A role for IκB kinase 2 in bipolar spindle assembly

Irelan et al. 10.1073/pnas.0706493104.

Supporting Information

Files in this Data Supplement:

SI Figure 6
SI Figure 7
SI Figure 8
SI Movie 1
SI Figure 9
SI Figure 10
SI Figure 11
SI Text

SI Figure 6

Fig. 6. Pathway-based analysis of siRNAs identified in genomewide screen for mitotic progression reveals an enrichment of genes associated with death receptor/TNF-a/NF-kB signaling. Approximately 300 genes identified in this genomic screen for inhibitors of mitotic progression were analyzed by Ingenuity Pathway Analysis (IPA) software (unpublished results). A pathway cluster containing overrepresentation of TNF-a signaling-related molecules is shown, with red nodes representing genes that were identified in the genomewide mitotic analysis. Significance was calculated by using the right-tailed Fisher's Exact Test.

SI Figure 7

Fig. 7. NF-kB signaling and mitosis. NF-kB is a major target of TNF/death receptor signaling, and we identified four proteins that inhibitied mitotic progression that could be also be directly mapped to these canonical signaling pathway (AKT3, NFKBIA, IKBKB, TNFRSF1A). We hypothesized that additional members may not have been identified in the analysis because of false negative activities, functional redundancies, assay sensitivity, or other technical issues. Thus, we sought to determine whether the functional link between TNF/NF-kB signaling and mitosis could be supported by physical interaction evidence. We interrogated a Human Interactome database (HyNet, Prolexys) which contains >120,000 nonredundant human yeast two hybrid protein-protein interactions. Significant internal connectivity (P < 0.005) was seen between known or inferred NF-kB pathway members (green circles) and genes that were identified in the genome-wide siRNA screen as regulators of mitosis (blue circles) (unpublished results). Similar functional overrepresentation was observed in an independent genomewide siRNA analysis of cell cycle progression (1). Taken together, these results potentially implicate components of the NF-kB signaling machinery as regulators of metaphase-to-anaphase transition.

1. Mukherji M, Bell R, Supekova L, Wang Y, Orth AP, Batalov S, Miraglia L, Huesken D, Lange J, Martin C, et al. (2006) Proc Natl Acad Sci USA 103:14819-14824.

SI Figure 8

Fig. 8. Mitotic delay and spindle defects in IKK-depleted cells. (A) (Left) Quantitation of mitotic delay in siIKK2-treated HeLa cells. Cells were synchronized by a double thymidine block, released, fixed at the indicated times, and analyzed for percentage of G₂/M phase cells by FACS analysis. (Right) Control for activation of IKK phosphorylation of IkBa during cell cycle analysis. NT; untreated; 0h, thymidine block; 8h, 8 h post release; TNF, TNF-a control (10 ng/ml for 15 min). The phosphor-specific IkBa antibody recognizes phosphorylated serines 32 and 36; cytochrome c serves as a loading control. (B) Quantitation of spindle defects in siIKK2-treated Hct116 and U2OS cells. (C) siRNA molecules targeting IKK1 or NEMO were assessed for production of abnormal spindle morphologies in HeLa cells; validation of siRNA mediated knockdown is in SI Fig.10A.

SI Figure 9

Fig. 9. Spindle defects in IKK2 null MEFs and validation of dominant-negative IKK2 activity. (A) Examples of spindle defects observed in IKK2^-/- immortalized MEFs. Cells were triple-labeled with anti-a-tubulin (green), propidium iodide (blue), and anti-phospho-histoneH3 (red) (B) Dominant negative IKK2 blocks NF-kB translocation and induces multipolar spindle morphology (arrows). HELA cells infected with the indicated adenoviral constructs were labeled with anti-p65relA (green) to follow nuclear translocation after TNF-a (10 ng/ml) stimulation (top two panels) or were triple-labeled with anti-a-tubulin (green), Hoechst dye (blue), and anti-g-tubulin (red) to assess spindle morphology.

SI Figure 10

Fig. 10. siRNA validation and assessment of Aurora A activation. (A) Validation of siRNA efficacy. Extracts from HeLa cells treated with the indicated siRNAs were Western blotted with the indicated antibodies. (B) Increased levels of total and phospho (T288) Aurora A in siIKK2-treated HeLa cells. Cells treated with the indicated siRNAs were synchronized by a double thymidine block and released, and extracts were assessed for Aurora A levels by immunoprecipitation with anti-Aurora A followed by Western blotting with the indicated antibodies.

SI Figure 11

Fig. 11. Measurement of Aurora A transcript levels in synchronized HELA cells. RNA from synchronized siIKK2 or control (siGL3)-treated HELA cells were isolated at indicated times postrelease of thymidine block. Gene specific primers for Aurora A (AURKA) and GAPDH (control) were used to measure relative mRNA levels for each gene by using semiquantitative RT-PCR.

SI Movie 1

Movie 1. Time-lapse analysis of IKK2-mediated spindle defect. HeLa cells stably expressing GFP-fused a-tubulin where transfected with an siRNA directed against IKK2 (see Materials and Methods). Image sequence of a representative spindle defect in IKK2-deficient cells is shown.

SI Text

Protein-Protein Interaction Mapping. We used the Prolexys Human Interactome Database, which contains >120,000 nonredundant proteins derived from yeast two-hybrid analysis for our biochemical network analysis (www.ariadnegenomics.com/products/prolexys). A total of 5,972 of 16,228) (36.8%) proteins in our siRNA screening library directly or indirectly (via another protein) interact with our siRNA hits with a minimum confidence score according to the Prolexys protein-protein interaction database. These proteins include 49 of 98 (50%) of NF-kB pathway members that are covered by the collection, therefore the enrichment of the associations between our siRNA hits and NF-kB members via protein-protein interaction network analysis is statistically significant (P = 0.005, calculated by using hypergeometric distribution). Pathway maps were visualized in Cytoscape.

Cell Cycle Analysis. Cells were harvested, washed in PBS, fixed with -20°C ethanol (70% final concentration), resuspended in PBS/1 mM EDTA, incubated with RNase A at 37C for 30 min, and stained with 10 mg/ml propidium iodide. The cells were analyzed with a LSR flow cytometer (Becton Dickinson) and cell cycle stage assessed with FlowJo (Tree Star) cell cycle analysis software.

Imaging of Living Cells. HeLa cells expressing GFP-a-tubulin where reverse-transfected with siRNA IKK2 no. 2. Briefly, 80 nmol of the siRNA was diluted in 500 ml of OptiMem (Invitrogen) and placed in a 35-mm glass bottom dish (MatTek). Two microliters of Lipofectamine 2000 (Invitrogen) diluted in 500 ml of OptiMem was added to the dish, and the complexes were incubated for 1 h. We plated 200,000 cells diluted in 1 ml of DMEM and 10% FBS, and 12 h later added more FBS to a final concentration of 10%. Forty-eight hours after the transfection, the cells were placed in L-15 media (Gibco) and imaged with an Ultraview RS spinning disk confocal microscope (PerkinElmer) with a custom-made environmental chamber allowing prolonged fluorescent analysis of live cells. Z-series of images through the entire cell were acquired and displayed for analysis as multiple intensity projections. For live cell movies, one z-series of images was acquired every 60 s.

Cell Culture, RNAi, Cell Cycle Analysis, and Spindle Morphology. The siRNA sequences used in this study were: IKK2 no. 1 r(AAUGCGGGAAUCUUAACGC)dTdT (Qiagen custom); IKK2 #2 r(UGUCGUGCUAACUCUGUCG)dTdT (Qiagen custom); Aurora A (Dharmacon STK6 Smartpool; M-003545-D9); GL3 control (Dharmacon D-001400-01-05). For spindle morphology assessment, 48 h before fixation, 48,000 cells were reverse-transfected in 24-well dishes with 1 ml of Lipofectamine 2000 (Invitrogen) and 25 nM final concentration for the siRNAs.

Immunoprecipitations and Western Blots. For analysis of Aurora A levels in siRNA-treated HeLa cells were lysed in 200 ml of Nonidet P-40 lysis buffer (1% Nonidet P-40; 50 mM Tris·HCl, pH 8.0), 150 mM NaCl, protease inhibitor mixture (Sigma P-8340), and phosphatase inhibitor cocktails (Sigma P-2805 and P-5726) for 30 min at 4°C with occasional vortexing and pelleted. Five hundred micrograms of total protein from each sample was precleared by incubating 30 min at 4°C with 50 ml of prewashed 50% slurry of Ezview Red Protein A Affinity Gel beads (Sigma P6486). Precleared lysates were immunoprecipitated by incubating at 4°C overnight with 4 mg/ml anti-Aurora A (Santa Cruz Biotechnology SC-25425) followed by 4-h incubation with Nonidet P-40 equilibrated beads, which were then pelleted, washed, and resuspended in sample buffer for Western blotting. Blots were probed for total Aurora by using the same antibody or phospho-Aurora (Abcam 1318). For coimmunoprecipitation experiments, HeLa cells were lysed in a buffer containing 0.5% Nonidet P-40, 170 mM NaCl, 50m M Tris (pH 8.0), 50 mM NaF, 10 mM DTT, and 1 Complete mini protease inhibitor mixture tablet (Roche) per 5 ml of buffer. Lysates were incubated on ice for 20 min with vortexing at 5-min intervals and centrifuged at 13,000 ´ g for 1 min. The supernatants were incubated with the appropriate antibodies and recombinant proteins overnight with protein G-Sepharose or Ezview Red FLAG M2 Affinity Gel (Sigma F-2426) by tumbling at 4°C. Immunoprecipitates were collected by centrifuging at 3,000 ´ g for 2 min at 4°C and washed four times in wash buffer containing 0.5% Nonidet P-40, 100 mM NaCl, 200 mM Tris (pH 8.0), and 1 Complete mini protease inhibitor mixture tablet (Roche) per 10 ml of buffer with tumbling for 10 min at 4°C. After the final wash, all of the liquid that adhered to the protein G beads were removed. Samples were then resuspended in sample loading buffer, subjected to SDS/PAGE, and transferred to PVDF membrane, and the immunoprecipitated proteins were then detected by Western blotting. Polyclonal antibodies against IKKa (M-280), IKKa/b (H-470), IKKg (FL-419), IkBa (C-21), p65 (C-20), p52 (K-27), p50 (NLS), p50 (H-119), ARK-1 (N-20), hnRNP U (C-15), and b-TrCP (N-15) were obtained from Santa Cruz Biotechnology. In addition, Aurora A (BL469) and TPX2 (BL1679) antibodies used in immunoprecipitation were purchased from Bethyl Laboratories. Aurora A (phosphor T288) antibodies were purchased from Abcam. Total and phospho KIF11 antibodies were purchased from Biolegend (627801 and 620501, respectively). Phospho IkBa was purchased from Cell Signaling (9246). All antibodies were used in immunoprecipitation or Western blotting in concentrations as suggested by the manufacturers.