A role for IκB kinase 2 in bipolar spindle assembly

Edited by Peter K. Vogt, The Scripps Research Institute, La Jolla, CA, and approved September 4, 2007
October 23, 2007
104 (43) 16940-16945

Abstract

IκB kinase 2 (IKK2 or IKKβ) is a component of the IKK complex that coordinates the cellular response to a diverse set of extracellular stimuli, including cytokines, microbial infection, and stress. In response to an external stimulus, the complex is activated, resulting in the phosphorylation and subsequent proteasome-mediated degradation of IκB proteins. This event triggers the nuclear import of the NF-κB transcription factor, which activates the transcription of genes that regulate a variety of fundamental biological processes, including immune response, cell survival, and development. Here, we define an essential role for IKK2 in normal mitotic progression and the maintenance of spindle bipolarity. Chemical and genetic perturbation of IKK2 promotes the formation of multipolar spindles and chromosome missegregation. Depletion of IKK2 results in the deregulation of Aurora A protein stability and coincident hyperactivation of a putative Aurora A substrate, the mitotic motor KIF11. These data support a function for IKK2 as an antagonist of Aurora A signaling during mitosis. Additionally, our results indicate a direct role for IKK2 in the maintenance of genome stability and underscore the potential for oncogenic consequences in targeting this kinase for therapeutic intervention.
The protein kinase IκB kinase 2 (IKK2) is associated with a 700- to 900-kDa multiprotein complex in the cytoplasm, which contains additional subunits, including the highly homologous IKK1/α (IKK1) and a nonenzymatic regulatory protein IKKγ (NEMO)(13). In response to a number of external stimuli, including TNF-α and LPS, the complex is activated, resulting in the phosphorylation of IκB proteins. The phosphorylated form of IκB is recognized and ubiquitinated by the SKP1/Cul1/F-box protein (SCF)–β-transducin repeat-containing protein (βTRCP) complex, targeting it for degradation by the proteasome. This event triggers the nuclear import of the NF-κB transcription factor, which activates the transcription of genes that regulate a variety of important physiological events (4, 5). A number of noncanonical activities for NF-κB pathway members have also been described. For example, NEMO coordinates a cellular response to genotoxic stress with the ataxia telangiectasia mutated (ATM) kinase independent of the IKK complex (6). Additionally, IKK1 has been shown to phosphorylate histone H3 and DNA-bound RelA, resulting in enhancement and repression of transcription, respectively (79). NF-κB-independent activities of IKK include the regulation of keratinocyte differentiation and tumor promotion through inhibition of the oncogenic transcription factor FOXO3a (10, 11).
Control of mitosis is achieved through the integration of complex signaling pathways, which ensures strict temporal and spatial regulation (12, 13). Aurora kinase A is a key regulator of several steps in mitosis, including centrosome maturation and bipolar spindle assembly (14). Activation of Aurora A triggers the phosphorylation of a number of substrates that regulate mitotic progression, including the kinesin-like motor protein KIF11/Eg5, an important component in generating poleward forces for the maintenance of bipolar spindles (1517). In a reciprocal regulatory circuit, Aurora A protein is rapidly degraded in an anaphase-promoting complex (APC/C)-dependent manner at the conclusion of mitosis (18). Here, we demonstrate that reduction of IKK2 activity results in the induction of multipolar spindles, accompanied by increased chromosomal segregation defects and cellular transformation. We further demonstrate that these spindle abnormalities result from IKK2-dependent misregulation of Aurora A protein stability and concomitant hyperactivation of KIF11/Eg5. These results define a role for IKK2 in regulation of the metaphase-to-anaphase transition and suggest that inhibition of this kinase promotes genetic instability and oncogenic transformation.

Results

We have previously conducted genomewide siRNA screens to identify proteins required for the proper regulation of cell-cycle progression (unpublished work and ref. 20). Interestingly, these analyses revealed that a significant number of genes associated with TNF-α/NF-κB signaling, including IKK2, may participate in the transition through the mitotic phase of the cell cycle [supporting information (SI) Figs. 6 and 7]. Specifically, reduction in IKK2 mRNA levels in HeLa cells resulted in a 50% increase of their mitotic index (unpublished work). To further investigate this observation, we transfected siRNA oligonucleotides targeting IKK2 into synchronized HeLa cells. FACS analysis demonstrated that depletion of IKK2 induced an accumulation of cells with 4N DNA content, confirming that the absence of this kinase results in a delayed transition through the cell cycle (Fig. 1A and SI Fig. 8A). Although the thymidine block/release procedure triggered a slight activation of the IKK complex, this induction was resolved before the cells transitioned through G2/M (SI Fig. 8A). Because these data indicated that IKK2 reduction may affect the metaphase-to-anaphase transition, we sought to determine whether defects in spindle architecture might underlie the observed phenotype. Mitotic spindles were visualized through immunofluorescent labeling of microtubules. Whereas 90–95% of control siRNA-treated cells displayed normal bipolar spindle structures, between 20% and 25% of cells with reduced IKK2 protein levels contained defective spindle morphology (Fig. 1 B and C). These structures primarily contained supernumerary spindle poles; however, monopolar spindles were occasionally observed. Additional human cell lines, including U2OS (osteosarcoma) and HCT116 (colorectal carcinoma), also displayed multipolar spindles when transfected with siRNA directed against IKK2 (SI Fig. 8B). To understand the temporal dynamics of this defect, we created time-lapse movies of spindle organization by tracking GFP-α-tubulin in IKK2 siRNA-treated HeLa cells. Visual analysis of this image sequence indicated that multipolar spindles in IKK2-deficient cells emerge after the microtubules are sorted into bipolar arrays, thus suggesting that the observed phenotypes are not a result of aberrant centriole duplication in interphase (SI Movie 1). Immortalized mouse embryo fibroblasts (MEFs) harboring targeted deletion of the IKK2 locus (IKK2−/−) also exhibited significant multipolar, and to a lesser extent monopolar, spindle abnormalities when compared with control wild-type MEFs (SI Fig. 9A). Additionally, we observed that primary MEFs collected from a fraction (3/4) of IKK2−/− embryos also presented approximately twice the number of multipolar spindles (≈25% of cells) in comparison to their wild-type littermates (≈13% of cells, 4/4 embryos; Fig. 1D). These observations suggest a hitherto undefined role for IKK2 in bipolar spindle maintenance.
Fig. 1.
Disruption of IKK2 signaling results in cell-cycle delay and altered spindle morphology. (A) siIKK2- and control-treated HeLa cells were synchronized by a double thymidine block, released, fixed at the indicated times, and analyzed for DNA content by FACS analysis. Percentages of cells in G1 or G2/M are indicated; the balance is in S phase. (B) Examples of spindle defects observed in IKK2-deficient cells. HeLa cells were double-labeled with anti-α-tubulin (green) and propidium iodide (red). (Scale bar: 10 μm.) (C) Quantitation of multipolar spindles shown in B. (D) Representative rates of spindle abnormalities in primary MEFs from wild-type and IKK2−/− littermates. Similar results were seen in cells derived from 4/4 wild-type and 3/4 IKK2 null embryos. Frequencies are presented as mean values (± 1 SD). P values were determined by using Student's t test. Validation of siRNA knockdown is shown in SI Fig. 10A.
To probe the functional domains of IKK2 required for this activity, we introduced adenoviral particles encoding mutants of IKK2, which inhibit TNF-α-induced p65 nuclear translocation (SI Fig. 9B), into HeLa cells (Fig. 2A). Expression of a kinase inactive mutant of IKK2 (K-M) promoted the appearance of abnormal spindles, similar to those observed in IKK2-depleted cells (Fig. 2A and SI Fig. 9B). Additionally, the introduction of another dominant negative IKK2 molecule, which harbors mutations in the activation loop (SS-AA), resulted in a similar phenotype (Fig. 2A). The overexpression of wild-type IKK2 at levels equivalent to the mutant versions did not induce spindle defects (Fig. 2A). We next incubated HeLa cells with [5-(p-fluorophenyl)-2-ureido]thiophene-3-carboxamide, a cell-permeable small-molecule antagonist of IKK2 activity (21). Inspection of mitotic cells after treatment with this compound demonstrated that the inhibition of IKK2 also results in the induction of aberrant spindle phenotypes in a dose-dependent manner (Fig. 2B). High concentrations of this inhibitor promoted the formation of multipolar spindles at frequencies significantly greater than those observed by siRNA-mediated knockdown (Fig. 1C vs. Fig. 2B). This finding may reflect nonspecific inhibition of other kinases at these doses. Additionally, comparatively lower levels of spindle abnormalities in IKK2 siRNA-treated cells may be caused by incomplete depletion of IKK2 protein (SI Fig. 10A). Taken together, however, we conclude that enzymatic activity of IKK2 contributes to proper cell cycle progression by ensuring the preservation of spindle bipolarity.
Fig. 2.
Inhibition of IKK2 activity induces mutlipolar spindle formation. (A) Increased multipolar spindles observed in dominant-negative IKK2-expressing cells (Ad-IKK2-KM, Ad-IKK2-SS/AA). Spindle defects are not altered in cells overexpressing the IκBα super repressor mutant (pBABE-IκBαM) or expressing wild-type IKK2 (Ad-IKK2-WT). (B) Dose–response curve for spindle defects in HeLa cells treated with the IKK2 inhibitor [5-(p-fluorophenyl)-2-ureido]thiophene-3-carboxamide. Frequencies are presented as mean values (±1 SD).
Intriguingly, we also found that siRNA molecules targeting the other members of the complex, IKK1 and the regulatory subunit NEMO, induced multipolar spindles in mitotic HeLa cells, albeit at a lower frequency (SI Fig. 8C). However, cells infected with adenovirus carrying the IκBα superrepressor molecule (IκBαM), which impaired TNF-α-induced p65 nuclear accumulation (SI Fig. 9B), did not display significant alterations of spindle morphology (Fig. 2A). These data indicate that, although additional members of the IKK complex also regulate the establishment of spindle bipolarity, the induction of NF-κB activity is not required for maintenance of spindle integrity.
Aurora kinase A (STK6, ARK1) is a major regulator of several processes in mitosis, including maintenance of spindle bipolarity. Alterations in Aurora A activity promote spindle abnormalities resembling those observed in IKK2-depleted cells (2224). We therefore assessed whether the IKK and Aurora kinases may be physically associated during mitosis. Aurora A was immunoprecipitated from mitotically enriched cells, and the eluate was probed by Western blot analysis. IKK2, as well as IKK1 and NEMO, coimmunoprecipitated with the Aurora kinase, whereas IκBα, p65, and p52 were not found to interact with this complex (Fig. 3A). The autocatalytic activity of Aurora A is regulated through interactions with TPX2, another component of the spindle machinery (14, 2527). Each of the IKK complex members was coimmunoprecipitated with TPX2, demonstrating that the IKK proteins are associated with the Aurora-A protein complex in mitotic cells (Fig. 3A).
Fig. 3.
Biochemical and functional interactions between members of the IKK complex and Aurora A. (A) Coimmunoprecipitation of members of the IKK signalsome, but not downstream pathway members, with Aurora A and TPX2 from mitotic HeLa cells. (B) Addition of importin α/β disrupts interaction between IKK2 and Aurora A. Aurora A was immunoprecipitated from mitotic HeLa cell extracts and incubated with BSA (−) or the indicated importin proteins. IKK2 interaction was assessed by Western blot analysis. (C) Bacterially expressed, constitutively active IKK2 and/or kinase-dead Aurora A were incubated with γ-P32 ATP and subjected to gel electrophoresis and autoradiography. Phosphorylation and autophosphorylation species of Aurora A and IKK2 (*), respectively, are shown. (D) HEK293T cells were transiently transfected with the indicated constructs, and equivalent quantities of Aurora A complexes purified by immunoprecipitation were subjected to Western blotting with the indicated antibodies. (E) HEK293T cells were transiently transfected with the indicated constructs, and equivalent quantities of total protein were subjected to Western blotting with the indicated antibodies.
TPX2 is prevented from binding Aurora A through its interaction with nuclear import receptors importin α/β, which ensures the proper temporal and spatial organization of microtubules into bipolar spindles (28). RAN-GTP, an activated form of the GTPase Ran, relieves Aurora A inhibition by liberating TPX2, and other spindle assembly factors, from the importin complex (29). We investigated whether importin α/β could also regulate IKK2 association with Aurora A. IKK2 was immunoprecipitated from HeLa cell mitotic nuclear extracts and treated with exogenous importin α, β, α/β, or BSA. We found that coincubation with importin α and α/β significantly inhibited the interaction between the two kinases, suggesting that the Ran signaling pathway governs Aurora A interaction with IKK2 (Fig. 3B).
Because we have observed that IKK2 kinase activity is required for maintenance of spindle bipolarity, we investigated whether Aurora A was a potential substrate of IKK2. We coincubated an activated form of IKK2 (SS-EE) with a mutant of Aurora A deficient of autophosphorylation activity (AurA-KD) and subjected them to an in vitro kinase assay in the presence of γ-P32ATP. After separation of the proteins by SDS/PAGE, we detected phosphorylated species by autoradiography (Fig. 3C). As expected, IKK2 possessed autophosphorylation activity in both the presence and absence of the Aurora A protein, and conversely, the mutant version of Aurora A was not able to autophosphorylate. Importantly, we observed substantial phosphorylation of Aurora A in the presence of activated IKK2, implicating this mitotic kinase as a substrate for IKK2 activity.
During canonical TNF-α/NF-κB signaling, the activated IKK complex phosphorylates IκB, which triggers β-TRCP-mediated degradation by the proteasome machinery (1, 2). Because the E3 ligase β-TRCP has been previously shown to recognize IKK-phosphorylated substrates, we sought to determine whether phosphorylation of Aurora A by IKK2 can result in the association of β-TRCP with the mitotic kinase. We cotransfected a dominant negative (F-box deleted) version of β-TRCP with Aurora A, in the presence of constitutively active IKK2 or a negative control. Complexes containing Aurora A were purified from the cellular extractions by immunoprecipitation and analyzed by SDS/PAGE and Western blot analysis. After normalizing for equivalent levels of Aurora A protein, evaluation of Aurora A-associated β-TRCP levels indicated that expression of kinase active IKK2 strongly promotes the recruitment of the E3 ligase to Aurora A (Fig. 3D). Furthermore, cotransfection of an activated form of IKK2 (SE) and DN β-TRCP resulted in a significant stabilization of the Aurora A protein (Fig. 3E). Taken together, these results suggest that IKK2 phosphorylation of Aurora A targets it for β-TRCP-mediated proteasomal degradation, a mechanism that parallels the IKK-directed targeting of IκB during canonical NF-κB signaling.
Consistent with this model, Western blot analysis of extracts derived from synchronized HeLa cells revealed that the reduction of IKK2 expression induced a sustained elevation in endogenous Aurora A protein levels throughout mitosis (Fig. 4A), and analysis of Aurora A mRNA levels suggests that this regulation occurs at the posttranscriptional level (SI Fig. 11). Furthermore, these cells also showed a commensurate increase of phosphorylated Aurora A (SI Fig. 10B). These results demonstrate that depletion of IKK2 results in an overall increase of activated Aurora A in mitosis and indicate that IKK2 acts as an antagonist of Aurora A activity.
Fig. 4.
IKK2 is an antagonist of AuroraA-KIF11 signaling. (A) HeLa cells treated with the indicated siRNAs were synchronized by a double thymidine block and released, and mitotic extracts were assessed for Aurora A levels by Western blot. Cytochrome c serves as a loading control. (B) Suppression of the spindle defect in IKK2-depleted cells by Aurora A depletion is shown. HeLa cells were treated with the indicated siRNA combinations and scored for multipolar spindle defects. (C) Suppression of the spindle defect in IKK2-deficient cells by Aurora A inhibition is shown. HeLa cells treated with the indicated siRNAs and the Aurora A-specific inhibitor VX-680 were scored for multipolar spindles. (D) siIKK2-treated cells contain elevated levels of phosphorylated KIF11. Protein extracts from cells prepared as described in A were probed with antibodies that detect phosphorylated and total KIF11. (E) Treatment of IKK2-depleted cells with the KIF11 inhibitor monastrol (25 μM) reduces multipolar spindles. Monastrol was added 48 h after transfection with the indicated siRNAs, and cells were fixed for spindle staining after a 4-h incubation period. Data points are indicated as mean values (±1 SD). P values were determined by using Student's t test.
To determine whether misregulation of Aurora A was sufficient to account for the spindle defects caused by the inhibition of IKK2, we cotransfected siRNAs targeting both IKK2 and Aurora A into cells. Quantification of the multipolar spindle phenotype revealed that depletion of Aurora A measurably reversed the mitotic defects caused by the IKK2 siRNAs (Fig. 4B). We also blocked Aurora kinase activity by using the small molecule inhibitor VX-680 and assessed spindle abnormalities in IKK2 siRNA-transfected cells (Fig. 4C and ref. 30). This Aurora kinase inhibitor shows 30- and 8-fold selectivity for Aurora A over Aurora B and Aurora C, respectively. We found that low concentrations (1 nM) of the inhibitor also reduced the occurrence of multipolar spindles in IKK2-deficient cells (Fig. 4C and ref. 30) Taken together, these results indicate that the role of IKK2 in bipolar spindle maintenance is, at least partially, mediated by regulation of Aurora A activity.
KIF11/Eg5/KSP is a putative target of Aurora A kinase and has a microtubule plus-end directed motor activity that is required for the assembly of bipolar spindles (1517). We therefore tested whether the absence of IKK2 altered the phosphorylation state of KIF11. Consistent with a role downstream of Aurora and IKK2 mitotic activity, phosphorylation of KIF11 is substantially increased in cells with IKK2 siRNA-induced elevation of Aurora A protein levels (Fig. 4D). Additionally, we used a small molecule inhibitor of KIF11, monastrol, to determine whether KIF11 activity is required for multipolar spindle formation in cells lacking IKK2. Forty-eight hours after transfection with a siRNA directed against IKK2, cells were treated with low concentrations (25 μM) of the KIF11 inhibitor, and then stained to determine spindle architecture. Antagonism of KIF11 in IKK2-depleted cells also reduced the fraction of cells with multipolar spindles to levels comparable to the negative control (Fig. 4E). These data indicate that KIF11 function is required for multipolar spindle formation in the absence of IKK2 and supports a function for IKK2 as an antagonist of an Aurora A-mediated pathway that regulates bipolar spindle assembly.
Spindle abnormalities caused by inappropriate Aurora A activation are often associated with chromosome segregation defects and aneuploidy (22, 31). Therefore, we tested whether IKK2 depletion induced chromosomal aberrations in human cell lines. Reduction of IKK2 consistently resulted in a 3-fold increase in the number of micronuclei, an indicator of chromosomal instability, in multiple cell lines (Fig. 5A and ref. 32). Additionally, treatment of HeLa cells with a small molecule inhibitor to IKK2 also resulted in a significant accumulation of micronuclei, suggesting that spindle defects caused by IKK2 inhibition may lead to chromosomal missegregation events (Fig. 5B). To ascertain whether sustained reduction of IKK2 promotes a transformed phenotype, we assessed the ability of IKK2-deficient MEFs to induce tumors in nude mice. As expected, s.c. injection of wild-type (IKK2+/+) cells did not manifest detectable tumor masses (4/4). However, the introduction of IKK2−/− MEFs into these immunocompromised animals promoted the growth of large tumor nodes (4/4; Fig. 5C). These data are consistent with a recent report (33) that concludes that IKK2 deficiency promotes cellular migration and growth.
Fig. 5.
Depletion of IKK2 results in aneuploidy and increased transformation potential. (A) Cells of the indicated type were transfected with control and IKK2 siRNAs, stained with Hoechst dye, and assessed for micronuclei, indicated as mean values (±1 SD). (B) HeLa cells treated with the IKK2 inhibitor [5-(p-fluorophenyl)-2-ureido]thiophene-3-carboxamide were scored as in A. (C) IKK2−/− and wild-type littermate control MEFs were expanded in culture for six passages after isolation, then cell suspensions were s.c.-injected into nude mice and assessed for tumor formation after 4 weeks.

Discussion

A number of studies have putatively identified several other components of the NF-κB pathway that are essential for mitotic progression, including IKKε, IRAK1, RIPK1, NFKBIA, and MEKK1 (SI Fig. 6 and refs. 34 and 35). In addition, we and others have observed that IKK1, IKK2, and NEMO individually associate with protein complexes containing Aurora A, but further analysis is required to determine whether the 700- to 900-kDa IKK signalsome complex is directly regulating bipolar spindle assembly (36). However, these findings implicate a potential intersection between a significant number of NF-κB signaling molecules and critical mitotic events.
Our data specifically indicate an essential role for IKK2 in mitotic spindle function. Mitotic cells depleted of IKK2 activity maintain increased Aurora A levels, and spindle abnormalities in these cells are partially reversed by inhibition of the mitotic kinase. Thus, these results support a model in which IKK2 acts as an antagonist of Aurora A during mitosis. Phosphorylation of Aurora A kinase by IKK2 likely targets it for β-TRCP-mediated degradation and serves to maintain appropriate levels of Aurora A activity to assure proper bipolar spindle assembly and mitotic progression. Recent work has suggested that members of the NF-κB pathway may serve both upstream and downstream of the mitotic regulator Aurora kinase A (36, 37). In the case of upstream regulation, it has been suggested that other members of the IKK complex may activate Aurora A activity, whereas our results indicate that IKK2 plays an antagonistic role to the Aurora A–KIF11 signal transduction pathway to maintain mitotic spindle integrity (Fig. 4 and ref. 37). With respect to a potential downstream regulation of the mitotic kinase by IKKs, it may be possible that IKK and Aurora A kinases play reciprocal regulatory roles in the complex that we and others describe (Fig. 3A and ref. 36).
The NF-κB pathway plays both pivotal and apparently paradoxical roles in tumorogenesis. Activating mutations in genes that encode NF-κB pathway members have previously been associated with various cancers, including colon cancer, liver cancer, melanoma, leukemias, and lymphomas (38). These genetic abnormalities include the amplification of c-rel in non-Hodgkin's B cell lymphomas, homozygous deletion of the IκBα locus in Hodgkin's lymphoma, and the constitutive activation of IKK activity in melanomas and breast cancer (39). Additionally, the NF-κB complex transcriptionally activates tumor-promoting proinflammatory cytokines, which has been shown to play a critical role in inflammation-associated cancers (39). Deletion of IKK2 in both intestinal epithelial and myeloid tissues reduced tumor formation in a mouse model for colitis-associated tumorigenesis, which can be attributed to the antiapo ptotic or proinflammatory roles for the IKK/NFκB pathway in each respective cell type (40). Conversely, IKK signaling can also induce the expression of proapoptotic genes, and inhibition of the pathway has been shown to promote cancer in certain cell types (41, 42). Selective ablation of the regulatory subunit, NEMO, in mouse hepatocytes leads to increased tumorigenesis through modulation of the apoptotic and oxidative stress responses, most likely caused by chronic induction of cell death, inflammatory response, and compensatory cell proliferation (43). Furthermore, MEFs derived from IKK2 knockout mice are known to exhibit increased proliferation and migration properties, consistent with a tumor suppressive role (33). Thus, it is likely this complex relationship between NF-κB pathway components and tumor suppression or promotion significantly depends on cellular context and microenvironments.
The present study reveals that the depletion of IKK2 activity in both tumor cell lines and MEFs leads to spindle anomalies that promote chromosome segregation defects, aneuploidy, and transformation. Previous work has shown that immortalized NF-κB relA −/− MEFs exhibit a transformed phenotype, but transforming ability varied in a cell line-dependent manner (44, 45). Although we did not detect p53 pathway abnormalities in the IKK2−/− MEFs used for the in vivo tumorigenesis studies described here (Fig. 5C and data not shown), a subpopulation of these MEFS likely acquired additional genetic or epigenetic alterations, either in vitro or in vivo, to promote the transformed phenotype. Identification of these cooperative mutations will provide further insight into the role of IKK2 as a tumor suppressor.
In addition to the established tumor-promoting and tumor-suppressive roles of IKK2 via modulation of apoptosis, proliferation, and the inflammatory response through NF-κB regulation, these results suggest a more direct role for IKK2 in tumor suppression via maintenance of genome integrity. Importantly, the spindle abnormalities described here are not likely the result of aberrant NF-κB transcriptional activation, because expression of a hyperinhibitory IκBα molecule did not affect spindle organization (Fig. 2A). Taken together, our data indicate that molecular components that modulate NF-κB function during interphase may also collectively regulate mitotic progression through mechanisms that are distinct from their canonical signaling activities. Inhibitors of NF-κB pathway signaling are used for a variety of autoimmune and inflammatory disorders, and several inhibitors of IKK activity are currently in development (19). The effects of IKK2 inhibition on spindle bipolarity, chromosomal stability, and cellular transformation underscore potential pitfalls in targeting members of this fundamental molecular pathway for therapeutic intervention.

Materials and Methods

Cell Culture.

Synchronous cultures were obtained by incubating cells for 15 h in 2 mM thymidine starting at 24 h posttransfection, followed by a 10-h incubation in normal medium and a second 15-h block with 2 mM thymidine. HeLa and MEF cells were grown in DMEM plus 10% FBS; U2OS and Hct116 cells were grown in McCoys's plus 10% FBS. TNF stimulations were performed by adding TNF-α (10 ng/ml) to the culture for 15 min at 37°C.

Cell Cycle Analysis, Immunoprecipitations and Western Blots.

See SI Text.

Spindle Morphology and Micronuclei Assays.

Cells were fixed in PBS containing 0.3% Triton X-100, 1% paraformaldehyde, 0.2% glutaraldehyde, and 100 nM Taxol for 20 min, followed by several PBS washes. Cells were incubated with primary antibodies recognizing α-tubulin and γ-tubulin (F2168 and T3559; Sigma, St. Louis, MO) in PBS containing 0.3% Triton X-100, and 5% BSA overnight, followed by PBS washes and incubation with secondary antibody (A11036; Molecular Probes, Carlsbad, CA). DNA was stained with Hoechst 33342, and cells were viewed on an Eclipse TE-2000-U microscope (Nikon, Tokyo). Spindles were scored as multipolar if they contained extraneous α-tubulin projections and/or more than two foci of γ-tubulin staining. A minimum of 100 cells was counted for each replicate. Micronuclei were scored as clear foci of Hoechst staining of subnuclear size associated with a normal nucleus. Treatment with VX-680 (gift of Nathanael Gray, Harvard Medical School, Boston, MA) and Monastrol (M8515; Sigma) was as described. All scores depict the mean value (±1 SD) of at least three independent experiments.

Kinase Assays.

One hundred-fifty nanograms of of bacterially expressed, constitutively active IKK2 (SSEE), 500 ng of bacterially expressed kinase-inactive AurkA KD (D274A), or both were incubated with 2.5 μM ATP and 0.5 μCi γ[32P]ATP for 30 min at 30C, followed by addition of sample buffer and SDS/PAGE.

Mouse Experiments.

IKK2−/− and wild-type littermate control MEFs at passage six postisolation were trypsinized to single cell suspensions of ≈3 × 105 in PBS, s.c. injected into nude mice, and assessed for tumor formation 4 weeks postinjection. Each experiment was repeated at least four times.

Abbreviations

IKK
IκB kinase
βTRCP
β-transducin repeat-containing protein
MEF
mouse embryonic fibroblast.

Acknowledgments

We thank Mridul Mukherjee, Charles Cho, Serge Batalov, Ed Manser, Zhuo-shen Zhao, and Quitang Li for important contributions to this study. This work was supported by the Novartis Research Foundation. V.T.'s laboratory is supported by Agency for Science, Technology, and Research, Singapore (A*STAR).

Supporting Information

Adobe PDF - 06493Fig6.pdf
Adobe PDF - 06493Fig6.pdf
Adobe PDF - 06493Fig7.pdf
Adobe PDF - 06493Fig7.pdf
Adobe PDF - 06493Fig8.pdf
Adobe PDF - 06493Fig8.pdf
Adobe PDF - 06493Fig9.pdf
Adobe PDF - 06493Fig9.pdf
Adobe PDF - 06493Fig10.pdf
Adobe PDF - 06493Fig10.pdf
Adobe PDF - 06493Fig11.pdf
Adobe PDF - 06493Fig11.pdf
06493Movie1.mov

References

1
CH Regnier, HY Song, X Gao, DV Goeddel, Z Cao, M Rothe Cell 90, 373–383 (1997).
2
F Mercurio, H Zhu, BW Murray, A Shevchenko, BL Bennett, J Li, DB Young, M Barbosa, M Mann, A Manning, et al. Science 278, 860–866 (1997).
3
JA DiDonato, M Hayakawa, DM Rothwarf, E Zandi, M Karin Nature 388, 548–554 (1997).
4
M Karin, Y Ben-Neriah Annu Rev Immunol 18, 621–663 (2000).
5
Q Li, IM Verma Nat Rev Immunol 2, 725–734 (2002).
6
TT Huang, SM Wuerzberger-Davis, ZH Wu, S Miyamoto Cell 115, 565–576 (2003).
7
Y Yamamoto, UN Verma, S Prajapati, YT Kwak, RB Gaynor Nature 423, 655–659 (2003).
8
V Anest, JL Hanson, PC Cogswell, KA Steinbrecher, BD Strahl, AS Baldwin Nature 423, 659–663 (2003).
9
T Lawrence, M Bebien, GY Liu, V Nizet, M Karin Nature 434, 1138–1143 (2005).
10
MC Hu, DF Lee, W Xia, LS Golfman, F Ou-Yang, JY Yang, Y Zou, S Bao, N Hanada, H Saso, et al. Cell 117, 225–237 (2004).
11
Y Hu, V Baud, T Oga, KI Kim, K Yoshida, M Karin Nature 410, 710–714 (2001).
12
T Wittmann, A Hyman, A Desai Nat Cell Biol 3, E28–E34 (2001).
13
A Musacchio, ED Salmon Nat Rev Mol Cell Biol 8, 379–393 (2007).
14
D Ducat, Y Zheng Exp Cell Res 301, 60–67 (2004).
15
KE Sawin, K LeGuellec, M Philippe, TJ Mitchison Nature 359, 540–543 (1992).
16
R Giet, D McLean, S Descamps, MJ Lee, JW Raff, C Prigent, DM Glover J Cell Biol 156, 437–451 (2002).
17
R Giet, R Uzbekov, F Cubizolles, K Le Guellec, C Prigent J Biol Chem 274, 15005–15013 (1999).
18
K Honda, H Mihara, Y Kato, A Yamaguchi, H Tanaka, H Yasuda, K Furukawa, T Urano Oncogene 19, 2812–2819 (2000).
19
J Strnad, JR Burke Trends Pharmacol Sci 28, 142–148 (2007).
20
M Mukherji, R Bell, L Supekova, Y Wang, AP Orth, S Batalov, L Miraglia, D Huesken, J Lange, C Martin, et al. Proc Natl Acad Sci USA 103, 14819–14824 (2006).
21
M Karin, Y Yamamoto, QM Wang Nat Rev Drug Discovery 3, 17–26 (2004).
22
S Dutertre, S Descamps, C Prigent Oncogene 21, 6175–6183 (2002).
23
T Marumoto, S Honda, T Hara, M Nitta, T Hirota, E Kohmura, H Saya J Biol Chem 278, 51786–51795 (2003).
24
C Roghi, R Giet, R Uzbekov, N Morin, I Chartrain, R Le Guellec, A Couturier, M Doree, M Philippe, C Prigent J Cell Sci 111, 557–572 (1998).
25
T Wittmann, M Wilm, E Karsenti, I Vernos J Cell Biol 149, 1405–1418 (2000).
26
PA Eyers, E Erikson, LG Chen, JL Maller Curr Biol 13, 691–697 (2003).
27
TA Kufer, HH Sillje, R Korner, OJ Gruss, P Meraldi, EA Nigg J Cell Biol 158, 617–623 (2002).
28
OJ Gruss, RE Carazo-Salas, CA Schatz, G Guarguaglini, J Kast, M Wilm, N Le Bot, I Vernos, E Karsenti, IW Mattaj Cell 104, 83–93 (2001).
29
MY Tsai, C Wiese, K Cao, O Martin, P Donovan, J Ruderman, C Prigent, Y Zheng Nat Cell Biol 5, 242–248 (2003).
30
EA Harrington, D Bebbington, J Moore, RK Rasmussen, AO Ajose-Adeogun, T Nakayama, JA Graham, C Demur, T Hercend, A Diu-Hercend, et al. Nat Med 10, 262–267 (2004).
31
H Zhou, J Kuang, L Zhong, WL Kuo, JW Gray, A Sahin, BR Brinkley, S Sen Nat Genet 20, 189–193 (1998).
32
H Norppa, GC Falck Mutagenesis 18, 221–233 (2003).
33
F Chen, Y Lu, V Castranova, Z Li, M Karin J Biol Chem 281, 37142–37149 (2006).
34
EM Eves, P Shapiro, K Naik, UR Klein, N Trakul, MR Rosner Mol Cell 23, 561–574 (2006).
35
J Moffat, DA Grueneberg, X Yang, SY Kim, AM Kloepfer, G Hinkle, B Piqani, TM Eisenhaure, B Luo, JK Grenier Cell 124, 1283–1298 (2006).
36
P Briassouli, F Chan, K Savage, JS Reis-Filho, S Linardopoulos Cancer Res 67, 1689–1695 (2007).
37
S Prajapati, Z Tu, Y Yamamoto, RB Gaynor Cell Cycle 5, 2371–2380 (2006).
38
SC Sun, G Xiao Cancer Metastasis Rev 22, 405–422 (2003).
39
Q Li, S Withoff, IM Verma Trends Immunol 26, 318–325 (2005).
40
FR Greten, L Eckmann, TF Greten, JM Park, ZW Li, LJ Egan, MF Kagnoff, M Karin Cell 118, 285–296 (2004).
41
M Dajee, M Lazarov, JY Zhang, T Cai, CL Green, AJ Russell, MP Marinkovich, S Tao, Q Lin, Y Kubo, et al. Nature 421, 639–643 (2003).
42
JL Luo, H Kamata, M Karin J Clin Invest 115, 2625–2632 (2005).
43
T Luedde, N Beraza, V Kotsikoris, G van Loo, A Nenci, R De Vos, T Roskams, C Trautwein, M Pasparakis Cancer Cell 11, 119–132 (2007).
44
ME Gapuzan, O Schmah, AD Pollock, A Hoffmann, TD Gilmore Oncogene 24, 6574–6583 (2005).
45
ME Gapuzan, PV Yufit, TD Gilmore Oncogene 21, 2484–2492 (2002).

Information & Authors

Information

Published in

The cover image for PNAS Vol.104; No.43
Proceedings of the National Academy of Sciences
Vol. 104 | No. 43
October 23, 2007
PubMed: 17939994

Classifications

Submission history

Received: July 11, 2007
Published online: October 23, 2007
Published in issue: October 23, 2007

Keywords

  1. Aurora A
  2. mitosis
  3. NF-κB
  4. spindle polarity

Acknowledgments

We thank Mridul Mukherjee, Charles Cho, Serge Batalov, Ed Manser, Zhuo-shen Zhao, and Quitang Li for important contributions to this study. This work was supported by the Novartis Research Foundation. V.T.'s laboratory is supported by Agency for Science, Technology, and Research, Singapore (A*STAR).

Notes

This article is a PNAS Direct Submission.
This article contains supporting information online at www.pnas.org/cgi/content/full/0706493104/DC1.

Authors

Affiliations

Jeffrey T. Irelan
Genomics Institute of the Novartis Research Foundation, 10675 John J. Hopkins Drive, San Diego, CA 92121;
Thomas J. Murphy
Genomics Institute of the Novartis Research Foundation, 10675 John J. Hopkins Drive, San Diego, CA 92121;
Paul D. DeJesus
Genomics Institute of the Novartis Research Foundation, 10675 John J. Hopkins Drive, San Diego, CA 92121;
Hsiangling Teo
Institute for Molecular and Cell Biology, 61 Biopolis Drive, Proteos, 3-02B, Singapore Singapore 138673;
DingYue Xu
Institute for Molecular and Cell Biology, 61 Biopolis Drive, Proteos, 3-02B, Singapore Singapore 138673;
Maria A. Gomez-Ferreria
Department of Physiology and Biophysics, Albert Einstein College of Medicine, 1300 Morris Park Avenue, Bronx, NY 10461; and
Yingyao Zhou
Genomics Institute of the Novartis Research Foundation, 10675 John J. Hopkins Drive, San Diego, CA 92121;
Loren J. Miraglia
Genomics Institute of the Novartis Research Foundation, 10675 John J. Hopkins Drive, San Diego, CA 92121;
Daniel R. Rines
Genomics Institute of the Novartis Research Foundation, 10675 John J. Hopkins Drive, San Diego, CA 92121;
Inder M. Verma
The Salk Institute for Biological Studies, 10010 North Torrey Pines Road, La Jolla, CA 92037
David J. Sharp
Department of Physiology and Biophysics, Albert Einstein College of Medicine, 1300 Morris Park Avenue, Bronx, NY 10461; and
Vinay Tergaonkar
Institute for Molecular and Cell Biology, 61 Biopolis Drive, Proteos, 3-02B, Singapore Singapore 138673;
The Salk Institute for Biological Studies, 10010 North Torrey Pines Road, La Jolla, CA 92037
Sumit K. Chanda [email protected]
Genomics Institute of the Novartis Research Foundation, 10675 John J. Hopkins Drive, San Diego, CA 92121;

Notes

To whom correspondence should be sent at the present address: Infectious and Inflammatory Disease Center, Burnham Institute for Medical Research, 10901 North Torrey Pines Road, La Jolla, CA 92037. E-mail: [email protected]
Author contributions: J.T.I., T.J.M., M.A.G.-F., D.J.S., V.T., and S.K.C. designed research; J.T.I., T.J.M., P.D.D., H.T., D.X., M.A.G.-F., and V.T. performed research; L.J.M. and I.M.V. contributed new reagents/analytic tools; J.T.I., T.J.M., Y.Z., D.R.R., V.T., and S.K.C. analyzed data; and J.T.I. and S.K.C. wrote the paper.

Competing Interests

The authors declare no conflict of interest.

Metrics & Citations

Metrics

Note: The article usage is presented with a three- to four-day delay and will update daily once available. Due to ths delay, usage data will not appear immediately following publication. Citation information is sourced from Crossref Cited-by service.


Citation statements




Altmetrics

Citations

Export the article citation data by selecting a format from the list below and clicking Export.

Cited by

    Loading...

    View Options

    View options

    PDF format

    Download this article as a PDF file

    DOWNLOAD PDF

    Login options

    Check if you have access through your login credentials or your institution to get full access on this article.

    Personal login Institutional Login

    Recommend to a librarian

    Recommend PNAS to a Librarian

    Purchase options

    Purchase this article to access the full text.

    Single Article Purchase

    A role for IκB kinase 2 in bipolar spindle assembly
    Proceedings of the National Academy of Sciences
    • Vol. 104
    • No. 43
    • pp. 16721-17240

    Figures

    Tables

    Media

    Share

    Share

    Share article link

    Share on social media