IκB Kinase β (IKKβ) Inhibits p63 Isoform γ (TAp63γ) Transcriptional Activity*

Background: The functional regulation of p63 Isoform γ (TAp63γ) by IκB Kinase β (IKKβ) remains unknown, although they were previously shown to bind to each other. Results: IKKβ, but not its kinase dead mutant, suppressed the transcriptional activity of TAp63γ by disrupting its interaction with p300. Conclusion: IKKβ negatively regulates the transcriptional activity of TAp63γ. Significance: IKKβ may favor cell proliferation by inhibiting TAp63γ activity. Previously, we reported that IκB kinase-β(IKKβ) phosphorylates and stabilizes TAp63γ. However, the effect of this phosphorylation on TAp63γ transcriptional activity remains unclear. In this study, we showed that overexpression of IKKβ, but not its kinase dead mutant and IKKα, can surprisingly inhibit TAp63γ transcriptional activity as measured by luciferase assays and real-time PCR analyses of p63 target genes. This inhibition was impaired by ACHP, an IKKβ inhibitor, and enhanced by TNFα that activates IKKβ. Consistently, IKKβ inhibited the binding between TAp63γ and p300, a co-activator of TAp63γ, and consequently counteracted the positive effect of p300 on TAp63γ transcriptional activity. Through phosphorylation site prediction and mass spectrometry, we identified that Ser-4 and Ser-12 of p63 are IKKβ-targeting residues. As expected, IKKβ fails to suppress the transcriptional activity of the S4A/S12A double mutant p63. These results indicate that IKKβ can suppress TAp63γ activity by interfering with the interaction between TAp63γ and p300.

Previously, we reported that IB kinase-␤(IKK␤) phosphorylates and stabilizes TAp63␥. However, the effect of this phosphorylation on TAp63␥ transcriptional activity remains unclear. In this study, we showed that overexpression of IKK␤, but not its kinase dead mutant and IKK␣, can surprisingly inhibit TAp63␥ transcriptional activity as measured by luciferase assays and real-time PCR analyses of p63 target genes. This inhibition was impaired by ACHP, an IKK␤ inhibitor, and enhanced by TNF␣ that activates IKK␤. Consistently, IKK␤ inhibited the binding between TAp63␥ and p300, a co-activator of TAp63␥, and consequently counteracted the positive effect of p300 on TAp63␥ transcriptional activity. Through phosphorylation site prediction and mass spectrometry, we identified that Ser-4 and Ser-12 of p63 are IKK␤-targeting residues. As expected, IKK␤ fails to suppress the transcriptional activity of the S4A/S12A double mutant p63. These results indicate that IKK␤ can suppress TAp63␥ activity by interfering with the interaction between TAp63␥ and p300.
The p63 transcriptional factor is a homolog of the p53 tumor suppressor and shares numerous target genes with the latter (1). It is considered the oldest ancestor among the p53 family members, including p63, p53, and p73 (2). While the transactivational (TA) 4 domain and DNA-binding domain (DBD) of p63 display a high similarity with those of p53 in amino acid sequence, p63 possesses an extended C-terminal domain compared with p53. Because of alternative C-terminal splicing, p63 has at least three isoforms: TAp63␣, TAp63␤, and TAp63␥ (1,3). The size of the C terminus affects the function of p63. For example, TAp63␥ with the shortest C terminus is the most transcriptionally active isoform among the three isoforms, whereas TAp63␣ is much less active due to the suppression by its own long C terminus (4). Also, an additional transcriptional start site was found in the p63 gene, and the product of this transcription is the TA-truncated isoform ⌬Np63, which often functions as a dominant-negative regulator of TAp63 isoforms (1,(5)(6)(7).
While p63 is rarely mutated in cancers (8), mutations of p63 have been found to be associated with ectodermal dysplastic syndromes in humans (9 -14). Consistent with this, animal studies have demonstrated that p63 plays a role in the development of squamous epithelia, as p63-deficient mice manifest severe limb truncations, absence of skin, hair, teeth, mammary, lachrymal, and salivary glands (15,16). Interestingly, similar phenotypes were also shown in IKK␣ knock-out mice (10,11). Recently, the IKK␣ gene was found to be a transcriptional target for p63 (17)(18)(19)(20), suggesting a possibly functional link between p63 and IKK␣ in regulating ectodermal development, although IKK␣ has been shown to also form a kinase complex with IKK␤ and IKK␥, which phosphorylates IB and inactivates it, consequently activating NFB in response to the pro-inflammatory cytokine TNF␣ (17)(18)(19). Previously, we also showed that IKK␤ can phosphorylate and stabilize TAp63␥ (20). However, the functional consequence of this phosphorylation remains unaddressed.
In an attempt to address this issue, we, surprisingly, found that IKK␤, but not its kinase dead mutant form, can suppress the transcriptional activity of TAp63␥ in response to TNF␣ as measured by luciferase reporter assays and detection of endogenous p63 targets including p21 and miR-34. Our previous study showed that p300 could act as a coactivator of TAp63␥ (6). Here, we further showed that IKK␤ could interfere with the interaction between p63 and p300, consequently impairing the p300 activation of TAp63␥. Hence, these results suggest that IKK␤ can inhibit the transcrip-tional activity of TA63␥ possibly by disrupting the association of p300 with TAp63␥.
Transient Transfection and Immunoblot-As described previously (22,23), cells were transfected with indicated plasmids as shown in each figure by using TransFectin (Bio-Rad), following the manufacturer's instructions. 48 h post-transfection, cells were harvested and lysed in lysis buffer. The total protein concentration for each sample was determined, and an equal amount of total protein was then subjected to SDS-PAGE, followed by immunoblotting (IB). Immunoprecipitation (IP) was conducted by using antibodies as indicated in the figure legends and described previously (24). Beads were washed three times with lysis buffer. Bound proteins were detected by IB with antibodies as indicated in the figure legends.
Reverse Transcriptase-Polymerase Chain Reaction and Quantitative (Q) Real-time PCR Analysis-RT-and Q-PCR for mature miRNA were done using the methods described previously (25). RT-and Q-PCR for other genes were carried out as described previously (26). Briefly, Q-PCR was performed on an ABI 7300 real-time PCR system (Applied Biosystems) using SYBR Green Mix (Applied Biosystems). Relative gene expression was calculated following the manufacturer's instruction. All reactions were carried out in triplicate.
Luciferase Reporter Assays-Cells were transfected with pCMV-␤-galactoside and indicated plasmids (total plasmid DNA 1 g/well) as indicated in figures. Luciferase activity was determined and normalized by a factor of ␤-gal activity in the same assay, as described previously (27).
Chromatin Immunoprecipitation (ChIP)-PCR-ChIP analysis was performed as described previously (28) using anti-p63 and anti-p300 for endogenous proteins. Immunoprecipitated DNA fragments were analyzed by real-time PCR amplification using primers for p21 and control genes.
Senescence-associated ␤-Galactosidase (SA-␤-Gal) Staining-To investigate senescence, the SA-␤-gal activity in cultured H1299 cells was detected using the Senescence ␤-Galactosidase Staining kit (Cell Signaling) according to the manufacturer's recommendations. Briefly, H1299 cells were transfected with indicated plasmids. Four days after transfection, Senescence ␤-Galactosidase Staining was performed.
Site-directed Mutagenesis-Mutagenesis was carried out as described before (29). Briefly, the GFP-TAp63␥ plasmid was used as a mutagenesis template. PCR amplification was carried out with Phusion high fidelity DNA polymerase as described in the manufacturer's manual (New England Biolabs) and verified by DNA sequencing.
Phosphorylation Analysis by Mass Spectrometry-H1299 cells were transfected with TAp63␥ or cotransfected with TAp63␥ and Myc-IKK␤ vector. Forty eight hrs after transfection, cells were harvested and subjected to SDS-PAGE. Putative bands of phosphorylation TAp63␥ in Coomassie Blue-stained gel were collected and sent to the Proteomics Shared Resource at Oregon Health & Science University for Mass spectrometry analysis.

RESULTS
TAp63, but Not ⌬Np63, Induces miR-34a Expression in Cells-To study the effect of IKK␤ on TAp63 activity, we initially wanted to employ the miR-34 promoter-driven luciferase reporter system. Although miR-34a was recently confirmed as a p63 transcriptional target (30), the effects of different p63 isoforms on miR-34a expression have not been characterized. To address this, we first checked the effects of different TAp63 isoforms on the miR-34a promoter-driving luciferase activity by co-transfecting p63 expression vectors with the miR-34a promoter-driven luciferase reporter vector in MEF p53 Ϫ/Ϫ cells. As shown in Fig. 1A, overexpression of Myc-TAp63␥ resulted in a 20-fold increase of the luciferase activity, but showed no significant effect on the empty and mutant promoter. Next, we tested which isoform of the p63 family members is more effective in induction of the luciferase activity by performing similar luciferase assays. As expected, TAp63␥ was much more effective than was TAp63␣ or ⌬Np63␥ in activation of the luciferase activity (Fig. 1B). This result is in line with previous reports showing that TAp63␣ activity is suppressed by its C terminus (4) and that ⌬Np63 cannot induce the expression of most of the p53 targets due to lack of the TA domain (1, 5-7). Consistently, TAp63␥ induced the expression of endogenous miR-34a in a dose-dependent fashion (Fig. 1C). These results indicate that the TAp63␥ is the most functional isoform for miR-34a transcription. Thus, we will use this isoform for the following experiments.
IKK␤, but Not IKK␣ and IKK␤ Mutant, Inhibits TAp63␥induced Expression of miR-34a-Previously, we showed that IKK␤ phosphorylates and stabilizes TAp63␥ (20). However, the functional outcome of this phosphorylation remains unknown. To address this remaining issue, we carried out the above mentioned miR-34a promoter luciferase assay in the presence of IKK␤. Unexpectedly, IKK␤ inhibited the TAp63␥-induced luciferase activity in a dose-dependent manner (Fig. 1D). This inhibition was dependent on the IKK␤ kinase activity, because the IKK␤ kinase dead mutant K44A failed to inhibit this luciferase activity (Fig. 1E). Next, the specificity of this IKK␤-induced inhibition was checked with IKK␣, another important component of the IKK complex. As shown in Fig. 1F, no significant inhibition was observed when Flag-IKK␣ was co-transfected with TAp63␥, indicating that the roles of IKK␤ and IKK␣ in p63 transcriptional activity are different. To further verify the inhibitory effect of IKK␤ on p63 transcriptional activity, the endogenous miR-34a level was assessed by Q-RT-PCR. The results showed that IKK␤ also inhibits the TAp63␥-induced endogenous miR-34a transcription (Fig. 1G). These findings indicate that IKK␤ inhibits both exogenous and endogenous miR-34a transcription induced by TAp63␥.
IKK␤ Inhibits TAp63␥-induced p21 Expression-To determine whether the inhibition of IKK␤ on TAp63␥ transcriptional activity is only specific to miR-34a, we also tested if IKK␤ affects the expression of p21, a well-known p53 target gene encoding a protein that inhibits cyclin-dependent kinases (31). When Flag-IKK␤ was co-expressed with Flag-TAp63␥, it led to TAp63␥ phosphorylation in a dose-dependent fashion, as expected (20). Overexpression of TAp63␥ drastically induced both of the p21 protein and mRNA levels ( Fig. 2A). However, this induction was markedly inhibited by overexpression of IKK␤ in a dose-dependent fashion ( Fig. 2A), which was consistent with the results in Fig. 1. This inhibition appeared to be phosphorylation dependent, as the kinase dead mutant form of IKK␤, which failed to phosphorylate TAp63␥ (20), was unable to inhibit its transcriptional activity toward p21 (Fig. 2B). Similar to the IKK␤ mutant, IKK␣ also showed no effect on the p21 expression that was induced by TAp63␥ (data not shown). Next, we tested if IKK␤ could also inhibit induction of p63 specific target genes, such as Rad51 (32) and sirt1 (33). As expected, IKK␤, but not IKK␤ kinase dead mutant, inhibited p63-induced Rad51 and sirt1 expression in a dose-dependent fashion (Fig. 2C). Together with the results in Fig. 1, these results indicate that IKK␤, but not IKK␤ kinase dead mutant and IKK␣, can inhibit the activation of p21 and miR-34a expression by TAp63␥ as well as the expression of at least two p63-specific target genes.
IKK␤ Inhibitor Affects TAp63␥ Transcriptional Activity-To further confirm the negative effect of IKK␤ on TAp63␥ activity, we next determined if IKK␤ inhibitor could affect the transcriptional activity of TAp63␥ by using an IKK␤ inhibitor, called ACHP (34). As shown in Fig. 2D, in the presence of ACHP, compared with DMSO, the supershifted band of phosphorylated TAp63␥ disappeared, and the p21 level was not affected by the overexpression of IKK␤. Also, the Q-RT-PCR results showed that ACHP at least partially rescues the expressions of p21 mRNA (Fig. 2D), which indicates that ACHP abrogates the inhibitory effect of IKK␤ on TAp63␥ transcriptional activity. IKK␤ Inhibits the Binding between p300 and TAp63␥ and Suppresses Acetylation of TAp63␥ by p300-Our previous work showed that p300 can bind to the N terminus of p63 and positively regulates p63 transcriptional activity by acetylating p63 (6). Interestingly, IKK␤ phosphorylates the same domain of p63 (20). Therefore we speculate that the phosphorylation of p63 by IKK␤ might affect the binding between p300 and p63 and thus regulate p63 acetylation by p300. To this end, we carried out a co-immunoprecipitation assay for the p300-TAp63␥ complex in the presence or absence of IKK␤. In line with previous works, GFP-TAp63␥ was only pulled down by the anti-Flag antibody in the cells overexpressed Flag-p300, but not in the cells without Flag-p300. By contrast, anti-Flag-p300 antibodies pulled down much less GFP-TAp63␥ in the presence of IKK␤, indicating that IKK␤ may inhibit the binding between p300 and TAp63␥ (Fig. 3A, left panel). Consistently, reciprocal co-IP with anti-GFP-p63 antibodies showed a similar result (Fig. 3A, right  panel). To further confirm the effect of IKK␤ on endogenous p300 and p63 binding, we checked the effect of TNF␣, an IKK␤ activator, on this binding in MEF IKK␤Ϫ/Ϫ cells. Interestingly, endogenous p300 was co-immunoprecipitated with more p63 in MEF IKK␤Ϫ/Ϫ cells than that in wild type MEF cells regardless of TNF␣ treatment (Fig. 3B), supporting the idea that IKK␤ plays a negative roles in p300 and p63 binding. Also, TNF␣ only impaired p300 and p63 binding in MEF cells, but not in MEF IKK␤Ϫ/Ϫ cells, indicating IKK␤ is indispensible for suppression of p300 and p63 binding by TNF␣ (Fig. 3B). These results demonstrate that IKK␤ inhibits the binding of p300 to TAp63␥. Next, we checked the effect of IKK␤ on TAp63␥ acetylation, for it was previously shown that p300 acetylates p63 (6). As expected, overexpressed p300 promoted acetylation of TAp63␥. Consistent with the result of Fig. 3A, IKK␤ suppressed this TAp63␥ acetylation in a dose-dependent fashion (Fig. 3C).
Because p300 enhances p63 activity on p21 transcription, it is logical to assume that IKK␤ would impair this enhancement. To test this idea, we first carried out ChIP analyses to check the binding of p300 to the p21 promoter. As shown in Fig. 3D, both p63 and p300 bound to the p21 promoter, but not an un-related control sequence. Intriguingly, activation of IKK␤ by TNF␣ abrogated the binding of p300 to the p21 promoter, but had no effect on the binding of p63 to the p21 promoter. In addition, overexpression of IKK␤ also excluded p300, but not p63, from the p21 promoter (Fig. 3E). These results suggest that IKK␤ could inhibit p63-activated p21 transcription by interrupting FIGURE 3. IKK␤ inhibits the interaction between p300 and p63␥, and impairs transcriptional activity of p63␥. A, H1299 cells were transfected with indicated plasmids, 48 h after transfection, cells were harvested. Flag-p300 (left panel) or GFP-TAp63␥ (right panel) was pulled down by anti-Flag antibodies or 4A4 antibodies, and then the co-IPed proteins were probed with indicated antibodies. The input is shown in the lower panel. B, TNF␣ impairs the binding of p300 to p63 in MEF cells, but not in MEF IKK␤Ϫ/Ϫ cells. MEF and MEF IKK␤Ϫ/Ϫ cells were treated with or without TNF␣ and harvested for co-IP with anti-p300 antibodies or IgG. p300 and 4A4 antibodies were used for IB as indicated. C, IKK␤ suppresses acetylation of p63 by p300. H1299 cells were transfected with indicated plasmids. Cell lysates were analyzed by WB with indicated antibodies. D, p300 is excluded from the p21 promoter by IKK␤ activator, TNF␣. Hacat cells were treated with DMSO or TNF␣. ChIP was performed by indicated antibodies. The DNA pulled down by indicated antibodies was determined by qPCR. E, p300 is excluded from the p21 promoter by IKK␤. Hacat cells were transfected with the pCDNA or Myc-IKK␤ plasmid. ChIP was performed with indicated antibodies. The DNA pulled down by indicated antibodies was determined by qPCR. F, H1299 cells were transfected with indicated plasmids, 48 h later, the cells were harvested, and the p21 mRNA level was detected by q-RT-PCR. The data were normalized with GAPDH mRNA. G, IKK␤ inhibits senescence induced by p63 and p300. H1299 cells were transfected with indicated plasmids. Four days after transfection, Senescence ␤-Galactosidase Staining was performed. Quantification was shown in the lower panel. Data are presented as means Ϯ S.E. *, p Յ 0.05. the p63-p300 binding at the promoter. This regulation was further confirmed in Fig. 3F, as this result showed that IKK␤ inhibits the effect of p300 on TAp63␥ transcriptional activity in a dose-dependent fashion. Since p21 can induce cellular senescence (35,36), we then checked the effect of IKK␤ on p63/p300 induced senescence. As shown in Fig. 3G, p300-enhanced senescence that was induced by TAp63␥, and this promotion was not observed when IKK␤ was co-expressed with these two transcriptional regulators in cells. Altogether, these results show that IKK␤-induced phosphorylation of TAp63␥ leads to the disruption of the p300-TAp63␥ binding, and consequently inhibits TAp63␥ transcriptional activity.
TNF␣, an IKK␤ Activator, Decreases p21 Level in an IKK␤and TAp63-dependent Manner-To verify the effect of endogenous IKK␤ on TAp63␥ transcriptional activity, H1299 cells was treated with TNF␣ to induce endogenous IKK␤ (17)(18)(19). Twenty minutes after TNF␣ treatment, the protein level of IB␣, a target of IKK␤, decreased dramatically, indicating that the endogenous IKK␤ is stimulated as expected. The total RNAs from the same samples were extracted, and the miR-34a level was determined by Q-RT-PCR. As shown in Fig. 4A, the expression of miR-34a was dramatically inhibited by TNF␣, which indicates that the endogenous IKK␤ also inhibits TAp63␥ transcriptional activity. These findings imply that IKK␤ does inhibit TAp63␥ transcriptional activity in certain physiological conditions. To check the specificity of this inhibition, H1299 cells were transfected with GFP, GFP-TAp63␥, or GFP-p53 plasmids and treated with or without TNF␣ to induce endogenous IKK␤. p21 expression was then determined by WB. As shown in the right panel of Fig. 4A, the induction of endogenous IKK␤ by TNF␣ only suppressed TAp63␥ activity, but not p53 activity. This result suggests that IKK␤ specifically targets TAp63␥, but not p53. Next, we checked the existence of this TNF␣-IKK␤-TAp63␥-p21 pathway in MCF7 cells. MCF7 cells were treated with TNF␣ and harvested at different time points. Endogenous IKK␤ was activated as indicated by reduced IKB␣ (Fig. 4B). As expected, TNF␣ treatment decreased p21 level in an IKK␤-dependent manner, for knocking down IKK␤ abrogated the reduction of p21 by TNF␣ (lower panel in Fig. 4B). To further confirm if IKK␤ is dispensable for TNF␣ to inhibit p63 transcriptional activity, we checked the effect of TNF␣ on p21 level in MEF IKK␤Ϫ/Ϫ cells. As shown in Fig. 4C, TNF␣ suppressed p21 expression in MEF cells, but not in MEF IKK␤Ϫ/Ϫ cells. Interestingly, after introducing ectopic Flag-IKK␤ into MEF IKK␤Ϫ/Ϫ cells, inhibition of p21 expression by TNF␣ was restored. These results suggest that TNF␣ suppresses p21 expression by activating IKK␤. We assumed that this reduction of p21 by TNF␣ also depends on TAp63␥ activ- ity, and the TNF␣-IKK␤-TAp63␥-p21 pathway may exit in MCF7 cells. To this end, MCF7 cells were treated with TNF␣ in the presence or absence of p63 siRNA. As shown in the left panel of Fig. 4D, the effect of TNF␣ on reduction of p21 was impaired by knocking down p63. To further confirm the dependence of TAp63␥, p21 expression in MEF TAp63 Ϫ/Ϫ cells after TNF␣ treatment was determined by Western blot. The results showed that TNF␣ can only decrease p21 level in wild type MEF cells by not in MEF TAp63 Ϫ/Ϫ cells (Fig. 4D,  right panel). These findings indicate that reduction of p21 by TNF␣ requires both IKK␤ and TAp63, and also suggest that this TNF␣-IKK␤-TAp63␥-p21 pathway exit in cells.
Ser-4 and Ser-12 of TAp63 Are the Targeting Residues of IKK␤-To identify the phosphorylation sites of TAp63 by IKK␤, we carried out both bioinformatics analysis and mass spectrometry analysis. By bioinformatics analysis, we identified a number of IKK␤ consensus phosphorylation sites in p63 sequence (data not shown). To confirm these results, H1299 cells were cotransfected with IKK␤ and TAp63, and subjected to SDS-PAGE. Compared with cells with TAp63 only in Coomassie staining gel, cells cotransfected with IKK␤ and TAp63 had two more bands that could be phosphorylated TAp63 (Fig.  5A). These two bands were then collected and subjected to mass spectrometry (MS). Indeed, this MS analysis showed that these two bands are TAp63, and a number of serines and threonines were identified as potential IKK␤ phosphorylation sites (Fig. 5A). We then generated more than a dozen of TAp63 mutants, including S4A, S4A/S12A, S12A/S29A, to further confirm these results and found that IKK␤ fails to phosphorylate the TAp63 S4A/S12A mutant, indicating that S4 and S12 of TAp63 are phosphorylation sites of IKK␤ (Fig. 5B). As TAp63 S4AS12A mutant displayed a similar transcriptional activity to that of TAp63 (Fig. 6, A and B), we checked the effect of IKK␤ on the activity of this TAp63 S4A/S12A mutant. As expected, the p21 level was induced by both wild type and mutant TAp63. However, the dual S4A/S12A mutations of p63 impaired the effect of IKK␤ on p21 expression (Fig. 6A), indicating that phosphorylation of TAp63 is required for the inhibition of TAp63-FIGURE 5. IKK␤ phosphorylates p63 at S4 and S12. A, H1299 cells were transfected with TAp63␥ or cotransfected with the TAp63␥ and Myc-IKK␤ vector as indicated. 48 h after transfection, cells were harvested and subjected to SDS-PAGE. The bands as indicated by arrows were cut and subjected to mass spectrometry. Mass spectrum of potential phosphorylation sites was shown here. B, H1299 cells were transfected with indicated plasmids. WB was carried out to detect TAp63␥ and its supershifted/phosphorylated forms. dependent p21 expression by IKK␤. Similarly, TNF␣ inhibited the transcriptional activity of TAp63, but not the TAp63 S4A/ S12A mutant (Fig. 6B). Furthermore, TNF␣ fails to impair the binding of p300 to the TAp63 S4A/S12A mutant (Fig. 6C), suggesting S4 and S12 are indeed the IKK␤-targeted phosphorylation sites that are indispensable for the effect of IKK␤ on TAp63 transcriptional activity.

DISCUSSION
In this study, we demonstrate that IKK␤ can inhibit TAp63␥dependent miR-34a promoter-driven luciferase activity and endogenous miR-34a expression. This negative effect was true to the p21 expression as well. We also showed that the kinase activity of IKK␤ is critical for this inhibition, as IKK␤ kinase dead mutant K44A showed no effect, and the IKK␤ kinase inhibitor, ACHP, blocks this inhibition. Phosphorylation has been shown to induce conformational changes in a number of proteins, some of which are critical for protein-protein binding (37). In the case of IKK␤ phosphorylation of TAp63␥, this phosphorylation led to the inhibition of the p300-TAp63␥ binding and thus impaired the TAp63␥ transcriptional activity because p300 bound to the same domain of TAp63␥ as that for IKK␤binding and mediated phosphorylation.
In cells, IKK␤ is often activated through various pathways, including the pro-inflammatory cytokine TNF␣ (17)(18)(19). Interestingly, in addition to inducing pro-survival factor NFB, TNF␣ also has anti-survival functions. IKK␤ had been surprisingly shown to compromise these anti-survival functions in cells (38 -40), indicating that IKK␤ not only activates NFB via degradation of IKB␣, but could also counteract anti-survival signals introduced by various stresses or cytokines to keep NFB active. Consistent with this notion, our study as described here shows that IKK␤ inhibits p63 transcriptional activity. Also, the TNF␣ responsive inhibition of TAp63␥ transcriptional activity suggests that IKK␤ may regulate TAp63␥ activity in response to this physiological signal.
Previously, we showed that IKK␤ phosphorylates and stabilizes TAp63␥ (20). Therefore, we previously expected that this kinase might promote TAp63␥ activity. However, surprisingly, we found out in this study that IKK␤ actually inhibits the activity of TAp63␥. In our initial attempt to identify IKK␤-targeted phosphorylation sites within TAp63␥, we learnt that there are multiple IKK␤ sites in the N terminus of this transcriptional factor. Our future mutagenesis study of these individual phosphorylation amino acids, such as serines or threonines, would help us further elucidate the mechanisms underlying the negative regulation of TAp63␥ activity by this cytokine responsive kinase. Indeed, our further analysis identified Ser-4 and Ser-12 as two main target residues within the N terminus of Tap63 by IKK␤ in response to TNF␣, as the double mutations at these residues impaired the inhibitory effect of this kinase on the transcriptional activity of TAp63. The development of antibodies specific to these phosphorylated sites in the near future would further validate this conclusion.
IKK␤ is also a major activator of another transcriptional factor NFB, which plays a key role in regulating the immune response to different stresses (17)(18)(19). A crosstalk of NFB with the p53 pathway has recently been noted (41). These two pathways are rarely activated simultaneously in the same cells. This is largely because the activation of NFB promotes cell proliferation and growth, while the activation of p53 induces apopto- FIGURE 6. IKK␤ fails to affect the activity of TAp63␥ S4A/S12A mutant. A, IKK␤ fails to phosphorylate p63 S4A/S12A and to affect its transcriptional activity. H1299 cells were transfected with indicated plasmids for 36 h. WB was carried out to determine p21 levels and p63 phosphorylation. B, TNF␣ fails to affect p63 S4A/S12A transcriptional activity. H1299 cells were transfected with indicated plasmids for 36 h. Before harvesting, cells were treated with TNF␣. WB was carried out to determine p21 levels. C, TNF␣ fails to affect the binding of p300 and p63 S4S12A. H1299 cells were transfected with indicated plasmids, and 48 h after transfection, cells were harvested. Flag-p300 was pulled down by anti-Flag antibodies, and then the co-IP proteins were probed with the indicated antibodies. The input is shown in the lower panel. D, model for the regulation of TAp63 by IKK␤. sis and cell cycle arrest, although both of the p53 and NFB pathways can respond to the same type of stresses, like UV irradiation (42,43). Our findings as presented in this study thereby unveil one new mechanism for how IKK␤ inactivates TAp63␥ that disfavors cell growth and proliferation, similar to p53, whereas it activates NFB that favors cell growth and proliferation, in response to TNF␣ (Fig. 6D).