Harry C. Blair

Variable effects of tyrosine kinase inhibitors on avian osteoclastic activity and reduction of bone loss in ovariectomized rats

We compared the effects of the tyrosine kinase inhibitor genistein, a naturally occurring isoflavone, to those of tyrphostin A25, tyrphostin A47, and herbimycin on avian osteoclasts in vitro. Inactive analogs daidzein and tyrphostin A1 were used to control for nonspecific effects. None of the tyrosine kinase inhibitors inhibited bone attachment. However, bone resorption was inhibited by genistein and herbimycin with ID50s of 3 μM and 0.1 μM, respectively; tyrphostins and daidzein were inactive at concentrations below 30 μM, where nonspecific effects were noted. Genistein and herbimycin thus inhibit osteoclastic activity via a mechanism independent of cellular attachment, and at doses approximating those inhibiting tyrosine kinase autophosphorylation in vitro; the tyrphostins were inactive at meaningful doses. Because tyrosine kinase inhibitors vary widely in activity spectrum, effects of genistein on cellular metabolic processes were compared to herbimycin. Unlike previously reported osteoclast metabolic inhibitors which achieve a measure of selectivity by concentrating on bone, neither genistein nor herbimycin bound significantly to bone. Osteoclastic protein synthesis, measured as incorporation of 3H‐leucine, was significantly inhibited at 10 μM genistein, a concentration greater than that inhibiting bone degradation, while herbimycin reduced protein synthesis at 10 nM. These data suggested that genistein may reduce osteoclastic activity at pharmacologically attainable levels, and that toxic potential was lower than that of herbimycin. To test this hypothesis in a mammalian system, bone mass was measured in 200 g ovariectomized rats treated with 44 μmol/day genistein, relative to untreated controls. During 30 d of treatment, weights of treated and control group animals were indistinguishable, indicating no toxicity, but femoral weight in the treated group was 12% greater than controls (P < 0.05). Our data indicate that the isoflavone inhibitor genistein suppresses osteoclastic activity in vitro and in vivo at concentrations consistent with its ID50s on tyrosine kinases, with a low potential for toxicity.

Tandem repeat of C/EBP binding sites mediates PPARγ2 gene transcription in glucocorticoid‐induced adipocyte differentiation

Bone marrow stromal stem cells differentiate into many different types of cells including osteoblasts and adipocytes. Long‐term glucocorticoid treatment decreases osteoblastic activity but increases adipocytes. We investigated the mechanism of glucocorticoid‐induced PPARγ2 transcription. Treatment of human bone marrow stromal cells with dexamethasone induced the differentiation of these cells into adipocytes as measured by oil‐red O staining, and Northern blot analysis showed that dexamethasone strongly induced PPARγ2 mRNA expression in cells cultured in adipocyte induction medium. Moreover, the mRNA of C/EBPδ, an adipocyte‐promoting transcription factor, was also induced by dexamethasone in the presence of induction medium. Gel mobility shift assays using purified GST‐C/EBPδ fusion protein showed that C/EBPδ specifically binds to a 40‐base pair DNA element from PPARγ2 promoter, which was found to contain a tandem repeat of C/EBP binding sites. Transfection studies in mouse mesenchymal C3H10T1/2 cells showed that it is the tandem repeat of the C/EBP binding site in PPARγ2 promoter region that regulates dexamethasone‐mediated PPARγ2 gene activation. We conclude that glucocorticoid‐induced adipogenesis from bone marrow stromal cells is mediated through a reaction cascade in which dexamethasone transcriptionally activates C/EBPδ; C/EBPδ then binds to PPARγ2 promoter and transactivates PPARγ2 gene expression. This activated master regulator, in turn, initiates the adipocyte differentiation. J. Cell. Biochem. 76:518–527, 2000.

Regulation of Avian Osteoclastic H+-ATPase and Bone Resorption by Tamoxifen and Calmodulin Antagonists EFFECTS INDEPENDENT OF STEROID RECEPTORS

We used highly purified avian osteoclasts and isolated membranes from osteoclasts to study effects of tamoxifen, 4-hydroxytamoxifen, calmodulin antagonists, estrogen, diethylstilbestrol, and the anti-estrogen ICI 182780 on cellular degradation of 3H-labeled bone in vitro and on membrane HCl transport. Bone resorption was reversibly inhibited by tamoxifen, 4-hydroxytamoxifen, and trifluoperazine with IC50 values of ∼1 μM. Diethylstilbestrol and 17-β-estradiol had no effects on bone resorption at receptor-saturating concentrations, while ICI 182780 inhibited bone resorption at concentrations greater than 1 μM. At these concentrations ICI 182780, like tamoxifen, inhibits calmodulin-stimulated cyclic nucleotide phosphodiesterase activity. Membrane HCl transport, assessed by ATP-dependent acridine orange uptake, was unaffected by 17-β-estradiol and diethylstilbestrol at concentrations up to 10 μM, while ICI 182780 inhibited HCl transport at concentrations greater than 1 μM. In contrast HCl transport was inhibited by tamoxifen, 4-hydroxytamoxifen, and the calmodulin antagonists, trifluoperazine and calmidazolium, with IC50 values of 0.25-1.5 μM. These results suggested the presence of a membrane-associated non-steroid receptor for tamoxifen in osteoclasts. Tamoxifen binding studies demonstrated saturable binding in the osteoclast particulate fraction, but not in the nuclear or cytosolic fractions. Membranes enriched in ruffled border by differential centrifugation following nitrogen cavitation showed binding consistent with one site, Kd ∼1 μM. Our findings indicate that tamoxifen inhibits osteoclastic HCl transport by binding membrane-associated target(s), probably similar or related to calmodulin antagonist targets. Further, effects of estrogens or highly specific anti-estrogens on bone turnover do not support the hypothesis of a direct effect on osteoclasts by these compounds in this species.

Proteinase expression during differentiation of human osteoclasts in vitro.

Osteoclasts are macrophage‐derived polykaryons that degrade bone in an acidic extracellular space. This differentiation includes expression of proteinases and acid transport proteins, cell fusion, and bone attachment, but the sequence of events is unclear. We studied two proteins expressed at high levels only in the osteoclast, cathepsin K, a thiol proteinase, and tartrate‐resistant acid phosphatase (TRAP), and compared this expression with acid transport and bone degradation. Osteoclastic differentiation was studied using human apheresis macrophages cocultured with MG63 osteosarcoma cells, which produce cytokines including RANKL and CSF‐1 that mediate efficient osteoclast formation. Immunoreactive cathepsin K appeared at 3–5 days. Cathepsin K activity was seen on bone substrate but not within cells, and cathepsin K increased severalfold during further differentiation and multinucleation from 7 to 14 days. TRAP also appeared at 3–5 d, independently of cell fusion or bone attachment, and TRAP activity reached much higher levels in osteoclasts attached to bone fragments. Two proteinases that occur in the precursor macrophages, cathepsin B, a thiol proteinase related to cathepsin K, and an unrelated lysosomal aspartate proteinase, cathepsin D, were also studied to determine the specificity of the differentiation events. Cathepsin B occurred at all times, but increased two‐ to threefold in parallel with cathepsin K. Cathepsin D activity did not change with differentiation, and secreted activity was not significant. In situ acid transport measurements showed increased acid accumulation after 7 days either in cells on osteosarcoma matrix or attached to bone, but bone pit activity and maximal acid uptake required 10–14 days. We conclude that TRAP and thiol proteinase expression begin at essentially the same time, and precede cell fusion and bone attachment. However, major increases in acid secretion and proteinases expression continue during cell fusion and bone attachment from 7 to 14 days. J. Cell. Biochem. 78:627–637, 2000.

Nitric oxide regulation of cGMP production in osteoclasts

Bone resorption by osteoclasts is modified by agents that affect cyclic guanosine monophosphate (cGMP), but their relative physiological roles, and what components of the process are present in osteoclasts or require accessory cells such as osteoblasts, are unclear. We studied cGMP regulation in avian osteoclasts, and in particular the roles of nitric oxide and natriuretic peptides, to clarify the mechanisms involved. C‐type natriuretic peptide drives a membrane guanylate cyclase, and increased cGMP production in mixed bone cells. However, C‐type natriuretic peptide did not increase cGMP in purified osteoclasts. By contrast, osteoclasts did produce cGMP in response to nitric oxide (NO) generators, sodium nitroprusside or 1‐hydroxy‐2‐oxo‐3,3‐bis(3‐aminoethyl)‐1‐triazene. These findings indicate that C‐type natriuretic peptide and NO modulate cGMP in different types of bone cells. The activity of the osteoclast centers on HCl secretion that dissolves bone mineral, and both NO generators and hydrolysis‐resistant cGMP analogues reduced bone degradation, while cGMP antagonists increased activity. NO synthase agonists did not affect activity, arguing against autocrine NO production. Osteoclasts express NO‐activated guanylate cyclase and cGMP‐dependent protein kinase (G‐kinase). G‐kinase reduced membrane HCl transport activity in a concentration‐dependent manner, and phosphorylated a 60‐kD osteoclast membrane protein, which immunoprecipitation showed is not an H+‐ATPase subunit. We conclude that cGMP is a negative regulator of osteoclast activity. cGMP is produced in response to NO made by other cells, but not in response to C‐type natriuretic peptide. G‐kinase modulates osteoclast membrane HCl transport via intermediate protein(s) and may mediate cGMP effects in osteoclasts. J. Cell. Biochem. 73:478–487, 1999.

Tyrosine kinase inhibitor effects on avian osteoclastic acid transport.

We found that tyrosine kinase pp60(c-src) coisolates with acid-transporting osteoclast membranes and hypothesized that this kinase regulates hydrochloric acid transport. We assayed the membrane acid transport and bone degradation effects of tyrosine kinase inhibitors in avian osteoclasts. Isoflavone, tyrphostin, and benzoquinonoid inhibitors were compared with inactive analogues to determine nonspecific effects. Acid-secreting membranes, isolated by nitrogen cavitation, were assayed as reconstituted vesicles by using acridine orange to indicate ATP-dependent hydrogen ion transport. The soy isoflavone genistein and the benzoquinonoid antibiotic herbimycin inhibited hydrochloric acid transport with 50% inhibition at approximately 10 and approximately 2 micromol/L, respectively; effects appeared in <2 min and were reversible. In membrane incubated with inhibitors, the herbimycin effect also inhibited Cl- transport by variable amounts, suggesting that this compound affects Cl- channel activity. However, genistein and tyrphostins did not produce chloride dependent effects. After 30 min with ATP, tyrphostin A47 irreversibly inhibited hydrochloric acid transport with 50% inhibition at approximately 10 micromol/L. Tyrphostin A25 and controls, tyrphostin A1 and daidzein (a genistein congener), were inactive despite preincubation. Osteoclastic bone resorption was more sensitive to the inhibitors over 3-5-d assays than was membrane acid transport, except for tyrphostins. Herbimycin and genistein inhibited bone resorption with half maximal effects at 0.5 and 10 micromol/L and complete inhibition at 3 d in 1 and 20 micromol/L, respectively. None of the tyrphostins, including A47, nor daidzein inhibited resorption to >20 micromol/L. We conclude that tyrosine kinase inhibition directly inhibits osteoclast membrane hydrochloric acid transport; differences among inhibitors may reflect chemical reactivity and permeability.

Prostaglandin E1-Induced Hyperostosis: Clinicopathologic Correlations and Possible Pathogenetic Mechanisms

Prostaglandin E1 (PGE1) causes skeletal hypertrophy, a phenomenon noted when it is administered for several weeks to maintain ductus arteriosus patency in neonates with congenital heart disease. This effect, a dose-dependent and reversible hyperostosis, was described radiologically as bone within bone, but skeletal histopathology was not studied. We compared postmortem gross, radiological, and histological bone findings for untreated controls and term gestation infants after 4, 27, and 56 days of continuous 0.1-0.2 microgram/kg/min PGE1. Bone was not significantly different from controls after 4 days of PGE1. Radiographs were negative after 27 days, but femoral cortex showed early periosteal osteoblast proliferation. At 56 days of PGE1, there was severe, radiologically apparent neocortex formation in tubular, rib, and scapular bones. Corresponding sections of femoral shaft revealed distinctive histopathology with thickened periosteum and fibrocartilage-like tissue covering an exuberant neocortex of closely aligned, gracile, woven bone trabeculae. Paratrabecular stroma contained ectatic capillaries orthogonally oriented to the periosteum, suggesting that a vascular reaction to PGE1 is important in the observed effect. The native cortex was partially resorbed; because it is stress shielded by the neocortex and no inflammation was present, this was interpreted as a secondary effect. We conclude that PGE1-associated paracortical bone hypertrophy is distinct from inflammatory processes and that its early stages may not be apparent radiologically. Moreover, the time course of PGE1-induced osteoblast proliferation and mineralization suggests that experimental use for 4-8 weeks may benefit conditions such as ununited fractures.

Differential effects of tamoxifen-like compounds on osteoclastic bone degradation, H+-ATPase activity, calmodulin-dependent cyclic nucleotide phosphodiesterase activity, and calmodulin binding

We studied effects of calmodulin antagonists on osteoclastic activity and calmodulin‐dependent HCl transport. The results were compared to effects on the calmodulin‐dependent phosphodiesterase and antagonist‐calmodulin binding affinity. Avian osteoclast degradation of labeled bone was inhibited ∼40% by trifluoperazine or tamoxifen with half‐maximal effects at 1–3 μM. Four benzopyrans structurally resembling tamoxifen were compared: d‐centchroman inhibited resorption 30%, with half‐maximal effect at ∼100 nM, cischroman and CDRI 85/287 gave 15–20% inhibition, and l‐centchroman was ineffective. No benzopyran inhibited cell attachment or protein synthesis below 10 μM. However, ATP‐dependent membrane vesicle acridine transport showed that H+‐ATPase activity was abolished by all compounds with 50% effects at 0.25–1 μM. All compounds also inhibited calmodulin‐dependent cyclic nucleotide phosphodiesterase at micromolar calcium. Relative potency varied with assay type, but d‐ and l‐centchroman, surprisingly, inhibited both H+‐ATPase and phosphodiesterase activity at similar concentrations. However, d‐ and l‐centchroman effects in either assay diverged at nanomolar calcium. Of benzopyrans tested, only the d‐centchroman effects were calcium‐dependent. Interaction of compounds with calmodulin at similar concentrations were confirmed by displacement of labeled calmodulin from immobilized trifluoperazine. Thus, the compounds tested all interact with calmodulin directly to varying degrees, and the observed osteoclast inhibition is consistent with calmodulin‐mediated effects. However, calmodulin antagonist activity varies between specific reactions, and free calcium regulates specificity of some interactions. Effects on whole cells probably also reflect other properties, including transport into cells. J. Cell. Biochem. 66:358–369, 1997.

Avian cathepsin B cDNA: sequence and demonstration that mRNAs of two sizes are produced in cell types producing large quantities of the enzyme.

Overlapping cDNA fragments encoding avian cathepsin B were cloned from an osteoclast cDNA library and sequenced. The primary structure of the prepro enzyme deduced from this sequence has 340 amino acids. The mature portion of the enzyme is 80% identical with murine cathepsin B; regions found in other papain superfamily enzymes are conserved. In osteoclasts and cultured macrophages, which produce large quantities of cathepsin B, mRNAs of 1.8 and 2.4 kb are produced in approximately equal quantities, while cells producing smaller quantities of the enzyme produce predominantly the 2.4 kb form. This variation in mRNAs suggests transcriptional differences related to production of large quantities of the enzyme.

Effects of cell culture time and bone matrix exposure on calmodulin content and ATP dependent cell membrane acid transport in avian osteoclasts and macrophages

Osteoclasts mediate bone resorption by secretion at the site of bone attachment. This process depends on calmodulin concentrated at a specialized acid‐secreting membrane. We hypothesized that increased calmodulin and bone attachment were required for acid secretion. We tested this by studying calmodulin, bone attachment, and membrane acid transport in osteoclasts and their precursor mononuclear cells. Osteoclasts and macrophages were isolated from medullary bone of hens; cell fractions were prepared after culturing cells with or without bone. Calmodulin was visualized by Western analysis; calmodulin mRNA was determined by Northern hybridization, and ATP‐dependent membrane acid transport was assayed by acridine orange uptake. Calmodulin decreased in osteoclasts cultured without bone. Calmodulin in isolated macrophages was ∼︁25% of osteoclast levels, but increased several fold by 5 days. Bone had no effect. Calmodulin mRNA was similar in osteoclasts with or without bone. However, only osteoclasts cultured with bone retained acid transport capacity. Macrophage calmodulin mRNA was not affected by bone, but increased three fold by day 5, paralleling protein production. Macrophages developed acid transport capacity at 3–5 days, but at lower levels than osteoclasts, and bone had no measurable effect. Chicken cells express 1.6 kb and inducible 1.9 kb calmodulin transcripts; in macrophages and osteoclasts, the 1.9 kb transcript predominated. We conclude that, following isolation, calmodulin levels decline in osteoclasts via a post‐transcriptional mechanism. In cultured macrophages, by contrast, calmodulin mRNA, protein, and acid secretion increase with time independently of bone substrate, possibly reflecting differentiation in vitro. Increased calmodulin correlated with membrane acid transport capacity in both cell types. The macrophage findings indicate that stimuli other than bone influence acid transport capacity in this family of cells.