1. Division of Hematology-Oncology, Department of Medicine, Harold Simmons Cancer Center, University of Texas Southwestern Medical Center, Dallas, Texas 75390-8594, USA

Correspondence to: Richard B. Gaynor Correspondence and requests for materials should be addressed to R.B.G. (Email: gaynor_richard@lilly.com).

First, we characterized cytokine-induced activation of the NF-B pathway in parental, IKK-^-/- and IKK-^-/- mouse embryo fibroblasts (MEFs). Tumour-necrosis factor- (TNF-) induced the rapid degradation of IB in both parental MEF cells and in IKK-^-/- cells but not in IKK-^-/- cells (Fig. 1a). However, defects in the expression of an NF-B reporter were noted after treatment with TNF- in both IKK-^-/- and IKK-^-/- cells without changes in the expression of a Rous sarcoma virus–-galactosidase reporter (Fig. 1b). Furthermore, quantitative real-time polymerase chain reaction (PCR) analysis confirmed that TNF- increased the transcription of the NF-B-regulated IB gene in both MEF and IKK-^+/+ cells, in which Myc-tagged IKK- was stably expressed in IKK-^-/- cells but not in IKK-^-/- and IKK-^-/- cells (Fig. 1c). These results indicated that IKK- is critical for the rapid cytokine-mediated induction of NF-B-responsive genes by a mechanism distinct from that of IKK--mediated increases in IB degradation.

Figure 1: Both IKK-^-/- and IKK-^-/- cells are defective in TNF--induced NF-B activation.

a, Cells were treated with TNF- for the specified durations, and western blot analysis of these extracts was performed with the indicated antibodies. b, TNF--induced NF-B activation was determined by luciferase assays after transfection of an NF-B-dependent reporter construct and normalized by a Rous sarcoma virus–-galactosidase reporter construct. c, The indicated cell lines were either untreated or treated with TNF- for 15 min and the change in IB mRNA expression was determined by quantitative real-time PCR and normalized by measuring the levels of 18S RNA. d, Immunofluorescence of HeLa cells and e, of MEF, IKK-^-/- and IKK-^-/- cells was performed with the indicated IKK--specific and IKK--specific antibodies.

High resolution image and legend (81K)

It has recently been noted that IKK- can shuttle between cytoplasm and nucleus¹⁰. To address whether the intracellular distribution of IKK- and IKK- was different, these endogenous proteins were immunostained in HeLa cells (Fig. 1d) in addition to both parental and IKK-deficient mouse embryo fibroblasts (Fig. 1e). IKK- had a different distribution from that of IKK-: IKK- was localized in both the cytoplasm and the nucleus, whereas IKK- was localized predominantly in the cytoplasm (Fig. 1d, e). These results indicate that IKK- might have an additional nuclear role in activating the NF-B pathway that is distinct from its previously described role in phosphorylating the IB proteins.

To investigate whether IKK- might function in the nucleus to regulate the cytokine-induced expression of the NF-B-responsive IB and interleukin-8 (IL-8) genes, chromatin immunoprecipitation (ChIP) assays were used. HeLa cells were treated with TNF- and ChIP assays were performed at various times after stimulation. In response to stimulation with TNF-, the recruitment of IKK-, p65 and CREB binding protein (CBP) to the IB and IL-8 promoters was detected 15 minutes after treatment and was present for at least 120 minutes (Fig. 2a, left and middle panels). In contrast, there was no association of either IKK- or IKK-/NEMO with these promoters unless several additional PCR cycles were used (Fig. 2a, left and middle panels). There was no cytokine-induced association of any of these factors with the -actin promoter (Fig. 2a, right panel). Quantitative real-time PCR analysis was used to determine the percentage of IB and IL-8 promoter input DNA that was bound by IKK-, p65 and CBP (Fig. 2b). Finally, the kinetics of binding of these factors to the IB and IL-8 genes was correlated with their increased transcription (Fig. 2c).

Figure 2: TNF- treatment leads to IKK- association with the IB and IL-8 promoters.

a, HeLa cells were treated with TNF-, and chromatin immunoprecipitation assays were performed with the indicated antibodies. The detection of the immunoprecipitated IB promoter (left panel), IL-8 promoter (middle panel) or -actin promoter (right panel) was analysed by PCR with promoter-specific primers. b, Quantitative real-time PCR was performed to determine the kinetics of factor binding to these promoters in comparison with input DNAs. Squares, IKK-; circles, p65; triangles, CBP. c, IB and IL-8 mRNA levels were determined by quantitative real-time PCR and normalized by measuring the levels of 18S RNA.

High resolution image and legend (71K)

Given the fact that interactions between CBP and p65 have been shown to be important for NF-B activation^{11, 12, 13, 14}, it was possible that IKK- might form a complex with these proteins on NF-B-regulated promoters to increase transcription. IKK- might then activate NF-B-regulated transcription by the phosphorylation of p65 (refs 15, 16) or potentially other nuclear targets. First, the interactions of epitope-tagged proteins including CBP and either IKK-, IKK- or p65 were analysed in extracts prepared from cells co-transfected with expression vectors encoding these proteins. Immunoprecipitation and subsequent western blot analysis demonstrated that there were strong interactions between IKK- and CBP as well as between p65 and CBP (Fig. 3a). In addition, the mammalian two-hybrid assay demonstrated that IKK- strongly interacted with the amino-terminal transactivation domain of CBP, which is also the binding site for p65 (Fig. 3b)¹⁷. However, no direct interaction was detected between IKK- and p65 in this analysis (data not shown). Consistent with the two-hybrid data is the observation that IKK- bound to a glutathione S-transferase (GST)–CBP fusion protein containing the N terminus of CBP (Fig. 3c). Similar data demonstrating the interaction of IKK- and p65, but not IKK-, with CBP were also seen with HeLa extracts containing either these transiently expressed proteins (Fig. 3d), or endogenous proteins (Fig. 3e).

Figure 3: IKK- interacts with the CBP transactivation domain.

a, Flag-tagged IKK-, IKK- or p65 was co-transfected with haemagglutinin (HA)-tagged CBP into HEK-293 T cells. After immunoprecipitation (IP) with anti-Flag antibody, the immunoprecipitates were analysed by immunoblotting (W) with anti-HA or anti-Flag antibody as indicated. b, The indicated regions of CBP fused to the GAL4 DNA-binding domain were co-transfected with a GAL4 luciferase reporter construct and IKK-, IKK- or p65 constructs containing the VP16 activation domain. c–e, GST–CBP fusion proteins were used to determine their interaction with Flag-tagged IKK-, IKK- or p65 in extracts prepared from transfected cells (c and d) or with endogenous proteins isolated from HeLa cells (e).

High resolution image and legend (49K)

It was important to determine whether p65 was required for the association of IKK- with the IB promoter. For these studies we used infection with adenoviruses encoding either the IB super-repressor^{18, 19}, which expresses an IB protein that is not degraded after stimulation by cytokine and prevents both p65 nuclear translocation and DNA binding, or -galactosidase as a control (Fig. 4a). ChIP assays demonstrated that the IB super-repressor prevented the association of p65, but not IKK- or CBP, with the IB promoter (Fig. 4a). Although these results suggested that p65 was not crucial for IKK- association, we found defects in the association of both IKK- and CBP, in addition to other factors that bound to the IB and IL-8 promoter, in ChIP assays performed in TNF--treated p65^-/- cells (data not shown). It remains to be determined whether the defect in IKK- and CBP promoter recruitment seen in p65^-/- cells was due only to the failure of p65 to bind to NF-B-regulated promoters or to other effects.

Figure 4: IKK- promoter association is not strictly dependent on p65 binding and results in increased CBP transcriptional activity.

a, HeLa cells infected with adenovirus encoding either the IB super-repressor (IBSR) or -galactosidase (-gal) were stimulated with TNF- for the specified durations. The association of IKK-, p65 and CBP with the IB promoter was analysed by ChIP assays followed by PCR analysis (left panel) or quantitative real-time PCR (right panels). Squares, IKK-; circles, p65; triangles, CBP. b, After transfection of either haemagglutinin (HA)-tagged IKK- (WT) or IKK-(SS/AA) into HeLa cells, ChIP assays (left panel) and quantitative real-time PCR (right panels) were performed. Squares, IKK-; circles, p65; triangles, CBP. Immunoblotting (W) of the epitope-tagged proteins is also indicated. c, An expression vector encoding a GAL4–CBP fusion protein was transfected with a GAL4 luciferase reporter construct and either 0.1 g (open bars) or 0.5 g (filled bars) of expression vectors encoding IKK-, IKK-(K/M) or IKK- tagged with Myc, and CBP-dependent transcriptional activity was determined by luciferase assays.

High resolution image and legend (46K)

Next we addressed whether intact IKK- kinase activity was required for association with the IB promoter. Both the epitope-tagged wild-type and mutant IKK- proteins were recruited to the IB promoter after treatment with TNF- (Fig. 4b). The ability of these IKK- proteins to alter CBP-mediated gene expression directly was next addressed. Transfection of wild-type IKK-, but not the kinase-dead IKK-(K/M) mutant or IKK-, dose-dependently enhanced GAL4–CBP-mediated transcriptional activity (Fig. 4c). These results suggested that although both wild-type and mutant IKK- proteins could associate with the IB promoter, IKK- kinase activity was important for modulating CBP-dependent transcription.

These observations raised the question of the mechanism by which IKK- could modulate CBP-dependent transcriptional activation. The histone acetyltransferase (HAT) activity of CBP acetylates the N-terminal tails of histones such as H3 during transcriptional activation^{20, 21, 22, 23}. One potential effect of histone H3 phosphorylation has been proposed to be the recruitment and activation of HAT^{22, 23}. For example, the phosphorylation of Ser 10 in histone H3 enhances the HAT-mediated acetylation of Lys 14. Such modifications of specific residues in the N-terminal tails of histones might serve as a signal for the binding of specific co-activators to result in increased gene expression.

Because the kinase that regulates the cytokine-induced phosphorylation of histone H3 has not been identified, we addressed whether IKK- might be involved in this process. The state of acetylation and phosphorylation of histone H3 in MEF, IKK-^-/- and IKK-^-/- cells that had been serum-starved for 16 h before treatment with TNF- was characterized. Treatment of MEF cells with TNF- resulted in an increased phosphorylation of Ser 10 and acetylation of Lys 14 in histone H3 (Fig. 5a). In IKK-^-/- cells there was a significantly decreased phosphorylation of Ser 10 and acetylation of Lys 14 in histone H3 both in the absence and in the presence of TNF- in comparison with that seen in MEF cells (Fig. 5a). IKK-^-/- cells contained similar levels of histone H3 phosphorylation to that detected in MEF cells, but exhibited marked decreases in the acetylation of Lys 14 in histone H3. Treatment of each of these cells with trichostatin A, an inhibitor of histone deacetylases, resulted in similar levels of histone acetylation. These results suggested that IKK- may play a role in phosphorylation of Ser 10, which might be important for the subsequent acetylation of Lys 14 in histone H3.

Figure 5: Requirement for IKK- in histone H3 phosphorylation.

a, Each of the indicated cell lines was serum-starved for 16 h, then treated with TNF- for the specified durations or with trichostatin A (TSA) for 60 min. Acid-extracted proteins from these cells were immunoblotted with the indicated antibodies. b, Flag-tagged IKK-, IKK-(SS/AA) or IKK-(K/M) constructs were transfected into HeLa cells and the kinase activity of these immunoprecipitated Flag-tagged proteins was determined with purified histone H3 as substrate. c, Extracts prepared from IKK-^-/-, IKK-^+/+ or MEF cells were analysed for IKK- expression and the acid-extracted proteins from these cells were immunoblotted with the indicated antibodies. d, The association of these factors with the IB promoter was also determined by ChIP assays (lower panel). e, MEF, IKK-^-/- and IKK-^-/- were treated with TNF- for the specified durations and ChIP assays were performed and analysed by PCR (top panel) or quantitative real-time PCR at 0 and 30 min after treatment with TNF- (lower panel).

High resolution image and legend (72K)

Next, kinase assays were performed in vitro to determine whether IKK- could directly phosphorylate histone H3. Kinase assays performed with purified histone H3 as substrate demonstrated that wild-type IKK-, but not the kinase-defective mutants IKK-(SS/AA) or IKK-(K/M), resulted in enhanced histone H3 phosphorylation (Fig. 5b). Consistent with these results was the observation from western blot analysis that the presence of IKK- resulted in increased levels of phosphorylated Ser 10 and acetylated Lys 14 in histone H3 (Fig. 5c). ChIP assays demonstrated that IKK- led to the increased association with the IB promoter of histone H3 that was phosphorylated on Ser 10 and acetylated on Lys 14 without changing the association of total histone H3 and phosphorylated histone H1 (Fig. 5d). These results indicated that IKK- was likely to be responsible for cytokine-induced phosphorylation and subsequent acetylation in histone H3.

Finally, we addressed the kinetics of TNF--mediated increases in the association of IKK-, p65, CBP and phosphorylated and acetylated histone H3 with the IB promoter (Fig. 5e). ChIP assays of MEF and IKK-^-/- cells demonstrated that TNF- treatment led to the recruitment of both p65 and CBP to the IB promoter (Fig. 5e). Whereas treatment of MEF cells with TNF- led to the enhanced promoter association of histone H3 phosphorylated on Ser 10 and acetylated on Lys 14, little or no promoter association of histone H3 modified in this manner was found in IKK-^-/- cells. ChIP assays of the IB promoter in IKK-^-/- cells required the addition of threefold more input DNA to facilitate the detection of factor binding than in the other cell lines, owing to the decreased association of IKK-, CBP and p65 with the IB promoter (Fig. 5e). Treatment of these cells with TNF- resulted in an increased association of histone H3 phosphorylated on Ser 10, but a markedly reduced association of both CBP and histone H3 acetylated on Lys 14 compared with that seen in MEF cells (Fig. 5e). These studies suggested that the binding of both IKK- and CBP to the IB promoter was important for the association of histone H3 phosphorylated on Ser 10 and acetylated on Lys 14.

Phosphorylation of histone H3 is associated with several distinct biological processes including the mitogen-induced and cytokine-induced activation of specific genes and the compaction of chromosomes at the onset of mitosis^{20, 21}. For example, the Aurora kinases have been shown to be important in the phosphorylation of Ser 10 in histone H3 during mitosis²¹. Stimulation of both the extracellular signal-regulated protein kinase (ERK) pathway and the stress-activated p38 pathway also result in an increase in phosphorylation of histone H3 (Ser 10) during the activation of immediate-early gene expression^{20, 21}. The Rsk-2 (ref. 24) and Msk-1 (ref. 25) kinases have been shown to be important in this process. Previous studies of the IB promoter indicated that an uncharacterized kinase distinct from Rsk-2 and Msk-1 could phosphorylate histone H3 during cytokine-mediated activation of this promoter²⁶. Our analysis and that of a companion paper²⁷ indicate that IKK-, but not IKK-, is likely to be the kinase that is critical in facilitating the rapid cytokine-induced expression of NF-B-regulated genes.

IKK- recruitment to NF-B regulated promoters is probably mediated by multiple promoter-bound factors including components of the basal transcription complex, co-activators such as CBP, and perhaps chromatin itself. The cytokine-induced phosphorylation by IKK- of Ser 10 in histone H3 seems to be especially important for the subsequent acetylation of Lys 14 by CBP. This result is consistent with previous studies indicating that phosphorylation of Ser 10 precedes the subsequent acetylation of Lys 14 in histone H3 and that the HAT activity for H3 phosphorylated on Ser 10 is enhanced in comparison with that for unmodified H3 (refs 22, 23). In addition to its previously described role on the processing of p100 to p52 (ref. 28), our analysis shows that IKK- is important in regulating a critical step required for NF-B-dependent gene expression. The studies presented here extend our knowledge of the multiple functions of the IB kinases in activating the NF-B pathway.

References

References

1.	Baldwin, A. S. The NF-B and IB proteins: new discoveries and insights. Annu. Rev. Immunol. 14, 649-681 (1996) | Article | PubMed | ISI | ChemPort |
2.	Ghosh, S., May, M. J. & Kopp, E. B. NF-B and Rel proteins: evolutionarily conserved mediators of immune responses. Annu. Rev. Immunol. 16, 225-260 (1998) | Article | PubMed | ISI | ChemPort |
3.	Silverman, N. & Maniatis, T. NF-B signaling pathways in mammalian and insect innate immunity. Genes Dev. 15, 2321-2342 (2001) | Article | PubMed | ISI | ChemPort |
4.	Zandi, E. & Karin, M. Bridging the gap: composition, regulation, and physiological function of the IB kinase complex. Mol. Cell. Biol. 19, 4547-4551 (1999) | PubMed | ISI | ChemPort |
5.	Takeda, K. et al. Limb and skin abnormalities in mice lacking IKK. Science 284, 313-316 (1999) | Article | PubMed | ISI | ChemPort |
6.	Hu, Y. et al. Abnormal morphogenesis but intact IKK activation in mice lacking the IKK subunit of IB kinase. Science 284, 316-320 (1999) | Article | PubMed | ISI | ChemPort |
7.	Li, Q. et al. IKK1-deficient mice exhibit abnormal development of skin and skeleton. Genes Dev. 13, 1322-1328 (1999) | PubMed | ISI | ChemPort |
8.	Sizemore, N., Lerner, N., Dombrowski, N., Sakurai, H. & Stark, G. R. Distinct roles of the IB kinase and subunits in liberating nuclear factor B (NF-B) from IB and in phosphorylating the p65 subunit of NF-B. J. Biol. Chem. 277, 3863-3869 (2002) | Article | PubMed | ISI | ChemPort |
9.	Li, X. et al. IKK, IKK and NEMO/IKK are each required for the NF-B mediated inflammatory response program. J. Biol. Chem. 277, 45129-45140 (2002) | Article | PubMed | ISI | ChemPort |
10.	Birbach, A. et al. Signaling molecules of the NF-B pathway shuttle constitutively between cytoplasm and nucleus. J. Biol. Chem. 277, 10842-10851 (2002) | Article | PubMed | ISI | ChemPort |
11.	Janknecht, R. & Hunter, T. Transcription. A growing coactivator network. Nature 383, 22-23 (1996) | Article | PubMed | ISI | ChemPort |
12.	Goodman, R. H. & Smolik, S. CBP/p300 in cell growth, transformation, and development. Genes Dev. 14, 1553-1577 (2000) | PubMed | ISI | ChemPort |
13.	Zhong, H., Voll, R. E. & Ghosh, S. Phosphorylation of NF-B p65 by PKA stimulates transcriptional activity by promoting a novel bivalent interaction with the coactivator CBP/p300. Mol. Cell 1, 661-671 (1998) | Article | PubMed | ISI | ChemPort |
14.	Zhong, H., May, M. J., Jimi, E. & Ghosh, S. The phosphorylation status of nuclear NF-B determines its association with CBP/p300 or HDAC-1. Mol. Cell 9, 625-636 (2002) | Article | PubMed | ISI | ChemPort |
15.	Sakurai, H., Chiba, H., Miyoshi, H., Sugita, T. & Toriumi, W. IB kinases phosphorylate NF-B p65 subunit on serine 536 in the transactivation domain. J. Biol. Chem. 274, 30353-30356 (1999) | Article | PubMed | ISI | ChemPort |
16.	Madrid, L. V., Mayo, M. W., Reuther, J. Y. & Baldwin, A. S. Jr Akt stimulates the transactivation potential of the RelA/p65 subunit of NF-B through utilization of the IB kinase and activation of the mitogen-activated protein kinase p38. J. Biol. Chem. 276, 18934-18940 (2001) | Article | PubMed | ISI | ChemPort |
17.	Gerritsen, M. E. et al. CREB-binding protein/p300 are transcriptional coactivators of p65. Proc. Natl Acad. Sci. USA 94, 2927-2932 (1997) | Article | PubMed | ChemPort |
18.	Wang, C. Y., Mayo, M. W. & Baldwin, A. S. J. TNF- and cancer therapy-induced apoptosis: potentiation by inhibition of NF-B. Science 274, 784-787 (1996) | Article | PubMed | ISI | ChemPort |
19.	Van Antwerp, D. J., Martin, S. J., Kafri, T., Green, D. R. & Verma, I. M. Suppression of TNF--induced apoptosis by NF-B. Science 274, 787-789 (1996) | Article | PubMed | ChemPort |
20.	Strahl, B. D. & Allis, C. D. The language of covalent histone modifications. Nature 403, 41-45 (2000) | Article | PubMed | ISI | ChemPort |
21.	Cheung, P., Allis, C. D. & Sassone-Corsi, P. Signaling to chromatin through histone modifications. Cell 103, 263-271 (2000) | Article | PubMed | ISI | ChemPort |
22.	Cheung, P. et al. Synergistic coupling of histone H3 phosphorylation and acetylation in response to epidermal growth factor stimulation. Mol. Cell 5, 905-915 (2000) | Article | PubMed | ISI | ChemPort |
23.	Lo, W.-S. et al. Phosphorylation of serine 10 in histone H3 is functionally linked in vitro and in vivo to Gcn5-mediated acetylation at lysine 14. Mol. Cell 5, 917-926 (2000) | Article | PubMed | ISI | ChemPort |
24.	Sassone-Corsi, P. et al. Requirement of Rsk-2 for epidermal growth factor-activated phosphorylation of histone H3. Science 285, 886-891 (1999) | Article | PubMed | ISI | ChemPort |
25.	Thomson, S. et al. The nucleosomal response associated with immediate-early gene induction is mediated via alternative MAP kinase cascades: MSK1 as a potential histone H3/HMG-14 kinase. EMBO J. 18, 4779-4793 (1999) | Article | PubMed | ISI | ChemPort |
26.	Saccani, S., Pantano, S. & Natoli, G. p38-dependent marking of inflammatory genes for increased NF-B recruitment. Nature Immunol. 3, 69-75 (2002) | Article | PubMed | ISI | ChemPort |
27.	Anest, V. et al. A nucleosomal function for IB kinase- in NF-B-dependent gene expression. Nature 423, 659-663 (2003) | Article | PubMed | ISI | ChemPort |
28.	Senftleben, U. et al. Activation by IKK of a second, evolutionary conserved, NF-B signaling pathway. Science 293, 1495-1499 (2001) | Article | PubMed | ISI | ChemPort |
29.	Yamamoto, Y. et al. IKK/NEMO facilitates the recruitment of the IB proteins into the IB kinase complex. J. Biol. Chem. 276, 36327-36336 (2001) | Article | PubMed | ISI | ChemPort |

Acknowledgements

We thank J. Guo for assistance, A. Herrera and M. Singh for preparation of the figures, J. Darrah for preparing the manuscript, and T. Collins for GAL4–CBP constructs.

Competing interests statement

The authors declare no competing financial interests.