Oscillating Fluid Flow Inhibits TNF-α-induced NF-κB Activation via an IκB Kinase Pathway in Osteoblast-like UMR106 Cells

Fluid flow plays an important role in load-induced bone remodeling. However, the molecular mechanism of flow-induced signal transduction in osteoblasts remains unclear. In endothelial cells, fluid flow alters activation of NF-κB resulting in changes in expression of cell adhesion molecules. To test the hypothesis that fluid flow alters NF-κB activation and expression of cell adhesion molecules in osteoblastic cells, we examined the effect of oscillating fluid flow (OFF) on tumor necrosis factor (TNF)-α-induced NF-κB activation in rat osteoblast-like UMR106 cells. We found that OFF inhibits NF-κB-DNA binding activities, especially TNF-α-induced p50-p65 heterodimer NF-κB activation and TNF-α-induced intercellular adhesion molecule-1 mRNA expression. The inhibitory effects of OFF on both TNF-α-induced NF-κB activation and intercellular adhesion molecule-1 mRNA expression were shear stress-dependent and also increased with OFF exposure duration, indicating that OFF has potent effects on mechanotransduction pathways. OFF also inhibited TNF-α-induced IκBα degradation and TNF-α-induced IκB kinase (IKK) activity in a shear stress-dependent manner. These results demonstrate that IKK is an initial target molecule for OFF effects on osteoblastic cells. Thus, OFF inhibits TNF-α-induced IKK activation, leading to a decrease in phosphorylation and degradation of inhibitory IκBα, which in turn results in the decrease of TNF-α-induced NF-κB activation and potentially the transcription of target genes.

Skeletal systems are maintained by continuous bone remodeling. Mechanical loading, as well as a number of biochemical factors, regulates this bone remodeling. Mechanotransduction in bone has been proposed to involve a variety of biophysical signals including electrical potentials (streaming potentials and piezoelectric effects) and direct transduction of matrix strain. Recent studies suggest that shear stress is an important biophysical signal in bone cell mechanotransduction (1)(2)(3). Indeed, experiments designed to discriminate between flow and strain effects suggest that fluid flow-induced shear stress is a more potent stimulator of bone cells than substrate deformation (4,5). As bone tissue is loaded in vivo, extracellular fluid in the canalicular network experiences a heterogeneous pressurization in response to the deformation of the mineralized bone matrix, resulting in generation of fluid flow along pressure gradients. When loading is removed, pressure gradients and flows are reversed. These fluid motions are dynamic and oscillatory in nature. Recently, Jacobs et al. (6) demonstrated that oscillating fluid flow (OFF), 1 similar to what a bone cell might experience in vivo, mobilizes cytosolic calcium in osteoblastic cells. This was the first study to examine the effect of OFF on bone cells. Other studies have demonstrated that steady or pulsating fluid flow regulates many biochemical factors such as cytosolic calcium (4,6), cAMP (1), prostaglandin E2 (2), inositol trisphosphate (2), nitric oxide (7), cyclooxygenase-2 mRNA (8,9), and c-Fos (9) in osteoblastic cells. However, the precise mechanism by which bone cells convert biophysical signals, such as fluid flow-induced shear stress, into these biochemical signals remains unclear.
A number of paracrine and autocrine factors have been identified that control bone remodeling. Tumor necrosis factor (TNF)-␣, a cytokine synthesized in the bone microenvironment, has been shown to exert pleiotropic effects on osteoblasts and osteoblast-like cells (10,11). It has also been shown that TNF-␣ increases the production of interleukin-6 and macrophage colony-stimulating factor in osteoblastic cells, thereby indirectly promoting differentiation of osteoclasts and enhancing bone resorption (12,13). In addition, the production of TNF-␣ in pathological conditions such as estrogen deficiency and rheumatoid arthritis has been suggested to result in osteopenia and bone destruction adjacent to areas of inflammation (14 -16).
We previously demonstrated that TNF-␣ induces the expression of intracellular adhesion molecule (ICAM)-1 through transcription factor NF-B activation in osteoblasts, leading to the promotion of bone resorption (17,18). The transcription factor NF-B was first identified as a protein that binds to a specific DNA site in the intrinsic enhancer of the Ig light chain gene. It is composed of homo-or heterodimers of members of the Rel family that control the expression of numerous genes involved in the immune and inflammatory responses, cell adhesion, and growth control. Moreover, NF-B plays a role as a primary regulator of the stress response. NF-B can be rapidly activated by many types of extracellular stimuli, including viral infection, bacterial products, oxidative stress, and physical stress (for reviews, see Refs. 19 -21). Additionally in endothelial cells, fluid flow alters activation of NF-B resulting in changes in expression of cell adhesion molecules (22)(23)(24). Therefore, we hypothesized that fluid flow-induced shear stress alters TNF-␣-induced NF-B activation and expression of cell adhesion molecules in osteoblastic cells. To explore our hypothesis osteoblastic UMR106 cells were exposed to OFF in the presence or absence of TNF-␣. Our results suggest that OFF inhibits TNF-␣-induced activation of NF-B.

EXPERIMENTAL PROCEDURES
Chemicals-All reagents were obtained from Sigma except as otherwise noted.
Fluid Flow Experiments-UMR106 cells were grown on glass slides in minimal essential medium (Life Technologies, Inc., Rockville, MD) supplemented with 10% fetal bovine serum (HyClone, Logan, UT). Nearly confluent cells were incubated in minimal essential medium without fetal bovine serum for 24 h before experiments. OFF was generated as previously described with some modification (6). In brief, cells on glass slides were mounted in a parallel plate flow chamber attached to a custom designed fluid pump via rigid wall tubing. Cells were exposed to OFF, in the absence or presence of 1 ng/ml TNF-␣, for various lengths of time at flow rates of 0, Ϯ4, Ϯ10, and Ϯ20 ml/min at a frequency of 1 Hz. In our flow chamber, these flow rates induce peak shear stresses of 0, 1.9, 4.7, and 9.3 dyne/cm 2 , respectively. Cells were then harvested for nuclear and cytosolic protein extraction or total RNA extraction.
Cell Viability and Adhesion Assay-Cells were subjected to OFF in the absence or presence of 1 ng/ml TNF-␣ at various flow rates for 2 h. After exposure to OFF, cells adhering to glass slides were washed twice with 10 ml of phosphate-buffered saline (PBS) without Ca 2ϩ and Mg 2ϩ (PBS (Ϫ); Life Technologies, Inc.) and were collected by trypsinization. Cell viability was evaluated by a trypan blue dye exclusion test.
Preparations of Nuclear and Cytosolic Protein Extracts-Nuclear and cytosolic protein extracts were prepared as described previously with some modification (17). After being washed twice with 10 ml of PBS (Ϫ), the cells were incubated on ice for 5 min in 1 ml of PBS (Ϫ) containing 2 mM EDTA (pH 8.0), harvested by scraping with a cell scraper, and then pelleted by centrifugation at 12,000 ϫ g for 15 s at 4°C. The cell pellet was resuspended in 100 l of Buffer A (25 g/ml aprotinin, 1 mM dithiothreitol (DTT), 10 mM HEPES-KOH (pH 7.9), 10 mM KCl, 10 g of leupeptin, 1.5 mM MgCl 2 , 100 nM pepstatin, 1 mM phenylmethylsuflonyl fluoride, 50 mM NaF, 0.5 mM Na 3 VO 4 , 1 mM sodium pyrophosphate, 5 g/ml N-tosyl-L-phenylalanine chromethyl ketone, and 0.4% Nonidet P-40 (Fluka, Milwaukee, WI)) and lysed by incubating on ice for 10 min and then centrifuged at 12,000 ϫ g for 5 min. The supernatant was used as cytosolic protein extract. The pellet was washed with PBS (Ϫ), resuspended in 100 l of Buffer C (1 mM DTT, 2 mM EDTA (pH 8.0), 20% glycerol, 20 mM HEPES-KOH (pH 7.9), 0.4 M KCl, 100 nM pepstatin, and 1 mM phenylmethylsuflonyl fluoride), and lysed by freezing and thawing. After centrifugation at 12,000 ϫ g for 5 min at 4°C, the supernatant was used as nuclear protein extract. Protein concentration was determined by a microassay kit (Bio-Rad, Hercules, CA) using bovine serum albumin as a standard. The nuclear and cytosolic protein extracts were aliquoted and stored at Ϫ80°C until analysis.
Electrophoretic Mobility Shift Assay-Nuclear extracts (20 g of protein) were used for electrophoretic mobility shift assay. The NF-B consensus oligonucleotide was obtained from Promega (Madison, WI). The NF-B oligonucleotide was labeled by T4 polynucleotide kinase in the presence of 20 Ci of [␥-32 P]ATP (Amersham Pharmacia Biotech) and used as a probe. To identify the NF-B subunits, supershift analysis was performed using antibodies directed against p50 and p65 (Santa Cruz Biotechnology, Santa Cruz, CA). The antibodies were added to the binding reaction mixture (40 mM HEPES-KOH (pH 7.9), 75 mM KCl, 0.5 M EDTA (pH 8.0), 0.5 mM DTT, and 10% glycerol) before the addition of the labeled probe and incubated for 1 h at 4°C. Samples were loaded on 4% polyacrylamide gels (29:1, acrylamide:bisacrylamide) containing 45 mM Tris-HCl (pH 8.0), 45 mM boric acid, and 1 mM EDTA (pH 8.0) for 3 h at 180 V. The gels were dried and autoradiographed to Kodak X-Omat AR films at room temperature.
Northern Blot Analysis-Total RNA was extracted using an RNeasy kit (Qiagen, Valencia, CA). 15 g of total RNA, as determined by a spectrophotometer, was fractionated in 1% agarose-formaldehyde gels. The RNA was transferred onto a nylon membrane (Gene Screen Plus; PerkinElmer Life Sciences) by capillary action. The membrane was prehybridized in a solution containing 30% deionized formamide, 50 mM sodium phosphate (pH 7.4), 1% SDS, and 1% bovine serum albumin for 10 min at 55°C. The heat-denatured probe for rat ICAM-1 cDNA was labeled with 50 Ci of [␣-32 P]dCTP (Amersham Pharmacia Biotech) using a random primed DNA labeling kit (Roche Molecular Biochemicals). The labeled cDNA probe was added to the solution, and the hybridization was performed for 20 h at 55°C. After the hybridization, the membranes were washed with 50 mM sodium phosphate (pH 7.4) and 1% SDS twice for 5 min at 55°C. Then the membranes were autoradiographed to films at Ϫ80°C. The mRNA levels were normalized to GAPDH mRNA levels. Radioactivity of the band for the respective mRNA was quantified by a PhosphorImager (Molecular Dynamics, Sunnyvale, CA).
Western Blot Analysis-The cytosolic extracts containing 60 g of protein were mixed with 1 volume of SDS loading buffer containing 5% mercaptoethanol. After denaturing in boiling water for 10 min, the samples and molecular weight markers (Low or High; Bio-Rad) were fractionated on 10 or 6% SDS polyacrylamide gels and electroblotted onto membranes (Trans-Blot; Bio-Rad) using the Mini-Protean II system (Bio-Rad). The membranes were soaked for 30 min in TBST (10 mM Tris-HCl (pH 8.0), 150 mM NaCl, and 0.05% Tween 20) containing 3% skim milk. Then the membranes were incubated for 3 h with anti-IB␣, IB kinase (IKK)␣, or IKK␤ antibody (Santa Cruz Biotechnology, Santa Cruz, CA) diluted 1:500 with TBST containing 3% skim milk. After washing three times with TBST, the membranes were incubated for 1 h with anti-rabbit IgG linked to horseradish peroxidase (Jackson Immu-noResearch, West Grove, PA) diluted 1:3000 with TBST containing 3% skim milk. After three additional washes with TBST, the membranes were soaked in enhanced chemiluminescence detection reagents (ECL; Amersham Pharmacia Biotech) according to the manufacturer's protocol. The membranes were then exposed to films.
IKK in Vitro Kinase Assay-The cytosolic extracts containing 200 g of protein were preincubated with 1 g of IKK␣ antibody (Santa Cruz Biotechnology) for 1 h and were then incubated for 20 h, together with 20 l of protein A/G-agarose (Santa Cruz Biotechnology) at 4°C. After washing 4 times with PBS (Ϫ), one-half of each of the immunocomplexes were subjected to IKK kinase assays. Kinase assays were performed as described previously with some modification (25). Briefly, each immunoprecipitate was resuspended in 25 l of kinase buffer (10 M ATP, 2 Ci of [␥-32 P]ATP, 25 g/ml aprotinin, 1 mM benzamidine, 1 mM DTT, 10 mM ␤-glycerophosphate, 20 mM HEPES-KOH (pH 7.9), 10 g/ml leupeptin, 2 mM MgCl 2 , 2 mM MnCl 2 , 10 mM p-nitrophenyl phosphate, 2 g/ml pepstatin, 0.5 mM phenylmethylsuflonyl fluoride, 10 mM NaF, and 0.5 mM Na 3 VO 4 ) in the presence of 2.5 g of IB␣ (1-317; Santa Cruz Biotechnology) as a substrate at 30°C for 30 min. Reactions were stopped by the addition of 6ϫ SDS loading buffer containing 30% mercaptoethanol. Samples and low molecular weight markers were fractionated on 10% SDS polyacrylamide gels. The gels were dried and autoradiographed to film at room temperature. Radioactivity of the band for the respective IKK activity was quantified by PhosphorImager. To confirm the presence of IKK, the remaining one-half of each of the immunocomplexes was subjected to Western blot analysis using 1:500 dilutions of IKK␣ or IKK␤ antibody.
Statistical Analyses-Results are expressed as mean Ϯ S.E. Statistical analysis was performed by analysis of variance with a Bonferroni test. A p value less than 0.05 was considered significant.

OFF Does Not Alter the Viability of UMR106 Cells
Adhering to Glass Slides-Trypan blue dye exclusion tests indicated that increasing flow rates up to Ϯ 20 ml/min, inducing shear stresses up to 9.3 dyne/cm 2 , tended to decrease the number of viable cells adhering to glass slides (Fig. 1). However, over 90% of cells were viable after exposure to OFF, which was not significantly different from untreated controls. Moreover, TNF-␣ did not affect cell viability relative to cells untreated with TNF-␣ at any flow rates examined. Thus, under our study conditions TNF-␣ was not cytotoxic.
OFF Inhibits TNF-␣-induced Activation of NF-B-We first examined the effect of OFF on NF-B-DNA binding activity, in the absence or presence of TNF-␣, by electrophoretic mobility shift assay using nuclear protein extracts obtained from UMR106 cells. A single weak NF-B-DNA complex was observed even in untreated control cells (Fig. 2A, lane 1). Exposure to OFF in the absence of TNF-␣ did not change this basal activation of the NF-B-DNA complex (lanes 2-4). In contrast, activation of two distinct complexes was observed in TNF-␣ treated cells, namely a fast-migrating complex, which exhibited the same mobility and weak activity as the complex observed in untreated cells, and an additional slow-migrating complex, which exhibited stronger activity (lane 5). Exposure to OFF decreased dramatically the TNF-␣-induced activation of the slow-migrating NF-B-DNA complex in a shear stress-dependent manner (lanes 6 -8). However, the activity of the fastmigrating NF-B-DNA complex was not altered by TNF-␣ or OFF. Time course studies (Fig. 2B) revealed that treatment with TNF-␣ resulted in a rapid (within 15 min) increase of the slow-migrating NF-B-DNA complex, which continued throughout the 120-min treatment period. Once again, TNF-␣ did not affect the activation of the fast-migrating NF-B-DNA complex. In contrast, exposure to OFF decreased markedly the TNF-␣-induced activation of the slow-migrating NF-B-DNA complex in a time-dependent manner, whereas it did not change the activation of the fast-migrating NF-B-DNA complex.
We next characterized the NF-B-DNA binding complexes in UMR106 cells utilizing supershift analysis with specific antibodies against p50 and p65, two of the more common members in the NF-B/Rel family that form a dimer in rat osteoblastic cells (17). A weak single NF-B-DNA complex in untreated cells (Fig. 2C, lane 1) was supershifted by anti-p50 antibody (lane 2) but not by anti-p65 antibody (lane 3), indicating that the complex represents a p50 homodimer NF-B. On the other hand, anti-p50 antibody supershifted both fast-and slow-migrating complexes induced by TNF-␣ (lane 5), whereas anti-p65 antibody supershifted only the slow-migrating complex (lane 6). Thus, the slow-and fast-migrating NF-B-DNA complexes represent a p50-p65 heterodimer and a p50 homodimer NF-B, respectively.
OFF Inhibits a TNF-␣-induced Increase in ICAM-1 mRNA Expression-ICAM-1 gene expression is mainly regulated by NF-B in rat osteoblastic cells (17,18). We therefore examined ICAM-1 mRNA expression to evaluate the effect of OFF on the expression of an endogenous NF-B target gene. Basal ICAM-1 mRNA expression was detected even in the absence of TNF-␣ (Fig. 3A). This basal mRNA level was not altered by a 2-h exposure to OFF at any shear stress level examined. In contrast, treatment with TNF-␣ markedly increased ICAM-1 mRNA expression within 2 h. Exposure to OFF inhibited this TNF-␣-induced ICAM-1 mRNA expression in a shear stressdependent manner with the inhibition reaching statistical sig-nificance at 4.3 dynes/cm 2 . Time course studies (Fig. 3B) revealed that TNF-␣ dramatically induced ICAM-1 mRNA expression during a 4-h exposure (lanes 1-5) while increasing  4 -9). GAPDH mRNA levels were not altered by OFF regardless of the presence of TNF-␣.
OFF Decreases TNF-␣-induced IB␣ Degradation-To evaluate the mechanism by which OFF affects NF-B activation, we examined protein levels of IB␣, which is the only endogenous inhibitor of NF-B activation thus far identified. IB␣ protein was highly expressed in the cytosol in the absence of TNF-␣ (Fig. 4, upper panel, lane 1). Treatment with TNF-␣ markedly decreased IB␣ protein levels within 60 min (lane 2), reflecting a degradation of IB␣. Exposure to OFF in the presence of TNF-␣ resulted in a substantial increase in IB␣ protein levels in a shear stress-dependent manner (lanes [3][4][5]. IB␣ protein levels in cells exposed to TNF-␣ and OFF at 9.3 dynes/cm 2 were similar to untreated control levels. These results suggest that OFF-induced shear stress decreases TNF-␣induced IB␣ degradation. Protein levels of IKK␣ and IKK␤, two kinase members of the IKK family that regulate IB␣ phosphorylation and thus degradation (21, 26 -28), were not affected by exposure to TNF-␣ or OFF (Fig. 4, middle and lower  panels, respectively).
Effect of OFF on IKK Activation-To evaluate the mechanism by which OFF decreases TNF-␣-induced IB␣ degradation and NF-B activation, we examined the activities of endogenous IKK utilizing IB␣ as a substrate in an in vitro kinase assay. The weak IKK activation observed in untreated cells (Fig. 5A, lane 1) dramatically increased after a 30-min exposure to TNF-␣ (lane 2). Exposure to OFF at 9.3 dyne/cm 2 significantly inhibited this TNF-␣-induced IKK activation (lane 3). OFF decreased the TNF-␣-induced IKK activation in a shear stress-dependent manner (Fig. 5B). Time course studies revealed that TNF-␣ markedly increased IKK activation throughout the 60-min treatment period (Fig. 5C, lanes 1-4). Additionally exposure to OFF decreased TNF-␣-induced IKK activation to a greater degree with increasing exposure duration eventually reaching control levels by 60 min (lane 5-7). Middle and lower panels in Fig. 5 represent the protein levels of IKK␣ and IKK␤ analyzed by Western blot analysis using a cytosolic immunocomplex precipitated by anti-IKK␣ antibody to confirm the presence of IKK. The immunoprecipitated protein levels of both IKK␣ and IKK␤ did not change following exposure to TNF-␣ or OFF. mRNA expression were shear stress-dependent and also increased with OFF exposure duration. These results suggest that OFF has a potent biophysical effect on mechanotransduction pathways, especially at the transcriptional level in osteoblastic cells stimulated by TNF-␣.
Theoretical models predict that the wall fluid shear stresses in the canaliculi of bone tissues are 6 -30 dynes/cm 2 (3). The maximum flow rate used for our study was Ϯ 20 ml/min, and frequency was 1 Hz. This fluid flow induces peak shear stresses of 9.3 dyne/cm 2 in our flow chamber. Thus, the fluid shear stress generated by our OFF system is within the range bone cells experience in vivo. Under the condition of our studies, fluid flow had no effect on cell viability or adhesion.
It has been demonstrated that the p50-p65 heterodimer NF-B is capable of transactivating gene expression (29). Indeed, we previously have shown that TNF-␣ induces interleukin-6 and ICAM-1 gene expressions in rat osteoblastic cells via activation of the p50-p65 heterodimer NF-B (17,18). The shear stress and time course for OFF inhibition of TNF-␣induced ICAM-1 mRNA expression were similar to the shear stress and time course for the inhibition of TNF-␣-induced NF-B activation. Taken together with our previous findings, this suggests the possibility that the effect of OFF on TNF-␣induced ICAM-1 gene expression may be via an NF-B-dependent pathway. However, a direct link between OFF, TNF-␣, and ICAM-1 expression cannot be made without rigorous mutational analysis. In any case, our data clearly show an interaction of OFF with TNF-␣ that has a strong potential to alter bone cell activity.
In this study, neither OFF nor TNF-␣ affected the basal p50 homodimer NF-B activation. Moreover, basal ICAM-1 mRNA expression, which was detected in the absence of TNF-␣, was not altered by OFF. It has been reported that moderate levels of p50 homodimer NF-B activation was conserved in other cells (30 -32). The role for the p50 homodimer NF-B activation in constitutive-type transcription is unclear, but it may provide low levels of transcriptional activity or it may serve as a transcriptional repressor protein (19,33,34).
In response to external stresses, mammalian cells rapidly translocate NF-B to the nucleus. Once there, this protein binds to 10-base pair B sites as a dimer within the DNA of specific genes, resulting in the regulation of transcription of these genes. The activity of NF-B is tightly regulated by interactions with inhibitory IB proteins in the cytoplasm, which block transport of NF-B into the nucleus in the absence of activating signals. Most extracellular signals such as TNF-␣ activate NF-B through a common pathway dependent on phosphorylation-induced degradation of IB (for reviews, see Refs. 19,20,and 35). In this study, we demonstrated that cytosolic IB␣ protein levels were dramatically decreased by TNF-␣ and that OFF attenuated the inhibitory effect of TNF-␣ on IB␣.
Recent evidence suggests that IB␣ degradation is regulated by phosphorylation via IKK (36). IKK is a protein complex the catalysis of which is generally carried out by a heterodimeric kinase consisting of IKK␣ and IKK␤ subunits (21, 26 -28). Indeed, we observed IKK␤, as well as IKK␣, by Western blot analysis using the immunocomplex precipitated by anti-IKK␣ antibody. This suggests that the IKK complex within the cytosol of UMR106 cells is largely a heterodimeric complex of IKK␣ and IKK␤. Both IKK␣ and IKK␤ activities are stimulated in response to TNF-␣ or interleukin-1 (21, 26 -28, 37). Furthermore, knockout mice studies indicate that IKK␣ plays an important role in skeletal development (38). Therefore, we examined IKK kinase activity to evaluate the mechanism by which OFF affects TNF-␣-induced IB␣ degradation and NF-B activation. Exposure to OFF inhibited TNF-␣-induced activation of IKK in a shear stress-dependent manner. The shear stress-dependent pattern of OFF inhibition of TNF-␣induced IKK activation was similar to the shear stress-dependent pattern of the inhibition of TNF-␣-induced NF-B activation, ICAM-1 mRNA expression, and IB␣ degradation, suggesting a link between these signaling pathways and OFF. Neither TNF-␣ nor OFF affected IKK␣ and IKK␤ levels, suggesting that the effects we observed were not because of an effect on protein synthesis or degradation. Our results suggest that IKK is an initial target molecule for OFF effects on osteoblastic cells. OFF inhibits TNF-␣-induced IKK activation, leading to a decrease in phosphorylation and degradation of inhibitory IB␣, which in turn results in the decrease of TNF-␣induced NF-B activation and the transcription of target genes.
The precise mechanism by which OFF decreases TNF-␣induced IKK activity remains unclear. Recently it has been demonstrated that IKK itself is also phosphorylated and regulated by one or more upstream kinases (37). One possibility is that OFF affects dephosphorylation events by inducing conformational changes in these upstream kinases of IKK. Shear stress might lead directly to dephosphorylation by deforming kinases and inactivating them (39). On the other hand, it has been demonstrated that fluid flow rapidly activates mitogenactivated protein kinases, including extracellular signal-regulated kinase and c-Jun N-terminal kinase, both recognized stress-activated protein kinases (40 -44), and focal adhesion kinase (45) in vascular endothelial cells. Therefore, another possibility is that OFF might activate upstream kinases, leading to phosphorylation of IKK itself and resulting in decreased IKK activation. Indeed, more recently it was discovered that phosphorylation of IKK␤ at C-terminal serines can also result in negative regulation of IKK activity by changing the conformation of the intrinsic kinase activator domain (46,47). Another possibility is an indirect effect through which OFF-induced shear stress may modify distinct signaling factors that compete for or couple with TNF-␣-induced IKK activation pathways.
An important point to consider is that our results were obtained with UMR106 cells, a rat osteogenic osteosarcoma cell line with osteoblastic characteristics. Although this cell line has provided many important insights into bone cell biology, especially as related to mechanotransduction, it could be argued that transcription factors in UMR106 cells are not the same as those in authentic osteoblasts. However, we believe that this is unlikely, because it has recently been shown that NF-B is required for TNF-␣ activity in primary culture human osteoblasts (48).
In summary, our results suggest that OFF inhibits TNF-␣induced NF-B activation in an osteoblastic cell line. Previous studies suggest that TNF-␣ increases accumulation of bone resorption stimulating cytokines by osteoblastic cells (12,13) and may also stimulate osteoblastic differentiation, both of which may be mediated by NF-B. (49) Thus, OFF may modulate bone turnover through its effect on bone cell activity.