Acinetobacter quorum sensing contributes to inflammation-induced inhibition of orthopaedic implant osseointegration.

Implant infection impairs osseointegration of orthopaedic implants by inducing inflammation. Acinetobacter spp. are increasingly prevalent multi-drug resistant bacteria that can cause osteomyelitis. Acinetobacter spp. can also cause inflammation and thereby inhibit osseointegration in mice. The purpose of the present study was to investigate the role of quorum sensing in this context. Therefore, wild-type bacteria were compared with an isogenic abaI mutant defective in quorum sensing in a murine osseointegration model. The abaI quorum- sensing mutant affected significantly less osseointegration and interleukin (IL) 1ß levels, without detectably altering other pro-inflammatory cytokines. Wild-type bacteria had fewer effects on IL1 receptor (IL1R)-/- mice. These results indicated that quorum sensing in Acinetobacter spp. contributed to IL1ß induction and the resultant inhibition of osseointegration in mice. Moreover, targeting the Gram-negative acyl-homoserine lactone quorum sensing may be particularly effective for patients with Acinetobacter spp. infections.

Characterization of a reproducible model of fracture healing in mice using an open femoral osteotomy.

PURPOSE: The classic fracture model, described by Bonnarens and Einhorn in 1984, enlists a blunt guillotine to generate a closed fracture in a pre-stabilized rodent femur. However, in less experienced hands, this technique yields considerable variability in fracture pattern and requires highly-specialized equipment. This study describes a reproducible and low-cost model of mouse fracture healing using an open femoral osteotomy. METHODS: Femur fractures were produced in skeletally mature male and female mice using an open femoral osteotomy after intramedullary stabilization. Mice were recovered for up to 28 days prior to analysis with microradiographs, histomorphometry, a novel µCT methodology, and biomechanical torsion testing at weekly intervals. RESULTS: Eight mice were excluded due to complications (8/193, 4.1%), including unacceptable fracture pattern (2/193, 1.0%). Microradiographs showed progression of the fracture site to mineralized callus by 14 days and remodelling 28 days after surgery. Histomorphometry from 14 to 28 days revealed decreased cartilage area and maintained bone area. µCT analysis demonstrated a reduction in mineral surface from 14 to 28 days, stable mineral volume, decreased strut number, and increased strut thickness. Torsion testing at 21 days showed that fractured femurs had 61% of the ultimate torque, 63% of the stiffness, and similar twist to failure when compared to unfractured contralateral femurs. CONCLUSIONS: The fracture model described herein, an open femoral osteotomy, demonstrated healing comparable to that reported using closed techniques. This simple model could be used in future research with improved reliability and reduced costs compared to the current options.

Multiple receptor tyrosine kinases promote the in vitro phenotype of metastatic human osteosarcoma cell lines.

The survival rate for osteosarcoma patients with localized disease is 70% and only 25% for patients with metastases. Therefore, novel therapeutic and prognostic tools are needed. In this study, extensive screening and validation strategies identified Axl, EphB2, FGFR2, IGF-1R and Ret as specific receptor tyrosine kinases (RTKs) that are activated and promote the in vitro phenotype of two genetically different metastatic osteosarcoma cell lines. Initial phosphoproteomic screening identified twelve RTKs that were phosphorylated in 143B and/or LM7 metastatic human osteosarcoma cells. A small interfering RNA (siRNA) screen demonstrated that siRNA pools targeting ten of the twelve RTKS inhibited the in vitro phenotype of one or both cell lines. To validate the results, we individually tested the four siRNA duplexes that comprised each of the effective siRNA pools from the initial screen. The pattern of phenotype inhibition replicated the pattern of mRNA knockdown by the individual duplexes for seven of the ten RTKs, indicating the effects are consistent with on-target silencing. Five of those seven RTKs were further validated using independent approaches including neutralizing antibodies (IGF-1R), antisense-mediated knockdown (EphB2, FGFR2, and Ret) or small molecule inhibitors (Axl), indicating that those specific RTKs promote the in vitro behavior of metastatic osteosarcoma cell lines and are potential therapeutic targets for osteosarcoma. Immunohistochemistry demonstrated that Axl is frequently activated in osteosarcoma patient biopsy samples, further supporting our screening and validation methods to identify RTKs that may be valuable targets for novel therapies for osteosarcoma patients.

Adherent endotoxin on orthopedic wear particles stimulates cytokine production and osteoclast differentiation.

Aseptic loosening of orthopedic implants is thought to be caused primarily by osteoclast differentiation induced by bone resorptive cytokines produced in response to phagocytosis of implant-derived wear particles. This study examined whether adherent endotoxin on the wear particles is responsible for inducing osteoclast differentiation as well as production of interleukin-1beta (IL-1beta), IL-6, and tumor necrosis factor a (TNF-alpha). Removal of adherent endotoxin almost completely inhibited the responses to titanium (Ti) particles by both murine marrow cells and human peripheral blood monocytes. In vivo experiments showed that endotoxin removal reduced particle-induced osteolysis by 50-70%. Addition of lipopolysaccharide (LPS) to the "endotoxin-free" particles restored their ability to induce cytokine production and osteoclast differentiation in vitro. Moreover, marrow cells from mice that are hyporesponsive to endotoxin because of mutation of Toll-like receptor 4 induced significantly less cytokine production and osteoclast differentiation in response to Ti particles with adherent endotoxin than did marrow cells from normoresponsive mice. This mutation also resulted in significantly less particle-induced osteolysis in vivo. Taken together, these results show that adherent endotoxin is involved in many of the biological responses induced by orthopedic wear particles and should stimulate development of new approaches designed to reduce the activity of adherent endotoxin in patients with orthopedic implants.

Rapid repair of titanium particle-induced osteolysis is dramatically reduced in aged mice.

Aseptic loosening is the most common cause of orthopaedic implant failure. This process is thought to be due to osteolysis induced by implant-derived wear particles. Teitelbaum and colleagues have recently developed a promising murine calvarial model of wear particle-induced osteolysis. However, prior to this study, this model had only been assessed qualitatively. We now report a reproducible, quantitative version of the calvarial model of wear particle-induced osteolysis, in which the extent of osteolysis (and repair) of entire parietal bones is assessed by histomorphometry of contact microradiographs. Using this model, we found that the osteolytic response is transient and rapidly repaired in one month old mice. The extent of osteolysis peaks 7 days after particle implantation and returns to baseline levels by 13 days. A similar amount of osteolysis and even more extensive repair is observed when particles are implanted repeatedly. In contrast, aged mice develop progressive osteolysis with no detectable repair. As a result, 26 month old mice have approximately 17-fold more osteolysis than one month old mice 21 days after particle implantation. Skeletally mature, adult mice (4-16 months old) show an intermediate pattern of response. Osteolysis in these mice peaks at 7 days after particle implantation but it is repaired more slowly than in the one month old mice. Taken together, these results underscore the role of an imbalance between bone resorption and bone formation in the development of aseptic loosening and suggest that agents that stimulate bone formation maybe useful in prevention or treatment of aseptic loosening.

Titanium particles stimulate bone resorption by inducing differentiation of murine osteoclasts.

BACKGROUND: Loosening of orthopaedic implants is mediated by cytokines that elicit bone resorption and are produced in response to phagocytosis of implant-derived wear particles. This accelerated bone resorption could be due to increased osteoclastic activity, survival, or differentiation. Although a number of in vitro studies have shown that wear particles increase osteoclastic activity, the increase was less than twofold in all cases. The objective of the current study was to test the hypothesis that wear particles stimulate bone resorption by inducing osteoclast differentiation. METHODS: Conditioned media were prepared from murine marrow cells or human peripheral blood monocytes incubated in the presence or absence of titanium particles. The effects of conditioned media on osteoclast differentiation were examined with use of a recently developed assay in which osteoclast precursors are co-cultured with mesenchymal support cells. RESULTS: The present study showed that titanium particles induced both murine marrow cells and human peripheral blood monocytes to produce factors that stimulated osteoclast differentiation. The mean increase in osteoclast differentiation was 29.3+/-9.4-fold. The stimulation of osteoclast differentiation led to a parallel increase in bone resorption. The amount of stimulation was regulated in a dose-dependent manner by the concentration of both titanium particles and conditioned media. The stimulation of osteoclast differentiation required interactions between the cells and the particles themselves and, therefore, was not due to metal ions, soluble contaminants released from the particles, or submicrometer particles. In contrast, conditioned media from control cells incubated in the absence of titanium particles had no detectable effect on any of the examined parameters. CONCLUSIONS: The present study showed that titanium particles stimulate in vitro bone resorption primarily by inducing osteoclast differentiation. In contrast, the titanium particles had only small effects on osteoclast activity or survival.

Regulation of osteoclast activity.

Osteoclasts are the primary cell type responsible for bone resorption. This paper reviews many of the known regulators of osteoclast activity, including hormones, cytokines, ions, and arachidonic acid metabolites. Most of the hormones and cytokines that inhibit osteoclast activity act directly on the osteoclasts. In contrast, most of the hormones and cytokines that stimulate osteoclast activity act indirectly through osteoblasts. Particularly interesting in this regard are agents that directly inhibit activity of highly purified osteoclasts yet stimulate activity of osteoclasts that are co-cultured with osteoblasts. Recent studies have demonstrated that the primary mechanism by which bone resorptive agents stimulate osteoclast activity indirectly is likely to be up-regulation of production of osteoclast differentiation factor/osteoprotegerin ligand (ODF/OPGL) by the osteoblasts. In addition to discussing regulators of osteoclast activity per se, this paper also reviews the role of osteoclast apoptosis to limit the extent of bone resorption.

Bone marrow cells produce soluble factors that inhibit osteoclast activity.

Cytokines that stimulate bone resorption are produced by cells found in bone marrow. However, marrow cells produce multiple factors, some of which may be inhibitors of osteoclast differentiation or activity. Thus, it is not possible to predict a priori whether the mixture of factors produced by marrow cells will have a net stimulatory or inhibitory effect on bone resorption. In this study, we showed that the net effect of whole marrow is to inhibit osteoclast activity induced by parathyroid hormone. Fractionation of the marrow revealed that the inhibitory activity was in the marrow fluid. However, conditioned media obtained from marrow cell cultures also inhibited osteoclast activity. Thus, it is likely that the inhibitory factors are produced in vivo by cells residing in the marrow. These inhibitory factors may represent a physiological regulatory process that plays an important role in maintaining the balance between bone resorption and formation. Because we have previously shown that interleukin-6 is one of the cytokines that parathyroid hormone induces in osteoblastic cells to stimulate osteoclast activity, one potential mechanism by which the marrow-derived inhibitory factors might act is by preventing this production of interleukin-6. However, we found that the marrow cell-conditioned media do not inhibit the production or activity of interleukin-6. Thus, the inhibitory factors appear to block osteoclast activity through a mechanism that does not involve interleukin-6. Taken together, these results demonstrate the importance of factors that inhibit bone resorption and emphasize that the presence of cytokines that stimulate bone resorption in conditions such as osteoporosis and orthopaedic implant loosening should be interpreted with caution unless evidence exists demonstrating their functional importance.

Measurement and removal of adherent endotoxin from titanium particles and implant surfaces.

Aseptic loosening is thought to be due primarily to osteolysis induced by cytokines and prostaglandins that are produced in response to implant-derived wear particles. Because endotoxin has many of the same effects as have been reported for wear particles, we hypothesized that adherent endotoxin may be responsible for the biological responses induced by wear particles. We demonstrated the presence of significant levels of adherent endotoxin on commonly used preparations of titanium particles as well as on titanium and titanium-alloy implant surfaces. In contrast, supernatants obtained by centrifugation of particle suspensions contained approximately 1% as much endotoxin as did the particles. Therefore, it is erroneous to assume that particles do not contain endotoxin on the basis of data that it cannot be detected in their supernatants or filtrates. These results emphasize the importance of considering the potential role of adherent endotoxin when examining the in vitro effects of wear particles and the in vivo performance of orthopaedic implants. We also developed a protocol that removed more than 99.94% of the adherent endotoxin from the titanium particles without detectably affecting their size or shape. The removal of adherent endotoxin will allow comparison of the biological responses induced by particles with or without adherent endotoxin.

Osteoclast differentiation requires ascorbic acid.

Osteoclast differentiation assays are usually conducted in alpha minimal essential medium (alpha-MEM). We reasoned that determining which components of this media are critical for osteoclast differentiation might provide insight into the mechanisms that regulate osteoclast differentiation. This study demonstrates that ascorbic acid is the crucial component of alpha-MEM that stimulates differentiation of murine osteoclasts in cocultures with murine mesenchymal support cells. Thus, supplementation with ascorbic acid allows osteoclast differentiation to occur in basal MEM media as well as in RPMI-1640 and basal media Eagle (BME) media. The conclusion that osteoclast differentiation is stimulated by ascorbic acid was obtained whether osteoclast differentiation was induced by 1,25-dihydroxyvitamin D3 or parathyroid hormone, whether ST2 or CIMC-2 cells were used as mesenchymal support cells, and whether osteoclast precursors were obtained from spleen or bone marrow. Time course studies revealed that although ascorbic acid only modestly increases the rate at which osteoclast precursors begin to express tartrate-resistant acid phosphatase, it strongly increases the rate at which precursors fuse into mature, multinucleated cells. Moreover, ascorbic acid strongly increases the life span of both osteoclasts and their precursors. The increases in precursor formation, fusion, and life span induced by ascorbic acid are together responsible for the stimulation of osteoclast differentiation by ascorbic acid. Given the known effects of ascorbic acid on differentiation of mesenchymal cells, it may stimulate osteoclast differentiation indirectly by regulating the differentiation state of the mesenchymal cells that support osteoclast differentiation.

In vivo demonstration that parathyroid hormone and parathyroid hormone-related protein stimulate expression by osteoblasts of interleukin-6 and leukemia inhibitory factor.

We have previously reported that parathyroid hormone (PTH) and PTH related protein (PTHrP) stimulate expression of interleukin-6 (IL-6) and leukemia inhibitory factor (LIF) in osteoblasts in vitro. In the current study, we have developed a model of hormone injection into the subcutaneous space overlying mouse parietal bones to demonstrate that similar processes occur in osteoblasts in vivo. Specifically, PTH and PTHrP rapidly and transiently induce expression of the mRNAs encoding IL-6 and LIF. The effects are dose-dependent, with a maximal stimulation of approximately 50-fold for each cytokine. Although PTH and PTHrP activate both adenyl cyclase and phospholipase C-dependent signal transduction pathways, stimulation of IL-6 and LIF depends on adenyl cyclase since it is not reproduced by PTH(3-34), a partial agonist that only activates phospholipase C. These results confirm our previous in vitro studies and support the hypothesis that IL-6 and/or LIF are physiologically important mediators of at least some of the actions of PTH and PTHrP.

Stimulation by parathyroid hormone of interleukin-6 and leukemia inhibitory factor expression in osteoblasts is an immediate-early gene response induced by cAMP signal transduction.

Parathyroid hormone and other agents that stimulate bone resorption function, at least in part, by inducing osteoblasts to secrete cytokines that stimulate osteoclast differentiation and activity. We previously demonstrated that parathyroid hormone induces expression by osteoblasts of interleukin-6 and leukemia inhibitory factor without affecting the 16 other cytokines that were examined. We also showed that stimulation of osteoclast activity by parathyroid hormone is dependent on activation of the cAMP signal transduction pathway and secretion of interleukin-6 by osteoblasts. In the current study, we demonstrate that the rapid and transient stimulation of interleukin-6 and leukemia inhibitory factor is inhibited by actinomycin D and superinduced by protein synthesis inhibitors, the classical characteristics of an immediate-early gene response. Moreover, activation of cAMP signal transduction by parathyroid hormone and parathyroid hormone-related protein is necessary and sufficient to induce both interleukin-6 and leukemia inhibitory factor. In addition, cAMP analogues as well as vasoactive intestinal peptide and isoproterenol, two neuropeptides that stimulate bone resorption by activating cAMP signal transduction in osteoblasts, also induce interleukin-6 and leukemia inhibitory factor in these cells. Taken together with our previous results, this study suggests that interleukin-6 is crucial for stimulation of bone resorption not only by parathyroid hormone, but also by parathyroid hormone-related protein, vasoactive intestinal peptide, and beta-adrenergic agonists, like isoproterenol.

Adenyl cyclase and interleukin 6 are downstream effectors of parathyroid hormone resulting in stimulation of bone resorption.

Parathyroid hormone and other bone resorptive agents function, at least in part, by inducing osteoblasts to secrete cytokines that stimulate both differentiation and resorptive activity of osteoclasts. We previously identified two potentially important cytokines by demonstrating that parathyroid hormone induces expression by osteoblasts of IL-6 and leukemia inhibitory factor without affecting levels of 14 other cytokines. Although parathyroid hormone activates multiple signal transduction pathways, induction of IL-6 and leukemia inhibitory factor is dependent on activation of adenyl cyclase. This study demonstrates that adenyl cyclase is also required for stimulation of osteoclast activity in cultures containing osteoclasts from rat long bones and UMR106-01 rat osteoblast-like osteosarcoma cells. Since the stimulation by parathyroid hormone of both cytokine production and bone resorption depends on the same signal transduction pathway, we hypothesized that IL-6 might be a downstream effector of parathyroid hormone. We found that addition of exogenous IL-6 mimics the ability of parathyroid hormone to stimulate bone resorption. More importantly, an antibody directed against the IL-6 receptor blocks moderate stimulation of osteoclast activity induced by the hormone. Interestingly, strong stimulation of resorption overcomes this dependence on IL-6. Thus, parathyroid hormone likely induces multiple, redundant cytokines that can overcome the IL-6 requirement associated with moderate stimulation. Taken together with studies showing that many other bone resorptive agents also stimulate IL-6 production, our results suggest that IL-6 may be a downstream effector of these agents as well as of parathyroid hormone.

Regulation of cytokine expression in osteoblasts by parathyroid hormone: rapid stimulation of interleukin-6 and leukemia inhibitory factor mRNA.

PTH and other hormones that stimulate resorption affect osteoclasts indirectly by modulating cytokine production by osteoblasts. However, the identity and role of the osteoblast-derived cytokines involved in this process are unclear. To examine which cytokines are regulated by PTH, we assessed cytokine mRNA levels in osteoblasts using the reverse transcription-polymerase chain reaction technique. Of the 16 cytokines we examined, unstimulated MC3T3-E1 osteoblastic cells expressed mRNA for interleukins 5, 6, and 7, macrophage and granulocyte-macrophage colony-stimulating factors, transforming growth factor beta 1, and leukemia inhibitory factor. PTH specifically increased expression of interleukin-6 (approximately 50-fold) and leukemia inhibitory factor (approximately 10-fold). Levels of both IL-6 and LIF mRNA peaked 30-60 minutes after addition of PTH and returned to baseline by 4-6 h. This rapid and transient mRNA response, which resembles that of immediate early genes, was also observed in primary rat osteoblasts. The transient mRNA response was accompanied by increased secretion of IL-6 protein. Lipopolysaccharide, another stimulator of resorption, increased mRNA levels of a group of cytokines that were not induced by PTH, namely interleukin-1 alpha, tumor necrosis factor alpha, and granulocyte-macrophage and granulocyte colony-stimulating factors. We conclude that osteoblasts produce complex networks of cytokines that (1) are regulated by bone-resorptive agents and (2) may be involved in controlling bone resorption.

Binding of osteopontin to the osteoclast integrin alpha v beta 3.

Occupancy of the chicken osteoclast alpha v beta 3 integrin stimulates immediate cell signals. Peptides from osteopontin containing Arg-Gly-Asp and peptides from the osteopontin and bone sialoprotein sequences containing Arg-Gly-Asp stimulated immediate reductions in osteoclast cytosolic Ca2+. The changes in cytosolic Ca2+ required the Arg-Gly-Asp sequence, and were blocked by LM609, a monoclonal antibody to the alpha v beta 3 integrin. Osteoclast stimulation by the proteins through the integrin did not require immobilization since soluble peptides produced changes in cytosolic Ca2+ and inhibited osteoclast binding to bone particles and bone resorption. The decrease in cytosolic Ca2+ stimulated by osteopontin and related peptides was due to activation of a plasma membrane Ca(2+)-ATPase. Thus, the data suggest that ligand binding to the osteoclast alpha v beta 3 integrin results in a reduction in cytosolic Ca2+ which participates in regulation of osteoclast function.

Avian osteoblast conditioned media stimulate bone resorption by targeting multinucleating osteoclast precursors.

Osteoblasts are thought to secrete factors that regulate the rate of osteoclastic bone resorption. We studied the effect of osteoblast conditioned medium on bone degradation by multinucleated osteoclast-like cells generated in vitro from mononuclear precursors and found that the medium stimulates bone degradation primarily through interactions with osteoclast precursors. The conditioned medium also stimulates expression of the osteoclast-specific antigen 121F. The increased bone degradation, but not increased 121F expression, is due to the conditioned medium maintaining activity of the osteoclast precursors. Although the osteoclast precursors exhibit the DNA fragmentation characteristic of apoptosis, the osteoblast conditioned medium does not prevent such fragmentation. Chicken macrophage growth factor neither mimics nor augments the ability of the conditioned medium to stimulate bone degradation. Studies of osteoclast generation or function should carefully consider whether the effects are dependent on the viability of the resorbing cells.

Osteoclast precursors circulate in avian blood.

The osteoclast is known to be derived from a marrow-residing precursor that is a member of the mononuclear phagocyte family, but the means by which this cell moves from marrow to bone is unknown. We herein demonstrate that mononuclear progenitors capable of differentiating, in vitro, into cells exhibiting the osteoclast phenotype circulate in chickens. The mononuclear fraction was isolated on a density gradient from blood drawn from calcium-deprived laying hens and the plastic-adherent population was obtained. These cells are members of the mononuclear phagocyte family, as demonstrated by nonspecific esterase and tartrate-resistant acid phosphatase (TRAP) activities, expression of the macrophage-specific mannose receptor, and their ability to phagocytose latex particles. When cultured in the presence of devitalized bone, these cells undergo progressive multinucleation and ultimately become essentially indistinguishable from isolated osteoclasts and those generated from bone marrow precursors. Specifically, the blood-derived polykaryons are TRAP-positive, exhibit characteristic ruffled membranes, and express the osteoclast antigens 121F and 23C6. When placed on bone slices, these cells form typical resorptive "pits." Moreover, when cultured with 3H-proline-labeled bone, the blood monocyte-generated osteoclasts mobilize matrix as effectively as those derived from marrow. Thus, osteoclast precursors circulate in the blood of laying hens and can be induced to differentiate in vitro.

Recognition of osteopontin and related peptides by an alpha v beta 3 integrin stimulates immediate cell signals in osteoclasts.

We have investigated the nature of immediate cell signals produced by occupancy of the chicken osteoclast alpha v beta 3 integrin. Synthetic osteopontin and peptides from the osteopontin and bone sialoprotein sequences containing Arg-Gly-Asp stimulated immediate reductions in osteoclast cytosolic Ca2+. The changes in cytosolic Ca2+ required the Arg-Gly-Asp sequence and were blocked by a monoclonal antibody to the alpha v beta 3 integrin, LM609. Osteoclast stimulation by the proteins through the integrin did not require immobilization since soluble peptides produced changes in cytosolic Ca2+ and inhibited osteoclast binding to bone particles and bone resorption. The decrease in cytosolic Ca2+ stimulated by osteopontin and related peptides appeared to be due to activation of a plasma membrane Ca(2+)-ATPase by calmodulin. Thus, the data suggest that ligand binding to the osteoclast alpha v beta 3 integrin results in calmodulin-dependent reduction in cytosolic Ca2+ which participates in regulation of osteoclast function.

Generation of avian cells resembling osteoclasts from mononuclear phagocytes.

Several lines of indirect evidence suggest that a monocyte family precursor gives rise to the osteoclast, although this hypothesis is controversial. Starting with a uniform population of nonspecific esterase positive, tartrate-sensitive, acid phosphatase-producing, mannose receptor-bearing mononuclear cells, prepared from dispersed marrow of calcium-deprived laying hens by cell density separation and selective cellular adherence, we generated multinucleated cells in vitro. When cultured with devitalized bone, these cells show, by electron microscopy, the characteristic osteoclast morphology in that they are mitochondria-rich, multinucleated, and, most importantly, develop characteristic ruffled membranes at the matrix attachment site. Moreover, as documented by scanning electron microscopy, these cells pit bone slices in a manner identical to freshly isolated osteoclasts. In addition, isoenzymes of acid phosphatase from generated osteoclasts, separated by 7.5% polyacrylamide gel electrophoresis at pH 4, are identical to those of mature osteoclasts in migration pattern and tartrate resistance, although the precursor cells from which the osteoclasts are generated produce an entirely different isoenzyme, which is tartrate-sensitive and migrates less rapidly at pH 4. The fused cells also exhibit a cAMP response to prostaglandin E2. Therefore, osteoclast-like cells can be derived by in vitro culture of a marrow-derived monocyte cell population.