Survivin withdrawal by nuclear export failure as a physiological switch to commit cells to apoptosis

Abstract

Apoptosis is a tightly controlled process regulated by many signaling pathways; however, the mechanisms and cellular events that decide whether a cell lives or dies remain poorly understood. Here we showed that when a cell is under apoptotic stress, the prosurvival protein Survivin redistributes from the cytoplasm to the nucleus, thus acting as a physiological switch to commit the cell to apoptosis. The nuclear relocalization of Survivin is a result of inefficient assembly of functional RanGTP-CRM1-Survivin export complex due to apoptotic RanGTP gradient collapse. Subsequently, Survivin undergoes ubiquitination, which not only physically prevents its diffusion back to the cytoplasm but also facilitates its degradation. Together, this spatial and functional regulation of Survivin abolishes its cytoprotective effect toward the apoptotic executors and thus commits a cell to apoptosis. Our data indicate that the withdrawal of Survivin is a novel and active physiological regulatory mechanism that tilts the survival balance and promotes the progression of apoptosis.

Figure 1

Apoptotic degradation of prosurvival protein Survivin (a) Total cell lysates collected at the stated time points after VP16 treatment and immunoblotted with anti-Survivin and anti-actin (loading control). (b) VP16 treatment-induced cell death as shown by immunoblots of cleaved Caspase-3, cleaved PARP and cleaved ICAD. (c) Total RNAs isolated from HeLa cells treated with or without VP16 at the indicated time points. Semiquantitative reverse transcription-PCR analysis was then performed using GAPDH as internal control. (d) Schematic representation of Survivin stability assay. (e) Experiments performed as shown by the schematics in (d). Endogenous Survivin level was probed at indicated time points after the resumption of the proteosome activity. Actin as loading control. (f) Quantified graphical data presented as mean±S.D. (error bar) of three independent experiments

Figure 2

RanGTP gradient collapse induces Survivin redistribution into the nucleus during apoptosis. (a) Survivin and mAb414 staining after VP16 treatment at indicated time points. Images acquired with auto-optimal exposure mode. (b) Quantification of cells with distinctive Survivin localization across the nucleocytoplasmic compartments. Results are presented as mean population percentage of HeLa cells having distinctive Survivin localization ±S.D. (error bar) of three independent experiments (n>300 cells). (c) Immunostaining of non-treated, and treated HeLa cells with LMB or VP16 against Survivin and mAb414. (d) FRET analysis of the nuclear RanGTP gradient by Rango biosensor (fusion protein of CFP and YFP flanking an importinβ binding domain). Rango undergoes FRET in the presence of RanGTP. Representative images (left) and quantified data (right) presented as mean±S.D. (error bar) of three independent experiments (n=120). ^*P<0.001 (Student's t-test). (e) Flag-Survivin was overexpressed, immunoprecipitated with anti-Flag from cells incubated with or without VP16 for 24 h. Bars, 10 μm. (right) Quantified Crm1 intensities normalized against Flag was presented as relative fold change ±S.D. (error bar) of three independent experiments

Figure 3

Restoration of RanGTP gradient reinstates Survivin export. (a) Immunoblotting against Mst1, Survivin, Ran and Actin was performed on lysates collected from mock, Control siRNA or Mst1 siRNA-transfected HeLa cells. (b) Rango FRET analysis was performed on Mst1 siRNA or Control siRNA-transfected HeLa cells, with or without VP16 treatment. Quantified data presented as mean±S.D. (error bar) of three independent experiments (n=120). ^**P<0.001 (Student's t-test) (c) HeLa cells transfected with Mst1 siRNA or Control siRNA before 24 h incubation with VP16. The cells were fixed and stained with anti-Survivin and mAb414. Bar, 10 μm. Images acquired with auto-optimal exposure mode

Figure 4

Ubiquitinated Survivin is sequestered and degraded in the nucleus upon apoptosis induction. (a) Nuclearcytoplasmic fractions were collected from CHX-treated cells at the indicated time points. The experiment was repeated on HeLa cells incubated with CHX and VP16 simultaneously. The samples were immunoblotted against Survivin, tubulin and GCN5. (b) Experiment was repeated as in (a) with the exception that CHX is replaced by MG132. (c) Survivin-WT-Myc was overexpressed and immunoprecipitated with anti-Myc antibody from HeLa cells incubated with or without VP16 for 24 h. The immunoprecipitated materials were immunoblotted against Survivin and ubiquitin

Figure 5

Apoptotic nuclear compartmentalized Survivin has abolished cytoprotectivity. (a) Nuclearcytoplasmic fractions collected from VP16-treated cells at the indicated time points. The samples were immunoblotted against Survivin, tubulin and GCN5. (b) Procaspase-3 and cleaved Caspase-3 are predominantly cytoplasmic in VP16-treated cells indicating early apoptosis. (c) Survivin depletion from HeLa cells by 24 h Survivin RNAi transfection and immunoblotted against Survivin and Actin. (d) HeLa cells were transfected with Survivin siRNA or Control siRNA before incubation with VP16 for 24 h. Cells were fixed and stained with anti-Survivin and mAb414. Bar, 10 μm. (e) Caspase-3 activity assay. Each experiment was performed in triplicates in three independent experiments. Data presented as mean±S.D. (error bar). ^*P<0.001 (Student's t-test). (f) Hypothetical model depicting the role of nuclear export system in regulating the commitment of cells during the onset of apoptosis