|
|
|
|
|
|
|
|
|||||
| Home Help Feedback Subscriptions Archive Search Table of Contents | |||||
| ||||||||||||||||||||
Files in this Data Supplement:
Fig. S1. RANBP1-specific and control siRNAs. (A) The position of siRNAs used in this work is shown relative to the human RANBP1 coding sequence. The RAN-binding domains and the nuclear export sequence are indicated. The following siRNAs were designed to interfere with the human RANBP1 gene: 202: GGAGCGAGGCACUGGUGAC; 459: AGUUUGAAGAAUGCAGGAA; 478: AGAGAUCGAAGAGAGAGAA. The broken line indicates the position of the region homologous to that targeted by siRNA 116, directed against the murine Ranbp1 gene (CGCUGGAGGAAGAUGAAGAA), highly related but not identical to the human sequence. (B) Western blot of RANBP1 protein after 48 hours of RNA interference with the indicated siRNAs (60 nM). Mix indicates that equimolar amounts of all three human siRNAs were used (20 nM each). Negative control siRNAs included: GFP siRNA, targeting the green fluorescent protein sequence; GL2, against the firefly luciferase gene; and murine-specific 116 siRNA. Most experiments were carried out using 202, Cy3-labeled 202 or 459 siRNAs. Systematic experiments were carried out in U20S cultures to compare: (a) their effect on the mitotic index (MI); (b) the induction of apoptosis by FACS; (c) the presence of resistant MTs to ice-induced depolymerization and to NOC-dependent inhibition of assembly; and (d) the induction of chromosome segregation defects. None of the negative control siRNAs interfered with RANBP1 protein abundance in western blot (B) or had any biological effect in the assays described above. Thereafter, GL2 was used as a negative control in most experiments.
Fig. S2. The AR12 antibody depicts variations in endogenous Ran-GTP levels. The AR12 antibody was characterized in HPLC assays using purified human RAN protein pre-loaded with either GTP or GDP nucleotide (Richards et al., 1995). AR12 was previously used in immunolocalization experiments in human mitotic cells (Keryer et al., 2003; Ciciarello et al., 2004) and revealed an accumulation at asters, spindle poles and MTs, consistent with that expected for RAN-GTP based on biochemical and functional data. To directly validate the specificity of AR12 further, we have experimentally altered endogenous Ran-GTP levels in rodent cell lines. (A) AR12 staining pattern in hamster tsBN2 cells, harbouring a temperature-sensitive RCC1 allele. Independent secondary antibodies, conjugated to rhodamine or to FITC, were used. The wild-type AR12 pattern in control cells cultured at 32°C is shown on the left. After three hours of culture at 39,5°C (right panels), mutant RCC1 is degraded, thus impairing Ran-GTP production: this is associated with significantly weakened AR12 signals, particularly evidently around the NE, where Ran-GTP is normally bound to Ranbp2. Bar, 10 μm. (B) As an independent strategy to impair Ran-GTP accumulation, Ranbp1 was overexpressed to increase GTP hydrolysis on RAN. Murine NIH/3T3 cells were used, in which pRanbp1 expression plasmids can be transfected with particularly high efficiency. IF panels: AR12 is detected by AMCA-conjugated secondary antibody (blue) in NIH/3T3 cells transfected with empty vector, or with pRanbp1 expression construct; in both cases cells were co-transfected with H2B-GFP to visualize nuclei. Bar, 10 μm. Bottom panel: the histograms show the distribution of AR12 signal intensities, measured by densitometry and expressed in fluorescence arbitrary units (a.u.), in nuclei of control cells (H2BGFP/empty vector, n=25) and of Ranbp1-overexpressing cells (H2BGFP/Ranbp1construct, n=32). Ranbp1-overexpressing cells are tendentially grouped in the region of decreased AR12 signal intensity.
Fig. S3. RANBP1 depletion does not alter the localization of total RAN in mitotic cells. RAN was detected using an antibody to the protein ‘core’ after mild solubilization of cell samples to render the experimental conditions comparable to those used for AR12 (see Table S1). In prophase (panel a) RAN essentially associates with the condensing chromatin and forming asters; from prometaphase (panel b) onwards (panel c), most of it colocalizes with the spindle MTs. RANBP1 interference (panels e,f) does neither alter the total amount nor the pattern of localization of RAN (n=380 controls and 290 RANBP1-interfered mitoses). Bar, 10 μm.
Fig. S4. RANBP1 depletion does not alter the behaviour of MAD2 and BUB1 at kinetochores. (A) IF pattern of MAD2 during mitotic progression; unaligned KTs are squared. Bar, 10 μm. (B) MAD2 localization was analyzed in a total of 331 control and 273 RANBP1-interfered mitotic cells (all stages); histograms represent the frequency of MAD2 association to KTs (visualized by CREST) in each stage. RANBP1 interference does not affect the overall timing of residency of MAD2 on KTs. Data from two experiments are shown. (C) MAD2 recruitment was found to be unaffected by RANBP1 levels when single unaligned KTs were examined (n=173 control and 143 RANBP1-interfered chromosomes in two experiments). (D) BUB1 pattern during mitotic progression; unaligned KTs are squared. Bar, 10 μm. (E) The localization of BUB1 was analyzed in 400 control and 280 RANBP1-interfered mitotic cells (all stages); histograms represent the frequency of BUB1 association to KTs (visualized by CREST) in each stage; RANBP1-interference does not affect the temporal pattern of BUB1 recruitment at, and release from, KTs. S.d. was calculated from three experiments. (F) The localization of BUB1 was examined after treatment with MT-damaging drugs. After 24 hours in 100 nM taxol, BUB1 localized at KTs in more than 80% of mitotic cells, regardless of RANBP1 levels. After 20 hours in 200 ng/ml NOC, BUB1 associated with KTs in about 95% of mitotic cells, regardless of RANBP1 levels. At least 200 mitotic cells per condition were counted in two experiments.
Fig. S5. RANBP1 depletion does not affect the spindle checkpoint response to NOC. GL2- and RANBP1-interfered cultures were treated with the indicated concentrations of NOC for 4 hours while in late G2 (i.e. starting 7 hours after G1/S block release), or directly in NOC for the indicated lengths of time. Control cultures were treated with DMSO for 20 hours. RANBP1 depletion does not increase the frequency of multinucleated (MN) cells, indicating that the checkpoint response to the lack of MT attachment is effective with and without RANBP1.
Movie 1. RANBP1(i)-dependent delay in early mitosis. Time-lapse recording of a representative RANBP1-interfered U20S cell delayed in early mitotic stages. At the onset of recording the cell begun to round up with condensed chromosomes and failed to progress from that state throughout the entire duration of the video. Bright-field frames were recorded every 2 minutes. Cy3 frames were taken every hour. Time-lapse data were processed using ImageJ software.
Movie 2. RANBP1(i)-dependent apoptosis in early mitosis. A representative RANBP1-interfered U20S cell: notice that the cell begins to round-up with condensed chromosomes at the onset of recording and fails to progress past that state; time-lapse recording reveals nuclear fragmentation, compatible with apoptosis, during prolonged delay in early mitosis. Bright-field frames were recorded every 2 minutes. Cy3 frames were taken every hour. Time-lapse data were processed using ImageJ software.
Movie 3. RANBP1(i) induces lagging chromosomes in anaphase after efficient alignment in metaphase. Time-lapse recording of a RANBP1-interfered HeLa cell stably expressing H2B-GFP while accomplishing mitosis. Note the appearance of the lagging chromosome in anaphase after full alignment was achieved in metaphase, compatible with merotelic attachment. Recording started 8 hours after thymidine release. GFP, Cy3 and bright-field frames were recorded every 2 minutes. H2B-GFP images are maximum intensity projections of seven z planes that were acquired 5 μm apart. Time-lapse data were processed using Metamorph software.
| ||||||||||||||||||||
