The role of geranylgeranylation in bone resorption and its suppression by bisphosphonates in fetal bone explants in vitro: A clue to the mechanism of action of nitrogen-containing bisphosphonates

Catabolic and anabolic signals in bone: therapeutic implications

Bone remodelling accounts for the continuous renewal of the adult skeleton, and its disturbances constitute the pathophysiological basis of most skeletal disorders. Recently identified bone signalling pathways explain the action of known therapeutic agents and provide targets for new developments in the management of bone diseases.

Effect of interleukin-17 on nitric oxide production and osteoclastic bone resorption: is there dependency on nuclear factor-kappaB and receptor activator of nuclear factor kappaB (RANK)/RANK ligand signaling?

Interleukin-17 (IL-17) is a proinflammatory cytokine produced exclusively by activated memory T cells and has recently been found to stimulate osteoclastic resorption. Like other proinflammatory cytokines, IL-17 may affect osteoclastic bone resorption indirectly via osteoblasts, possibly by mechanisms previously reported for chondrocytes that respond in very similarly to osteoblasts. As in chondrocytes, but only in combination with tumor necrosis factor-alpha (TNF-alpha), IL-17 induced nitric oxide (NO) production in osteoblastic cells and fetal mouse metatarsals by a nuclear factor-kappaB (NF-kappaB)-dependent mechanism. This effect was associated with elevated mRNA levels of the NF-kappaB isoforms RelA and p50. In fetal mouse metatarsals, IL-17 stimulated osteoclastic bone resorption only in combination with TNF-alpha. The pathway by which the cytokine combination exerts this effect was examined using inhibitors of NO synthesis and NF-kappaB activation. Although both inhibitors used abolished NO production, they did not prevent the stimulatory effect of the cytokine combination on osteoclastic resorption. In contrast, the inhibitors slightly increased osteoclastic resorption, suggesting a suppressive rather than stimulatory effect of NO on cytokine-induced bone resorption. In addition, we showed that IL-17 + TNF-alpha stimulated osteoclastic resorption independent of NF-kappaB signaling. To further examine the pathway by which osteoclastic resorption was stimulated, we used osteoprotegerin, a specific inhibitor of the receptor activator of NF-kappaB (RANK)/receptor activator of the NF-kappaB ligand (RANKL) pathway. Osteoprotegerin partially inhibited IL-17 + TNF-alpha-stimulated osteoclastic resorption only at the high concentration of 1000 ng/mL, whereas it completely blocked parathyroid hormone-related peptide-stimulated resorption at 300 ng/mL. In conclusion, IL-17 stimulated NO production by an NF-kappaB-dependent pathway in osteoblastic cells and fetal mouse metatarsals only in combination with TNF-alpha. Neither NO production nor NF-kappaB signaling, and only partly the RANK/RANKL pathway, were involved in the stimulatory effect of the cytokine combination on osteoblastic bone resorption in these long bones, suggesting the existence of other pathways by which osteoclastic resorption can be stimulated.

New advances in the biology and treatment of myeloma bone disease

The bisphosphonates provide effective therapy for the skeletal complications of multiple myeloma (MM). Although the earliest bisphosphonates had poor bioavailability and relatively low potency, newer compounds such as pamidronate and zoledronic acid have greater potency. Bisphosphonates block the development of monocytes into osteoclasts and are thought to promote apoptosis of osteoclasts. These agents prevent osteoclasts from moving to the bone surface and seem to inhibit the production of bone-resorbing cytokines such as interleukin-6 (IL-6) by bone marrow stromal cells. In addition, bisphosphonates seem to have a direct antimyeloma effect by inducing apoptosis of malignant plasma cells. The beneficial effects of pamidronate have been demonstrated in a clinical trial setting. Patients who failed to respond to chemotherapy had a slight prolongation of survival and better performance status and quality of life. Ongoing clinical trials with ibandronate and zoledronic acid indicate the latter is 100 to 1,000 times more potent than pamidronate. Biochemical effects of zoledronic acid continue for as long as 8 weeks after a single administration. In a new trial comparing pamidronate and zoledronic acid, 90% of the patients who received zoledronic acid were normocalcemic, compared with 69% of those who received pamidronate at 10 days. In addition, the time to relapse or development of hypercalcemia was shorter for patients receiving pamidronate compared with zoledronic acid.

Nitrogen-bisphosphonates block retinoblastoma phosphorylation and cell growth by inhibiting the cholesterol biosynthetic pathway in a keratinocyte model for esophageal irritation

The surprising discovery that nitrogen-containing bisphosphonates (N-BPs) act via inhibition of the mevalonate-to-cholesterol pathway raised the possibility that esophageal irritation by N-BPs is mechanism-based. We used normal human epidermal keratinocytes (NHEKs) to model N-BP effects on stratified squamous epithelium of the esophagus. The N-BPs alendronate and risedronate inhibited NHEK growth in a dose-dependent manner without inducing apoptosis. N-BPs (30 microM) caused accumulation of cells in S phase and increased binucleation (inhibited cytokinesis). Consistent with N-BP inhibition of isoprenylation, geranylgeraniol or farnesol prevented accumulation in S phase. Binucleation was also induced by the 3-hydroxy-3-methylglutaryl-coenzyme A reductase inhibitor lovastatin and by the squalene synthase inhibitor zaragozic acid A and was prevented by adding low-density lipoprotein. At 300 microM, N-BPs reduced expression of cyclin-dependent kinase (cdk) 2 and cdk4 and enhanced expression of p21(waf1) and p27(kip1) and their binding to cdks with corollary hypophosphorylation of retinoblastoma. Lovastatin and zaragozic acid A produced similar effects, except that p21(waf1) expression and binding to cdks was not induced. Growth inhibition, but not binucleation, was also caused by the geranylgeranyl transferase I inhibitor, GGTI-298, which also enhanced cdk2 and cdk4 association with p27(kip1). These findings are consistent with suppression of epithelial cell growth by N-BPs via inhibition of the mevalonate pathway and the consequent reduction in cholesterol synthesis, which blocks cytokinesis, and in geranylgeranylation, which interferes with progression through the cell cycle.

The palliative management of skeletal metastases in prostate cancer: use of bone-seeking radionuclides and bisphosphonates

In prostate cancer, the development of skeletal metastases is associated with a significant increase in morbidity, mainly because of severe bone pain, which eventually becomes refractory to conventional analgesia. Androgen ablation is the treatment of choice, but the majority of patients relapse within 2 to 3 years from initiation of treatment. After failure of hormone therapy, external-beam irradiation therapy is effective in the palliation of pain, but radionuclides represent an attractive and cost-effective alternative. Strontium 89 is currently the most commonly used radionuclide in the palliative management of prostate cancer metastatic to the skeleton. The rationale for the use of bisphosphonates in metastatic prostate cancer is not immediately obvious, given the predominantly osteoblastic nature of the metastatic process. The clinical use of these agents rests on a number of basic and clinical observations that provide ample evidence that, in prostate cancer, the metastatic process is associated with increased bone resorption. Evidence regarding the beneficial effects of bisphosphonates in reducing morbidity from metastatic prostate cancer is reasonably solid, although the choice of optimal bisphosphonate, mode of administration, dose, and duration of treatment must be determined in large, controlled studies before their widespread clinical use can be advocated. Available therapeutic modalities that use either radionuclides or bisphosphonates can effectively and safely be used in the palliative management of metastatic prostate cancer. Neither radionuclides nor bisphosphonates have been shown to prolong survival, but the potential of both agents to beneficially alter the metastatic process in prostate cancer is intriguing.

Inhibitors of prenylation of Ras and other G-proteins and their application as therapeutics

Anchoring of small G-proteins to cellular membranes via a covalently bound lipophylic prenyl group is essential for the functioning of these proteins. For example, the farnesylation of Ras by the action of the enzyme protein:farnesyl transferase (PFT) is pivotal for its signalling function in cell growth and differentiation. The development of inhibitors of PFT was triggered by the role of mutated Ras in certain types of cancer and by the observation that non-farnesylated Ras is inactive. Besides the screening of existing compounds for PFT inhibition, rational drug design has also led to new inhibitors. Our research is in the field of atherosclerosis and concerns the development of inhibitors of the growth of vascular smooth muscle cells. The latter process gives rise to reocclusion of the coronary artery (restenosis) after balloon angioplasty. We and others have developed several analogues of the two substrates of PFT, i.e. farnesyl pyrophosphate (FPP) and the so-called CAAX peptide consensus sequence, which were tested in vitro for the inhibition of PFT and of other enzymes involved in protein prenylation, such as protein:geranylgeranyl transferase-1 (PGGT-1). The FPP analogue TR006, a strong inhibitor of PFT (IC(50) of 67 nM), blocked the proliferation of cultured human smooth muscle cells and inhibited platelet-derived growth factor- and basic fibroblast growth factor-induced DNA synthesis. Similar but more highly charged compounds failed in this respect, probably because of an impaired uptake in the cells. Less charged derivatives were designed to circumvent this problem. The effect on the GF-induced activation of intermediates in signal transduction pathways was investigated in order to gain insight into the mechanism of action within the cells. TR006 decreased the bFGF activation of extracellular signal-regulated kinase 1 (ERK1), suggesting its involvement in inhibiting Ras activity. Although other analogues inhibited DNA synthesis, they affected neither ERK1 activation nor p38/stress-activated protein kinase 2 or Jun N-terminal kinase 1 activation. Since some of these compounds were also shown to be inhibitors of in vitro PGGT-1 activity, the geranylgeranylation of other G-proteins may be decreased by these compounds. Rho seems to be a good candidate as a target for inhibitors of PGGT-1. This uncertainty as to the mechanism of action within non-transformed as well as transformed cells applies to all prenylation inhibitors, but is not holding back their further development as drugs. Their current and possible future application as therapeutics in cancer, restenosis, angiogenesis, and osteoporosis is briefly discussed.

Involvement of interleukin-1 in the inflammatory actions of aminobisphosphonates in mice

Aminobisphosphonates (aminoBPs) are potent inhibitors of bone resorption. However, they cause undesirable inflammatory reactions, including fever, in humans. Intraperitoneal injection of aminoBPs into mice also induces inflammatory reactions, including a prolonged elevation of the activity of the histamine-forming enzyme, histidine decarboxylase (HDC). Because interleukin-1 (IL-1) is a typical pyrogen and a strong inducer of HDC, we examined whether aminoBPs induce inflammatory reactions in mice deficient in genes for both IL-1alpha and IL-1beta (IL-1-KO mice). In control mice, aminoBPs induced an elevation of HDC activity and other inflammatory reactions (enlargement of the spleen, atrophy of the thymus, exudate in the thorax and increase in granulocytic cells in the peritoneal cavity). These responses were all weak or undetectable in IL-1-KO mice. We have previously shown that lipopolysaccharides (LPSs) from Escherichia coli and Prevotella intermedia (a prevalent gram-negative bacterium both in periodontitis and endodontal infections) are capable of inducing HDC activity in various tissues in mice. In control mice treated with an aminoBP, the LPS-induced elevations of serum IL-1 (alpha and beta) and tissue HDC activity were both markedly augmented. However, such an augmentation of HDC activity was small or undetectable in IL-1-KO mice. These results, taken together with our previous findings (i) suggest that IL-1 is involved in the aminoBP-induced inflammatory reactions and (ii) lead us to think that under some conditions, inflammatory reactions induced by gram-negative bacteria might be augmented in patients treated with an aminoBP. In this study, we also obtained a result suggesting that IL-1-deficiency might be compensated by a second, unidentified, mechanism serving to induce HDC in response to LPS when IL-1 is lacking.

Protein geranylgeranylation is required for osteoclast formation, function, and survival: inhibition by bisphosphonates and GGTI-298

Bisphosphonates are the important class of antiresorptive drugs used in the treatment of metabolic bone diseases. Although their molecular mechanism of action has not been fully elucidated, recent studies have shown that the nitrogen-containing bisphosphonates can inhibit protein prenylation in macrophages in vitro. In this study, we show that the nitrogen-containing bisphosphonates risedronate, zoledronate, ibandronate, alendronate, and pamidronate (but not the non nitrogen-containing bisphosphonates clodronate, etidronate, and tiludronate) prevent the incorporation of [14C]mevalonate into prenylated (farnesylated and geranylgeranylated) proteins in purified rabbit osteoclasts. The inhibitory effect of nitrogen-containing bisphosphonates on bone resorption is likely to result largely from the loss of geranylgeranylated proteins rather than loss of farnesylated proteins in osteoclasts, because concentrations of GGTI-298 (a specific inhibitor of geranylgeranyl transferase I) that inhibited protein geranylgeranylation in purified rabbit osteoclasts prevented osteoclast formation in murine bone marrow cultures, disrupted the osteoclast cytoskeleton, inhibited bone resorption, and induced apoptosis in isolated chick and rabbit osteoclasts in vitro. By contrast, concentrations of FTI-277 (a specific inhibitor of farnesyl transferase) that prevented protein farnesylation in purified rabbit osteoclasts had little effect on osteoclast morphology or apoptosis and did not inhibit bone resorption. These results therefore show the molecular mechanism of action of nitrogen-containing bisphosphonate drugs in osteoclasts and highlight the fundamental importance of geranylgeranylated proteins in osteoclast formation and function.

HMG-CoA reductase inhibitors increase BMD in type 2 diabetes mellitus patients

Recently, it was reported that 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase inhibitors increased bone mineral density (BMD) in mice. We studied the effect of HMG-CoA reductase inhibitors on BMD of type 2 diabetes mellitus by a retrospective review of medical records. Sixty-nine type 2 diabetic patients were included. The control group (n = 33) did not take HMG-CoA reductase inhibitors. The treatment group (n = 36) was administered either lovastatin, pravastatin, or simvastatin. BMD of the spine, femoral neck, femoral trochanter, and total hip were measured by dual-energy X-ray absorptiometry. There were no significant differences between control and treatment groups in age, sex, body mass index, glycemic control, and serum insulin levels. In the control group, BMD of the spine significantly decreased (from 1.116 +/- 0.165 to 1.081 +/- 0.178 g/cm2) after 14 months. In the treatment group, BMD of the femoral neck significantly increased (from 0.853 +/- 0.139 to 0.878 +/- 0.147 g/cm2) after 15 months. In male subjects treated with HMG-CoA reductase inhibitors, there was a significant increase in BMD of the femoral neck and femoral trochanter (from 0.899 +/- 0.139 to 0.934 +/- 0.139 and from 0.801 +/- 0.145 to 0.833 +/- 0.167 g/cm2, respectively), but in female subjects, only BMD of the femoral neck increased (from 0.819 +/- 0.132 to 0.834 +/- 0.143 g/cm2). Percentage increments of BMD of the femoral neck, femoral wards triangle, femoral trochanter, and total hip in the treatment group were significantly higher than in the control group (2.32% vs. -0.99, 1.77% vs. -1.25%, 1.40% vs. -1.21%, 0.88% vs. -1.03%, respectively). The proportion of subjects who had an increase in BMD of the spine and total hip more than two percentages was significantly larger in the treatment group than in the control group (30.6% vs. 15.2% and 30.6% vs. 9.1%, respectively). The increased increment in BMD of the treatment group was significantly greater than those in the control group after adjustment for age and body mass index (P < 0.05). These results suggest that HMG-CoA reductase inhibitors may increase BMD of the femur in male patients with type 2 diabetes mellitus.

Prevention and treatment of osteoporosis: efficacy of combination of hormone replacement therapy with other antiresorptive agents

Osteoporosis is a debilitating disease characterized by decreased bone mineral density (BMD) leading to fractures. It primarily affects postmenopausal women and elderly men. Prevention of osteoporosis is very important because present therapies do not have the potential to mend damage to the bone microarchitecture caused by osteoporosis. The first line of prevention and treatment of osteoporosis is hormone replacement therapy (HRT). All of the approved drugs for the prevention and treatment of osteoporosis act as inhibitors of bone resorption; these drugs include HRT, selective estrogen receptor modulators, calcitonin, and bisphosphonates. The latter two drugs have also been shown to prevent fractures. This article discusses data from nine controlled prospective clinical studies. Study 1 was designed to assess the efficacy of combined HRT and bisphosphonate in preventing osteoporosis during the early stages of menopause. This combined therapy increased the lumbar spine BMD by 10.9% and femoral BMD by 7.3% over 4 yr, compared with 6.8 and 4.0% with HRT alone, and 6.8 and 1.2% with bisphosphonate alone. Study 2 was conducted on postmenopausal women with established osteoporosis. These results showed a 10.4 and 7.0% increase in BMD in vertebrae and femora, respectively, compared with 7.3 and 4.8% increases in the HRT group, and 6.8 and 0.9% in the bisphosphonate group. Data from study 3 demonstrated similar findings in that the combination of alendronate and HRT also enhanced BMD values. Studies 4 and 5 assessed the efficacy of the combined therapy of HRT and calcitonin in the prevention of early postmenopausal bone loss. Both studies demonstrated a significant increase in BMD over and above that observed with either HRT or calcitonin alone. Studies 6, 7, and 8 demonstrated that the addition of testosterone to estrogen therapy further increased BMD when compared to estrogen therapy alone, and also prevented the expected decreases in markers of bone formation in early postmenopausal women. Study 9 demonstrated a synergistic effect on BMD in postmenopausal women, when HRT was coadministered with monofluorophosphate. Other combination therapies may also enhance BMD (e.g., the combination of alendronate and parathyroid hormone [PTH]). However, some agents either lose their efficacy or have no added effects on BMD when they are coadministered (e.g., tiludronate and PTH, calcitonin and PTH, calcitonin and anabolic steroids). These studies illustrate that in a subgroup of patients (i.e., patients with high bone turnover and/or severe osteoporosis), specific combination treatments such as HRT with bis-phosphonates, calcitonin, or androgens (and perhaps also with PTH, fluoride, nitric oxide donors) provide additional beneficial effects over a single-drug therapy. Whether these combination therapies are more effective than individual drugs in reducing fractures still needs to be determined.

Farnesyl pyrophosphate synthase is the molecular target of nitrogen-containing bisphosphonates

Bisphosphonates (Bps), inhibitors of osteoclastic bone resorption, are used in the treatment of skeletal disorders. Recent evidence indicated that farnesyl pyrophosphate (FPP) synthase and/or isopentenyl pyrophosphate (IPP) isomerase is the intracellular target(s) of bisphosphonate action. To examine which enzyme is specifically affected, we determined the effect of different Bps on incorporation of [(14)C]mevalonate (MVA), [(14)C]IPP, and [(14)C]dimethylallyl pyrophosphate (DMAPP) into polyisoprenyl pyrophosphates in a homogenate of bovine brain. HPLC analysis revealed that the three intermediates were incorporated into FPP and geranylgeranyl pyrophosphate (GGPP). In contrast to clodronate, the nitrogen-containing Bps (NBps), alendronate, risedronate, olpadronate, and ibandronate, completely blocked FPP and GGPP formation and induced in incubations with [(14)C]MVA a 3- to 5-fold increase in incorporation of label into IPP and/or DMAPP. Using a method that could distinguish DMAPP from IPP on basis of their difference in stability in acid, we found that none of the NBps affected the conversion of [(14)C]IPP into DMAPP, catalyzed by IPP isomerase, excluding this enzyme as target of NBp action. On the basis of these and our previous findings, we conclude that none of the enzymes up- or downstream of FPP synthase are affected by NBps, and FPP synthase is, therefore, the exclusive molecular target of NBp action.

Nitrogen-containing bisphosphonates as carbocation transition state analogs for isoprenoid biosynthesis

Nitrogen-containing bisphosphonates are potent bone antiresorptive agents as well as having herbicidal and antiparasitic activity, and are thought to act by inhibiting enzymes of the mevalonate pathway. Using molecular modeling and ab initio quantum chemical calculations, we show that bisphosphonates can act as aza-isoprenoid transition state analogs, thereby inhibiting isoprenoid biosynthesis. The two phosphonate groups of the 1,1-bisphosphonates readily dock into the diphosphate-Mg(2+) binding site in farnesyl diphosphate synthase, while the charged ammonium (or pyridinium or imidazolium) groups act as carbocation transition state analogs, whose binding is stabilized by a cluster of oxygen atoms in the active site cleft, and an overall negative electrostatic potential in this region. Enhanced activity is shown to correlate with increasing van der Waals stabilization due to N-alkylation, or the presence of a charged, planar (sp(2)-hybridized) aromatic residue in the carbocation binding site. These results are of general interest since they suggest a rational approach to bisphosphonate drug design.