Inhibition of IkappaB kinase activity by sodium salicylate in vitro does not reflect its inhibitory mechanism in intact cells.

Sodium salicylate inhibits activation of the transcription factor NF-kappaB by blocking the phosphorylation and degradation of the NF-kappaB inhibitor IkappaBalpha. We previously demonstrated that salicylate inhibits IkappaBalpha degradation induced by tumor necrosis factor (TNF) but not by interleukin-1 (IL-1) and implicated p38 mitogen-activated protein kinase activation by salicylate in the inhibition of TNF-induced IkappaBalpha phosphorylation. Both TNF and IL-1 rapidly activate the IkappaB kinase (IKK) complex, containing the catalytic subunits IKKalpha and IKKbeta, which directly phosphorylates IkappaB proteins. Others have recently suggested that salicylate inhibits NF-kappaB activation by directly binding to IKKbeta. To clarify the mechanism whereby salicylate inhibits IKK activity, we examined its effects upon cytokine-induced IKK activity in intact cells and in vitro. Treatment of intact cells with salicylate inhibited TNF-induced but not IL-1-induced IKK activity, and this inhibition was prevented by the p38 inhibitor SB203580. In contrast, inhibition of IKK activity by salicylate in vitro was neither selective for TNF nor affected by SB203580. In vitro, salicylate treatment comparably inhibited the kinase activity of overexpressed IKKalpha and IKKbeta and also decreased p38 kinase activity. Therefore, direct inhibition of IKK activity in vitro does not reflect the inhibitory mechanism of salicylate in intact cells, which involves interference with TNF signaling.

Sodium salicylate inhibits activation of the transcription factor NF-B by blocking the phosphorylation and degradation of the NF-B inhibitor IB␣. We previously demonstrated that salicylate inhibits IB␣ degradation induced by tumor necrosis factor (TNF) but not by interleukin-1 (IL-1) and implicated p38 mitogen-activated protein kinase activation by salicylate in the inhibition of TNF-induced IB␣ phosphorylation. Both TNF and IL-1 rapidly activate the IB kinase (IKK) complex, containing the catalytic subunits IKK␣ and IKK␤, which directly phosphorylates IB proteins. Others have recently suggested that salicylate inhibits NF-B activation by directly binding to IKK␤. To clarify the mechanism whereby salicylate inhibits IKK activity, we examined its effects upon cytokine-induced IKK activity in intact cells and in vitro. Treatment of intact cells with salicylate inhibited TNF-induced but not IL-1-induced IKK activity, and this inhibition was prevented by the p38 inhibitor SB203580. In contrast, inhibition of IKK activity by salicylate in vitro was neither selective for TNF nor affected by SB203580. In vitro, salicylate treatment comparably inhibited the kinase activity of overexpressed IKK␣ and IKK␤ and also decreased p38 kinase activity. Therefore, direct inhibition of IKK activity in vitro does not reflect the inhibitory mechanism of salicylate in intact cells, which involves interference with TNF signaling.
The transcription factor NF-B 1 is an important regulator of genes involved in immune and inflammatory responses. The NF-B pathway is activated by a variety of stimuli, including the cytokines tumor necrosis factor (TNF) and interleukin-1 (IL-1) and bacterial lipopolysaccharide. In most mammalian cells, the NF-B dimer is sequestered in the cytoplasm by an inhibitory IB isoform such as IB␣, IB␤, or IB⑀ (1). Stimulus-induced phosphorylation of IB proteins on two conserved N-terminal serine residues leads to IB degradation via the ubiquitin-proteasome pathway allowing for the release of NF-B and its nuclear translocation. TNF-induced NF-B activation is initiated by oligomerization of the type 1 or type 2 TNF receptors. This leads to the recruitment of cytoplasmic signaling proteins, such as TNF receptor-associated death domain, TNF receptor-associated factor 2 and receptor interacting protein, and activation of kinases including MAP/ERK kinase kinase 1 (MEKK1) and the NF-B-inducing kinase (NIK) (2)(3)(4)(5). Both NIK and MEKK1 can activate a multiprotein IB kinase (IKK) complex of 700 -900 kDa that is responsible for directly phosphorylating IB isoforms (3, 6 -8). The IKK complex consists of two catalytic components, IKK␣ and IKK␤, as well as a regulatory IKK␥ subunit, which may facilitate interaction with upstream signaling factors (9 -11). Other proteins, such as IKK complex associated protein (12), have been shown to interact with the IKK complex, and their precise roles in the activation of the IKK complex remain to be defined.
Salicylate and its acetylated derivative, aspirin (acetylsalicylic acid), represent the oldest known nonsteroidal anti-inflammatory drugs. A well characterized mechanism of action for aspirin involves the acetylation of cyclooxygenase isoforms, leading to irreversible inhibition of prostaglandin synthesis (13). In the intact organism, aspirin is rapidly deacetylated to salicylate, which is a relatively weak inhibitor of cyclooxygenases (14). Nonetheless, both aspirin and salicylate are potent anti-inflammatory agents, used in the treatment of chronic inflammatory disorders such as rheumatoid arthritis (14,15).
Sodium salicylate (NaSal) and aspirin can inhibit the activation of NF-B by preventing the phosphorylation and degradation of IB␣ (16,17), and inhibition of NF-B may explain some of the clinically documented anti-inflammatory effects seen with high concentrations of salicylates (14,15). We have demonstrated that NaSal rapidly and persistently activates p38 MAP kinase and that NaSal-induced p38 activation is important for its ability to inhibit TNF-induced IB␣ phosphorylation and degradation (18,19). Furthermore, we have demonstrated that p38 activation may play a more general role in the inhibition of TNF-induced NF-B activation (20). A recent report has suggested that NaSal and aspirin may inhibit NF-B by an alternative mechanism, via direct binding to the IKK complex (21). In this study, we sought to clarify the mechanism by which NaSal inhibits IKK activity. We demonstrate that treatment of intact cells with NaSal selectively inhibits IKK activity induced by TNF, yet not by IL-1, and this inhibition is prevented by the p38 inhibitor SB203580. However, we find that inhibition of IKK activity in vitro is not selective and cannot be prevented by SB203580, suggesting that direct inhibition of the IKK complex is not likely to be the mechanism by which NaSal exerts its inhibitory effect upon NF-B activation in vivo. and were serum-starved for 18 -24 h in DMEM with 0.5% FBS. Normal human diploid FS-4 fibroblasts were cultured in Eagle's minimal essential medium (MEM) supplemented with 5% FBS and were serumstarved for 2-3 days in MEM with 0.25% FBS before use in experiments.
NaSal, sodium meta-arsenite, and myelin basic protein (MBP) were purchased from Sigma. NaSal and arsenite were dissolved in distilled water, unless otherwise indicated. Recombinant human TNF-␣ was provided by Masafumi Tsujimoto of the Suntory Institute for Biomedical Research, Osaka, Japan. Recombinant human IL-1␣ was obtained from the National Cancer Institute. The p38 inhibitor SB203580 (22) was purchased from Calbiochem and dissolved in Me 2 SO. The rabbit polyclonal anti-IB␣ and anti-IKK␣ (H-744) antibodies were from Santa Cruz Biotechnology. The M2 monoclonal anti-FLAG antibody was from Kodak Scientific Imaging Systems.
Plasmids and Transfections-Full-length FLAG-tagged IKK␣ and IKK␤ cloned into the pRK5 vector were provided by David Goeddel (6,23). FLAG-p38␣ cloned into the pcDNA3 vector was provided by Jiahuai Han (24). Full-length NIK cloned into the pcDNA3 vector was a gift from David Wallach (2). Full-length IB␣ in the pGEX-2T vector was provided by John Hiscott (25). A fragment encoding the first 62 amino acids of IB␣ was amplified from this construct by polymerase chain reaction and cloned into the EcoRI and NotI sites of the pGEX-4T1 vector. Glutathione S-transferase (GST)-IB␣ (amino acids 1-62) was subsequently purified from Escherichia coli DH5␣ with glutathioneagarose for use in immunokinase assays.
Transfections were performed in COS-1 cells plated at a density of approximately 2.5 ϫ 10 5 cells per well in 6-well plates. Cells were transfected the following day using LipofectAMINE (Life Technologies, Inc.) in a total volume of 1 ml of serum-free DMEM. 5 h post-transfection, 1 ml of serum-free DMEM was added to each well. Cells were incubated for an additional 24 h, prior to stimulation as indicated.
Immunokinase Assays-Whole cell lysates were generated using a buffer consisting of 1% Igepal, 50 mM Hepes (pH 7.5), 100 mM NaCl, 2 mM EDTA, 1 mM pyrophosphate, 10 mM sodium orthovanadate, 1 mM phenylmethylsulfonyl fluoride, and 100 mM NaF (18,20). To immunoprecipitate the endogenous IKK complex, equal amounts of lysates were incubated for 1-2 h at 4°C with anti-IKK␣ antibody followed by collection of immune complexes for 2 h at 4°C using protein A/G PLUSagarose beads (Santa Cruz Biotechnology). For immunoprecipitation of transfected FLAG-IKK␣, FLAG-IKK␤, or FLAG-p38␣, equal amounts of lysates were incubated with anti-FLAG M2 antibody followed by collection of immune complexes using protein G-agarose beads (Life Technologies, Inc.). The immunoprecipitates were subsequently washed three times with lysis buffer and once with kinase buffer without ATP. Assays for IKK activity were performed for 30 min at 30°C in kinase buffer consisting of 20 mM Hepes (pH 7.6), 20 mM MgCl 2 , 20 mM ␤-glycerophosphate, 10 mM NaF, 0.2 mM sodium orthovanadate, 0.2 mM dithiothreitol, 10 mM ATP, 10 mCi of [␥-32 P] ATP, and 5 g of GST-IB␣ (1-62) as substrate. Assays for p38 kinase activity were performed under identical conditions in kinase buffer containing 5 g of MBP as substrate. Kinase reactions were terminated by the addition of 2ϫ protein sample buffer, and phosphorylated GST-IB␣ or MBP was visualized following SDS-polyacrylamide gel electrophoresis (PAGE) and autoradiography.
To determine the effect of NaSal treatment in vitro upon IKK activity, aliquots of immunoprecipitates were incubated with various concentrations of NaSal in kinase buffer for 30 min on ice. Kinase assays were then performed as described above in the continuous presence of NaSal.
Immunoblotting-Immunoblot analysis was performed as described (20). Briefly, whole cell lysates were fractionated by SDS-PAGE, transferred to Immobilon-P membranes (Millipore), and blocked for 1 h at room temperature in TBS (10 mM Tris (pH 7.5), 150 mM NaCl) containing 0.5% Tween 20 and 5% nonfat milk. Membranes were then incubated overnight at 4°C with primary antibody. Antibody-antigen complexes were detected with the aid of horseradish peroxidase-conjugated secondary antibody (Bio-Rad) and a chemiluminescent substrate development kit (Kirkegaard and Perry Laboratories).

NaSal Inhibits TNF-induced but Not IL-1-induced IKK Activity in Intact
Cells-We and others have reported that NaSal inhibits IB␣ degradation induced by TNF but not by IL-1 (19,20,26). A preponderance of evidence suggests that both TNF and IL-1 induce IB␣ phosphorylation and degradation via activation of the IKK complex (23,27). To determine whether blockage of IB␣ degradation correlates with inhibition of IKK activity, we treated COS-1 cells with various concentrations of NaSal prior to stimulation with TNF or IL-1. An antibody to IKK␣ was used to immunoprecipitate endogenous IKK complexes from whole cell lysates (21), which were then assayed for in vitro kinase activity using GST-IB␣ as substrate. NaSal inhibited TNF-induced IKK activity in a dose-dependent manner with an IC 50 of approximately 5 mM (Fig. 1A, top panel). In contrast, NaSal did not appreciably affect IL-1-induced IKK activity. This profile of IKK inhibition is consistent with the selective inhibition of TNF-induced IB␣ degradation by NaSal (19) as demonstrated by immunoblot analysis of whole cell lysates using an antibody against IB␣ (Fig. 1A, bottom panel). TNF-induced IB␣ degradation was dose-dependently inhibited by NaSal treatment. Although IL-1 induced a less complete degradation of IB␣, this effect was not inhibited at the concentrations of NaSal tested.
Yin et al. (21) reported that NaSal and aspirin inhibit IKK activity in vitro by directly binding to IKK␤ and postulated this as the mechanism whereby salicylates inhibit NF-B activation. It seemed difficult to reconcile the notion that NaSal directly binds and inhibits the IKK complex with our finding of selective inhibition of TNF-induced but not IL-1-induced IKK activation by NaSal. We therefore examined the effect of NaSal upon IKK activity in vitro. COS-1 cells were treated with either TNF or IL-1 for 10 min followed by immunoprecipitation of the endogenous IKK complex. The immune complexes were divided into equal fractions, which were incubated with various concentrations of NaSal for 30 min, followed by an in vitro kinase assay. In contrast to the selective inhibition of TNF-induced IKK activity observed in intact cells treated with NaSal, in vitro treatment with NaSal caused a comparable dose-dependent inhibition of both TNF-and IL-1-activated IKK activity (Fig. 1B).
SB203580 Reverses the Inhibition of TNF-induced IKK Activation by NaSal in Intact Cells but Not in Vitro-NaSal leads to a rapid and persistent activation of p38 MAP kinase (18,19,28). We have shown that inhibition of TNF-induced IB␣ phos- Aliquots of whole cell lysates were immunoprecipitated with an antibody against IKK␣. Immunoprecipitates (IP) were assayed for in vitro kinase activity using GST-IB␣ as substrate followed by 12% SDS-PAGE and autoradiography (top panel). Separate aliquots of lysates were subjected to immunoblot analysis (IB) with an antibody against IB␣ (bottom panel). B, COS-1 cells were left untreated or treated with TNF (20 ng/ml) or IL-1␣ (4 ng/ml) for 10 min. IKK␣ was immunoprecipitated from whole cell lysates, and immunoprecipitates were divided into five equal fractions for incubation with the indicated concentrations of NaSal for 30 min, as described under "Experimental Procedures." These fractions were then assayed for in vitro kinase activity using GST-IB␣ as substrate followed by 12% SDS-PAGE and autoradiography. phorylation and degradation by NaSal is preventable by the p38 inhibitor SB203580 (19). To determine whether SB203580 can reverse the inhibition of TNF-induced IKK activity, COS-1 cells were incubated with SB203580 or Me 2 SO vehicle prior to the addition of NaSal. Cells were harvested following a 10-min treatment with TNF, and the IKK complex was immunoprecipitated from whole cell lysates and assayed for in vitro kinase activity with GST-IB␣ as substrate. NaSal (20 mM) strongly inhibited TNF-induced IKK activity, and this inhibition was significantly reversed by SB203580 ( Fig. 2A, top panel). In addition, immunoblot analysis of cell lysates with an antibody against IB␣ demonstrated that TNF-induced IB␣ degradation was inhibited by NaSal, and SB203580 prevented this inhibition ( Fig. 2A, bottom panel).
To determine whether SB203580 can reverse the inhibitory effect of NaSal on IKK activity in vitro, IKK complexes were immunoprecipitated from untreated or TNF-treated COS cells and incubated with either SB203580 or Me 2 SO prior to treatment with 5 or 20 mM NaSal. Immunoprecipitated IKK was then assayed for in vitro kinase activity with GST-IB␣ as substrate. NaSal dose-dependently inhibited TNF-induced IKK activity in vitro. However, contrary to the reversal observed in intact cells ( Fig. 2A), SB203580 treatment failed to reverse the inhibition by NaSal in vitro (Fig. 2B).
Effect of NaSal on IKK Activity in Normal Human FS-4 Fibroblasts-Although NaSal strongly inhibits IB␣ degradation in numerous cell systems (17), it fails to appreciably inhibit this step in human FS-4 fibroblasts (20). Consistent with this observation, treatment of intact FS-4 cells with 5 or 20 mM NaSal did not markedly affect TNF-induced IKK activity (Fig.  3A, top panel). Immunoblot analysis of these lysates with an antibody against IB␣ confirmed the inability of NaSal to inhibit TNF-induced IB␣ degradation (bottom panel). In contrast, treatment of immunoprecipitated IKK with NaSal in vitro inhibited IKK activity, especially at a concentration of 20 mM (Fig. 3B).

NaSal Inhibits Both IKK␣ and IKK␤ Kinase Activity in
Vitro-To determine whether the inhibitory effect of NaSal in vitro is selective for either the IKK␣ or IKK␤ catalytic components of the IKK complex, we transfected COS-1 cells with expression vectors encoding FLAG-IKK␣ or FLAG-IKK␤. Because the kinase activity of overexpressed IKK␤ is at least 20 times higher than that of IKK␣ (23), we cotransfected NIK along with IKK␣ to augment its in vitro kinase activity. FLAGtagged IKKs were immunoprecipitated from whole cell extracts and used in an in vitro kinase assay with GST-IB␣ as substrate in the presence of 0, 5, or 20 mM NaSal. NaSal dose-dependently inhibited IKK␣ activity in the presence or absence of cotransfected NIK in addition to inhibiting IKK␤ activity in vitro (Fig. 4, top panel). Levels of immunoprecipitated IKK␣ and IKK␤ were equivalent in all lanes as demonstrated by immunoblot analysis (Fig. 4, bottom panel).
NaSal Inhibits p38 Kinase Activity in Vitro-To determine whether the in vitro inhibitory effect of NaSal is specific for IKKs or whether other kinases are also affected, we analyzed the effect of NaSal treatment upon in vitro p38 kinase activity. COS-1 cells were transfected with FLAG-tagged p38␣ and either left untreated or treated with arsenite for 15 min to activate p38. FLAG-tagged p38 was immunoprecipitated from whole cell extracts and used in an in vitro kinase assay with MBP as substrate in the presence of 0, 5, or 20 mM NaSal. Arsenite stimulated p38 kinase activity above baseline, and both baseline and arsenite-stimulated p38 activity were inhibited dose-dependently by NaSal treatment in vitro (Fig. 5, top  panel). Expression of immunoprecipitated FLAG-p38 was equivalent in all lanes (Fig. 5, bottom panel). We also observed that in vitro treatment with 5 and 20 mM NaSal inhibited the kinase activity of endogenous p38 activated by TNF treatment (data not shown). Because treatment of cells with NaSal has been shown to activate p38 MAP kinase (18,19), the inhibitory effect of NaSal upon p38 activity supports the argument that the effect of NaSal upon kinase activity in vitro does not reflect the action of NaSal in intact cells.

FIG. 2. SB203580 reverses the inhibition of TNF-induced IKK activation by NaSal in vivo but not in vitro.
A, COS-1 cells were incubated for 1.5 h in the presence of SB203580 (10 M) or Me 2 SO vehicle followed by incubation for 30 min in the presence or absence of NaSal (20 mM). Cells were then either left untreated or treated with TNF (20 ng/ml) for 10 min. Whole cell lysates were immunoprecipitated with an antibody against IKK␣. Immunoprecipitates (IP) were assayed for in vitro kinase activity using GST-IB␣ as substrate followed by 12% SDS-PAGE and autoradiography (top panel). Separate aliquots of lysates were subjected to immunoblot analysis (IB) with an antibody against IB␣ (bottom panel). B, COS-1 cells were left untreated or treated with TNF (20 ng/ml) for 10 min. IKK␣ was immunoprecipitated from whole cell lysates, and immunoprecipitates were divided into six equal fractions for incubation with either SB203580 (10 M) or Me 2 SO for 10 min followed by incubation with the indicated concentrations of NaSal for 30 min. These fractions were then assayed for in vitro kinase activity using GST-IB␣ as substrate followed by 12% SDS-PAGE and autoradiography.

FIG. 3. Effects of in vivo and in vitro treatment with NaSal upon TNF-induced IKK activity in FS-4 fibroblasts. A, FS-4 cells
were treated with the indicated concentrations of NaSal for 1 h and then either left untreated or treated with TNF (20 ng/ml) for 10 min. Cell lysates were immunoprecipitated with an antibody against IKK␣, and immunoprecipitates (IP) were subjected to an in vitro kinase assay using GST-IB␣ as substrate followed by 12% SDS-PAGE and autoradiography (top panel). Separate aliquots of lysates were subjected to immunoblot analysis (IB) with an antibody against IB␣ (bottom panel). B, FS-4 cells were left untreated or treated with TNF (20 ng/ml) for 10 min. IKK␣ was immunoprecipitated from whole cell lysates, and immunoprecipitates from each group were divided into five equal fractions for incubation with the indicated concentrations of NaSal for 30 min. These fractions were then assayed for in vitro kinase activity using GST-IB␣ as substrate followed by 12% SDS-PAGE and autoradiography.

DISCUSSION
It has been shown that NaSal and aspirin inhibit NF-B activation by preventing the phosphorylation and subsequent degradation of IB␣ (16,17). NaSal inhibits IB␣ phosphorylation and degradation induced by TNF but not by IL-1 (19,26). The ability of NaSal to inhibit TNF-induced IB␣ phosphorylation is dependent upon NaSal-induced p38 MAP kinase activation, and p38 may play a more general role in the inhibition of TNF-induced NF-B activation (19,20). Recently, Yin et al. (21) have reported that NaSal and aspirin may inhibit the IKK complex by directly binding to IKK␤, thereby interfering with ATP binding. Because both TNF and IL-1 activate the IKK complex (23,27), such a mechanism of NF-B inhibition by NaSal is inconsistent with the observations that NaSal inhibits TNF-induced but not IL-1-induced IB␣ degradation (19,26) and that the inhibition by NaSal is p38-dependent (19,20). To clarify the mechanism by which NaSal inhibits IB␣ phospho-rylation, we analyzed the effects of NaSal upon cytokine-induced IKK activity in intact cells and in vitro. We conclude that the direct inhibitory effect of NaSal upon IKK activity in vitro does not reflect the mechanism whereby NaSal inhibits IKK activity in intact cells.
Studies of IKK␣ and IKK␤ knockout mice indicate that only IKK␤ is required for cytokine-induced IB␣ degradation, whereas IKK␣ appears to be important for epidermal differentiation and skeletal morphogenesis (reviewed in Refs. 29 and 30). Because both TNF-and IL-1-induced IKK activation is impaired in IKK␤Ϫ/Ϫ cells (31)(32)(33), direct inhibition of IKK␤ by NaSal would be expected to impair IKK activation induced by either TNF or IL-1. However, our results demonstrate that treatment of intact cells with NaSal inhibits TNF-induced IKK activity but does not appreciably affect IL-1-induced IKK activity (Fig. 1A). This observation, in concert with our previous data (19,20), suggests that NaSal may inhibit NF-B activation by selectively targeting a component of the TNF signaling pathway leading to IKK activation. On the other hand, treatment of immunoprecipitated IKK with NaSal in vitro dose-dependently inhibited IKK activity from either TNF-or IL-1treated cells (Fig. 1B). In addition, we observed that NaSal did not appreciably inhibit either TNF-induced IKK activity or IB␣ degradation in intact FS-4 fibroblasts (Fig. 3A), whereas in vitro treatment of IKK immunoprecipitated from FS-4 cells with 20 mM NaSal led to a strong inhibition of IKK activity (Fig. 3B). The inhibitory effect of NaSal in vitro is therefore not indicative of its inhibitory profile in whole cells in at least two distinct cell systems.
Because the p38 MAP kinase inhibitor SB203580 can prevent the ability of NaSal to inhibit TNF-induced IB␣ phosphorylation and degradation (19,26), we analyzed the effect of this inhibitor on the ability of NaSal to affect TNF-induced IKK activity. Consistent with previous studies, we observed that incubation of cells with SB203580 significantly prevented the NaSal-mediated inhibition of TNF-induced IKK activity ( Fig.  2A). SB203580 is a specific inhibitor of p38␣ and p38␤ isoforms and fails to inhibit the activity of numerous other kinases (34). However, some reports indicate that high concentrations of SB203580 may exert effects unrelated to inhibition of p38 (35)(36)(37). It seemed possible that SB203580, a low molecular weight imidazole compound, might directly interact with the IKK complex and thereby block NaSal binding. We have ruled out a direct effect of SB203580 on the IKK complex by demonstrating that incubation of immunoprecipitated IKK with SB203580 in vitro, prior to incubation with NaSal, did not reverse the inhibition by NaSal (Fig. 2B). This finding further supports the conclusion that the direct inhibition of IKK activity in vitro does not reflect the mechanism of IKK inhibition by NaSal in vivo.
It is likely that the effect of SB203580 in vivo reflects a genuine role for NaSal-induced p38 activation in the inhibition of IKK activity. In support of this notion, diverse stimuli that persistently activate p38, such as hyperosmolarity, H 2 O 2 , staurosporine, and vitamin C, similarly inhibited TNF-induced IB␣ phosphorylation and degradation in a manner preventable by SB203580 (20,26,38). Furthermore, we demonstrated that a constitutively active MAP kinase kinase 6 mutant, which activates p38, decreased TNF-induced IB␣ phosphorylation and NF-B reporter activity (20), and this inhibition occurs at the level of receptor interacting protein or TNF receptor-associated factor 2 in the TNF signaling pathway (data not shown). Other cellular effects of NaSal have been attributed to p38 activation, such as the induction of apoptosis in fibroblasts and induction of adipocyte differentiation (18, 28),  1 g), or FLAG-IKK␤ (0.5 g) as indicated. The total amount of DNA transfected was equalized to 1.1 g using pcDNA3 vector. Approximately 30 h post-transfection, cells were harvested and lysates were immunoprecipitated with an antibody against the FLAG epitope. Immunoprecipitates (IP) were each divided into three equal fractions for incubation with the indicated concentrations of NaSal for 30 min. These fractions were then assayed for in vitro kinase activity (KA) using GST-IB␣ as substrate, and aliquots of the kinase reactions were analyzed by 12% SDS-PAGE and autoradiography (top panel). A significantly shorter exposure for IKK␤ activity is shown, because of its high activity relative to IKK␣. Separate aliquots of the kinase reactions were subjected to immunoblot analysis (IB) with an antibody against the FLAG epitope to demonstrate equal levels of immunoprecipitated FLAG-IKK␣ and FLAG-IKK␤ (bottom panels).
FIG. 5. NaSal inhibits p38 kinase activity in vitro. COS-1 cells were transfected with FLAG-p38␣ (1.0 g). Approximately 30 h posttransfection, cells were either left untreated or treated with sodium meta-arsenite (0.5 mM) for 15 min, and whole cell lysates were immunoprecipitated with an antibody against the FLAG epitope. Control (Ctrl) and arsenite-activated immunoprecipitates (IP) were each divided into three equal fractions for incubation with the indicated concentrations of NaSal for 30 min. These fractions were then assayed for in vitro kinase activity (KA) using MBP as substrate, and aliquots of the kinase reactions were either analyzed on a 4 -15% gradient gel (Bio-Rad) followed by autoradiography (top panel) or subjected to immunoblot analysis (IB) with an antibody against the FLAG epitope to demonstrate equal levels of immunoprecipitated FLAG-p38␣ (bottom panel).
suggesting that NaSal may exert multiple biological effects through activation of p38.
Although it has been reported that NaSal and aspirin directly inhibit IKK␤ but fail to affect IKK␣ (21), we observed that NaSal inhibited the activity of both immunoprecipitated IKK␣ and IKK␤ in vitro (Fig. 4). We observed significant in vitro and in vivo inhibitory effects of NaSal only at millimolar concentrations, although Yin et al. (21) have demonstrated an IC 50 for inhibition of the IKK complex, and IKK␤ specifically, at concentrations ranging from 50 -100 M. We cannot explain this discrepancy; however, numerous groups have documented significant effects of NaSal upon NF-B function, MAP kinase activity, induction of apoptosis, and gene expression only at concentrations of NaSal in the 5-20 mM range (16 -18, 39 -41). Although these concentrations exceed the 1-2 mM serum levels that may be achieved in patients undergoing high dose salicylate therapy (15), they may nonetheless be consistent with a role for p38-mediated NF-B inhibition in the anti-inflammatory effects of salicylates. As organic acids, salicylates accumulate at the mildly acidic environments prevailing at sites of inflammation (14,15,42). Salicylates are uncharged at low pH and can readily cross membranes, yet deprotonate and become trapped as anions in the more neutral environment found within cells (42). Therefore, it is likely that local concentrations of salicylates at sites of inflammation may reach levels sufficient for NF-B inhibition. Furthermore, studies employing catalytically inactive IKK mutants indicate that slight reductions in IKK activity may correlate with significant decreases in NF-B-mediated gene expression (8,9), suggesting that the reduction in TNF-induced IKK activity observed at 1-5 mM NaSal (Fig. 1A) is likely to translate into biologically relevant NF-B inhibition.
Interestingly, we noted that millimolar concentrations of NaSal significantly inhibited p38 MAP kinase activity in vitro (Fig. 5). However, treatment of numerous cell types with millimolar concentrations of NaSal leads to p38 activation rather than to its inhibition (18,19,28). This finding further underscores the notion that the effect of NaSal upon in vitro kinase activity does not reflect the mechanism by which NaSal affects various kinases within the cell. Furthermore, the inhibitory action on p38 in vitro indicates that the effect of NaSal upon in vitro kinase activity is not limited to the IKKs. In agreement with this conclusion, it was recently reported that 20 mM NaSal inhibits RSK2 kinase activity in vitro (43). Overall, our results emphasize the need to exercise caution when attributing the actions of pharmacological agents in intact cells to effects observed in vitro without thoroughly considering potential interactions with various signal transduction components.