JCS03425 Supplementary Material

Files in this Data Supplement:

• Supplemental Figure 1 -

  Fig. S1. Aurora B is more sensitive to Hesperadin treatment in HeLa cells than aurora A activity. Human HeLa cells were treated with DMSO, 12.5 nM, 25 nM, 50 nM, 100 nM, 200 nM or 400 nM Hesperadin for 1 hour, fixed and subsequently stained for phosphorylated Histone H3 on Ser10 (H3S10ph; staining the corona of chromosomes) and autophosphorylated Aurora A on Thr288 (Aurora A T288ph; staining spindle poles as indicated by white arrows). Prophase cells and metaphase cells were analysed for each condition and representative images are shown. While in prophase cells staining of H3S10ph which is dependent on Aurora B activity was reduced at 25 nM and absent at 50 nM Hesperadin, the signal was only abolished at a concentration of 100 nM in metaphase cells. The signal of Aurora A T288ph was still clearly visible at a concentration of 100 nM Hesperadin in all cells and only treatment with 400 nM Hesperadin abolished both signals the H3S10ph and Aurora A T288ph signal. Chromosomes were counterstained with DAPI (1 μg/ml).

• Supplemental Figure 2 -

  Fig. S2. Loss of Aurora B Activity reduces chromatin-bound condensin I throughout mitosis. (A-C) Cells stably expressing EGFP-Kleisin-γ (condensin I) were synchronously released from S-phase and the levels of EGFP-Kleisin-γ on mitotic chromosomes were monitored throughout mitosis. Hoechst 33342 (0.2 μg/ml) was used to visualise chromatin. Control cells (A), Hesperadin treated (100 nM, 1 hour before the bulk of cells entered mitosis) (B) and Aurora B RNAi transfected cells (C) are shown. Movies were taken on an Olympus DeltaVision microscope (40× objective, five slices every 5 minutes, 1×1 binning). Projected images are shown. (D,E) Quantification of EGFP-Kleisin-γ on mitotic chromatin in early mitosis (D) and late mitosis (E) was done by normalisation to the Hoechst signal. Values of control cells (n=15) are compared with those of cells treated with Hesperadin (n=31) and transfected with siRNAs against Aurora B (n=17). (F) Aurora B protein levels after transfection with Aurora B siRNAs were analysed by western blotting. The same membrane was probed for α-tubulin to demonstrate equal loading.

• Supplemental Figure 3 -

  Fig. S3. Condensin II levels are independent of Aurora B activity throughout mitosis. (A,B) Levels of condensin II were monitored during mitosis in cells stably expressing EGFP-Kleisin-β in the presence (B) or absence (A) of Hesperadin (100 nM). Hoechst 33342 (0.2 μg/ml) was used to visualise chromatin. Movies were taken on an Olympus DeltaVision microscope (40× objective, five slices every 5 minutes, 1×1 binning). Projected images are shown. (C,D) Quantification of the EGPF-Kleisin-β in early (NEBD) (C) and late mitosis (anaphase) (D) revealed no difference in the association of Condensin II with mitotic chromatin when cells were treated with Hesperadin (red, n=15) or left untreated (blue, n=15). EGFP-Kleisin-β intensities were normalised to the Hoechst signal.

• Supplemental Figure 4 -

  Fig. S4. Centromeric enrichment and chromosome arm association of condensin II (Cap-D3) is independent of Aurora B. (A) Nocodazole-arrested HeLa cells were spread on glass slides under hypotonic conditions, fixed and stained with antibodies against condensin II (Cap-D3). CREST autoimmune serum was used to detect centromeric regions. Aurora B kinase activity was inactivated by Hesperadin treatment (100 nM) and monitored by staining for phosphorylated Histone H3 Ser10 (H3S10ph). Chromosomes were counterstained with DAPI (1 μg/ml). Bar, 10 μm. The quantification can be found in Fig. 3C. (B) A magnified chromosome of the same experiment is shown stained with DAPI, antibodies against Cap-D3 and CREST autoimmuneserum. Bar, 2 μm.

• Supplemental Figure 5 -

  Fig. S5. High resolution gel electrophoresis reveals that Aurora B does not phoshorylate Smc2 but Condensin I and II regulatory subunits in vitro. (A) The condensin I complex was immunoprecipitated from EGFP-FLAG-Kleisin-γ expressing HeLa cells lines by using anti-FLAG-antibody-coupled beads. Immunoprecipitates were incubated with recombinant Aurora B and radioactive ATP in vitro, either in the presence of DMSO (left lane) or 2 μM Hesperadin (right lane). Afterwards the eluted complexes were resolved on a gel under conditions that allowed separation of EGFP-Kleisin-γ and Smc2 (80 V for 4 hours on a 6% polyacrylamide gel). Silver staining (Silver), exposure (³²P) and mass spectrometry confirmed that the in vitro target of Aurora B is Kleisin-γ and not Smc2. (B) Under the same conditions as described in A, we were unable to resolve the Cap-D3 and Smc4 subunits of condensin II immunoprecipitated from EGFP-FLAG-Kleisin-β expressing cells. Kleisin-β as well as Cap-G2 however displays incorporation of ³²P indicating that they are substrates of Aurora B in vitro.