IKKα Regulates Estrogen-induced Cell Cycle Progression by Modulating E2F1 Expression*

The IκB kinase (IKK) complex consists of the catalytic subunits IKKα and IKKβ and a regulatory subunit, IKKγ/NEMO. Even though IKKα and IKKβ share significant sequence similarity, they have distinct biological roles. It has been demonstrated that IKKs are involved in regulating the proliferation of both normal and tumor cells, although the mechanisms by which they function in this process remain to be better defined. In this study, we demonstrate that IKKα, but not IKKβ, is important for estrogen-induced cell cycle progression by regulating the transcription of the E2F1 gene as well as other E2F1-responsive genes, including thymidine kinase 1, proliferating cell nuclear antigen, cyclin E, and cdc25A. The role of IKKα in regulating E2F1 was not the result of reduced levels of cyclin D1, as overexpression of this gene could not overcome the effects of IKKα knock-down. Furthermore, estrogen treatment increased the association of endogenous IKKα and E2F1, and this interaction occurred on promoters bound by E2F1. IKKα also potentiated the ability of p300/CBP-associated factor to acetylate E2F1. Taken together, these data suggest a novel mechanism by which IKKα can influence estrogen-mediated cell cycle progression through its regulation of E2F1.

The mammalian cell cycle is controlled by a series of highly regulated processes, and its dysregulation is frequently associated with growth abnormalities including the development of cancer (reviewed in Refs. [1][2][3]. The Rb/E2F pathway, which governs the G 1 to S phase transition, is one of the most important pathways that regulate the cell cycle. A major function of Rb is to sequester E2F family transcription factors and repress E2F-regulated promoters (4). The Rb family of proteins, which includes Rb, p107, and p130, forms different complexes with the E2F family members (5,6). In the early G 1 phase, the activated cyclin D-CDK4 complex phosphorylates Rb, leading to its degradation and the release of E2F. This process in turn activates the expression of cyclin E and other genes required for DNA replication (1). Cyclin E then binds to CDK2 to further phosphorylate Rb, thus forming a positive feedback loop to promote the entry of cells into the S phase. Abnormalities in this pathway and in the p53 tumor suppressor pathway are seen in almost all human tumors (4,7).
The E2F family, which is critical for cell cycle progression from the late G 1 into S phase, comprises seven members, designated E2F1-7, and can be further classified into three subfamilies based on sequence homology and function: E2F1-3, E2F4 -5, and E2F6 -7 (5,6,8).
Although E2F1-3 are positive regulators of gene expression, E2F4 -5 are transcriptional repressors when bound to Rb family proteins, and E2F6 functions as a transcriptional repressor given the fact that it lacks a transactivation domain (1,5). The E2F family members form complexes with the DP proteins (DP-1 and DP-2) to activate the transcription of genes important for DNA replication (such as thymidine kinase 1 (TK1), 3 proliferating cell nuclear antigen (PCNA), and cdc6) and cell cycle progression (such as cyclin E and cdc25A). In addition to its association with different regulators, E2F activity can be regulated by posttranslational modifications, including phosphorylation (9 -12) and acetylation (13)(14)(15)(16). Phosphorylation of E2F1 by cyclin A-CDK2 and TFIIH leads to the inhibition of E2F1 DNA-binding activity and its rapid degradation, respectively. In contrast, acetylation of lysine residues that lie outside of the DNA-binding domain of E2F1 increases its DNAbinding properties and protein half-life (13).
The IB kinase (IKK) complex, which consists of IKK␣ and IKK␤ and the regulatory subunit IKK␥/NEMO, is a critical activator of the NF-B pathway (17)(18)(19)(20)(21). IKK␤ is the major kinase required to activate the canonical NF-B pathway (22,23), whereas IKK␣ is critical for activating the noncanonical pathway that leads to the processing of the NF-B2/p100 precursor to the p52 subunit (24). Knock-out studies indicate that IKK␣ is important in other processes including the development of the epidermis, mammary gland, and B cell maturation (25,26). Recent studies have demonstrated that IKK␣ is present in the nucleus and increases transcription by binding to the promoter regions of both estrogen and cytokine-regulated genes (27)(28)(29)(30). IKK␣ recruitment to specific promoter regions leads to the phosphorylation of critical regulatory factors including histone H3 (27,29), estrogen receptor ␣ (ER␣), the co-activator SRC3 (28), and the corepressor SMRT (30). Recently, IKK␣ has also been shown to repress the expression of specific factors involved in regulating innate immunity in macrophages (31,32). Both IKK␣ and IKK␤ have also been shown to be involved either directly or indirectly in the regulation of cellular proliferation (17,28,(33)(34)(35)(36)(37)(38). For example, IKK␣ has been shown to be involved in regulating the proliferation of mammary epithelium by effects on cyclin D1 expression (35).
Recently, our group demonstrated that IKK␣ is required for the expression of a variety of estrogen-induced genes including cyclin D1 (28). This finding led us to address the role of the IKKs in regulating estrogen-mediated increases in cell cycle progression. Previous studies have demonstrated that estrogen induces a 30 -50% increase in the S phase fraction in the breast cancer cell line, MCF7 (39,40). Several lines of evidence have indicated that estrogen stimulation can induce the expression of multiple cell cycle genes, including c-myc, cyclin D1, and E2F1, to potentiate cell cycle progression (4,(41)(42)(43). In the current study, we have demonstrated that IKK␣, but not IKK␤, is important for estrogen-induced cell cycle progression in the MCF7 cells. Treatment of these cells with IKK␣ siRNA diminished estrogen-induced transcription of E2F1, as well as other E2F1-responsive genes. The defect in E2F1 expression could not be compensated by overexpression of cyclin D1. Estrogen treatment increased the association of promoter-bound IKK␣ and E2F1, and IKK␣ potentiated PCAF-mediated acetylation of E2F1. These findings demonstrate a novel mechanism by which IKK␣ can influence the cell cycle through regulation of E2F1 expression.
Immunoprecipitation and Western Blotting-MCF7 cells were plated in 5% DCC-FBS, serum-starved for 48 h, and then either untreated or treated with 100 nM E2 and assayed at different time points. Cells were lysed, and extracts were incubated with antibodies against E2F1 (sc-22820), E2F3 (sc-879), E2F4 (sc-1082), or normal IgG overnight and further incubated for 2 h following the addition of 30 l of protein A beads (Zymed Laboratories Inc.). Samples were separated by SDS-PAGE and transferred to a polyvinylidene difluoride membrane, and endogenous IKK␣ and IKK␤ were detected by Western blot using monoclonal antibodies directed against either IKK␣ (BD Pharmingen) or IKK␤ (Upstate). Aliquots of samples were also used in Western blots to analyze the expression levels of these proteins. For E2F1 acetylation, MCF7 cells maintained in 10% FBS were transfected with the indicated expression vectors. Forty-eight h later, the cells were lysed, and extracts were incubated overnight with anti-acetyl-lysine antibody (Cell Signaling). Western blot analysis was performed using antibody against E2F1 (Upstate).
Luciferase Assays-MCF7 cells were plated overnight in phenol redfree DMEM containing 5% DCC-FBS at 40% confluence in 6-well plates, and transfections were performed. After 24 h, the medium was replaced with fresh DMEM without serum, and E2 (100 nM) was added as described. Luciferase assays were performed using the Luciferase Assay System (Promega) as described in the manufacturer's instructions, and the luciferase activity was measured with a Victor 3-V 1420 multilabel counter from PerkinElmer Life Sciences.
Cell Cycle Analysis-MCF7 cells were plated overnight in phenol red-free DMEM containing 5% DCC-FBS at 40% confluence, and siRNA transfection was performed. After 24 h, the medium was replaced with fresh DMEM without serum, and E2 (100 nM) and 5% DCC-FBS were added as described. The cells were harvested at the indicated time points and fixed in 70% ethanol. For cell cycle analysis, the fixed cells were rehydrated in PBS for 30 min on ice and then resuspended in cold PBS containing 50 g/ml propidium iodide, 0.1% Triton X-100, and 0.2 mg/ml RNase A. After 2 h of incubation at 4°C, flow cytometry was performed using the BD FACSArray, and the distribution of cells in each cell cycle stage was examined using ModFit LT 3.0.
Statistical Analysis-The statistical difference among different groups was determined by analysis of variance and Student's t test. The data were presented as means Ϯ S.D.

IKK␣ Regulates Estrogen-induced Cell Cycle Progression-IKKs
have been demonstrated to be involved in regulating cellular growth properties in both normal cells and cancer cells, but the mechanisms involved in this process need to be better defined (17,18,28,37,38,49). Because IKK␣ has been found to be important for estrogen-induced proliferation of the breast cancer cell line, MCF7 (28), we first tested whether siRNA knock-down of either IKK␣ or IKK␤ affected the cell cycle progression of these cells. These cells, which were maintained in the absence of estrogen, were transfected with control, IKK␣, or IKK␤ siR-NAs and, following serum starvation for 24 h, either not treated or treated with 100 nM E2. In the absence of E2 treatment, a majority of the cells (Ͼ82%) were arrested at G 0 /G 1 stage of the cell cycle regardless of the siRNA transfection ( Fig. 1, left panels). After 20 h of incubation with E2, 45% of the MCF7 cells transfected with control siRNA were in S phase, and 47% of cells were in G 0 /G 1 phase (Fig. 1, upper right panel). In contrast, MCF7 cells transfected with IKK␣ siRNA resulted in only 28% of cells in S phase (17% lower than control) and 66% in G 0 /G 1 (Fig. 1, middle right panel). IKK␤ siRNA-treated cells showed only a slight decrease (4%) in the percentage of cells in S phase (41%) as compared with the control group (Fig. 1, lower right panel). These data demonstrate that the siRNA targeting of IKK␣, but not IKK␤, can interfere with estrogen-mediated entry into the S phase.
IKK␣ Is Involved in Transcription of E2F1-responsive Genes-IKK␣ has been shown to regulate the transcription of a variety of genes in response to distinct stimuli (27)(28)(29)(30). To address whether IKK␣ regulates the cell cycle by modulating transcription of specific genes that are involved in this process, we determined the effects of IKK␣ knock-down on the expression of certain cell cycle genes involved in regulating the G 1 /S transition. MCF7 cells were transfected with a control siRNA or either IKK␣ or IKK␤ siRNAs. After the cells were synchronized by serum starvation, they were either not treated or treated with 100 nM E2 for different time periods, and RNA was analyzed by real-time PCR. First, we examined E2F1 expression, which is critical for G 1 /S transition. Estrogen stimulation induced a 4-fold increase in E2F1 mRNA level, and this effect was reduced by IKK␣ siRNA treatment ( Fig. 2A). MCF7 cells treated with IKK␤ siRNA showed a pattern of estrogen-mediated E2F1 inducibility similar to that seen in the control, except at the 24-h time point, where a slight decrease was noted (Fig. 2B).
Next, we determined whether the IKK␣-mediated decrease in E2F1 mRNA levels resulted in reduced transcription of E2F1-responsive genes that are important for S phase entry including TK1, PCNA, cdc25A, and cyclin E. As expected, MCF7 cells transfected with IKK␣ siRNA exhibited lower mRNA levels of TK1, PCNA, cdc25A, and cyclin E mRNAs compared with the control (Fig. 2A). IKK␤ siRNA-transfected cells showed no significant change from the control (Fig. 2B). The IKK␣ and IKK␤ siRNAs reduced the levels of their respective mRNAs (Fig. 2, A and B). As a negative control, the estrogen-induced transcription of the SKP2 gene was not affected by transfection of the either IKK␣ or IKK␤ siRNAs (Fig. 2, A and B). These data suggest that IKK␣ is involved in the transcriptional regulation of E2F1 and E2F1-responsive genes.
IKK␣ Regulates the Expression of Estrogen-induced Cell Cycle Genes in a Pattern Similar to That of E2F1-The results described above suggest that IKK␣ may exert its cell cycle effects through changing the level of E2F1. Next, we investigated whether E2F1 siRNA knock-down resulted in effects on gene expression similar to that seen with the IKK␣ siRNA. To address this question, MCF7 cells were transfected with either the control siRNA or siRNAs targeting IKK␣, IKK␤, or E2F1. After serum starvation for 48 h, the cells were treated with E2 at different time points Western blots were performed on cell lysates to quantitate the amount of the different cell cycle-related gene products. In IKK␣ siRNA-treated MCF7 cells, the levels of the E2F1 protein and the E2F1-regulated gene products TK1 and Rb were dramatically decreased as was cyclin D1 (Fig. 3, A and D), whereas IKK␤ and the other E2F family members, E2F3 and E2F4, were not affected (Fig. 3A). In IKK␤ siRNA-treated cells, no significant change in the levels of these genes could be detected (Fig. 3B). Importantly, cells treated with E2F1 siRNA demonstrated a similar pattern of gene expression to that of the cells treated with IKK␣ siRNA with the expression of E2F1, TK1, cyclin D1, and Rb being decreased, whereas E2F3 and E2F4 levels were not changed (Fig. 3C). The expression of IKK␣ and IKK␤ was not affected by the E2F1 siRNA (Fig. 3C). These data suggest that IKK␣ may exert its effects on cell cycle progression through effects on E2F1.
E2F Binding Sites in the E2F1 Promoter Are Critical for Estrogeninduced E2F1 Expression-To address the mechanism by which IKK␣ regulates E2F1 expression, we analyzed the promoter region of E2F1, which is regulated by two E2F binding sites extending from Ϫ31 to Ϫ2 relative to its transcription initiation site (50,51). Previous reports demonstrated that SP1 binding sites, located between Ϫ169 and Ϫ116, and NFY binding sites, located between Ϫ122 and Ϫ54, were involved in estrogen-mediated stimulation of the E2F1 promoter via an ER␣-regulated process (43,50,(52)(53)(54). The co-activator SRC3, which is involved in estrogen receptor activation, has also been found in ChIP assays to bind to a region of the E2F1 promoter extending from Ϫ489 to Ϫ269 (47). Based on these studies, we divided the E2F1 promoter into two regions to study its function, one that includes the E2F binding sites and the other, an estrogen-response region (ERR), that contains SP1 and NFY binding sites and can mediate interaction with SRC3.
To dissect the function of these promoter regions in estrogen-induced E2F1 expression, luciferase constructs were constructed by inserting either a portion of the wild type E2F1 promoter (WT, Ϫ1389 to ϩ11, numbers based on sequence ID S74230) or mutants deleted in either the E2F binding sites (⌬E2F, Ϫ1389 to Ϫ49) or both the E2F and the ERRs (⌬E2F⌬ERR, Ϫ1389 to Ϫ590) upstream of the TATA-containing promoter in the pTAL-Luc vector (Fig. 4A). MCF7 cells were transfected with these constructs, serum-starved for 48 h, and either not treated or treated with 100 nM E2. As shown in Fig. 4B, E2 treatment induced WT luciferase activity and in the case of ⌬E2F, this effect was abolished (Fig. 4B). When both the E2F and ERR sites were deleted in the ⌬E2F⌬ERR construct, there was both dramatically lowered basal activity and E2 inducibility (Fig. 4B). This data indicates that E2F binding sites are critical for E2 inducibility and that the ERR is important for the basal expression of the E2F1 promoter.
We next examined the effects of IKK␣ knock-down on the WT E2F1 luciferase construct. IKK␣ siRNA-treated MCF7 cells were co-transfected with the WT luciferase construct and the expression vectors indicated in Fig. 4C. E2 led to increased activity of this construct, and cells treated with IKK␣ siRNA resulted in significantly decreased luciferase activity (Fig. 4C). Reconstitution of IKK␣ knock-down cells with mIKK␣, which prevents IKK␣ siRNA-mediated degradation because of sequence differences from human IKK␣, partially rescued the E2 inducibility of the luciferase construct (Fig. 4C). A similar effect was not seen with the kinase-dead mIKK␣ (K/M) construct. Overexpression of E2F1 or cyclin D1 in IKK␣ knock-down cells only slightly increased the basal luciferase activity as compared with IKK␣ siRNA-treated cells but could not compensate for the loss of E2 inducibility (Fig. 4C). Co-expression of mIKK␣ and E2F1 in IKK␣ siRNA-treated cells led to increased basal and E2-induced luciferase activity (Fig. 4C). Interestingly, transfection of the kinase-inactive mIKK␣ (K/M) and E2F1 enhanced the basal luciferase activity but did not affect E2 inducibility (Fig. 4C). This experiment demonstrated that overexpression of cyclin D1 and E2F1 is not sufficient to overcome the effects of IKK␣ knock-down and that IKK␣ kinase activity is required for E2 inducibility of the E2F1 promoter. IKK␣ Interacts with E2F1 and Potentiates E2F1 Acetylation-Next we asked whether the effects of IKK␣ are mediated through its direct interaction with E2F1. MCF7 cells were either untreated or treated with E2 for different time periods, cell lysates were immunoprecipitated with antibodies directed against E2F1, E2F3, E2F4, or normal IgG, and the association of IKK␣ and IKK␤ was assayed by Western analysis. Endogenous IKK␣ and E2F1 bound in an E2-dependent manner (Fig. 5A). This interaction is specific in that IKK␤ did not bind to E2F1 nor did IKK␣ bind to the E2F family members E2F3 and E2F4 (Fig. 5A).
Acetylation of E2F1 by PCAF has been shown to increase its DNAbinding properties and protein stability (13). To investigate whether IKK␣ can affect post-translational modifications of E2F1, we next examined the acetylation of E2F1 in MCF7 cells transfected with an IKK␣ expression vector. MCF7 cells were co-transfected with constructs expressing E2F1, PCAF, and either wild-type IKK␣ or a kinaseinactive IKK␣ (SS/AA). Lysates were prepared at 48 h post-transfection and immunoprecipitated with antibody directed against acetyllysine, and the presence of E2F1 in the immunoprecipitates was examined by Western blot. As shown in Fig. 5B, the cells expressing E2F1 alone revealed a low basal level of acetylation (lane 1), and the co-expression of PCAF increased E2F1 acetylation (Fig. 5B, lane 2). Intriguingly, IKK␣ overexpression increased E2F1 acetylation in the presence of PCAF, whereas the kinase-inactive IKK␣ (SS/AA) had no such effect (Fig. 5B,  lanes 3 and 4). Consistent with this finding, endogenous IKK␣ was found to interact with PCAF in MCF7 cells (data not shown). This data suggested that IKK␣ can potentiate PCAF acetylation of E2F1 and that the kinase activity of IKK␣ is required for this process.
Estrogen Induces IKK␣ Recruitment to the Promoter Regions of the E2F1 and TK1 Genes-Next we investigated the possibility that IKK␣ mayberecruitedtoE2FbindingsitesintheE2F1promoterinanestrogendependent manner to regulate its transcription. This effect would be similar to the role of IKK␣ in regulating the transcription of estrogenand cytokine-dependent genes (27)(28)(29)(30). To investigate whether IKK␣ The open boxes denote E2F binding sites, and solid boxes denote ERRs. B, MCF7 cells grown in 5% DCC-FBS were transfected with reporter gene constructs as indicated, and 24 h later serum-free medium was added for an additional 24 h. The cells were either untreated or treated with 100 nM E2 for 24 h, and luciferase activity was then measured and normalized to ␤-galactosidase. C, MCF7 cells were transfected with either control or IKK␣ siRNA, and 12 h later the cells were co-transfected with pTal-Luc-WT and CMV expression vectors encoding murine IKK␣, murine IKK␣ (K/M), cyclin D1, or E2F1. At 24 h posttransfection, serum-free medium was added for 24 h, and the cells were then either untreated or treated with 100 nM E2 for 24 h. Luciferase activity was measured and normalized to ␤-galactosidase. Aliquots of cell lysates were used for Western blot (WB) to show protein expression. A representative result from one of three independent experiments is shown. c, denotes control siRNA; ␣, IKK␣ siRNA. *, indicate statistical significance (p Ͻ 0.001). can be recruited to E2F1 promoter binding sites, MCF7 cells were either untreated or treated with E2 for 20 h, and ChIP assays were performed using primers that amplify either E2F (Ϫ105 to ϩ18) or ERR (Ϫ489 to Ϫ269) sites. As shown in Fig. 6A, IKK␣, PCAF, and E2F1 were found to bind to E2F binding sites in an E2-dependent manner, whereas no ER␣ and SRC3 binding was detected. In contrast, E2-inducible binding to the ERR was seen with ER␣ and SRC3 as well as with IKK␣, PCAF, and E2F1 (Fig. 6A). Estrogen also induced the binding of phospho-polymerase II (serine 2) to these promoter regions (Fig. 6A, p-Pol II).
Because other E2F1-responsive genes were also affected by IKK␣ siRNA knock-down, we also examined the promoter recruitment of IKK␣ to the E2F binding sites of TK1. Estrogen induced the promoter binding of IKK␣, PCAF, and E2F1 to this promoter (Fig. 6A). The use of primers for U6 promoter did not result in the amplification of any fragment in these assays (Fig. 6A). The fold change in the ChIP assays in response to estrogen stimulation is shown in the table in Fig. 6A (lower  panel). The mRNA levels of both the E2F1 and TK1 genes were increased following E2 stimulation (Fig. 6B). These results demonstrate that E2 can induce the binding of IKK␣ to the promoters of E2F1responsive genes.

DISCUSSION
IKKs have been implicated in the control of cell growth (17, 28, 34 -38). Several potential mechanisms may be involved in this process (17,21,55), including the regulation of the expression and localization of cyclin D1 (28,34,35,56) and the increase in the phosphorylation and proteolysis of the Forkhead transcription factor, FOXO3a (38). In the current study, we describe a role for IKK␣ in regulating the cell cycle by modulating the expression of E2F1 and its responsive genes in MCF7 cells following E2 treatment.
A previous study from our group demonstrated that IKK␣ is required for the expression of multiple estrogen-responsive genes including cyclin D1 (28). Estrogen stimulation induces the rapid formation of a complex including IKK␣, ER␣, and SRC3 on the promoter region of these genes. IKK␣ is also capable of phosphorylating both ER␣ and SRC3 to result in increases in their transcriptional activity (28). Here, we find that estrogen-induced E2F1 expression is dramatically decreased by IKK␣ siRNA treatment in MCF7 cells to reduce the expression of E2F1-regulated genes including TK1, PCNA, cyclin E, and cdc25A. The cyclin D1-CDK4 complex can affect E2F1 pathway by phosphorylating Rb, leading to Rb degradation (5,6). However, our data suggest that the effects described here with IKK␣ knock-down do not result exclusively from reductions in cyclin D1 expression but rather from IKK␣ effects on E2F1 expression. First, overexpression of cyclin D1 cannot reconstitute the effects of IKK␣ knock-down. Second, E2 induces IKK␣ recruitment and interaction with E2F1 on E2F1-responsive promoters, and IKK␣ potentiates E2F1 acetylation. Third, IKK␣ and E2F1 are required to control the expression of E2F1 containing promoters. These data indicate that the regulation of E2F1 by IKK␣ is direct and is not exclusively due to IKK␣ effects on cyclin D1. Finally, the effect of IKK␣ on E2F1 and the cell cycle is not dependent on an IKK␤-dependent, process because knock-down of IKK␤ did not affect the expression of E2F1-regulated genes.
E2F1 overexpression in IKK␣ knock-down cells could not reverse the decrease in E2F-mediated gene expression. These data suggest that IKK␣ is essential not only for estrogen-induced E2F1 expression but also for E2F1 activity. Consistent with this finding, we demonstrate that IKK␣ binds to E2F1 in an estrogen-inducible manner and potentiates E2F1 acetylation by PCAF. The interaction of endogenous IKK␣ and PCAF also supports this notion. PCAF is the most potent enzyme found FIGURE 6. Estrogen induces IKK␣ recruitment to the promoter regions of E2F1 and TK1. A, MCF7 cells were plated as described under "Materials and Methods" and either untreated or treated with 100 nM E2 for 20 h. ChIP assays (IP, immunoprecipitation) were performed using antibodies directed against IKK␣, E2F1, PCAF, SRC-3, and phospho-polymerase II (p-pol II; serine 2) followed by PCR amplification with primers derived from promoters of E2F1, TK1, and U6. The amount of input DNA (1%) is also shown. The -fold changes in ChIP assays in response to estrogen stimulation are shown in the table (lower panel); -, indicates not applicable. B, total RNA was extracted from aliquots of cells described in A, and the mRNA levels of E2F1 and TK1 were determined by real-time PCR. *, indicates statistical significance (p Ͻ 0.05).
to acetylate E2F1 in in vitro assays, and this acetylation increases E2F1 DNA-binding ability and protein half-life (13). However, it is unclear how E2F1 acetylation is regulated in vivo, especially in response to various stimuli. Our data suggest that IKK␣ may play a role in transmitting estrogen-induced signals to E2F1 to increase its acetylation. A role for IKK␣ in regulating acetylation has been demonstrated previously in histone H3, where it can increase the phosphorylation of serine 10 and its interaction with CBP results in the subsequent acetylation of lysine 14 on this histone during the activation of estrogen and cytokine-dependent genes (27,29). The requirement of IKK␣ kinase activity in this process suggests that phosphorylation is involved. Further investigation is needed to address potential IKK␣ phosphorylation of E2F1, PCAF, or other factors to regulate the acetylation of E2F1.
Recently, the nuclear role of IKK␣ in regulating transcription of a variety of genes has been described (27)(28)(29)(30). IKK␣ interacts with different factors on various promoters in response to diverse stimuli. For example, in response to TNF␣ stimulation, IKK␣ is recruited to the promoter regions of cytokine-regulated genes and mediates the phosphorylation and subsequent acetylation of histone H3 (27,29). In contrast, in response to laminin attachment, IKK␣ forms a complex with and phosphorylates the co-repressor SMRT on the NF-B-regulated promoters to mediate the chromatin-dissociation and nuclear export of SMRT (30). IKK␣ also been shown to form a complex with and phosphorylate ER␣ and the co-activator SRC3 on the ER␣-responsive promoters in response to estrogen stimulation (28). Here, we add to this growing list another example of the role of IKK␣ in regulating E2Fresponsive promoters in response to estrogen. The E2F and ERR binding sites in the E2F1 promoter function to modulate the estrogen response by the recruitment of E2F1, SRC3, ER␣, PCAF, and IKK␣ to this promoter. In summary, given the fact that IKK␣ is also indispensable for estrogen-induced cyclin D1 expression, it is likely that IKK␣ regulates the estrogen-induced cell cycle at multiple levels.