nodule formation in MOB were inhibited by DEX. 1,25(OH)₂D₃ enhanced and DEX lowered the amount of osteocalcin synthesized by MOB. 1,25(OH)₂D₃ did not affect the number of nodules, but increased their sizes. Treating the cells for 2 days with only DEX at the beginning of the culture enhanced the effect of 1,25(OH)₂D₃ on ALP. We found that in murine calvarial cells, DEX inhibits and 1,25(OH)₂D₃ enhances ALP activity, osteocalcin synthesis, and calcified nodule formation. This is in contrast to previous reports of rat calvarial cells where DEX is a positive and 1,25(OH)₂D₃ can be a negative regulator of the osteoblastic phenotype. These results suggest that profound species-specific differences exist between mice and rats in the regulation of the osteoblastic phenotype. Received: 15 October 1997 / Accepted: 16 June 1998     

Effect of Polyethylene on Osteocalcin, Alkaline Phosphatase and Procollagen Secretion by Human Osteoblastic Cells

We have studied the direct effects of polyethylene particles on osteoblastic function in primary human bone cell cultures. The cells were obtained from trabecular bone fragments of patients undergoing knee reconstructive surgery. When the cells reached confluency, they were subcultured into two flasks, one untreated (control culture) and the other treated with polyethylene particles, and incubated until confluency. Osteoblastic function was evaluated by assaying osteocalcin, alkaline phosphatase, and C-terminal procollagen type I, with and without 1,25(OH)₂D stimulation, in the cell-conditioned medium. We found that addition of polyethylene to these osteoblastic cell cultures induced higher levels of secreted osteocalcin after 1,25(OH)₂D stimulation. Alkaline phosphatase levels increased whereas C-terminal procollagen type I levels decreased in the cell conditioned medium after polyethylene was added to the cultures. Treatment of the control cultures with 1,25(OH)₂D stimulated alkaline phosphatase levels and decreased C-terminal procollagen type I. However, these osteoblastic markers in 1,25(OH)₂D-treated cells did not change in cultures with polyethylene. This study demonstrates that polyethylene particles have a direct effect on osteoblastic markers in human bone cells in culture. Received: 2 May 1996 / Accepted: 17 June 1997     

睾酮对离体胎鼠头盖骨成骨细胞影响的研究

目的 探讨雄激素对成骨细胞代谢的影响。方法 胎鼠头盖骨成骨细胞培养于含不同浓度睾酮的培养基中．观察细胞胸腺嘧啶核苷掺入、碱性磷酸薛(ALP)活性、骨钙素基因表达及骨钙素合成等细胞增殖及分化等指标的变化情况。结果 胎鼠头盖骨成骨细胞的增殖不受睾酮影响,但ALP活性、骨钙素基因表达及其合成在不同程度上受睾酮的调节。结论 雄激素对成骨细胞的分化具有促进作用,但对其增殖无影响。     

Parathyroid Hormone (1-34) Receptor-Binding and Second-Messenger Response in Rat Incisor Odontoblasts

Even though indirect evidence indicates that PTH exerts an anabolic effect on dentinogenesis, the existence of PTH receptors and any second-messenger response in odontoblasts have not been demonstrated. The aim of this study was to investigate whether rat incisor odontoblasts express PTH receptors, and to identify which second messenger pathway the hormone may activate. Odontoblasts were dissected from rat incisors. Amino-terminal (1-34) fragment rat PTH [rPTH(1-34)] conjugated to fluorescein isothiocyanate visualized receptor sites on the cell surface. Upon incubation of odontoblasts with rPTH(1-34), cAMP formation was increased. However, no fluctuations in intracellular calcium activity were observed upon rPTH(1-34) stimulation when using Fura-2 as a Ca²⁺ probe. In long-time incubations, stimulation with PTH(1-34) upregulated APase activity. The results demonstrate that rPTH(1-34) evokes an anabolic response in dentinogenically active odontoblasts, and that this may be mediated through the protein kinase A/cAMP pathway, whereas no indications for Ca²⁺ as a second messenger were evident. Received: 12 March 1997 / Accepted: 26 June 1997     

A New fluorometric assay for determination of osteoblastic proliferation: Effects of glucocorticoids and insulin-like growth factor-I

A novel fluorometric proliferation assay, AlamarBlue (AB), was used to study the proliferative capacity of isolated human osteoblasts (hOBs). AB is an oxidation-reduction indicator that yields a fluorescent signal in response to metabolic activity. The assay was performed by replacing the experiment media in a microtiter plate with a 10% AB solution and measuring fluorescence after a 3–8-hour incubation. The assay was optimized with respect to incubation time, cell density, and AB concentration. When the results of the AB assay were compared with cell counting in a Bürker chamber there were consistently good correlations (r > 0.9), regardless of the agonist with which the cells were treated. The mean intraassay coefficient of variance (CV) values were 9.9–11.8% in experiments where osteoblasts were treated for 12 days with insulin-like growth factor-I (IGF-I; 100 nM), or dexamethasone (1 μM). IGF-I dose dependently, at and above 1 nM, stimulated proliferation of hOBs. This effect was detectable after 3 days and reached 130–140% of untreated controls after 12 days in culture. The effects of dexamethasone (DEX) on the proliferation rate of hOBs were more complex. In short-term cultures, 3 days, DEX dose dependently stimulated proliferation. However, at and above 6 days, DEX exerted a biphasic effect, with stimulation seen at 1–10 nM and a marked inhibition of cell proliferation at and above 100 nM. dexamethasone, hydrocortisone, prednisolone, and deflazacort had almost identical biphasic effects on osteoblastic proliferation in 12 day cultures with a stimulation seen at 1–10 nM, and a marked inhibition down to 50–60% of untreated controls at and above 100 nM. When IGF-I (0.1–100 nM; 12 day culture) was combined with different doses of DEX, IGF-I still dose dependently stimulated the proliferation rate in hOBs regardless of the amount of DEX added. The stimulatory effect of DEX (10 nM, 12 days culture) was additive to the effect of 100 nM IGF-I. We conclude that AB is an easy and reliable assay for osteoblastic cell proliferation, well suited for large scale studies of cell growth using small amounts of cells, and that IGF-I partly reverses the glucocorticoid-induced inhibition of osteoblastic proliferation. Received: 12 March 1996 / Accepted: 17 June 1996     

Bone Sialoprotein (BSP) is a Crucial Factor for the Expression of Osteoblastic Phenotypes of Bone Marrow Cells Cultured on Type I Collagen Matrix

In this study, we demonstrated that type I collagen matrix induced the expression of osteoblastic phenotypes of bone marrow cells, and that antibone sialoprotein (BSP) monoclonal antibody suppressed the expression of these phenotypes. On the other hand, BSP accelerated the expression of osteoblastic phenotypes of bone marrow cells. The adherent bone marrow cells were harvested from rat femur and cultured on type I collagen matrix gels in medium containing 15% fetal calf serum, neither β-glycerophosphate nor glucocorticoid. Cells showed osteoblastic phenotypes (high alkaline phosphatase activity, osteocalcin synthesis, and responsiveness against parathyroid hormone) on collagen matrix gels at week 3 after the inoculation, and simultaneously, BSP was detected in the conditioned medium by Western blotting using an anti-BSP monoclonal antibody. However, cells in the conventional culture dishes did not show osteoblastic phenotypes during the experimental period. To investigate the physiological function of BSP in osteoblastic differentiation, bone marrow cells were cultured on collagen matrix with an anti-BSP monoclonal antibody for 3 weeks. This treatment suppressed the expression of the osteoblastic phenotypes, and the effect of the antibody was abolished by the addition of bovine bone BSP. Furthermore, bovine bone BSP stimulated the expression of osteoblastic phenotypes of bone marrow cells. Our results indicate that BSP plays a crucial role in the expression of osteoblastic phenotypes of bone marrow cells. Received: 17 February 1999 / Accepted: 14 December 1999     

Bone Morphogenetic Protein 2 (BMP-2) Enhances BMP-3, BMP-4, and Bone Cell Differentiation Marker Gene Expression During the Induction of Mineralized Bone Matrix Formation in Culturesof Fetal Rat Calvarial Osteoblasts

Normal bone formation is a prolonged process that is carefully regulated and involves sequential expression of growth regulatory factors by osteoblasts as they proliferate and ultimately differentiate. Since this orderly sequence of gene expression by osteoblasts suggests a cascade effect, and BMP-2 is capable of initiating and maintaining this effect, we examined the effects of BMP-2 on expression of other BMPs and compared these effects with the expression pattern of bone cell differentiation marker genes in primary cultures of fetal rat calvarial (FRC) osteoblasts. To examine the gene expression profile during bone cell differentiation and bone formation, we also examined the effects of rBMP-2 on bone formation in vivo and in vitro. rBMP-2 stimulated bone formation on the periosteal surface of mice when 500 ng/day rBMP-2 was injected subcutaneously. When rBMP-2 was added to primary cultures of FRC osteoblasts, it accelerated mineralized nodule formation in a time and concentration-dependent manner (10–40 ng/ml). rBMP-2 (40 ng/ml) enhanced BMP-3 and -4 mRNA expression during the mineralization phase of primary cultures of FRC osteoblasts. Enhancement of BMP-3 and -4 mRNA expression by rBMP-2 was associated with increased expression of bone cell differentiation marker genes, alkaline phosphatase (ALP), type I collagen, osteocalcin (OC), osteopontin (OP), and bone sialoprotein (BSP). These results suggest that BMP-2 enhances expression of other BMP genes during bone cell differentiation. BMP-2 may act in a paracrine fashion in concert with other BMPs it induces to stimulate bone cell differentiation and bone formation during remodeling. Received: 27 November 1995 / Accepted: 19 July 1996     

Na+/Ca2+ Exchanger Isoforms of Rat Odontoblasts and Osteoblasts

In odontoblasts as well as osteoblasts, a number of mechanisms for the inflow and extrusion of Ca²⁺ have been demonstrated. The entrance of Ca²⁺ ions into odontoblasts occurs mainly through voltage-gated calcium channels. Extrusion of Ca²⁺ is found to be an ATP-dependent process and, in addition, Na⁺/Ca²⁺-antiports exist, which are provoked by extracellular Na⁺. The aim of this study was to identify the Na⁺/Ca²⁺-antiport isoforms expressed in dentinogenically active rat incisor odontoblasts and to make a comparison with different osteoblastic cells. Using RT-PCR and RNAse protection assay, we demonstrated the expression of three different isoforms, NaCa 3, 7, and 10, of the NCX1-encoded antiport in odontoblasts and osteoblastic cells. When incubated in the presence of Na⁺, dissected rat incisor odontoblasts as well as the osteoblastic cells extruded Ca²⁺ ions, as detected by chlorotetracycline and Fura-2 fluorometry, thus supporting a physiological role for the detected isoform expression. Odontoblasts and rat calvarial osteoblasts, as well as osteoblast-like cell lines UMR-106.01 and Saos-2, were shown to exhibit identical phenotypes of Na⁺/Ca²⁺-antiport isoform expression, different from the expression patterns of other tissues. The significance of this specific expression pattern is unknown, but there is a possibility that it is in some way related to the unique demands on these cell types to produce mineralized connective tissue. Received: 8 May 1999 / Accepted: 21 January 2000     

The Role of Osteoblast Density and Endogenous Interleukin-6 Production in Osteoclast Formation From the Hemopoietic Stem Cell Line FDCP-MIX C2GM in Coculture With Primary Osteoblasts

Osteoclast formation from the hemopoietic stem cell line FDCP-mix C₂GM was shown to be strongly dependent on osteoblast density. In cocultures of C₂GM cells with fetal mouse osteoblasts seeded at high density (i.e., 2.5 × 10⁴ cells/cm²), we found a significantly lower osteoclast formation compared with cocultures with osteoblasts seeded at low density (i.e., 1 × 10⁴ cells/cm²). The differentiation state of osteoblasts in high-density cultures resembled more than that of osteoblasts in low-density cultures, the differentiation state of mature osteoblasts, since the cells in the former cultures showed higher alkaline phosphatase (APase) activity than the cells in the latter cultures, and nodules were formed in high-density cultures but not in low-density cultures. Endogenous interleukin-6 (IL-6) production was found to be significantly lower in high-density cultures, which may partly explain the impaired osteoclast formation in high-density cocultures. Addition of IL-6 to the high-density cocultures indeed restored osteoclast formation. There appeared to be no overt difference in IL-6 receptor mRNA expression between high-density and low-density cultures. In conclusion, this paper suggests that mature, highly differentiated osteoblasts are not directly involved in osteoclastogenesis. In contrast, osteoblast-like cells lacking mature osteoblast markers induce osteoclast formation. Whether these low-density osteoblast-like cells represent an immature differentiation state or the lining cell phenotype is unclear. Received: 26 June 1997 / Accepted: 14 November 1997     

Measurement of Intact Rat Osteocalcin in Osteoblast (ROS17/2.8) Cells and in Ovariectomized Rats with a Sandwich Enzyme Immunoassay

We developed a sandwich enzyme immunoassay system for intact rat osteocalcin to improve the region specificity for the detection of this molecule. We synthesized two peptides of N-terminal 20 residues and C-terminal 10 residues of rat osteocalcin. After conjugating these peptides with carrier protein, we obtained anti-N- and anti-C-terminal rat osteocalcin antibodies in rabbits raised against these two peptides, respectively. By using these antibodies, we measured intact rat osteocalcin levels in a two-site immunoassay manner. These antibodies did not show the cross-reactivity to human osteocalcin. The immunoreactive peak corresponding to the intact molecules was detected by our intact osteocalcin method after high-performance liquid chromatographic fractionation of osteocalcin fragments in plasma from uremic rats. Furthermore, the intact rat osteocalcin was stable over 8 hours at 25°C. Intact rat osteocalcin levels extracellularly secreted from ROS 17/2.8 cells were measured by this method, showing time- and dose-dependent significant increases when administered 1,25(OH)₂D₃. The inhibition for the secretion of intact osteocalcin by actinomycin D was also detected quantitatively with this method. In ovariectomized rats, intact osteocalcin levels in plasma were acutely elevated after ovariectomy, and its elevation was significantly depressed by 17β-estradiol administration. These data suggest that this sandwich method is able to measure the intact form of osteocalcin secreted by osteoblasts. As the antibodies identify the specific regions of osteocalcin molecule, this method would be useful for sensitive estimation of bone turnover for various experimental conditions in rats. Received: 12 August 1995 / Accepted: 28 November 1995     

Parathyroid Hormone,Prostaglandin E2, and 1,25-Dihydroxyvitamin D3 Decrease the Level of Na+-Ca2+ Exchange Protein in Osteoblastic Cells

We previously described Na⁺-Ca²⁺ exchange in osteoblastic rat osteosarcoma cells (UMR-106) and demonstrated that Na⁺-dependent Ca²⁺ transport was inhibited by 24-hour treatment of cells with parathyroid hormone (PTH), prostaglandin E₂ (PGE₂), or 1,25(OH)₂D₃. To determine whether this inhibition of Na⁺-Ca²⁺ exchange is at the level of exchanger protein synthesis we have examined exchanger protein levels using immunoblot analysis. UMR-106 cells were treated for 24 hours with or without PTH, PGE₂, or 1,25(OH)₂D₃. Plasma membrane fractions (7500 g) were obtained and proteins were separated by SDS-PAGE, transferred to nylon membranes, and immunoblotted with a polyclonal antibody to the canine cardiac Na⁺-Ca²⁺ exchanger. In rat cardiac membranes, we detected 125 and 75 kD bands, similar to findings for the canine exchanger. In the osteoblastic UMR cell membranes, a specific band was detected at 90 kD that decreased 65% after treatment of cells with PTH. Inhibition by PTH was dose dependent, was maximal with 10⁻⁷ M PTH, and required 16–24 hour treatment time. Similar inhibition was observed after a 24 hour treatment with 10⁻⁶ M PGE₂ or 10⁻⁸ M 1,25(OH)₂D₃. These results demonstrate the presence of a specific protein in UMR cells that cross-reacts with antibody directed against the cardiac Na⁺-Ca²⁺ exchanger. Thus, the previously reported inhibition of Na⁺-Ca²⁺ exchange activity by calcemic agents in osteoblasts appears to be due to regulation of exchanger protein levels in these osteoblastic cells. Received: 5 February 1996 / Accepted: 18 October 1996     

Influence of Skeletal Site of Origin and Donor Age on Osteoblastic Cell Growth and Differentiation

Bone loss with aging may be due, at least in part, to inadequate bone formation. Moreover, the process of bone aging is known to follow a different pattern throughout the skeleton. In this study, we examined the cell proliferation rate (area under the cell growth curve, AUC) and the secretion of C-terminal type I procollagen (PICP), alkaline phosphatase (ALP), and osteocalcin (OC) in primary cultures of osteoblastic cells from human trabecular bone. Osteoblastic cells were obtained for 168 donors (100 women and 68 men). Ninety-eight bone samples were obtained from subjects undergoing knee arthroplastia, 52 aged 50–70 years (64 ± 5) and 46 over age 70 (73 ± 2). Another 70 bone samples were obtained from subjects undergoing hip arthroplastia; 51 were 50–70 years old (64 ± 4) and 19 were over 70 (75 ± 5). Osteoblastic cells from the older donors had a lower proliferation rate and OC secretion than those from younger subjects. However, ALP secretion was higher in the former subjects, whereas PICP secretion was unchanged. Osteoblastic cells from hip had a lower proliferation rate than those from knee. PICP secretion was also lower and ALP secretion was higher in the former cells. In age-matched cell cultures, osteoblastic cells from the knee had higher proliferation rate and PICP secretion than osteoblastic cells from the hip. However, ALP secretion was lower in knee osteoblastic cells than those from hip only in the younger group. With aging, ALP secretion was found to increase in knee osteoblactic cells, whereas OC secretion decreased in osteoblastic cell cultures from the hip. Our findings suggest that bone loss with aging may be accounted for, at least in part, by a decreased osteoblastic cell proliferation and an increased osteoblastic maturation. In addition, our data indicate that these changes with aging do not occur similarly at different skeletal sites. Received: 25 March 1998 / Accepted: 1 October 1998     

Collagenase and Tissue Plasminogen Activator Production in Developing Rat Calvariae: Normal Progression Despite Fetal Exposure to Microgravity

Exposure to zero gravity has been shown to cause a decrease in bone formation. This implicates osteoblasts as the gravity-sensing cell in bone. Osteoblasts also are known to produce neutral proteinases, including collagenase and tissue plasminogen activator (tPA), which are thought to be important in bone development and remodeling. The present study investigated the effects of zero gravity on development of calvariae and their expression of collagenase and tPA. After in utero exposure to zero gravity for 9 days on the NASA STS-70 space shuttle mission, the calvariae of rat pups were examined by immunohistochemistry for the presence and location of these two proteinases. The ages of the pups were from gestational day 20 (G20) to postnatal (PN) day 35. Both collagenase and tPA were found to be present at all ages examined, with the greatest amount of both proteinases present in the PN14 rats. At later ages, high amounts were maintained for tPA but collagenase decreased substantially between ages PN21 to PN35. The location of collagenase was found to be associated with bone-lining cells, osteoblasts, osteocytes, and in the matrix along cement lines. In contrast, tPA was associated with endothelial cells lining the blood vessels entering bone. The presence and developmental expression of these two proteinases appeared to be unaffected by the exposure to zero gravity. The calvarial thickness of the pups was also examined; again the exposure to zero gravity showed little to no effect on the growth of the calvariae. Notably, from G20 to PN14, calvarial thickness increased dramatically, reaching a plateau after this age. It was apparent that elevated collagenase expression correlated with rapid bone growth in the period from G20 to PN14. To conclude, collagenase and tPA are present during the development of rat calvariae. Despite being produced by the same cell in vitro, i.e., the osteoblast, they are located in distinctly different places in bone in vivo. Their presence, developmental expression, and quantity do not seem to be affected by a brief exposure to zero gravity in utero. Received: 6 May 1997 / Accepted: 27 January 1998     

Ultrastructure and Cytochemical Detection of Alkaline Phosphatase in Long-Term Cultures of Osteoblast-like Cells from Rat Calvaria

Two methods of collecting osteoblast-like cells from newborn rat calvaria were tested, either placing individual glass fragments or tipping dense glass beads onto the endocranial surface of periosteum-free bone. Inoculated at high density, cells collected by using these two methods form large mineralized plates after three weeks of culture. The main purpose of our investigation was to analyze the progressive formation of this mineralized structure and to localize alkaline phosphatase activity. At the beginning of the culture, flattened cells gathered into multilayers and synthesized collagen fibers. Cells in the upper layer became rapidly cuboidal in shape and continued to secrete collagen at their basal pole, whereas other cells became progressively embedded in the extracellular matrix. The upper cells featured ultrastructural characters of osteoblasts, whereas the embedded cells resembled osteocytes. After two weeks, the matrix began to mineralize: crystals appeared on collagen fibers, on matrix vesicles, and on cell debris. During the first days of the culture, the alkaline phosphatase activity was localized on the plasma membranes and on the collagen fibers. Thereafter, only the upper cells and collagen fibers that were juxtaposed to these cells showed alkaline phosphatase activity. In addition, the presence of mineralized matrix prevented the reaction product from being visualized on collagen fibers. The ultrastructural analysis reveals large mineralized plates with a structure resembling that of bone in vivo. This culture appears to be an appropriate model to study bone formation and regulation. Received: 30 September 1995 / Accepted: 3 May 1996     

Increase in the Potential of Osteoblasts to Support Bone Resorption by Osteoclasts In Vitro in Response to Roughness of Bone Surface

The development of the potential of osteoblasts to support bone resorption by osteoclasts in response to roughness on bone slices was examined in the co-incubation cell system of immature osteoclasts and osteoblastic cells. The immature osteoclasts, which need alkaline phospatase (ALP)-positive osteoblastic cells for bone resorption, were generated in mouse spleen cultures with 1, 25-dihydroxyvitamin D₃ and prostaglandin E₂. ALP-negative osteoblastic cells from mouse calvaria were incubated on rough surfaced bone slices for 3 days. The number of ALP-positive cells increased greatly on the rough surface, but little on the smooth surface. When immature osteoclasts were added and incubated for 1 more day, the resorption pit number and the total pit areas on the smooth surface were not much different from those before incubation but were approximately four times higher on the rough surface. Received: 21 July 1998 / Accepted: 12 March 1999     

A Novel Human Bone Marrow Stroma-Derived Cell Line TF274 Is Highly Osteogenic In Vitro and In Vivo

A novel, immortalized, human bone marrow stroma-derived cell line TF274 is described which has the ability to form bone both in vitro and in vivo. Under basal conditions these cells expressed alkaline phosphatase (ALP) and type I collagen genes which are characteristic of the osteoblast phenotype. ALP levels were upregulated in the presence of osteotropic agents such as parathyroid hormone (PTH), transforming growth factor beta (TGF-β), and BMP-2. In addition, PTH also increased cAMP levels in these cells. The capacity of these cells to form bone in vitro was evaluated by culturing them in the presence of L-ascorbic acid and β-glycerophosphate. Matrix mineralization in these cultures was assessed by Alizarin Red staining and increased ⁴⁵Ca uptake. Under these conditions mineralized nodule formation was observed in less than 2 weeks. Northern analysis of TF274 cells at various times during the mineralization process indicated a temporal expression of the osteocalcin gene that is typically associated with differentiating osteoblasts. The osteogenic nature of TF274 cells was confirmed in vivo using the severe combined immunodeficient (SCID) mouse model. Antibodies to human leukocyte antigens (HLA), class I antigens, and human OK^a blood group antigen were used to demonstrate that the lesions formed were of human origin. By 21 days, the lesion consisted of a homogeneous focus of ALP-positive cells containing areas of mineralized bone lined with tartarate-resistant acid phosphatase (TRAP) positive osteoclasts. Thus, the TF274 cells exhibit osteogenic potential both in vitro and in vivo. This immortalized cell line represents a consistent source of cells that can be used to study human osteoblast differentiation both in vitro and in vivo. Received: 30 July 1997 / Accepted: 23 January 1998     

The role of glucocorticoids and prostaglandin E2 in the recruitment of bone marrow mesenchymal cells to the osteoblastic lineage: Positive and negative effects

The role of glucocorticoids in bone formation presents a problem because although pharmacological doses in vivo give rise to osteoporosis, physiological concentrations are required for osteoblast (OB) differentiation in vitro. To try and rationalize this dichotomy, we investigated the effect of dexamethasone on the recruitment of OB precursors present in bone marrow. Using the CFU-f assay, we can measure (1) total colony formation; (2) the osteoblastic differentiation of the colonies defined as their ability to express alkaline phosphatase, synthesize collagen, and to calcify; and (3) colony expansion as either average colony surface area or average colony number. In control cultures and in the presence of 10⁻¹⁰–10⁻⁹ M dexamethasone, colony formation and total cell number was maximal, but the addition of PGE₂ had no effect on colony number and very few colonies expressed the OB phenotype. In the presence of 10⁻⁸–10⁻⁷ M dexamethasone, colony numbers and total cell numbers were reduced but were increased by the addition of PGE₂, the average colony cell number and surface area were relatively unchanged and a proportion of the colonies expressed APase, calcified and synthesized collagen. In cultures containing 10⁻⁶–10⁻⁵ M dexamethasone, colony numbers were further reduced but were stimulated by the addition of PGE₂ and some colonies differentiated; however, colony expansion was dramatically reduced by up to 80%. These results suggest that physiological levels of glucocorticoids are necessary for OB differentiation and allow the control of OB recruitment by PGE₂. High levels of glucocorticoids drastically reduce proliferation of the OB precursors leading to glucocorticoid-induced osteoporosis. Received: 8 May 1995 / Accepted: 23 February 1996