Calmodulin-dependent Kinase II Mediates T Cell Receptor/CD3- and Phorbol Ester-induced Activation of IκB Kinase*

Numerous fundamental biological processes involve the NFκB family of transcription factors. The mechanisms by which this family of proteins is regulated are therefore of widespread importance. In most cells, NFκB is bound to inhibitory IκB proteins and sequestered in the cytoplasm. NFκB-inducing signals result in activation of a large multisubunit kinase complex, IKK, which phosphorylates IκB. IκB is subsequently degraded, releasing NFκB, which translocates to the nucleus. We previously reported that inhibitors of the calcium-binding protein calmodulin (CaM) prevent phorbol ester-induced phosphorylation of IκB. Here we show that KN93, an inhibitor of CaM-dependent kinases (CaMKs), also inhibits the phosphorylation of IκB. The effect of both CaM and CaMK inhibitors on IκB phosphorylation is due to the inhibition of the activity of CaMK II because neither drug has any effect when a derivative of CaMK II that is insensitive to these inhibitors is expressed. When CaMK II is inhibited, phorbol ester is no longer able to activate IKK, placing CaMK II in the signaling pathway that leads to IKK activation. CaM and CaMK inhibitors also block T cell receptor/CD3-induced activation but have no effect on the ability of the cytokine tumor necrosis factor α or the phosphatase inhibitor calyculin A to induce degradation of IκB. Finally we show that expression of a constitutively active CaMK II results in the activation of NFκB. The results identify CaMK II as a mediator of IKK activation specifically in response to T cell receptor/CD3 and phorbol ester stimulation.

NFB is a family of eukaryotic transcription factors that is expressed in virtually all cell types and implicated in the regulation of numerous genes (for review, see Ref. 1). Fundamental processes such as cell growth, apoptosis, and development are regulated by NFB, and NFB is a central mediator of immune, inflammatory, and stress responses (1)(2)(3)(4)(5). NFB is primarily regulated by a family of inhibitory IB proteins (6,7). IB binds to NFB and masks its nuclear localization sequence, preventing it from being transported to the nucleus. NFBactivating signals result in the rapid destruction of IB, exposing the nuclear localization sequence of NFB, which directs NFB to the nucleus where it can act on its target genes. The destruction of IB is initiated by its phosphorylation on specific serine residues, labeling it for degradation through the ubiquitin-proteasome pathway (7). This initiating phosphorylation of IB is mediated by a large kinase complex termed IKK. 1 IKK is composed of a heterodimer of two kinases, IKK␣ and IKK␤, an undefined number of the non-kinase protein IKK␥ (also denoted NEMO or IKKAP1), and possibly other components (7)(8)(9). IKK␥ is required for the assembly of the large complex and is indispensable for its activity (10). IKK␣ and IKK␤ can both directly phosphorylate IB (11), although there is genetic and biochemical evidence that these two kinases have distinct roles. IKK␤ mediates NFB activation in response to proinflammatory cytokines (12), a process that does not require IKK␣ (13,14). IKK␣, on the other hand, is involved in various aspects of embryonic development (13)(14)(15).
IKK␣, -␤, and -␥ are all phosphorylated in response to the cytokine tumor necrosis factor ␣ (TNF␣) (16), and treatment of IKK purified from TNF␣-stimulated cells with protein phosphatase 2A results in a loss of kinase activity (17). Inducibly phosphorylated sites of IKK␣ and IKK␤ have been mapped to serine residues in the activation loop of their kinase domains and have been shown to be essential for the activity of each kinase (16,18,19). With a diversity of signaling pathways leading to the activation of NFB (1), it is not surprising that a number of proteins have been identified that either directly or indirectly activate IKK (for reviews, see Refs. 8 and 9).
The calmodulin-dependent kinases (CaMKs) are a large family of structurally related proteins that are dependent on the calcium-binding protein calmodulin (CaM) for their activation (20,21). The catalytic domain of a CaMK is followed by a CaM regulatory domain, which is comprised of an autoinhibitory and a CaM binding domain. The catalytic site is normally sequestered by the autoinhibitory domain through an intramolecular interaction, keeping the kinase in an inactive state. When intracellular calcium (Ca 2ϩ ) levels rise, Ca 2ϩ binds to and induces a conformational change in CaM that allows it to bind to, among a diversity of other targets, the CaM binding domain of the CaMK. This disrupts the interaction between the autoinhibitory and catalytic domains, activating the kinase. One of the best characterized CaMKs is the multifunctional CaMK II (20 -22). CaMK II contains in its C terminus an association domain through which it forms multimers of 8 -12 kinase subunits (20,23). Upon activation by Ca 2ϩ /CaM, CaMK II phosphorylates not only exogenous substrates but also a threonine residue (Thr-286) in the autoinhibitory domain of the neighboring subunit of the CaMK II multimer (20 -22). This phosphorylation prevents the autoinhibitory domain from reassociating with the kinase domain and results in Ca 2ϩ /CaM-independent kinase activity. CaMK II is eventually inactivated by removal of this phosphate by a CaMK-dedicated phosphatase (24,25).
We previously reported that activation of transcription by NFB in response to phorbol ester is blocked by inhibitors of CaM and that this inhibition is due to the prevention of IB␣ phosphorylation (26). This prompted us to investigate whether a CaMK is involved in the signals leading to IB phosphorylation. We report here that this is indeed the case and identify CaMK II as a mediator of IKK activation specifically in response to T cell receptor/CD3 and phorbol ester stimulation.
Plasmids-The eukaryotic expression plasmids for human wild type and T286D mutated CaMK II ␥B and the parental expression plasmid pSR␣.BKS have been described previously (27). For expression in Escherichia coli, the IB␣ cDNA (28) was subcloned into pET-20bϩHis (29).
Cell Culture and Transient Transfections-Jurkat T cells were grown in RPMI medium supplemented with 5% fetal calf serum and antibiotics. Cells were transiently transfected with 10 g of expression plasmid by electroporation as described previously (26). 24 h after transfection, cells were harvested or treated with drugs for the indicated times and then harvested.
Analysis of IKK-Immunoprecipitation and analysis of IKK activity was based on the protocol of Trushin et al. (30). Cells were resuspended in lysis buffer (40 mM Tris (pH 8.0), 0.3 M NaCl, 0.1% Nonidet P-40, 6 mM EDTA, 6 mM EGTA, 10 mM NaF, 10 mM sodium pyrophosphate, 10 mM ␤-glycerophosphate, 0.5 mM Na 3 VO 4 , 1 mM dithiothreitol, and protease inhibitor mixture tablets without EDTA (Roche Molecular Biochemicals)). Lysates were adjusted to 0.5 M NaCl, and 200 g of protein were incubated with 2 g of anti-MKP1 M-18 antibody (Santa Cruz Biotechnologies, Inc.) for 1 h at 4°C. This antibody has been used to purify IKK (18) and was more efficient at immunoprecipitating IKK than the anti-IKK␣ M-280 antibody (data not shown). The mixture was then incubated with protein A-Sepharose for 1 h at 4°C, washed three times with lysis buffer containing 0.5 M NaCl, and washed once with 50 mM Tris (pH 7.4) buffer containing 40 mM NaCl. The Sepharose was resuspended in 20 mM HEPES (pH 7.4), 2 mM MgCl 2 , 2 mM MnCl 2 , 10 M ATP, 10 mM NaF, 10 mM sodium pyrophosphate, 10 mM ␤-glycerophosphate, 0.5 mM Na 3 VO 4 , 1 mM dithiothreitol, and protease inhibitor mixture tablets without EDTA. 18 Ci of [␥-32 P]ATP and 5 g of IB␣ were added to each reaction incubated at 30°C for 30 min. The reaction was stopped by adding sample buffer and boiling the samples at 95°C. Unincorporated [␥-32 P]ATP was removed using Micro Bio-Spin 6 chromatography columns (Bio-Rad). Reactions were separated by SDS-polyacrylamide gel electrophoresis. The bottom part of the gel was Coomassie-stained to confirm an equal amount of IB in each lane (data not shown), dried, and exposed to x-ray film. The top part of the gel was analyzed for IKK␣ by Western blot analysis.
Western Blot Analysis-Western blot analysis of IB␣ from cytoplasmic extracts was as described previously (26). Immunoprecipitated IKK␣ was detected using the anti-IKK␣ M-280 antibody (Santa Cruz Biotechnologies, Inc.) and the SuperSignal chemiluminescence substrate (Pierce).
Immunohistochemical Analysis-After transient transfection, cells were incubated for 4 h and then subjected to Lymphoprep (Nycomed Pharma) to remove dead cells. After 20 h of further incubation, cells were harvested onto slides by cytocentrifugation or treated with PMA for 30 min and then harvested. Cells were fixed with ice-cold methanol, permeabilized with 0.1% Triton X-100, and blocked with 1 mg/ml bovine serum albumin in phosphate-buffered saline. Cells were incubated over night at ϩ4°C with the primary antibodies anti-IB␣ (C-15) and anti-CaMKII␥ (C-18) (both from Santa Cruz Biotechnologies, Inc.) diluted 1:50 in phosphate-buffered saline with 1 mg/ml bovine serum albumin. Cells were rinsed and incubated for 4 h at room temperature with the secondary antibodies fluorescein isothiocyanate-conjugated donkey anti-rabbit IgG and Rhodamine Red X-conjugated donkey anti-goat IgG (both from Jackson Immunoresearch Laboratories) diluted 1:50 in phosphate-buffered saline with 1 mg/ml bovine serum albumin. Cells were rinsed and mounted in a medium containing Citifluor (Chemical Laboratory, The University of Kent, Kent, UK) as an antifading agent. Cells were visualized by confocal laser scanning microscopy using a Leica SP2 confocal microscope equipped with an argon and a HeNe laser (Leica Laser Technik). Images were acquired sequentially using the 488 and 546 nm laser lines to excite fluorescein isothiocyanate and Rhodamine Red dyes, respectively, with an ϫ63 oil immersion PL APO objective. Data presented in the same figure were registered with the same laser and multiplier settings.

Phorbol Ester-induced Phosphorylation of IB Requires
Calmodulin-dependent Kinase II-We previously reported that phosphorylation and subsequent degradation of IB␣ induced by the phorbol ester mitogen PMA was blocked by a number of CaM inhibitors including W7 (26) (Fig. 1A). To investigate whether this was due to the involvement of a CaMK, we analyzed the effect of the CaMK II inhibitor KN93 on the ability of PMA to activate NFB. Like W7, KN93 prevented the phosphorylation and subsequent degradation of IB␣ in Jurkat T cells ( Fig. 1B and data not shown). An inactive analogue of KN93, KN92, had no effect (Fig. 1B), indicating that the inhibitory effect of KN93 is due to its interaction with CaMK II. These results suggest that CaM inhibitors block IB␣ phosphorylation because CaMK II is involved.
CaMK II is normally present in an inactive conformation due to an intramolecular interaction between its catalytic domain and its autoinhibitory domain ( Fig. 2A). Ca 2ϩ /CaM activates the kinase by binding near the autoinhibitory domain and releasing the catalytic domain ( Fig. 2A). W7 binds directly to CaM and prevents it from interacting with its targets (31) (Fig.  2B, top left). KN93 binds to the CaM binding domain of CaMK II and prevents Ca 2ϩ /CaM from binding to and activating the kinase (32, 33) (Fig. 2B, top left). However, once CaMK II is activated it becomes autophosphorylated on threonine 286 and is no longer dependent on CaM for its activity (34, 35) ( Fig. 2A) and will therefore not be inhibited by KN93 or W7. A constitutively active kinase resembling the autophosphorylated form is generated by mutating Thr-286 to aspartic acid (T286D) (27) (Fig. 2B, top right). We used this information to address the specificity of KN93 and W7. We argued that if KN93 and W7 are acting on CaMK II, then transient expression of CaMK II T286D would override their ability to inhibit IB␣ degradation. A sufficiently high proportion of the cells would have to be transfected to detect any effect on IB␣ degradation in the cell extract. Fluorescence-activated cell sorting analyses of Jurkat cells transfected with different green fluorescent protein fusion constructs revealed that indeed most of the cells (Ͼ80%) had taken up DNA and expressed the fluorescent proteins, although at varying levels (data not shown). When the Jurkat cells were transiently transfected with CaMK II T286D expression plasmid, we found that KN93 and W7 no longer had any effect on PMA-induced degradation of IB␣, whereas the inhibitors displayed the expected effect in cells transfected with an empty expression vector or wild type CaMK II plasmid (Fig.  2B). We conclude that expression of CaMK II T286D is sufficient to override the effect of the inhibitors and thus that CaMK II is a critical component of the pathway leading to IB degradation.
Calmodulin-dependent Kinase II Is Specifically Required by Stimuli That Activate NFB through Protein Kinase C-dependent Pathways-There are numerous stimuli that activate NFB, and although most of these result in the activation of IKK and consequent phosphorylation of IB, the early events of their signaling pathways are often quite different. To determine the specificity of the requirement of CaMK II, we examined the effect of W7 and KN93 on stimuli that activate NFB through distinct signaling pathways. Neither drug had any effect on IB␣ degradation induced by the cytokine TNF␣ or the phosphatase inhibitor calyculin A (Fig. 3). This suggests that CaMK II acts in the phorbol ester signaling pathway before a step that is common to different NFB inducers. W7 and KN93 also blocked PMA-but not TNF␣-induced degradation of IB␣ in the early erythroleukemia cell line K562 (data not shown) suggesting that CaMK II is a general, rather than cell-type specific, requirement of phorbol ester-induced activation of NFB.
PMA binds to and activates protein kinase C (PKC), but PMA can also have PKC-independent effects in some systems. To analyze if PMA induction of IB degradation in Jurkat cells was dependent on PKC activity, we analyzed the effect of the specific PKC inhibitor GF109203X (36). Fig. 4A shows that PMA-induced degradation of IB␣ in Jurkat cells is blocked by the PKC inhibitor. Since inhibition occurs already at 50 nM, a concentration at which GF109203X has been reported not to act on any other kinase (36), we conclude that PMA induction of phosphorylation and degradation of IB␣ is dependent on a PKC-initiated pathway in Jurkat cells. GF109203X had no effect on the ability of TNF␣ or calyculin A to induce degradation of IB␣, showing that these stimuli activate NFB independently of PKC (Fig. 4B). Taken together, these data suggest that CaMK II is specifically required for the activation of NFB in response to mitogenic stimulation and that it acts downstream of PKC.
Stimulation of the T cell receptor (TCR)/CD3 complex activates PKC-and Ca 2ϩ -dependent pathways that synergistically activate NFB by inducing the phosphorylation and degradation of IB (30,(37)(38)(39)(40). We therefore asked if induction of degradation of IB␣ by stimulation of the TCR/CD3 complex is blocked by inhibitors of CaM or CaMK II. Jurkat T cells were stimulated by cross-linking the TCR/CD3 complex with antihuman CD3 antibody in the absence or presence of W7 or KN93. Both the CaM and CaMK II inhibitor resulted in a complete block of TCR/CD3-induced degradation of IB␣ (Fig.  5A). Induction of degradation of IB␣ by cross-linking the TCR/ CD3 complex was as sensitive as PMA induction to the PKC inhibitor GF109203X (Fig. 5B).
Phorbol Ester-induced Activation of IKK Requires Calmodulin-dependent Kinase II-Like most analyzed NFB-activating signals, PMA-induced phosphorylation of IB is mediated by IKK (30,41). The activity of IKK can be measured by immunoprecipitating it from cells and analyzing its ability to phosphorylate exogenous IB in an in vitro kinase assay. When immunoprecipitated from cells stimulated with PMA, IKK showed an increased ability to phosphorylate IB␣ compared with IKK immunoprecipitated from unstimulated cells (Fig.  6A). However, both W7 and KN93 inhibited this induction of IKK activity (Fig. 6A). None of these drugs affected the efficiency of IKK immunoprecipitation because each sample contained the same amount of IKK␣, one of the components of the IKK complex (measured by Western blot analysis, Fig. 6B). CaMK II is therefore part of a PMA-induced PKC-dependent signaling pathway that leads to activation of IKK.
Expression of Constitutively Active CaMK II Results in NFB Activation-One of the target genes activated by NFB is IB␣. Newly synthesized IB␣ can enter the nucleus, remove NFB

FIG. 2. KN93 and W7 inhibition of IB␣ degradation is prevented by inhibitor-resistant CaMK II.
A, schematic diagram of how CaMK II is activated. The autoinhibitory domain sequesters the catalytic site by an intramolecular interaction, which is disrupted upon binding of Ca 2ϩ /CaM. B, top, a schematic diagram explaining the effects of W7 and KN93 on wild type CaMK II and their lack of effects on a constitutively active derivative of CaMK II (CaMK II T286D). See "Results" for details. Bottom, expression of CaMK II T286D abolishes the ability of KN93 and W7 to inhibit IB␣ degradation. Jurkat cells were transiently transfected with the indicated expression plasmids followed by stimulation with PMA (ϩ) for 30 min in the absence or presence of increasing concentrations of KN93 (5, 10, 15, and 30 M) or W7 (2.5 and 10 g/ml). IB␣ was detected by Western blot analysis. from DNA, and export the complex back to the cytoplasm to restore the original inactive state of NFB in the cell (42,43). To further highlight the role of CaMK II in the signaling pathway leading to NFB activation, the intracellular localization of IB␣ in CaMK II-overexpressing cells was assessed by immunohistochemical analysis. Jurkat cells were transiently transfected with different CaMK II expression vectors or empty vector and stained for IB␣ (green) and CaMK II (red). In cells transfected with empty expression vector, IB␣ was localized mainly to the cytoplasm (Fig. 7A). Stimulation with PMA resulted in a dramatic decrease in cytoplasmic IB␣ and the appearance of IB␣ in the nucleus (Fig. 7B). This is presumably the result of PMA-induced degradation of IB␣, activation of NFB, NFB-induced resynthesis of IB␣, and the subsequent transport of the newly synthesized protein to the nucleus. Overexpression of wild type CaMK II resulted in a more even distribution of IB␣ throughout the cell with a slight predominance in the nucleus (Fig. 7C). This could be because overexpression of a wild type CaMK II, although Ca 2ϩ /calmodulindependent, can enhance the ability of the cell to respond to present amounts of Ca 2ϩ and calmodulin (and perhaps other inducing factors). When expressing the constitutively active T286D mutant of CaMK II, we observed a dramatic increase in nuclear IB␣ (Fig. 7D) that was even more pronounced than that observed in PMA-stimulated cells (Fig. 7B). This nuclear redistribution can be blocked by treatment with the proteasome inhibitor lactacystin (data not shown), supporting that the action of CaMK II is through direct activation of NFB. The effect of CaMK II T286D expression was present in most cells in the transfected culture and not only in the cells that are most heavily expressing the constitutively active CaMK (Fig. 7D). As mentioned above, most (Ͼ80%) of the cells in the transiently transfected cell cultures were expressing protein from the transfected plasmid, albeit at varying levels, possibly explaining why an effect is seen in most cells. Furthermore, Jurkat cells treated with conditioned medium from CaMK II T286D-express-ing cells showed a nuclear distribution of IB␣ similar to that seen in Fig. 7D (data not shown), implying that the constitutively active kinase leads to expression and secretion of NFB-activating products. A likely explanation is the known NFB induction of a number of genes whose products also are NFB activators (1). This interpretation is supported by the inhibition of the conditioning of the medium by lactacystin, an inhibitor of IB degradation (data not shown). We conclude that overexpression of wild type CaMKII and in particular expression of constitutively active CaMKII results in the activation of NFB.

DISCUSSION
It is well established that NFB is involved in a plethora of biological systems, justifying the need to understand the mechanisms by which this family of proteins is regulated. The phosphorylation and subsequent degradation of the NFB inhibitor IB is a key control point. The IB kinase IKK has therefore become a subject of intense interest, and attention is now focusing on how IKK is regulated. Initiated by our finding that CaM inhibitors prevent phorbol ester-induced activation of NFB by blocking phosphorylation of IB␣ (26), we here investigated whether this NFB activation involves a CaMK. We conclude that CaMK II mediates IKK activation specifically in response to TCR/CD3 and phorbol ester stimulation based on the following observations: (i) phosphorylation and degradation of IB␣ in response to PMA is blocked by the CaMK II inhibitor KN93 but not by its inactive analogue KN92, (ii) expression of a constitutively active derivative of CaMK II that is insensitive to CaM and CaMK II inhibitors abolishes the effect of these drugs on IB␣ degradation, (iii) CaM and CaMK II inhibitors also block TCR/CD3-but not TNF␣-and calyculin A-induced degradation of IB␣, (iv) CaM and CaMK II inhibitors prevent the activation of IKK in response to PMA, and (v) expression of constitutively active CaMK II results in NFB activation.
CaMK II is in itself a large family of proteins. There are four genes encoding CaMK II in vertebrates (␣, ␤, ␥, and ␦), and each of these is subject to extensive alternative splicing (20,21). CaMK II ␣ isoforms are specifically expressed in the brain, and ␤ isoforms are expressed in the brain and a few other tissues, whereas ␥ and ␦ are more broadly expressed. The ␥ isoform used in our studies was in fact cloned from Jurkat T cells (27) and is so far the only CaMK II known to be expressed in lymphocytes. We cannot, however, exclude the possibility that the PMA-induced IKK activation is mediated by another CaMK II isoform and that we see an effect when expressing the mutant ␥ isoform because of a functional redundancy between this isoform and a hypothetical other CaMK II isoform.
We have found that PMA-induced activation of NFB is dependent on CaMK II in the absence of a Ca 2ϩ signal. Upon co-stimulation with the Ca 2ϩ ionophore ionomycin, NFB activation is also blocked by calmodulin inhibitors (26) and the CaMK II inhibitor KN93 (data not shown). It is an intriguing question how an enzyme activated by Ca 2ϩ -loaded CaM can also be required in the absence of a co-stimulus increasing the Ca 2ϩ level. The fact that we observe inhibition with CaM inhibitors argues against this being due to the well established mechanism of CaM-independent kinase activity through autophosphorylation of Thr-286 (20 -22). It rather indicates that, in this case, CaM may either function independently of Ca 2ϩ , in line with many previously reported Ca 2ϩ -independent CaMregulated processes, or that CaM/CaMK II may have a higher affinity for Ca 2ϩ than that in typical CaMK II-regulated processes. The latter possibility has actually been demonstrated for a CaMK, myosin light chain kinase, where the affinity of CaM for Ca 2ϩ is increased upon binding to the kinase and further increased when the CaMK binds its substrate (44,45). Increased Ca 2ϩ affinity or Ca 2ϩ independence could perhaps be facilitated through specific association with another protein(s) in the signaling pathway.
So what is the signaling pathway that leads to IKK activation via CaMK II? We have shown that stimuli that are dependent on PKC (TCR/CD3 cross-linking and PMA) are also dependent on CaMK II, whereas stimuli independent of PKC (TNF␣ and calyculin A) are also independent of CaMK II. It has recently been shown that bradykinin, a proinflammatory peptide ligand, induces transcription of interleukin-6 in an NFBdependent manner (46). The authors showed that interleukin-6 production is blocked by both KN93 and inhibitors of PKC, indicating that, as we have shown here, a signaling pathway involving both PKC and CaMK II leads to activation of NFB. These correlations suggest a link between PKC and CaMK II, and we show here that the PKC-and CaMK II-dependent step is in the pathway to activation of IKK. Some of the PKC isoforms have been shown to bind IKK and possibly directly phosphorylate IKK␤ in its activation loop (47). With the data presented here, an intermediate step between PKC and IKK involving CaMK II has to be envisaged. Interestingly PKC has been shown to directly phosphorylate CaMK II in vitro (48). The authors provided evidence that this phosphorylation occurs on Thr-286 of CaMK II, and it is possible that this results in kinase activation in vivo. One possible model would therefore be that PKC directly activates CaMK II, which then activates IKK.
We have shown that CaMK II is required for TCR/CD3 and PMA signaling but not for TNF␣ or calyculin A signaling to IB. With over 150 ways of activating NFB (1), it is understandable that different stimuli utilize different proteins in their signaling pathways. Other proteins specifically involved in mitogenic activation of NFB have recently been identified.
In fibroblasts, the kinase Akt was shown to bind to and activate IKK␤ in response to platelet-derived growth factor but not TNF␣ or the phorbol ester mitogen 12-O-tetradecanoylphorbol-13-acetate (49). In Jurkat cells, inactivation of Akt by Ly294002, an inhibitor of the Akt-activating phosphatidylinositol 3-kinase, delayed but did not inhibit PMA-induced IB␣ degradation (data not shown). Although elusive, this may indicate that Akt is also involved in the CaMK II-dependent signaling pathway described herein. Another kinase shown to mediate the activation of NFB in response to T cell co-stimulation, but not TNF␣ or interleukin-1, is the mixed-lineage group kinase 3 (50), providing another candidate with which CaMK II may cooperate in this signaling pathway. TBK1 (51), also cloned as NAK (41) and T2K (52), is yet another kinase that is implicated in the activation of NFB. This kinase has been reported to be required for the activation of IKK␤ specifically in response to PMA and platelet-derived growth factor (41), although a subsequent study argues that it instead acts at the level of regulating NFB-dependent transactivation (52). If TBK1/NAK/T2K does act in the pathway leading to the activation of IKK, it remains to be determined whether it acts in concert with CaMK II or if these kinases are cell-type specific mediators of mitogenic activation of NFB. Another recently cloned IKK-related kinase, IKK⑀ (53), also cloned as IKKi (54), is required for NFB activation in response to PMA and TCR stimulation but not cytokines. This is an IB kinase that exists together with another, as yet unidentified IB kinase in a complex distinct from the "classical" IKK consisting of IKK␣, IKK␤, and IKK␥. The authors also showed that both the classical IKK and IKK⑀ are necessary for PMA and TCR activation of NFB, but how the actions of these kinases are linked is unknown. Since Jurkat cells were also used in their study, CaMK II has to be placed somewhere in this apparently complex signaling network. It has also been shown that both PMA stimulation and overexpression of Raf, which is an effector kinase of Ras, activates IKK␤ through the membrane shuttle kinase MEKK1 (55). These authors also used Jurkat T cells in their study. There are therefore numerous alternative pathways from PKC to IKK that could be CaMK II-dependent. Characterizing the influences of these kinases on each other is an obvious topic of future investigations.
It is becoming clear that the signaling pathways that lead to the activation of IKK involve quite a diversity of proteins. Both the nature of the stimuli and the particular type of cell is likely to govern which of these proteins are used. Here we have identified CaMK II as a critical component of mitogenic signaling to IKK in at least some cell types. This knowledge will aid our understanding of not only the regulation of this key kinase complex but also of how the important family of NFB transcription factors is differentially regulated.