Inhibition of interleukin-1beta -induced NF-kappa B activation by calcium/calmodulin-dependent protein kinase kinase occurs through Akt activation associated with interleukin-1 receptor-associated kinase phosphorylation and uncoupling of MyD88.

Calcium/calmodulin-dependent protein kinase kinase (CaMKK) and Akt are two multifunctional kinases involved in many cellular responses. Although Akt and Ca(2+) signals have been implicated in NF-kappaB activation in response to certain stimuli, these results are still controversial, and the mechanism(s) involved remains unknown. In this study, we show the roles that CaMKK and Akt play in regulating interleukin-1beta (IL-1beta)-induced NF-kappaB signaling. In human embryonic kidney 293 cells, IL-1beta induces IkappaB kinase beta (IKKbeta) activation, IkappaBalpha degradation, NF-kappaB transactivation, and weak Akt activation. A CaMKK inhibitor (KN-93) and phosphatidylinositol 3-kinase inhibitors (wortmannin and LY294002) do not inhibit IL-1beta-induced NF-kappaB activation. However, IL-1beta-induced NF-kappaB activity is attenuated by increased intracellular calcium in response to ionomycin, UTP, or thapsigargin or by overexpression of CaMKKc and/or Akt. Ionomycin and CaMKKc overexpression increases Akt phosphorylation on Thr(308) and enzyme activity. Under these conditions or upon overexpression of wild type Akt, IL-1beta-induced IKKbeta activity is diminished. Furthermore, a dominant negative mutant of Akt abolishes IKKbeta inhibition by CaMKKc and ionomycin, suggesting that Akt acts as a mediator of CaMKK signaling to inhibit IL-1beta-induced IKK activity at an upstream target site. We have also identified a novel interaction between CaMKK-stimulated Akt and interleukin-1 receptor-associated kinase 1 (IRAK1), which plays a key role in IL-1beta-induced NF-kappaB activation. CaMKKc and Akt overexpression decreases IRAK1-mediated NF-kappaB activity and its association with MyD88 in response to IL-1beta stimulation. Furthermore, CaMKKc and Akt overexpression increases IRAK1 phosphorylation at Thr(100), and point mutation of this site abrogates the inhibitory effect of Akt on IRAK1-mediated NF-kappaB activation. Taken together, these results indicate a novel regulatory mechanism for IL-1beta signaling and suggest that CaMKK-dependent Akt activation inhibits IL-1beta-induced NF-kappaB activation through interference with the coupling of IRAK1 to MyD88.

The transcription factor NF-B is a key regulator of immune and stress responses in mammals, and NF-B activity increases in response to a variety of stimuli (1,2). Different receptors are known to use distinct combinations of intracellular proteins to initiate NF-B activation; however, some signaling pathways converge downstream into a common pathway that leads to activation of the IB kinase (IKK) 1 complex and the phosphorylation and degradation of IB (inhibitor of NF-B) (3)(4)(5). Although several serine/threonine kinases have been suggested to activate IKK, the NF-B-inducing kinase (NIK) has been identified as the upstream kinase (6).
Interleukin (IL)-1 is a major cytokine responsible for the induction of a number of proteins associated with inflammation (7,8). Many of these responses are activated by the rapid activation of the transcription factor NF-B following signal transduction by IL-1␤ bound to the type I IL-1 receptor (9). Activation of the type I IL-1 receptor leads to recruitment of IRAK to the receptor complex via its association with the IL-1 receptor accessory protein (10) and an adaptor protein MyD88 (11)(12)(13). Upon recruitment, IRAK is highly phosphorylated and subsequently dissociates from the receptor complex to interact with tumor necrosis factor receptor-associated factor 6 (14), which in turn is involved in NIK and NF-B activation (6,15). Phosphatidylinositol 3-kinase (PI3K), a lipid kinase controlled by membrane phospholipid inositol 1,4,5-trisphosphate, and its downstream activating target, Akt, have been identified in many cell types and are implicated in a wide variety of biological responses (16). At present, conflicting reports exist as to the role of PI3K in NF-B activation. The pharmacological effects of PI3K inhibitors, including wortmannin and LY294002, suggest that the activity of NF-B induced by some ligands requires PI3K signaling. These include IL-1 (17)(18)(19), insulin growth factor II (21), bradykinin (22,23), fMet-Leu-Phe (24), tumor necrosis factor ␣ (TNF-␣) (25,26), platelet-derived growth factor (27), and interferon ␣/␤ (28). This suggestion is further supported by evidence that overexpression of constitutively activated forms of either the p110 catalytic subunit of PI3K or Akt results in enhancement of NF-B transactivation induced by IL-1 in HepG2 cells (17,29) and by phorbol 12myristate 13-acetate plus ionomycin in Jurkat T cells (30). Alternatively, overexpression of kinase-dead Akt abrogated platelet-derived growth factor-and TNF-␣-induced IKK and NF-B activation (25,27). One conclusion of these studies is that Akt does not act alone to stimulate NF-B, and signals from other pathways are required (17,18,25,30). In contrast to the positive effects of PI3K/Akt signal cascades on NF-B transcription activity, Park et al. (31) and Pahan et al. (32) showed that wortmannin can increase lipopolysaccharide (LPS)-or cytokine-induced expression of the inducible nitric-oxide synthase gene, whose regulation definitely requires NF-B activation (33). Diaz-Guerra et al. (34) showed a more sustained activation of NF-B by wortmannin in LPS-stimulated macrophages. Despite both positive and negative control on NF-B activity having been proposed, some studies have ruled out the involvement of the PI3K signal cascade in NF-B activation by TNF-␣, IL-1 (22,(35)(36)(37), platelet-derived growth factor (38), and epidermal growth factor (39). Thus, the regulatory role of the PI3K/Akt signal cascade in NF-B activity appears to be cell-and ligand-specific.
Another possible explanation to account for the discrepancies discussed above might be the existence of multiple targeting molecules within the signaling cascades for transduction of NF-B activity being controlled by the PI3K/Akt pathway in different manners. Thus far, the most clearly characterized target for the positive regulatory role of PI3K/Akt is IKK␣. The results from Ozes et al. (25) revealed that IKK␣, but not IKK␤, can serve as a phosphorylation target for Akt at the Thr 23 residue and that this phosphorylation-dependent activation of IKK␣ contributes to NF-B activation by TNF-␣. The results FIG. 1. IL-1␤-induced IKK activation, IB␣ degradation, and NF-B activation in HEK 293 cells. A, 293 cells were incubated with IL-1␤ (10 ng/ ml) for 0 -60 min, and the cell lysates were then immunoprecipitated with antibody specific for IKK␤. One set of immunoprecipitates was subjected to kinase assay (KA) as described under "Experimental Procedures" using GST-IB␣ (5-55) as a substrate (top panel). The other set of immunoprecipitates was subjected to SDS-PAGE and analyzed by immunoblotting (IB) with anti-IKK␤ antibody (bottom panel). The presence of equal amounts of the immunoprecipitated kinase complex in each kinase assay was confirmed by immunoblotting for IKK␤. B, following incubation for different periods with IL-1␤ (10 ng/ml), IB␣ degradation was determined by immunoblotting with IB␣-specific antibody. The results shown are representative of three experiments with similar results. C, 293 cells were transiently transfected with 0.5 g of pGL2-ELAM-Luc and 1 g of pBK-CMV-Lac Z for 24 h, and then the cells were incubated with 0.1-10 ng/ml of IL-1␤ for another 24 h. Luciferase activities were determined as described under "Experimental Procedures." The level of induction of luciferase activity was compared with that of cells without IL-1␤ treatment. The data represent the means Ϯ S.E. of three to seven experiments with all of the reactions performed in duplicate. *, p Ͻ 0.05 as compared with the control without IL-1␤ treatment. D, 293 cells were transiently transfected with 1 g of WT NIK, DN IKK␣, DN IKK␤, or DN NIK for 24 h and then stimulated with IL-1␤ (10 ng/ml) for another 24 h. The NF-B reporter gene assays were performed as in C. The data are representative of three independent experiments. *, p Ͻ 0.05 as compared with the control IL-1␤ or NIK response. from Sizemore et al. (29) also showed that IKK␣ is solely required for IL-1 and TNF-␣-induced phosphorylation and activation of the p65 subunit of NF-B: effects that are mediated by the PI3K/Akt pathway. However, another recent report from Madrid et al. (19) demonstrated that Akt, functioning through IKK␣, IKK␤, and p38 mitogen-activated protein kinase, stimulates the p65 subunit of NF-B, in turn stimulating the transcriptional activity of NF-B. This action of Akt is independent of NF-B translocation, IB phosphorylation, and degradation (18,29). However, a recent finding implies that the involvement of the PI3K pathway in IB-independent NF-B activation is not related to stimulation of IKKs (40).
In addition to the action of PI3K/Akt, calcium-mediated signaling pathways, in particular those involving the calcium/ calmodulin-dependent protein kinases (CaMKs) and calcineurin (PP2B), have been implicated in NF-B regulation. In the case of the CaMKs, we demonstrated previously that increased calcium mobilization in mouse macrophages enhanced the responses of LPS in IKK activation (41,42) and NF-Binitiated gene regulation (42)(43)(44). These potentiating effects initiated by the calcium signal are susceptible to inhibition by the general CaMKs inhibitor, KN-93. Supporting this notion is the finding by Hughes et al. (45) that CaMKII mediates T cell receptor/CD3-and phorbol ester-induced IKK activation. Another indication of positive control by the calcium signal is the observation that histamine-stimulated increase in NF-B reporter gene activity is dependent on the oscillating intracellular calcium frequency (46). As well as acting through IKK activation (41,45), Jang et al. (47) recently showed that CaMKIV can target the p65 subunit of NF-B and stimulate NF-B transactivation. In the case of calcineurin, a serine/ threonine phosphatase controlled by cellular calcium, although its role in regulating NF-B-dependent transcription has been studied, conflicting data have been obtained suggesting that it may act in a cell-specific manner (48 -50).
It was recently demonstrated that Akt is also activated following the phosphorylation of Thr 308 by CaMKK␣, a CaMK kinase, whose activity also depends on the presence of calcium/ calmodulin (51). In this study, we evaluate the roles that PI3K/ Akt and CaMKK play in IL-1␤-induced signaling. We present evidence showing that upon intracellular calcium up-regulation by ionomycin, UTP, or thapsigargin, IL-1␤-induced NF-B activation was concomitantly decreased in human embryonic kidney (HEK) 293 cells. This involves the sequential activation of CaMKK and Akt through a Thr 308 phosphorylation step, which leads to phosphorylation of IRAK1 and a reduction in its association with MyD88 in response to IL-1␤.

EXPERIMENTAL PROCEDURES
Reagents-Dulbecco's modified Eagle's medium, fetal bovine serum, penicillin, and streptomycin were obtained from Invitrogen. IL-1␤ was obtained from PeproTech (London, UK). KN-93 and calcineurin substrate (RII phosphopeptide) (catalog number 207008) were purchased from Calbiochem (La Jolla, CA). [␥-32 P]ATP (6,000 Ci/mmol) and the enhanced chemiluminescence detection agent were purchased from PerkinElmer Life Sciences. Rabbit polyclonal antibodies specific for IB␣, IKK␤, NIK, Akt, IRAK1, MyD88, CaMKK␣, protein A/G beads, and horseradish peroxidase-conjugated anti-mouse and anti-rabbit antibodies were from Santa Cruz Biotechnology (Santa Cruz, CA). Rabbit polyclonal antibodies specific for Akt phosphorylated at Thr 308 and Ser 473 were from Cell Signaling & Neuroscience (St. Louis, MO). Plasmid pGEX-IB␣ (amino acids 5-55) was provided by Dr. Frank S. Lee (Pennsylvania Medical Center). Plasmid pGEX-CaMKIV was a kind gift from Dr. A. R. Means (Duke University, Durham, NC). Plasmid PGEX-p65 (amino acids 354 -551) was provided by Dr. H. Sakurai (Tanabe Seiyaku, Osaka, Japan). FK506 was a kind gift from Fujisawa Pharmaceuticals (Osaka, Japan). Histone H2B was obtained from Roche Molecular Biochemicals. All of the materials for SDS-PAGE were obtained from Bio-Rad. All of the other chemicals were obtained from Sigma.
Cell Culture-HEK 293 cells were cultured at 37°C in a humidified atmosphere of 95% air and 5% CO 2 in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum and antibiotics (100 units/ml of penicillin and 100 g/ml streptomycin).
Expression Vectors-pGL2-ELAM-Luc, which is under the control of one NF-B-binding site, and pBK-CMV-Lac Z were provided by Dr. . Wild type and kinase-dead Akt were provided by Dr. James R. Woodgett (Toronto, Canada). Constitutively active calcineurin catalytic subunit (CNc), which is designed to mimic proteolyzed forms and have calcium independence, was a gift from Dr. G. R. Crabtree (Stanford, CA).
Transfection and B Luciferase Assays-For these assays, 5 ϫ 10 5 293 cells were seeded into 6-well (35-mm) plates. The cells were transfected on the following day by the calcium phosphate precipitation method with 0.5 g of pGL2-ELAM-Luc and 1 g of pBK-CMV-LacZ and, when needed, other expression constructs at the amounts indicated. After 24 h, the medium was aspirated and replaced with fresh Dulbecco's modified Eagle's medium containing 10% fetal bovine serum. The cells were then stimulated with IL-1␤ (10 ng/ml) for another 24 h before harvesting. To assess the effects of several inhibitors, drugs were added to the cells 20 min before IL-1␤ stimulation. Luciferase activity was determined with a luciferase assay system (Promega) and was normalized on the basis of LacZ expression. The level of induction of luciferase activity was determined as a ratio in comparison with cells with no stimulation.
Measurement of Intracellular Calcium-293 cells grown on glass slides were loaded with 3 M Fura-2/acetoxymethylester (AM) and pluronic F-127 (0.02% v/v) in Dulbecco's modified Eagle's medium at 37°C for 45 min. The fluorescence was monitored on a PTI M series spectrofluorometer with dual excitation wavelengths of 340 and 380 nm and an emission wavelength of 510 nm. Intracellular calcium ([Ca 2ϩ ] i ) was calculated from the ratio of the fluorescence at the two excitation wavelengths, using a K d value of 224 nM for the Fura-2/Ca 2ϩ equilibrium, as described by Grynkiewicz et al. (52).
Immunoblot Analysis-After treatment with the indicated agents or after transfection with the indicated plasmids, the cells were washed twice in ice-cold phosphate-buffered saline and then solubilized in buffer containing 20 mM Tris-HCl, 0.5 mM EGTA, 2 mM EDTA, 2 mM DTT, 0.5 mM PMSF, and 10 g/ml leupeptin, pH 7.5. To detect the increased phosphorylated form of Akt, the cells transfected with wild type Akt and with or without CaMKKc were allowed to sit for 1 day before treatment with IL-1␤ or ionomycin for the indicated times. The samples of equal amounts of protein (60 g) were subjected to SDS-PAGE and then transferred onto a nitrocellulose membrane, which was FIG. 2. IL-1␤ induced a weak Akt activation. After the cells were treated with 10 ng/ml IL-1␤ for different intervals, Akt phosphorylation at either Thr 308 or Ser 473 was determined by immunoblotting (IB) with antibody specific for phosphorylated Akt. For Akt kinase activity (KA), the cell lysates were immunoprecipitated with Akt-specific antibody. One set of immunoprecipitates was assayed for KA using Histone H2B as a substrate. The other set of immunoprecipitates was subjected to 8% SDS-PAGE and analyzed by immunoblotting with anti-Akt antibody. The presence of equal amounts of the immunoprecipitated kinase complexes in each kinase assay was confirmed by immunoblotting for Akt. The results shown are representative of four independent experiments. then incubated in TBST buffer (150 mM NaCl, 20 mM Tris-HCl, 0.02% Tween, pH 7.4) containing 1% nonfat milk. The proteins were visualized by specific primary antibodies followed by horseradish peroxidase-conjugated second antibodies. Immunoreactivity was detected by enhanced chemiluminescence following the manufacturer's instructions.
Immunoprecipitation and Protein Kinase Assays-293 cells grown in 60-mm dishes were washed twice with ice-cold phosphate-buffered saline, lysed in 1 ml of lysis buffer containing 20 mM Tris-HCl, pH 7.5, 1 mM MgCl 2 , 125 mM NaCl, 1% Triton X-100, 1 mM PMSF, 10 g/ml leupeptin, 10 g/ml aprotinin, 25 mM ␤-glycerophosphate, 50 mM NaF, and 100 M sodium orthovanadate, and centrifuged. The supernatant was then immunoprecipitated with polyclonal antibody against IKK␤, Akt, CaMKK␣, or IRAK1 in the presence of A/G-agarose beads overnight. The beads were washed three times with lysis buffer and two times with kinase buffer (20 mM HEPES, pH 7.4, 20 mM MgCl 2 , 2 mM DTT). The kinase reactions were performed by incubating immunoprecipitated beads with 20 l of kinase buffer supplemented with 20 M ATP and 3 Ci of [␥-32 P]ATP at 30°C for 30 min. For the IKK␤ kinase assay, 2.5 g of bacterially expressed GST-IB␣ (amino acids 5-55) was added as a substrate. For CaMKK␣ kinase assay, 2 g of bacterially expressed GST-CaMKIV was added as a substrate. For Akt and IRAK1 kinase assays, 100 g/ml histone H2B and 50 g/ml of myelin basic protein were added as the substrates, respectively. The reaction mixtures were analyzed by 12% (IKK␤), 8% (CaMKK␣), or 15% (Akt or IRAK1) SDS-PAGE followed by autoradiography.
In Vitro IRAK1 Phosphorylation by Akt-To assess IRAK1 as the Akt phosphorylation target, IRAK1 and Akt proteins were obtained by immunoprecipitation from 293 cells overexpressing IRAK1 and Akt, respectively. Kinase assays were performed with IRAK1-and Aktconjugated beads in 20 l of kinase buffer supplemented with 20 M ATP and 3 Ci of [␥-32 P]ATP at 30°C for 30 min. The reaction mixtures were analyzed by 8% SDS-PAGE followed by autoradiography.
Co-immunoprecipitation-293 cells plated on 60-mm dishes were transfected with the indicated amounts of expression plasmids. 24 h post-transfection, the cells were stimulated with 10 ng/ml IL-␤, 100 nM ionomycin, or both for 20 min. The cells were then harvested, lysed in 1 ml of PD buffer (40 mM Tris-HCl, pH 8.0, 500 mM NaCl, 0.1% Nonidet P-40, 6 mM EGTA, 10 mM ␤-glycerophosphate, 10 mM NaF, 300 M sodium orthovanadate, 2 mM PMSF, 10 g/ml aprotinin, 1 g/ml leupeptin, and 1 mM DTT), and centrifuged. The supernatant was then immunoprecipitated with polyclonal antibody against Akt or MyD88 in the presence of A/G-agarose beads overnight at 4°C. The immunoprecipitated beads were then washed three times with PD buffer. The samples were fractionated on 8% SDS-PAGE, transferred to a nitrocellulose membrane, and subjected to immunoblot analysis with IRAK1 antibody.
Calcineurin Activity Assay-Calcineurin activity was measured by the malachite green assay system, according to the manufacturer's instructions (Calbiochem). Briefly, confluent cells on 35-mm dishes were washed three times with physiological saline solution (118 mM NaCl, 4.7 mM KCl, 1.8 mM CaCl 2 , 1.2 mM MgCl 2 , 1.2 mM KH 2 PO 4 , 11 mM glucose, and 10 mM HEPES, pH 7.4) and incubated at 37°C for 20 min. After this preincubation, the cells were incubated with or without FK506 (100 ng/ml) for 20 min and then treated with 100 nM ionomycin for another 5 min. The cells were washed three times with physiological saline solution without KH 2 PO 4 , collected on ice, and disrupted by sonication at 4°C in lysis buffer (50 mM Tris, pH 7.5, 0.1 mM EGTA, 1 mM EDTA, 0.5 mM DTT, 50 g/ml PMSF, 5 g/ml leupeptin, and 5 g/ml aprotinin). The homogenate was centrifuged at 12,000 ϫ g for 30 min, and the supernatant was immediately used for calcineurin activity assay. Free phosphate released from calcineurin substrate was determined in the reaction buffer (200 mM Tris, 12 mM MgCl 2 , 1 mM DTT, 0.05% Nonidet P-40, pH 7.5, and 500 nM okadaic acid) at 30°C for 30 min. The reaction was terminated by adding malachite green/Tween solution for another 20 min, and then the OD 620 was measured. Calcineurin activity was determined as pmol/min/mg protein using different concentrations of KH 2 PO 4 as standard.
Statistical Analysis-The values are expressed as the means Ϯ S.E. of at least three experiments. Analysis of variance was used to assess the statistical significance of the differences, and a p value of less than 0.05 was considered statistically significant.

IL-1␤ Induction of IKK Activity, IB␣ Degradation, and
NF-B Activation-In HEK 293 cells, treatment with 10 ng/ml IL-1␤ induced IKK␤ activity in a time-dependent manner, beginning after 5-min incubation and reaching a maximum after 30 min (Fig. 1A). IB␣ is targeted for IKK-specific phosphorylation followed by ubiquitination and proteasome-dependent degradation and thus results in the dissociation of NF-B from IB and its transcriptional activation. As shown in Fig. 1B, in parallel with the rapid onset of IKK activation, IB␣ degradation was apparent after 5 min of incubation with IL-1␤ and displayed a time-dependent reduction within 60 min. To directly determine NF-B activation after IL-1␤ treatment, HEK 293 cells were transiently transfected with pGL2-ELAM-B-Luc as an indicator of NF-B activation. As shown in Fig. 1C, when incubating HEK 293 cells with IL-1␤ (0.1-10 ng/ml) for 24 h, a concentration-dependent increase in B luciferase activity was seen with about 4.5-fold (n ϭ 7) increase at 10 ng/ml IL-1␤. Our results also showed that the three signal transducers currently identified for the activation of NF-B in response to many stimuli, i.e. IKK␣, IKK␤, and NIK, appeared to perform distinct roles in IL-1␤ signaling. In cells transfected with negative mutants of IKK␣, IKK␤, and NIK, IL-1␤-induced B luciferase activity were inhibited by 32 Ϯ 7%, 70 Ϯ 11%, and 33 Ϯ 8%, respectively (Fig. 1D). Compared with the distinct selectivity for IL-1␤ action, IKK␣ and IKK␤ mutants reduced the NIK-induced B response to a similar extent (Fig. 1D).
Slightly IL-1␤-induced Akt Activity Does Not Involve NF-B Activation-To explore whether PI3K-Akt might mediate IL-1␤-induced NF-B activation, we directly measured Akt phosphorylation and activity in response to IL-1␤. Probing of immunoblots with antibodies specific for phosphorylated Akt showed that 10 ng/ml of IL-1␤ slightly increased Akt Thr 308 and Ser 473 phosphorylation within 60 min (Fig. 2). Using histone H2B as an Akt substrate, a slight increase in Akt kinase activity was noted within the same time interval as Akt phosphorylation (Fig. 2). When 293 cells were pretreated for 20 min with the PI3K inhibitors wortmannin (100 nM) and LY294002 (10 M), IL-1␤-induced NF-B activity was unaltered (Fig. 3A). B-luciferase activities induced by overexpression of IKK␤ and NIK were also unchanged (Fig. 3B). In contrast, 10% seruminduced B-luciferase activity in quiescent cells was inhibited by wortmannin (100 nM) and LY294002 (10 M) with 78 Ϯ 8 and 68 Ϯ 7% inhibition, respectively (data not shown). These results suggest that the extent of IL-1␤-stimulated PI3K signaling has no regulatory role in NF-B transactivation.
To determine whether the endogenous activities of CaMKs and calcineurin are involved in NF-B activation, KN-93 (10 M), a CaMK inhibitor, and FK506 (100 ng/ml), a calcineurin inhibitor, were used. As shown in Fig. 3, neither inhibitor altered NF-B activation by IL-␤, IKK␤, or NIK.
Calcium Increase and Akt Overexpression Inhibit IL-1␤-induced NF-B Activation-To investigate the roles of enhanced calcium and Akt signals on NF-B activation, the cells were treated with calcium-increasing agents and/or caused to overexpress Akt. In 293 cells, 100 nM ionomycin (a calcium ionophore), 100 M UTP (a P2Y receptor agonist), and 30 nM thapsigargin (an inhibitor of the endoplasmic reticulum Ca 2ϩ pump) but not IL-1␤, induced a rapid and prominent increase in intracellular calcium, followed by a decrease to a [Ca 2ϩ ] i level that was subsequently maintained (Fig. 4). At the concentrations used, ionomycin, UTP, and thapsigargin elevated [Ca 2ϩ ] i from 121 Ϯ 15 nM to 4102 Ϯ 475, 398 Ϯ 30, and 758 Ϯ 28 nM (n ϭ 5), respectively. Fig. 5 shows that although ionomycin (10 -100 nM), UTP (100 M), and thapsigargin (30 nM) themselves did not affect basal activity of NF-B, their combination with IL-1␤ dramatically reduced NF-B activation. The maximal levels of inhibition achieved were 52 Ϯ 5% with 100 nM ionomycin, 31 ؎ 4% with 100 M UTP, and 59 Ϯ 7% with 30 nM thapsigargin (n ϭ 3). To investigate the downstream mechanism of NF-B inhibition caused by [Ca 2ϩ ] i elevation, the effects of the CaMKK inhibitor KN-93 and the calcineurin inhibitor FK506 were tested. The data revealed that KN-93 (10 M) pretreatment almost completely reversed ionomycin-induced inhibition (Fig. 5A). In contrast, the effect of ionomycin was unchanged by the presence of FK506 (100 ng/ml) (Fig. 5A). These results suggest that both Ca 2ϩ -dependent and KN-93-sensitive signaling elements contribute to ionomycin activation.
CaMKK and Akt Involvement in Inhibition of NF-B Activation-To identify a Ca 2ϩ -mediated and KN-93-sensitive signal element in the inhibition of NF-B activation, we tested whether CaMKK might play a crucial role in the ionomycin response. Fig. 5B shows that CaMKKc alone did not affect basal NF-B activity, but it significantly inhibited IL-1␤-induced NF-B activation by 27 Ϯ 7% (n ϭ 3). The effects of CaMKKc and ionomycin were shown to be nonadditive. On the contrary, DN CaMKK did not change the IL-1␤-induced response. Consistent with the effect of FK506 on ionomycin, overexpression of the constitutively active form of calcineurin, a downstream signal transducer following calcium/calmodulin stimulation, did not significantly affect IL-␤-induced NF-B activation (Fig. 5B). These results suggest that CaMKK plays the role of a primary response element in calcium/calmodulin signaling to inhibit IL-1␤-induced NF-B activation.
Akt Transduction of CaMKK Signaling to Inhibit NF-B and IKK Activity-Akt activation occurs as a result of multiple phosphorylation events on specific residues, including Thr 308 and Ser 473 . Phosphorylation of both residues is prerequisite for full activation of this enzyme. We examined whether CaMKK can phosphorylate Akt on Thr 308 and hence increase Akt activity, as reported previously (51). Immunoblotting analysis indicated that ionomycin and CaMKKc, either alone or together, can increase Akt phosphorylation at Thr 308 but not at Ser 473 , whereas overexpression of wild type Akt markedly increased Akt phosphorylation at both sites (Fig. 7A). Treatment of Aktoverexpressing cells with ionomycin, CaMKKc, or both further increased Akt Thr 308 phosphorylation and left Akt Ser 473 phos- phorylation at a level similar to that seen without ionomycin and/or CaMKKc addition (Fig. 7A). Akt kinase activity assayed in vitro was up-regulated in correlation with increased phosphorylation. Overexpression of CaMKKc and Akt increased Akt activity by 247 Ϯ 16 and 274 Ϯ 12%, respectively, compared with controls (n ϭ 4), and following transfection of both of these genes, Akt enzyme activity was further increased by 400 Ϯ 30% compared with the control (Fig. 7B). In contrast, IL-1␤ treatment alone for 30 min increased Akt activation by only 61 Ϯ 3% (n ϭ 4) (Fig. 7B).
To determine the sites of action for the processes that give rise to the inhibition of NF-B activation via the Ca 2ϩ /CaMKK/ Akt cascade, we first investigated IKK␤, which is known to be a crucial element in NF-B signaling. Fig. 8 shows that IL-1␤induced IKK␤ activity was inhibited by ionomycin, CaMKKc, and Akt and that the inhibitory effects of ionomycin and CaMKKc were reversed by the dominant negative mutant of Akt. These results suggest that the site of Akt action is located in the upstream signaling cascade prior to IKK activation.
CaMKK and Akt Inhibition of IRAK1-induced B Luciferase Activity-We next investigated IRAK1, which has been established as a key signal transducer proximately coupled to the stimulated IL-1 receptor. Fig. 9A shows that IL-1␤ at 10 ng/ml increased IRAK1 activity, which reached a maximum after 10 min and was sustained for at least 90 min. IL-1␤-induced IRAK1 activity was inhibited by treatment with ionomycin or overexpression of CaMKKc and Akt. In contrast, DN CaMKK or DN Akt did not influence IRAK1 activity following IL-1␤ stimulation (Fig. 9B). The possible change in quantity of IRAK1 in cells overexpressing CaMKKc or Akt was ruled out by immunoblotting the immunoprecipitated complex with IRAK1 antibody (Fig. 9, A and B, lower panels). In addition, overexpressed Akt, either the wild type or negative mutant form, did not affect cell viability, which was assessed by both MTT and propidium iodide staining (data not shown). Fig. 9C shows that co-expression of CaMKKc or Akt with IRAK1 inhibited IRAK1-elicited B luciferase activity by 42 Ϯ 5 and 24 Ϯ 7% (n ϭ 4), respectively. Although IRAK1 is necessary for IL-1-induced NF-B activity, some reports have demonstrated that the catalytic activity of IRAK1 is not required for IL-1-dependent signaling (53)(54)(55). To test this notion and elucidate the linkage with Akt inhibition, we determined the B-luciferase response of kinase-dead IRAK1 (K239A). In support of previous findings, kinase-dead IRAK1 (K239A) transfection efficiently triggered B luciferase activity at a level equivalent to that of wild type IRAK1, and this activity was similarly reduced by CaMKKc and Akt (Fig. 9C). These results suggest that the inhibition of IRAK1-dependent signaling by Akt is unrelated to the kinase activity of IRAK1.
Phosphorylation and Uncoupling of IRAK Signaling by Akt-To investigate whether IRAK1 can be a phosphorylation target of Akt, we analyzed the protein sequence of IRAK1 and found a putative site for Akt mediated phosphorylation at Thr 100 , which is within the consensus sequences (RXRXXX(S/ T)X) specific for Akt action. An in vitro kinase assay was carried out on Akt and IRAK1 immunocomplexes (Fig. 10A). The IRAK1 complex was obtained from IRAK1-overexpressing cells, whereas Akt complexes were obtained from vector (first lane), wild type Akt-transfected (second lane), or DN Akt-transfected (third lane) cells. When the IRAK1 complex was mixed with mouse IgG precipitated complex (fourth lane), the low level phosphorylation of IRAK1, as observed previously (54), was attributed to the autophosphorylated state of IRAK1. Apart from this autophosphorylation, we found that phosphorylation of IRAK1 was increased in the presence of endogenous Akt (first lane) and, to a greater extent, in the presence of overexpressed Akt (second lane). In contrast, the Akt immunocomplex containing DN Akt displayed IRAK1 phosphorylation at levels similar to those of the controls (third lane compared with first lane). These results suggest that the active form of Akt might target IRAK1 in vitro.
Next, we wanted to determine whether interaction between IRAK1 and Akt occurs in vivo. An antibody to Akt was used to generate Akt immunocomplex from 293 cells, which were divided into four groups: vehicle treatment, ionomycin treatment, CaMKKc overexpression, and wild type Akt overexpression. Western blots of immunoprecipitated proteins were prepared and probed with antibody to IRAK1 or antibody to Akt. As shown in Fig. 10B, a moderate amount of IRAK1 was co-precipitated together with Akt (first lane). This Akt-associated IRAK1 protein, both of the phosphorylated and nonphosphorylated form, was increased in ionomycin-stimulated cells (third lane), CaMKKc-overexpressing cells (second lane), and wild type Akt-overexpressing cells (fourth lane). These results confirm that IRAK1 and Akt can interact in vivo, particularly following Akt activation.
MyD88 is the adaptor molecule that links the IL-1 receptor intracellular domain with IRAK1 (56). We next examined whether CaMKK/Akt activation interferes with MyD88 binding to IRAK1. Upon stimulation with 10 ng/ml IL-1␤ for 15 or 30 min, IRAK1 co-immunoprecipitation with MyD88 was enhanced, whereas it was reduced by the overexpression of Akt and CaMKKc (Fig. 10, C and D).
After observing the ability of Akt to phosphorylate IRAK1, both in vitro and in vivo, we investigated the putative Akt phosphorylation site of IRAK1 and its significance in IL-1␤ signaling. By using site-directed mutagenesis to substitute Thr 100 of IRAK1 with Ala and performing in vitro kinase assays, we showed that endogenous or transfected Akt induced phosphorylation of WT IRAK1 (Fig. 11A, second and fifth lanes  compared with first and fourth lanes, respectively), whereas the mutant IRAK1 (T100A) failed to be phosphorylated by Akt (third and sixth lanes compared with first and fourth lanes, respectively). In other words, WT IRAK1 but not IRAK1 (T100A) can serve as a target for Akt.
Next, we wanted to determine the effect of IRAK1 (T100A) mutation on Akt-mediated inhibition of the IL-1␤ signaling pathway. Because a high dose of IRAK1 plasmid (1 g) itself induced prominent B luciferase activity (about 6.5-fold) (Fig.  9C) and would therefore mask the effect of IL-1␤, we used a low dose of IRAK1 plasmid (10 ng). Fig. 11B shows that expression of WT IRAK1 and IRAK1 (T100A) not only increased basal B luciferase activity to similar extents (about 80%) but also slightly potentiated IL-1␤-induced B luciferase activity. Under these conditions, the inhibitory effect of Akt overexpression on IL-1␤induced B luciferase activity was still detected in cells expressing WT IRAK1 but not those expressing IRAK1 (T100A).

DISCUSSION
The proinflammatory cytokine IL-1 mediates its effects primarily by reprogramming gene expression in inflamed tissues. It is therefore likely that clarifying the molecular regulation of IL-1-induced gene expression will yield novel molecular targets for anti-inflammatory therapy. Because IL-1-activated NF-B is a key molecule for up-regulating the expression of many inflammatory mediators, it would be interesting to understand the mechanism of signal restriction that occurs proximal to the IL-1 receptor complex, which might involve several different signaling pathways. The contribution of the two IKK isoforms in signalsome activation appears to vary depending on the inducing molecule type of inducers. In accordance with previous results (57), our data confirm that IL-1␤ predominantly stimulates IKK␤ rather than IKK␣. On the other hand, when signalsome activation is induced directly by activating the upstream kinase NIK, IKK␣ and IKK␤ are both equally important (5). Indeed, previous studies have shown the differential effects of LPS and TNF-␣ on signalsome activation, with a greater effect of TNF-␣ on IKK␣ than on IKK␤ and a greater effect of LPS on IKK␤ than on IKK␣ (58,59). All of these studies indicate that activation of the IKK complex by a variety of different inducers might proceed in different manners and receive differential preferences for signaling cross-talk.
NF-B activation is distinctly regulated in response to specific cytokines in different cell types. Several studies have demonstrated that PI3K/Akt plays a positive role in NF-B activation in response to certain stimuli (17,18,(22)(23)(24)29), whereas other studies have proposed conflicting results (31)(32)(33)(34)(35)(36)(37)(38)(39). In IL-1␤-treated HepG2 cells and mouse embryo fibroblasts, previous reports have shown that the PI3K/Akt pathway plays a crucial role in phosphorylating and transactivating the p65 subunit of NF-B and that this action requires IKK␣ (18,29). However, in this study using HEK 293 cells, we ob-tained quite distinct results. We found that in HEK 293 cells IL-1␤ induced only weak Akt phosphorylation and kinase activity, which did not appear to contribute to NF-B activation. This conclusion is based on the ineffectiveness of the PI3K inhibitors wortmannin and LY294002 on IL-1␤-induced activation of NF-B. However, we did find that overexpression of Akt negatively regulated IL-1␤-induced NF-B activity in HEK 293 cells. These somewhat contradictory results indicated that multiple signaling molecules involved in NF-B activation are targeted and regulated by Akt in different manners (19,25). In this context, we identified a novel negative regulatory mechanism that inhibits the IKK activity induced by IL-1␤. Overexpression of Akt in HEK 293 cells markedly increased Akt Thr 308 and Ser 473 phosphorylation and Akt kinase activity, and under these conditions significant inhibition of IL-1␤-induced IKK activity and NF-B activation was detected. These results suggest that Akt may act in a negative feedback loop to inhibit IL-1␤-induced NF-B activation, especially when cells are exposed to IL-1␤ and other inflammatory stimuli, which together trigger Akt activity. Certainly, previous and present results indicate that the exact role and mechanism of action of Akt in IL-1␤-mediated NF-B signaling cascades varies depending on cell type. It appears that cell type determines the endogenous potency of the Akt signal pathway induced by IL-1␤ and the Our results also showed that intracellular Ca 2ϩ signaling triggered by ionomycin, UTP, and thapsigargin leads to inhibition of IL-1␤-induced NF-B activation. Nucleotides, including ATP and UTP, are inflammatory modulators, because they can be released from damaged cells at inflammatory sites (60,61). In parallel with the effect on NF-B activity, ionomycin inhibits IL-1␤-induced IKK activity. Pharmacological analysis revealed a requirement for KN-93-sensitive CaMKs, but not calcineurin, in this signal regulation. This notion is further supported by molecular evidence showing the inhibition of NF-B and IKK by CaMKKc.
It was recently reported that not only PDK1, but also CaMKK, can activate Akt via phosphorylation of Thr 308 , which is located in the activation loop of Akt (51). Our results confirm the downstream signaling role of Akt in CaMKK-mediated inhibition of NF-B on the basis of the following evidence. First, ionomycin and CaMKKc can induce Akt phosphorylation at Thr 308 , but not Ser 473 , and in turn produce increased Akt activity. Second, IKK and NF-B inhibition by CaMKKc mimics that by ionomycin and is abrogated by kinase-dead Akt. Third, wild type Akt overexpression provides more available signal transducers for CaMKKc action, thus creating additive interaction in Akt phosphorylation, kinase activation, and NF-B inhibition.
IRAK is a central element in IL-1-mediated signaling. Permanent overexpression of IRAK1 induces NF-B activation, whereas IRAK-deficient murine embryonic fibroblasts show a dramatically reduced response to IL-1 (62). Activation of the type I IL-1 receptor leads to the recruitment of IRAK to the receptor complex via MyD88, which is an adapter molecule that links the IL-1 receptor intracellular domain with IRAK1 (13,56). Upon recruitment, IRAK is highly phosphorylated either as a result of autophosphorylation (11) or by an unidentified kinase (63). Phosphorylated IRAK then leaves the receptor complex to interact with tumor necrosis factor receptor-associated factor 6, propagating to the sequential signaling molecules NIK and IKK for NF-B activation (11,64). Recent studies exploring the importance of IRAK phosphorylation in signal transduction have indicated that the role of IRAK in IL-1mediated signaling does not require its kinase activity (55,63,64). Instead, phospho-IRAK might play a negative regulatory role. It was shown that MyD88 has a high affinity for hypophosphorylated IRAK (5). In addition, the phosphorylation of IRAK appears to direct degradation through proteolysis, thereby providing a possible negative feedback mechanism for regulating the IL-1 signaling pathway (51,62).
Our investigation of the interaction between Akt and the MyD88-DRAK complex showed that IRAK1 can be phosphorylated at Thr 100 by Akt both in vivo and in vitro. The association between Akt and IRAK1 was demonstrated in vivo, and it was observed that when Akt is activated, IRAK1 undergoes phosphorylation. We further showed that this association and interaction with Akt might lead to less IRAK1 being recruited to MyD88 upon IL-1␤ stimulation, thus limiting its capacity and/or ability to mediate downstream NF-B signaling. Furthermore, experiments conducted with the IRAK1 (T100A) mutant confirmed the requirement of this phosphorylation site for Akt-mediated inhibition of IL-1␤ signaling. This mutation did not affect the ability of IRAK1 to up-regulate NF-B signaling pathways or its potentiating effect on the IL-1␤ response. Moreover, although LPS (20) and IL-1␣ (64) activation have been shown to reduce the quantity of IRAK1, our present study did not detect any changes in IRAK1 protein levels following IL-1␤ stimulation and Akt transfection (data not shown).
It should be noted that transfection with CaMKKc/Akt was found to inhibit B luciferase activity in HEK 293 cells. Using an IL-1␤-induced B luciferase reporter assay, simultaneous CaMKKc and Akt overexpression was shown to result in about 50% inhibition, whereas ionomycin and thapsigargin treatment give rise to 50 -60% inhibition. This partial antagonism or the absence of total B inhibition by CaMKK/Akt/IRAK1 may be due to the requirement for other molecules in this inhibitory response. There is increasing evidence for IB-independent control of NF-B activation, which involves phosphorylation of the p65 subunit of NF-B. Phosphorylation of the p65 subunit leads to enhanced transactivation ability, and a number of kinases, including CaMKIV (47), have been shown to exert this action. Currently, although there is no evidence for a direct interaction between Akt and p65, there is strong evidence that the PI3K/Akt pathway is involved in IB-independent positive regulation of NF-B through IKK␣/␤ and p38 mitogen-activated protein kinase (19). To clarify the possibility that this p65 phosphorylation mechanism might be involved in CaMKK and Akt action, thus limiting their inhibitory effects on NF-B activity, we performed in vitro kinase assays. In experiments using GST-p65 (354 -551) as a substrate, we found that wild type CaMKK can directly phosphorylate the p65 subunit, but Akt cannot (data not shown). At present, we still cannot rule out the involvement of Akt downstream signaling (such as p38 mitogen-activated protein kinase, GSK-3, protein kinase C, etc.) in the modulation of p65 transactivation either directly or indirectly. Nonetheless, the effects of p65 phosphorylation by CaMKK at least might explain its partial inhibition of IL-1␤-induced NF-B activity.
Taken together, the results of the present study indicate the existence of a novel regulatory signaling pathway in which increased intracellular calcium and activated Akt may exert