Nuclear Role of IκB Kinase-γ/NF-κB Essential Modulator (IKKγ/NEMO) in NF-κB-dependent Gene Expression*

The IκB kinase (IKK) complex, which is composed of the two kinases IKKα and IKKβ and the regulatory subunit IKKγ/nuclear factor-κB (NF-κB) essential modulator (NEMO), is important in the cytokine-induced activation of the NF-κB pathway. In addition to modulation of IKK activity, the NF-κB pathway is also regulated by other processes, including the nucleocytoplasmic shuttling of various components of this pathway and the post-translational modification of factors bound to NF-κB-dependent promoters. In this study, we explored the role of the nucleocytoplasmic shuttling of components of the IKK complex in the regulation of the NF-κB pathway. IKKγ/NEMO was demonstrated to shuttle between the cytoplasm and the nucleus and to interact with the nuclear coactivator cAMP-responsive element-binding protein-binding protein (CBP). Using both in vitro and in vivo analysis, we demonstrated that IKKγ/NEMO competed with p65 and IKKα for binding to the N terminus of CBP, inhibiting CBP-dependent transcriptional activation. These results indicate that, in addition to the key role of IKKγ/NEMO in regulating cytokine-induced IKK activity, its ability to shuttle between the cytoplasm and the nucleus and to bind to CBP can lead to transcriptional repression of the NF-κB pathway.

The nuclear factor-B (NF-B) 1 proteins are critical for activating the expression of cellular genes that are involved in the control of the immune and inflammatory response and in protecting cells from apoptosis in response to a variety of stress stimuli (1)(2)(3). NF-B is sequestered in the cytoplasm of most cells, where it is bound to a family of inhibitory proteins known as IB (2,4,5). Cytokines (including tumor necrosis factor-␣ (TNF␣) and interleukin-1) increase the activity of IB kinase (IKK)-␣ and IKK␤, resulting in enhanced phosphorylation of the IB proteins (6 -10). IKK␤ is the critical kinase required for cytokine-induced phosphorylation of the IB proteins, leading to their ubiquitination and subsequent degradation by the proteasome (6 -10). In contrast, IKK␣ is not required for IB␣ degradation, but is involved in an alternative pathway of NF-B activation that leads to the processing of p100 to p52 (11). In addition to IKK␣ and IKK␤, an additional factor known as IKK␥/NEMO is also critical for the activity of the IKK complex. IKK␥/NEMO was initially identified in a genetic complementation assay as a factor that can restore NF-〉 activation in cells that are resistant to a variety of stimuli that normally induce the NF-B pathway (12). IKK␥/NEMO was also identified independently in biochemical studies that demonstrated its essential role in the formation of the high molecular mass IKK complex (13,14) and additionally as a factor designated as FIP-3, which binds to the adenovirus E3 protein and can inhibit TNF␣-induced apoptosis (15).
IKK␥/NEMO interacts with IKK␣ and IKK␤ and is a component of the high molecular mass IKK complex, which migrates between 600 and 900 kDa following gel filtration chromatography (12)(13)(14)(15)(16)(17)(18)(19)(20). Although IKK␥/NEMO itself does not have kinase activity, it is essential for NF-B activation likely via a scaffold function that is required for IKK activity (12)(13)(14). IKK␥/NEMO has a molecular mass of 48 kDa and contains several domains (21,22), including an N-terminal domain, which is involved in its interactions with IKK␤ (22); a coiledcoil domain, which mediates its oligomerization, which is critical in stimulating IKK activity (23); and a C-terminal domain, which is involved in the recruitment of upstream factors such as receptor-interacting protein that are involved in IKK activation (15,24). Thus, the structure of IKK␥/NEMO is consistent with its role as a scaffold that is critical for function.
Genetic studies have also demonstrated an important role of IKK␥/NEMO in regulating the NF-B pathway (25)(26)(27). Disruption of the IKK␥/NEMO gene (which is located on the X chromosome) in male mice and homozygous deletion in female mice result in embryonic lethality due to TNF␣-induced hepatocyte apoptosis (25,27). Female mice with a deletion of a single copy of IKK␥/NEMO develop granulocytic infiltration and both hyperproliferation and increased apoptosis of keratinocytes (25,27). Fibroblasts isolated from IKK␥/NEMO-null mice are defective in activating the NF-B pathway in response to a variety of stimulators of this pathway. In humans, mutation of a single copy of the IKK␥/NEMO gene is associated with a syndrome known as incontinentia pigmenti, an X-linked defect that results in lethality in males and granulocytic infiltration of the skin in females (28). Another syndrome has been described in humans that is due to mutations in the putative zinc finger domain in the C terminus of IKK␥/NEMO that impair, but do not eliminate, NF-B function, resulting in an X-linked immunodeficiency syndrome characterized by hyper-IgM production and hypohydrotic ectodermal dysplasia (29 -31). Thus, both biochemical and genetic studies indicate a critical role for IKK␥/NEMO in regulating NF-B activation.
Recently, we (32) and others (33) demonstrated that, in addition to the cytoplasmic role of the IKK complex, one of its components (IKK␣) can also function in the nucleus to stimulate cytokine-induced expression of NF-B-responsive genes. IKK␣ was found to interact with CBP and, in conjunction with p65, is recruited in a cytokine-dependent manner to NF-Bresponsive promoters, where it is critical for the phosphorylation and subsequent acetylation of specific residues in histone H3 to activate gene expression (32,33). The nuclear levels of IKK␣ result from its ability to shuttle between the cytoplasm and the nucleus in a CRM1-dependent fashion (34). In this study, we demonstrate that IKK␥/NEMO is present in both the nucleus and cytoplasm of HeLa cells and that leptomycin B treatment increases its nuclear localization. This observation suggests that IKK␥/NEMO constitutively shuttles between cytoplasmic and nuclear compartments in a CRM1-dependent manner, as has been demonstrated for other proteins involved in the regulation of the NF-B pathway (35)(36)(37)(38)(39). Mammalian two-hybrid and in vitro binding assays demonstrated that nuclear IKK␥/NEMO bound to the N terminus of CBP, repressing NF-B-regulated genes. These studies indicate that IKK␥/ NEMO, like IKK␣, can regulate NF-B-dependent gene expression via its interactions with factors in both the cytoplasm and the nucleus.

MATERIALS AND METHODS
Cell Lines and Reagents-HeLa and 293 cells were purchased from American Type Culture Collection (Manassas, VA). Mouse embryo fibroblasts (MEFs) and IKK␥/NEMO Ϫ/Ϫ cells were gifts from Drs. Xiaodong Wong and Michael Karin, respectively. These cell lines were maintained in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum (Hyclone Laboratories), 2 mM L-glutamine, and penicillin/streptomycin.
Plasmid Constructs-The pCMV5 expression vectors encoding FLAG-or Myc-tagged IKK␣, IKK␤, IKK␥, or p65 or HA-tagged CBP were described previously (32,40,41). Gal4-CBP constructs were kindly provided by Dr. Tucker Collins, and fusions of the VP16 activation domain with IKK␣, IKK␤, or IKK␥/NEMO were constructed in pCMV5 as described previously (32). Glutathione S-transferase (GST)-CBP fusions were constructed by generating the desired fragments from fulllength CBP using PCR and subsequently cloning these fragments with GST into the pGEX vector (42). All PCR products were verified by DNA sequencing.
Expression and Purification of GST-CBP Fusion Proteins-Recombinant GST-CBP fusion proteins were expressed in bacterial strain BL21 and lysed in HMK buffer (50 mM Tris-HCl (pH 7.5), 100 mM NaCl, and 1 mM phenylmethylsulfonyl fluoride), and the bacterial lysates were incubated with 0.5 ml of packed glutathione-conjugated agarose beads (Sigma) overnight at 4°C. After three washes, the GST fusion proteins bound to the glutathione beads were stored at 4°C in HMK buffer. Protein expression was assessed by SDS-PAGE and Coomassie Blue staining.
Luciferase Reporter Assays-293T or HeLa cells were plated at 50% confluence in 6-well tissue culture plates. After 24 h, the cells were transfected using Genejuice transfection reagent (Novagen) with the indicated DNA constructs and a Gal4-luciferase reporter. A Rous sarcoma virus-␤-galactosidase expression vector was included in the transfection assays to control for differences in transfection efficiency, and a pCMV5 plasmid was added to standardize DNA quantities. Between 30 and 36 h post-transfection, the cells were washed twice with cold phosphate-buffered saline (PBS), and the reporter activity was measured using the luciferase assay system (Promega). All transfections were performed in duplicate and repeated at least three times.
Protein Association and Western Blot Analysis-For GST pull-down analysis with endogenous proteins, whole cell lysates were prepared from HeLa cells in lysis buffer A (40 mM Tris-HCl (pH 8.0), 500 mM NaCl, 6 mM EDTA, 6 mM EGTA, 10 mM ␤-glycerophosphate, 10 mM NaF, 1 mM Na 3 VO 4 , 1 mM dithiothreitol, 0.1% Nonidet P-40, and protease inhibitors (Roche Applied Science)), and equal protein amount of cell lysate were mixed with beads containing equal protein amount of GST-CBP or -GST and incubated overnight at 4°C on a rotatory shaker. Following incubation, the beads were extensively washed with HMK buffer, with HMK buffer with 500 mM NaCl, and with HMK buffer with 0.1% Triton X-100, respectively, before a final wash with HMK buffer. Proteins bound to the beads were eluted by adding protein loading buffer and heating to 100°C for 5 min and resolved on a 10% SDSpolyacrylamide gel; transferred to nitrocellulose membranes (Amersham Biosciences); and probed with antibodies to IKK␣, IKK␤, IKK␥/ NEMO, and p65. For GST pull-down analysis with transiently expressed proteins, expression plasmids encoding FLAG-tagged IKK␣, IKK␤, IKK␥/NEMO, and p65 were transfected into 293T cells. Thirtysix hours post-transfection, whole cell lysates were prepared in PD buffer and incubated with GST-CBP or GST beads, and assays were performed as described above, with the Western blots being probed with anti-FLAG antibody.
For immunoprecipitation and Western blot analysis, 293T cells were transfected with expression vectors encoding HA-tagged CBP in combination with FLAG-tagged IKK␣, IKK␤, IKK␥/NEMO, or p65 or Myctagged IKK␥/NEMO as indicated. Cell lysates were prepared in lysis buffer B (50 mM HEPES (pH 7.5), 150 mM NaCl, 10% glycerol, 1% Triton X-100, 1.5 mM MgCl 2 , 1 mM EGTA, 10 mM sodium pyrophosphate, 10 mM NaF, and 10 mM Na 3 VO 4 ). Immunoprecipitation was performed with anti-FLAG antibody or mouse IgG as a control, and the precipitates were captured on protein A-agarose beads and extensively washed. The bound proteins were resolved by SDS-PAGE; transferred to nitrocellulose membrane; and probed with anti-HA antibody 12CA5, anti-FLAG antibody M2, or anti-Myc antibody as indicated. Western blots were analyzed by enhanced chemiluminescence (Amersham Biosciences) after labeling with horseradish peroxidase-conjugated antimouse or anti-rabbit secondary antibody.
Immunofluorescence and Confocal Microscopy-The cellular localization of IKK␣, IKK␤, IKK␥/NEMO, and p65 was analyzed using both endogenous as well as transiently expressed proteins. For these experiments, HeLa cells and MEFs alone or transfected with expression vectors encoding the indicated Myc epitope-tagged constructs were cultured on coverslips and either untreated or treated with leptomycin B (Sigma) at final concentration of 10 ng/ml for 2 h. Coverslips were washed two times with PBS, and the cells were fixed with 3.7% formaldehyde for 10 min, followed by a brief permeabilization with 0.5% Triton X-100 in PBS. The cells were blocked for 30 min with 3% normal donkey serum in PBS and then incubated for 1 h with primary antibodies as indicated in the figures (diluted 1:50 to 1:200 in 1% normal donkey serum in PBS). The coverslips were washed three times with PBS and then incubated for 1 h with secondary antibodies conjugated to FITC or rhodamine Red-X (diluted 1:400 in 1% normal donkey serum in PBS). Nuclei were visualized by staining for lamin B. Samples were washed three times and then treated with Aquamount (Polyscience), and the results were analyzed using an MRC 1000 laser scanning confocal microscope (Bio-Rad).

Cellular Localization of IKK␥/NEMO-Immunofluorescence
studies were performed with untreated HeLa cells to analyze the localization of IKK␣, IKK␤, IKK␥/NEMO, and p65. As demonstrated previously (32), IKK␣ was localized predominantly in the nucleus, whereas both IKK␤ and p65 were localized predominantly in the cytoplasm (Fig. 1A). IKK␥/NEMO was found to be localized in both the cytoplasm and the nucleus (Fig. 1A). As a positive control, TNF␣ stimulation was found to induce the nuclear translocation of p65, but did not significantly change the cytoplasmic or nuclear distribution of IKK␥/ NEMO or IKK␣ (data not shown).
Next, we addressed whether IKK␥/NEMO can constitutively shuttle between the cytoplasm and the nucleus, as do other components of the NF-B pathway, including IKK␣ (34), NF-B-inducing kinase (34), and IB␣ and p65 (35)(36)(37)(38)(39). HeLa cells were treated with leptomycin B, a selective inhibitor of CRM1dependent nuclear export (43,44), and the localization of IKK␥/ NEMO, IKK␣, IKK␤, and p65 was then determined. In the presence of leptomycin B, both IKK␥/NEMO and p65 became predominantly localized in the nucleus, without marked changes in the distribution of IKK␣ and IKK␤ (Fig. 1B). To confirm these results, we next evaluated the distribution of Myc-tagged IKK␣, IKK␤, and IKK␥/NEMO expressed in HeLa cells in both the presence and absence of leptomycin B. Similar to the results seen with endogenous IKK␥/NEMO, transfected Myc-tagged IKK␥/NEMO became predominantly localized to the nucleus following leptomycin B treatment (Fig. 2B). No significant change in the distribution of IKK␣ or IKK␤ was noted following leptomycin B treatment (Fig. 2B).
Next, we characterized the cellular localization of several Myc-tagged IKK␥/NEMO constructs in the presence and absence of leptomycin B treatment. Expression vectors encoding Myc-tagged wild-type IKK␥/NEMO, an N-terminal truncation mutant (IKK␥/NEMO-(101-412)), and two C-terminal truncation mutants (IKK␥/NEMO-(1-358) and IKK␥/NEMO-(1-306)) were analyzed following their transfection into HeLa cells. All of these IKK␥/NEMO constructs, except IKK␥/NEMO-(1-306), were predominantly localized in the cytoplasm and become localized to the nucleus following leptomycin B treatment (Fig.  3). In contrast, IKK␥/NEMO-(1-306) was present predominantly in the nucleus in both the presence and absence of leptomycin B (Fig. 3). Classical nuclear export sequences contain leucine-rich domains with variations of the motif LXXX-LXXLX(L/I) (39,43,44). IKK␥/NEMO contains the sequence LLXXXLXXL between residues 328 and 336, which is similar, although not an exact match, to a consensus nuclear export sequence. These experiments suggest that a region in the C terminus of IKK␥/NEMO is involved in regulating its nuclear export.
IKK␥/NEMO Interacts with CBP-The results of the immunofluorescence studies indicate that IKK␥/NEMO translocated to the nucleus either constitutively or in response to a yet unidentified stimulus. However, no previous studies have indicated a nuclear role of IKK␥/NEMO. Recently, we (32) and others (33) demonstrated that IKK␣ can interact with the N terminus of the coactivator CBP, resulting in the phosphorylation and subsequent acetylation of histone H3, leading to increases in NF-B-dependent gene expression. Because the N terminus of CBP is also the site of p65 interaction, we hypothesized that IKK␥/NEMO might also interact with CBP to alter NF-B-regulated gene expression in a manner similar to IKK␣ and p65.
First, the interaction of IKK␥/NEMO and CBP was characterized using the mammalian two-hybrid system. HeLa cells were transfected with a Gal4-luciferase reporter and Gal4-CBP constructs in conjunction with VP16 fusions with IKK␥/NEMO, IKK␣, IKK␤, or p65. We have demonstrated previously that the fusion of VP16 with either IKK␣ or p65 can interact with Gal4-CBP to stimulate luciferase reporter activity (32). Luciferase reporter activity assayed at 30 -36 h post-transfection demonstrated strong interactions of IKK␣, IKK␥/NEMO, and p65 with the N terminus of CBP and only a minimal association of IKK␣ and IKK␥/NEMO with the C terminus of CBP (Fig.  4A). IKK␤ did not interact with CBP (Fig. 4A). It is interesting to note that p65, IKK␣, and IKK␥/NEMO could all interact with the N terminus of CBP.
Next, the interaction of IKK␥/NEMO with CBP was characterized using in vitro binding assays of HeLa cell lysate with either GST or GST-CBP. HeLa cell lysate was incubated with either GST-CBP or GST immobilized on glutathione beads, and the bound proteins were subjected to Western blot analysis with antibody directed against IKK␣, IKK␤, IKK␥/NEMO, or p65. There was strong interaction between IKK␥/NEMO and CBP, which was comparable with those seen with p65 and IKK␣ (Fig. 4B). Consistent with our previous results (32), there was no interaction between IKK␤ and CBP (Fig. 4B). Similar experiments were also performed with 293T cell extracts prepared from cells transfected with expression vectors encoding FLAG-tagged IKK␣, IKK␤, IKK␥/NEMO, or p65. Whole cell lysates prepared from these cells were incubated with either GST or GST-CBP, and Western blot analysis was performed with anti-FLAG antibody M2. Similar to the results seen with the endogenous proteins, FLAG-tagged IKK␥/NEMO was associated with GST-CBP, as were both FLAG-tagged p65 and IKK␣, but not FLAG-tagged IKK␤ (Fig. 4C).
Finally, we further defined the domains of CBP that are involved in its in vitro binding to IKK␥/NEMO. For these experiments, GST fusions with CBP-(1-446), CBP-(447-776), CBP-(777-1099), CBP-(1100 -1458), CBP-(1459 -1891), and CBP-(1892-2441) were utilized. These GST-CBP fusion proteins were purified, immobilized on glutathione beads, and incubated with lysates prepared from 293T cells transfected with FLAG-tagged IKK␥/NEMO. Western blot analysis performed with anti-FLAG antibody demonstrated that IKK␥/ NEMO associated with the N-terminal 446 amino acids of CBP, and only minimal interactions were seen with other domains of CBP (Fig. 4D). These results are consistent with the in vivo data obtained in the mammalian two-hybrid analysis demonstrating interactions between the N terminus of CBP and IKK␥/NEMO.

Role of IKK␥/NEMO in Regulating CBP-dependent Transcription-
The previous data show that IKK␥/NEMO translocated to the nucleus and bound to the same region of CBP as IKK␣ and p65. Transfection of expression vectors encoding IKK␥/NEMO has previously been demonstrated to inhibit cytokine-induced NF-B-mediated gene expression (15). This result raised the possibility that IKK␥/NEMO may repress IKK␣and p65-induced gene expression by preventing their binding to CBP. Alternatively, it is possible that IKK␥/NEMO could activate IKK␣-and p65-mediated gene expression by facilitating their binding to CBP. To address these possibilities, we utilized IKK␥/NEMO-deficient MEFs to assay the ability of IKK␥/NEMO to modulate IKK␣ and p65 activation of CBPregulated gene expression. First, it was important to determine Myc-tagged IKK␣, IKK␤, and IKK␥/NEMO constructs were transfected into MEFs in the absence and presence of leptomycin B, followed by immunofluorescence and confocal microscopy analysis utilizing antibody directed against the Myc epitope and FITC-conjugated secondary antibody. As a control, we also evaluated the localization of endogenous p65 in the absence and presence of leptomycin B. This analysis indicated that IKK␣ nuclear localization was slightly increased by treatment with leptomycin B, whereas IKK␤ was not affected by this treatment (Fig. 5). In contrast, the nuclear localization of both IKK␥/NEMO and p65 was markedly increased by leptomycin B treatment (Fig. 5). This analysis further substantiates a nuclear role for IKK␥/NEMO in both non-transformed and transformed cells.
Given these results, IKK␥/NEMO-deficient MEFs were transfected with a Gal4-luciferase reporter and Gal4-CBP in conjunction with increasing quantities of expression vectors encoding p65 (Fig. 6A) or IKK␣ (Fig. 6B) alone or these constructs in conjunction with a fixed amount of IKK␥/NEMO. Both IKK␣ and p65 increased Gal4-CBP-dependent luciferase reporter activity in a dose-dependent manner, whereas the cotransfection of IKK␥/NEMO prevented p65-and IKK␣-mediated increases in CBP-regulated reporter activity (Fig. 6, A and  B). These results suggest that IKK␥/NEMO may be involved in negatively regulating IKK␣-and p65-mediated transcriptional activity.
To extend this analysis, we performed immunoprecipitation and Western blot analysis to characterize the ability of IKK␥/ NEMO to prevent IKK␣ and p65 binding to CBP. For these experiments, 293T cells were transfected with HA-tagged CBP and either FLAG-or Myc-tagged IKK␥/NEMO in conjunction with either FLAG-tagged IKK␣ or p65. Immunoprecipitation was performed on whole cell lysates with either anti-FLAG antibody or normal mouse IgG, and Western blot analysis was performed with anti-HA antibody to detect HA-tagged CBP. This analysis demonstrated that IKK␣, IKK␥/NEMO, and p65 all interacted with CBP (Fig. 6C). When Myc-tagged IKK␥/ NEMO was transfected in combination with FLAG-tagged IKK␣ and HA-tagged CBP, the amount of HA-tagged CBP that was associated with FLAG-tagged IKK␣ was significantly less than that seen when only FLAG-tagged IKK␣ and HA-tagged CBP were cotransfected (Fig. 6C). These results suggest that IKK␣ and IKK␥/NEMO can compete for CBP binding. Similar results were seen when Myc-tagged IKK␥/NEMO was transfected with FLAG-tagged p65 and HA-tagged CBP (Fig. 6C), indicating that IKK␥/NEMO can also compete with p65 for binding to CBP. These results indicate that IKK␥/NEMO can compete with both p65 and IKK␣ for binding to the N terminus of CBP, resulting in IKK␥/NEMO-mediated inhibition of IKK␣ and p65 transcriptional activity. DISCUSSION Until recently, the role of IKK in regulating NF-B activity has been thought to be due exclusively to the assembly and activation of the high molecular mass IKK complex, leading to the stimulus-induced phosphorylation of IB (6-10). However, additional studies have defined novel roles of IKK␣ both in the cytoplasm by an alternative pathway involved in NF-B activation that leads the processing of p100 to p52 (11) and in the nucleus by phosphorylating histone H3 to increase NF-B-dependent gene expression (32,33). IKK␥/NEMO is known to play a key role in regulating the NF-B pathway as a cytoplasmic protein that is a component of the IKK complex and is critical in modulating its activity (12)(13)(14)(15)(25)(26)(27). In this study, we explored the nucleocytoplasmic shuttling of IKK␥/NEMO and addressed its nuclear function.
The results presented here suggest that IKK␥/NEMO can shuttle between the nucleus and the cytoplasm in a CRM1-de- exogenous stimuli (35)(36)(37)(38)(39). It is possible that signals such as phosphorylation (46) lead to increases in IKK␥/NEMO oligomerization, resulting in its increased nuclear translocation. For example, phosphorylation-induced homodimerization of the mitogen-activated protein kinase ERK2 (extracellular signal-regulated kinase-2) regulates its nuclear transport (47). Recently, IKK␥/NEMO has been shown to interact with the deubiquitination enzyme CYLD to regulate IKK activity through the ability of this enzyme to deubiquitinate TNF receptor-associated factor-2 (48 -50). This process provides another potential mechanism to regulate IKK␥/NEMO nuclear translocation. Finally, either IKK␥/NEMO may have an atypical nuclear localization sequence, or its nuclear localization may be mediated by its interaction with other proteins that traffic between the cytoplasm and the nucleus.
A region in the C terminus of IKK␥/NEMO between residues 306 and 358, containing leucine-rich sequences with homology to other nuclear export signal domains, was found to be critical for its nuclear export. Although IKK␥/NEMO does not have a consensus nuclear export sequence, only 36% of the proteins in the data base that are known to undergo nucleocytoplasmic shuttling have classical leucine-rich export sequences (45). Thus, the nucleocytoplasmic transport of IKK␥/NEMO is likely regulated by leucine-rich sequences, although interactions of IKK␥/NEMO with additional proteins may be involved in regulating its nuclear export.
In addition to IKK␣ and IKK␥/NEMO, other regulators of the NF-B pathway, including IKK␣, NF-B-inducing kinase, IB␣, and p65, have been demonstrated to shuttle between the nucleus and the cytoplasm in a CRM1-dependent manner (34 -39). For example, IB␣ has been demonstrated to shuttle between the cytoplasm and the nucleus, and its nuclear export is dependent on a nuclear export signal located within its Nterminal domain between residues 45 and 55 (35)(36)(37)(38). The p65 component of NF-B contains nuclear localization sequences that facilitate its translocation to the nucleus upon treatment of cells with a variety of stimuli, including TNF␣ (51,52), and also contains a nuclear export sequence that has homology to nuclear export sequences found in a variety of proteins, including the Rev protein of human immunodeficiency virus type 1 (39). The regulation of p65 nuclear export is complex, as demonstrated by the fact that acetylation of p65 decreases its interaction with IB␣, facilitating its DNA binding properties, whereas deacetylation of p65 by HDAC3 increases its associa-tion with IB␣ and enhances the nuclear export of this complex (53). Recently, p100 processing to p52 was demonstrated to be dependent on its nucleocytoplasmic shuttling (54). It was postulated that p100 transport to the nucleus might lead to its interaction with enzymes that ubiquitinate this protein, resulting in its subsequent processing. It is interesting to note that IKK␥/NEMO is ubiquitinated and that this process is critical for its ability to activate the IKK complex, rather than resulting in its degradation (55). It is possible that IKK ubiquitination may occur in the nucleus.
Our studies with mammalian two-hybrid analysis combined with in vitro and in vivo binding of both endogenous and exogenous proteins indicate that IKK␥/NEMO binds to the N terminus of CBP. The IKK␥/NEMO-binding site on CBP corresponds to that previously noted for IKK␣ and p65. It remains to be seen how the recruitment of IKK␥/NEMO to the transcriptional complex regulates NF-B-dependent gene expression. IKK␥/NEMO inhibits IKK␣-and p65-dependent activation of gene expression by CBP likely by IKK␥/NEMO-mediated competitive inhibition of their binding to CBP. Alternatively, IKK␥/NEMO may be involved in recruitment of proteins such as histone deacetylases, resulting in decreased NF-B-dependent gene expression. Several histone deacetylases have been demonstrated to be associated with the p65-histone acetyltransferase complex (56), and it is conceivable that IKK␥/NEMO could be involved in the recruitment of these histone deacetylases.
Mice heterozygous for IKK␥/NEMO deficiency display skin abnormalities that are similar to the phenotype of the human disease incontinentia pigmenti, which is caused by heterozygous mutations of IKK␥/NEMO (27,30). These mice exhibit a severe proliferative disorder of skin keratinocytes, which contain high levels of NF-B-inducible genes, including chemokines (RANTES (regulated on activation normal T cell expressed and secreted); macrophage inflammatory protein-1␣, -1␤, and -2; IP10; and monocyte chemotactic protein) and cytokines (interleukin-1␣ and -1␤, TNF␣, interferon-␥, and transforming growth factor-␤1 and -␤2) (25). Most of these cytokine genes are induced by NF-B, in contrast to cells with homozygous mutations in IKK␥/NEMO, which do not have NF-Binducible activity. The explanation for the up-regulation of these cytokines has been attributed to surrounding cells that have an intact copy of the IKK␥/NEMO gene due to random X chromosome inactivation. However, an alternative explanation for these results is that IKK␥/NEMO may serve as a transcrip-FIG. 6. Binding of IKK␣ and p65 to CBP is inhibited by IKK␥/NEMO. IKK␥/NEMO Ϫ / Ϫ cells were transfected with Gal4-CBP, a Gal4-luciferase reporter, and an expression vector encoding increasing quantities of either p65 (A) or IKK␣ (B) either alone or in the presence of an expression vector encoding IKK␥/ NEMO (0.1 g), followed by determination of luciferase activity. In C, 293T cells were transfected with HA-tagged CBP (2.0 g) and 1.0 g of FLAG-tagged IKK␣, IKK␥/NEMO, or p65 or Myc-tagged IKK␥/ NEMO in the indicated combinations. Immunoprecipitation (IP) of extracts (300 g) was performed with anti-FLAG antibody, and Western blot analysis was performed with anti-HA antibody. tional repressor and that the increases in the levels of the various cytokines seen in mice with heterozygous mutations in IKK␥/NEMO may be due to abrogation of the repressive activity of IKK␥/NEMO. Further studies with mice containing gene disruptions of IKK␥/NEMO will be necessary to address this possibility.
In summary, we have demonstrated that IKK␥/NEMO is translocated to the nucleus and strongly interacts with CBP to repress IKK␣-and p65-induced transcriptional activation. These results may provide a mechanism to help explain the down-regulation of NF-B activity under basal conditions or following cytokine-induced activation of the NF-B pathway. These data are consistent with recent observations on the multiple roles (11,32,33) that the IKK proteins play in regulating the NF-B pathway.