Nucleophosmin is a novel Bax chaperone that regulates ... - Nature

Oncogene (2007) 26, 2554–2562

& 2007 Nature Publishing Group All rights reserved 0950-9232/07 $30.00 www.nature.com/onc

ORIGINAL ARTICLE

Nucleophosmin is a novel Bax chaperone that regulates apoptotic cell death LE Kerr, J-LA Birse-Archbold, DM Short, AL McGregor1, I Heron, DC MacDonald, J Thompson, GJ Carlson, JS Kelly, J McCulloch and J Sharkey Astellas CNS Research in Edinburgh, The University of Edinburgh, Edinburgh, UK

The proapoptotic B-cell lymphoma-2 family protein Bax is a key regulatory point in the intrinsic apoptotic pathway. However, the factors controlling the process of Bax activation and translocation to mitochondria have yet to be fully identified and characterized. We performed affinity chromatography using peptides corresponding to the mitochondrial-targeting region of Bax, which is normally sequestered within the inactive structure. The molecular chaperone nucleophosmin was identified as a novel Bax-binding protein by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. Reciprocal co-immunoprecipitation and proximity assays confirmed the Bax-nucleophosmin protein–protein interaction and verified that nucleophosmin only bound to activated conformationally altered Bax. Confocal microscopy in a cell-based apoptosis model, demonstrated that nucleophosmin translocation from nucleolus to cytosol preceded Bax movement. Specific knockdown of nucleophosmin expression using RNAi attenuated apoptosis as measured by mitochondrial cytochrome c release and activation of the caspase cascade. In a mouse model of ischaemic stroke, subcellular fractionation studies verified that nucleophosmin translocation occurred within 3 h, at a time before Bax translocation but after Bax conformational changes have occurred. Thus, we have elucidated a novel molecular mechanism whereby Bax becomes activated and translocates to the mitochondria to orchestrate mitochondrial dysfunction and apoptotic cell death, which opens new avenues for therapeutic intervention. Oncogene (2007) 26, 2554–2562. doi:10.1038/sj.onc.1210044; published online 30 October 2006 Keywords: Bcl-2 family; translocation; apoptosis; focal cerebral ischaemia; protein–protein interaction

Correspondence: Dr LE Kerr, Astellas CNS Research in Edinburgh, The University of Edinburgh, The Chancellor’s Building, 49 Little France Crescent, Edinburgh EH16 4SB, UK. E-mail: [email protected] 1 Current address: Department of Pharmacology, Faculty of Medical and Health Sciences, The University of Auckland, Auckland, New Zealand. Received 19 June 2006; revised 17 August 2006; accepted 7 September 2006; published online 30 October 2006

Introduction The balance of pro- and anti-apoptotic members of the Bcell lymphoma (Bcl)-2 family of proteins is a key determinant of cell survival and is essential for normal development and organ homeostasis. Disturbances to this balance have been implicated in the pathogenesis of numerous disease states, including cancer, autoimmune diseases, stroke and a variety of neurodegenerative disorders (reviewed by Akhtar et al., 2004). Bax, the first proapoptotic member of the Bcl-2 family to be identified, is essential for neuronal apoptotic cell death (White et al., 1998; Chiesa et al., 2005). Bax exists as a monomeric cytosolic protein in normal, healthy cells (Nechushtan et al., 1999), but following exposure to a proapoptotic stimulus, Bax undergoes a conformational change and translocates to the mitochondria where (inter alia) it can induce the release of cytochrome c, resulting in the activation of caspases (Wolter et al., 1997; Narita et al., 1998). Nuclear magnetic resonance spectroscopy indicates that cytosolic Bax consists of nine a helices (ha1–a9) (Suzuki et al., 2000). The last 21 amino acids of Bax at the C terminus (Thr172-Gly192), which form most of helix a9, are essential for the targeting of Bax to the mitochondria as deletion of this hydrophobic sequence abrogates the ability of Bax to insert into the mitochondria during apoptosis (Wolter et al., 1997). Substituting the serine residue at position 184 with a valine (Bax S184V) or deleting the serine altogether (Bax DS184) results in constitutive mitochondrial localization of Bax (Nechushtan et al., 1999; Suzuki et al., 2000). Furthermore, fusion of the 20/21-amino-acid C-terminal sequence of Bax DS184 or S184V to the reporter gene green fluorescent protein (GFP) is sufficient to constitutively target GFP to mitochondria, whereas fusion proteins containing the wildtype (WT) C-terminal peptide do not translocate to mitochondria (Nechushtan et al., 1999), suggesting that the mutant peptide somehow mimics the active conformation of Bax. The present studies have employed WT and S184V mutant Bax C-terminal peptides to identify factors that may regulate Bax activation and translocation to the mitochondria and thus control apoptotic cell death. We identify nucleophosmin as a novel Bax C-terminal-binding protein by affinity chromatography and demonstrate that it regulates Bax-mediated cell death in vitro (following staurosporine treatment) and in vivo (following transient focal cerebral ischaemia).

Nucleophosmin is a Bax chaperone LE Kerr et al

2555

Results Identification of nucleophosmin as a Bax-binding molecule Peptides corresponding to the C-terminal sequence of WT and S184V mutant Bax were used to pull out proteins, which regulate the translocation of Bax from whole-cell lysates of a human neuroblastoma cell line (SH-SY5Y cells). Bound proteins were analysed by sodium dodecyl sulphate–polyacrylamide gel electrophoresis (SDS–PAGE) and identified by matrix-assisted laser desorption/ionization time-of-flight (MALDITOF) mass spectrometry (Figure 1a). Three proteins that bound to both peptides were identified as heat shock protein 90 (HSP90), HSP60 and b-actin. A further two proteins bound only to the S184V mutant peptide and one of these was identified as nucleophosmin. Successive attempts failed to identify the other lowabundance protein, which had a molecular mass of approximately 26 kDa. The interaction between Bax and nucleophosmin was confirmed by co-immunoprecipitiation of endogenous proteins from SH-SY5Y cell lysates (Figure 1b and c) and using in vitro expressed recombinant proteins (Figure 1d). Binding of the mutant Bax peptide to nucleophosmin was also demonstrated using an amplified luminescence proximity assay (Figure 2a and b). Moreover, this protein–protein interaction was competitively inhibited by an antibody directed against the C terminus of Bax but not by one raised against the N terminus of Bax (Figure 2c). This technology was also used to investigate the interaction between nucleophosmin and full-length Bax. No binding was observed unless the conformation of Bax was altered by including the non-ionic detergent Nonidet P40 (NP-40, 0.1%) in the reaction buffer (Figure 2d).

Figure 1 Nucleophosmin (NPM) binds to the C terminus of Bax. (a) Affinity chromatography was performed with wild-type (WT) and S184V Bax C-terminal peptides and bound proteins analysed by SDS–PAGE and identified by MALDI-TOF mass spectrometry. (b, c) NPM, SMN and Bax were immunoprecipitated from SH-SY5Y cell lysates and analysed for co-immunoprecipitation of Bax, NPM and GRB2 by Western blotting. (d) Bax and NPM were expressed in vitro using a cell-free expression system and analysed for reciprocal co-immunoprecipitation as indicated.

Knockdown of nucleophosmin expression attenuates apoptosis Staurosporine is a broad-spectrum protein kinase inhibitor that induces apoptosis via the intrinsic mitochondrial pathway. Staurosporine treatment of SH-SY5Y cells causes activation of Bax and translocation of this proapoptotic molecule to mitochondria with subsequent release of cytochrome c, cleavage of procasapse-3 and ultimately cell death. Therefore, the role of nucleophosmin in Bax-mediated apoptotic cell death was evaluated using this model of apoptosis. Nucleophosmin expression levels did not change over the 5 h time course of staurosporine treatment (Figure 3a), whereas cytochrome c was progressively released from mitochondria into the cytosol (Figure 3b) and levels of cleaved caspase-3 steadily increased (Figure 3c). Confocal microscopy revealed that Bax is primarily cytosolic in control SH-SY5Y cells and following 2 h exposure to staurosporine. However by 4 h, Bax translocation to the mitochondria was evident (Figure 4). In contrast, nucleophosmin was detected principally in the nucleolus of control cells, but had a more widespread distribution throughout the cytoplasm of apoptotic cells at 2 and 4 h of staurosporine treatment (Figure 4). To establish whether nucleophosmin is involved in Bax-mediated apoptosis, nucleophosmin expression was knocked down in SHSY5Y cells using short hairpin RNA (shRNA) (Figure 5a) and apoptosis induced using staurosporine. A specific reduction in nucleophosmin protein levels of B50% was associated with concomitant reductions of B50% in cytochrome c released and caspase-3 cleaved (Figure 5b and c). Roles of Bax and nucleophosmin in ischaemic cell death Mice were subjected to transient occlusion of the middle cerebral artery (MCA) for 30 min and brains processed for histological analysis at 3, 6, 12 and 24 h. Thionin staining confirmed the classical MCA territory lesion encompassing cortex and striatum at 6, 12 and 24 h (Figure 6a). No lesion was evident at 3 h, in shamoperated animals or hemispheres contralateral to occlusion of the MCA. Activated Bax was immunoprecipitated from the ipsilateral cortex and striatum 3 h after the onset of the ischaemic insult (Figure 6b). A small amount of activated Bax was also evident in contralateral samples, but no active Bax was seen in the cerebellum or samples from sham-operated animals. Alterations in the subcellular localizations of Bax and nucleophosmin following focal cerebral ischaemia were analysed using enriched nuclear, mitochondrial and cytosolic fractions prepared from samples of ischaemic cortex (3 h after the onset of ischaemia) and the corresponding cortical area of sham-operated mice (n ¼ 6/group). The relative subcellular distribution of Bax was unchanged following the ischaemic challenge (Figure 6c), whereas nucleophosmin expression was reduced in the nuclear-enriched fraction and increased in the mitochondrial and cytosolic fractions in the ischaemic group (Figure 6d). Oncogene


2556

Figure 2 Verification of Bax–NPM interaction by proximity assay. (a) Schematic representation of the proximity assay designed to quantify real-time binding. Biotinylated (B) Bax C-terminal peptide was directly attached to streptavidin (SA)-coated donor beads (D). His-tagged NPM was attached to the acceptor beads (A) via a NPM monoclonal antibody (Sigma). The donor bead is excited by a laser beam (680 nm) and produces singlet oxygen. If the two proteins interact, the acceptor beads are brought into close proximity to the excited donor bead and the singlet oxygen released upon laser excitation sets off a chemical cascade in the acceptor bead resulting in the emission of light at 520–620 nm. This emitted light is a measure of the binding between the two proteins of interest. In experiments to study the interaction between the whole proteins, the C-terminal Bax peptide was replaced by biotinylated full-length Bax. (b) Binding curve for the interaction between S184V Bax C-terminal peptide and NPM demonstrating increased signal at higher concentrations of the proteins indicative of binding. The characteristic hook seen at the maximum protein concentrations is due to nonspecific quenching of the signal. Negative control reactions containing only one protein are indicated Bax and NPM. Error bars represent s.e.m.; n ¼ 3. (c) S184V Bax C-terminal peptide-NPM binding was inhibited by an antibody raised against the Bax C terminus () but not by one targeted to the N terminus (n). Data are expressed as percentage of control (Bax peptide-NPM binding). (d) Binding between NPM and full-length Bax protein was observed only when NP-40 was included in the reaction buffer (BaxNPM NP-40), whereas no interaction was detectable in buffer containing CHAPS (BaxNPM CHAPS) or using each protein individually (Bax, NPM). Error bars represent s.e.m.; n ¼ 3.

Discussion Although there is some consensus regarding the method of Bax activation during apoptosis (whereby Bax undergoes a conformational change and subsequently translocates to mitochondria), the factors controlling this process have yet to be fully identified and characterized. Controversy in the literature also exists about which specific region of Bax is pivotal to its apoptosis-inducing function with data from some researchers, suggesting that the N terminus contains the mitochondrial targeting sequence (Tremblais et al., 1999; Cartron et al., 2003), whereas others argue that this activity resides in the sequence located at the C terminus (Wolter et al., 1997; Nechushtan et al., 1999; Schinzel et al., 2004). Nonetheless, all published data agree that using the native 21 amino-acid C terminus of Bax alone never results in mitochondrial translocation, whereas the same construct with the serine (corresponding to position 184 in full-length Bax) mutated to a valine (S184V) is constitutively targeted to mitochondria (Nechushtan et al., 1999; Mar Martinez-Senac et al., Oncogene

2001; Cartron et al., 2003; Schinzel et al., 2004). Therefore, to identify proteins that may be involved in regulating the translocation of Bax, we used a Bax Cterminal S184V mutant peptide, which artificially mimics the correct conformation for targeting Bax to the mitochondria and the WT Bax sequence as control. HSP90, HSP60 and b-actin bound to both peptides. We have also demonstrated the binding of Bax to HSP60 and b-actin by co-immunoprecipitation and proteomic analysis; however, the interaction with HSP90 could not be confirmed (MacDonald et al., submitted). In the present studies, HSP60 and b-actin bound to the Bax C-terminal peptides regardless of their ability to target to the mitochondria suggesting that these proteins are not involved in the translocation process per se, but may serve to help retain Bax in the cytosol. Studies on cardiac myocytes support this theory at least for HSP60, as cytosolic HSP60 translocation to the plasma membrane following hypoxia or ATP depletion coincides with the release of cytochrome c from mitochondria (Gupta and Knowlton, 2002) and knockdown of HSP60 expression precipitates apoptosis (Kirchhoff et al., 2002).


2557

Figure 4 Bax and NPM translocate during apoptosis. SH-SY5Y cells were treated with staurosporine (STS, 2 or 4 h) or dimethylsulphoxide vehicle (control, 4 h) and the subcellular localizations of Bax (top panels, green) and NPM (bottom panels, red) analysed by confocal microscopy. Data are representative of three essentially identical experiments.

Figure 3 NPM expression does not change during apoptosis. Apoptosis was induced in SH-SY5Y cells using 500 nM staurosporine (0–5 h) and analysed for: (a) changes in NPM expression level by Western blotting, (b) levels of cytosolic cytochrome c by ELISA and (c) the appearance of cleaved caspase-3 in the cytosol by ELISA. Cleaved caspase-3 data are expressed as a percentage relative to the positive control (camptothecin-treated Jurkat cell lysate). Data are presented as mean of three separate experiments performed with individual data points shown.

In addition to identifying HSP60 and b-actin as Bax retention factors, we have crucially demonstrated that nucleophosmin is a novel activated Bax-binding protein. Nucleophosmin is a multifunctional nucleolar phosphoprotein, which has been shown to have nucleic acid binding, ribonuclease and molecular chaperone activities (Hingorani et al., 2000). The observations that nucleophosmin only bound to the mutant C-terminal peptide and only bound to the full-length protein after conformational change was induced by detergent treatment (Hsu and Youle, 1997) would be consistent with a role in the translocation process. Such a role is not unprecedented, as nucleophosmin has previously been shown to regulate the movement of proteins containing nuclear localization sequences such as retinoblastoma

protein (Takemura et al., 2002) and the HIV Tat protein (Li, 1997), although in each case nucleophosmin has been implicated in the translocation of protein into the nucleolus. However, nucleophosmin has been reported to shuttle constantly between the nucleus and cytoplasm (Borer et al., 1989) and translocation of nucleophosmin (although only to the nucleoplasm) in response to a number of apoptosis-inducing drugs (such as daunomycin, actinomycin D, camptothecin and toyocamycin) has been demonstrated (Chan et al., 1999). There is also evidence that treatment of HeLa cells with staurosporine resulted in the dephosphorylation of nucleophosmin and its translocation to the cytosol (Lu et al., 1996). The interaction between Bax and nucleophosmin was confirmed by reciprocal co-immunoprecipitation studies and proximity assays. Further evidence for a role for nucleophosmin in regulating Bax translocation comes from studies detailing the localization of the proteins within the cell. Under normal conditions, nucleophosmin and Bax are physically separate as they are primarily located in different compartments within the cell: nucleophosmin within the nucleolus and Bax in the cytoplasm. However, following an apoptotic stimulus, nucleophosmin accumulated within the cytoplasm before Bax translocation to the mitochondria. Subcellular fractionation analysis of ischaemic cortex from mice confirmed our in vitro observations. Unfortunately, dual-labelling studies proved to be unfeasible as the respective antibody fixation procedures were mutually incompatible. A number of other Bax-binding proteins have recently been identified including Ku70 (Sawada et al., 2003), humanin (Guo et al., 2003), apoptosis repressor with caspase recruitment domain (ARC; Gustafsson et al., 2004) and several 14-3-3 isoforms (Nomura et al., 2003). These proteins have been proposed as putative Bax regulators as knockdown studies sensitize cells to death stimuli, thus suggesting that they function as cytosolic Bax-retention factors (Guo et al., 2003; Nomura et al., Oncogene


2558

Figure 5 Knockdown of NPM expression attenuates apoptosis in vitro. SH-SY5Y cells were treated with scrambled (control) or NPM shRNA or transfection reagent alone and apoptosis induced 24 h later with staurosporine treatment. (a) NPM expression levels were analysed by Western blotting. b-Actin was used as loading control and to ensure specificity of protein knockdown. (b) Levels of cytosolic cytochrome c and (c) levels of cleaved caspase-3 were measured by ELISA. Data are representative of at least three essentially identical experiments.

2003; Sawada et al., 2003; Nam et al., 2004). In contrast to the aforementioned retention factors, knockdown of nucleophosmin expression resulted in attenuation of staurosporine-induced apoptosis. Similar roles for regulating Bax have been suggested for the apoptosisassociated speck-like protein (ASC; Ohtsuka et al., 2004), Bax-interacting factor-1 (Bif-1; Cuddeback et al., 2001) and the modulator of apoptosis protein via the tumour suppressor RASSF1A (Baksh et al., 2005). Knockdown of ASC expression reduced the apoptotic response to the tumour suppressor p53 or a genotoxic insult (etoposide; Ohtsuka et al., 2004) and knockdown Oncogene

of Bif-1 expression abrogated apoptosis induced by various intrinsic death signals (Takahashi et al., 2005). Interestingly, knockdown of RASSF1A resulted in reduced apoptotic response to death receptor activation (e.g. by tumour necrosis factor-a), but not intrinsic apoptotic stimuli such as staurosporine or etoposide (Baksh et al., 2005). Taken together, these data suggest that different factors/proteins may positively or negatively regulate the activation and translocation of Bax depending on the apoptotic stimulus and intracellular pathway utilized. Although knockdown of nucleophosmin expression in our hands attenuated staurosporineinduced apoptosis, others have reported that this potentiates apoptosis induced by hypoxia (Li et al., 2004) or NGF deprivation (Ahn et al., 2005). Whether these conflicting data reflect the different apoptotic stimuli or are due to another reason is not currently clear. However, similar data exist for clusterin, a recently identified activated Bax-binding protein, which appears to both inhibit (Zhang et al., 2005) and promote (Gleave and Jansen, 2003) cell death. In addition, recent data have suggested that calpain proteases both prevent and promote cell death in response to different stimuli (Tan et al., 2006). Nucleophosmin has also been shown to interact directly with and be a crucial regulator of the tumour suppressor p53 (Colombo et al., 2002). Although conflicting data have also been reported for the role of nucleophosmin in regulating p53-mediated cell death following ultraviolet radiation where both knockdown (Colombo et al., 2002; Kurki et al., 2004) and overexpression (Maiguel et al., 2004; Li et al., 2005) of nucleophosmin are protective, recent data suggest that nucleophosmin is a critical molecule in the regulation of p53, which sets a threshold for p53 activation (Maiguel et al., 2004). Caution should be taken when interpreting data obtained using immortalized cell lines such as the SHSY5Y cells used in this study, as data produced in vitro are not always replicated in vivo. For example, in vitro experiments suggested that C-terminal modifications of the tumour suppressor protein, p53 were critical for its regulation, although these data were not replicated in vivo (Krummel et al., 2005). We therefore examined the role of nucleophosmin in vivo using a mouse model of stroke. Previous data from our laboratory (Kerr et al., 2004b) and many others (Matsushita et al., 1998; Cao et al., 2001) have demonstrated Bax involvement in the apoptotic cell death that occurs following experimental transient occlusion of the MCA. Bax translocation has also been reported in both rats (Cao et al., 2001) and mice (Gao et al., 2005) using this model. These studies provide a backdrop for the present observation that Bax conformational change (6A7 reactivity) and nucleophosmin translocation were detected but no Bax translocation was evident in the ischaemic cortex at a time point before the development of a visible lesion (3 h post insult). Taken together, these data suggest a role for nucleophosmin in the regulation of Bax both in vitro and in vivo and as such it represents a novel therapeutic target for the numerous diseases associated with dysregulated apoptosis.


2559

Figure 6 Bax conformational change and NPM translocation occur in vivo following transient focal cerebral ischaemia. (a) Thioninstained sections of mouse brain (Bregma þ 0.74) at 3, 6, 12 and 24 h following 30 min of ischaemia. (b) Tissue samples (cortex and striatum from both contralateral and ipsilateral regions and cerebellum) from sham-operated and ischaemic (30 min occlusion followed by 2.5 h reperfusion) animals were analysed for the presence of activated Bax by immunoprecipitation with the conformational specific antibody 6A7. Inactive (I) and active (A) Bax were identified by Western blotting with a pan-reactive Bax antibody. (c, d) Samples of sham-operated (Sh) and ischaemic (MCAo) cortex (n ¼ 6/treatment group) were separated into enriched nuclear, mitochondrial and cytosolic fractions and analysed for the relative levels of Bax and NPM.

Materials and methods Cell culture SH-SY5Y cells were grown and apoptosis induced (500 nM staurosporine, 0–5 h) as described (Birse-Archbold et al., 2005). Affinity chromatography and mass spectrometry Columns of biotinylated peptides corresponding to the Cterminal 21 amino acids of WT (biotin-TVTIFVAGVLTASLTIWKKMG) and S184V mutant (biotin-TVTIFVAGVLTAVLTIWKKMG) Bax (CSS-Albachem, Gladsmuir, East Lothian, UK) coupled to NeutrAvidin–agarose beads (Pierce, Cramlington, UK) were equilibrated in a mobile phase of 10 mM piperazine-N,N0 -bis[2-ethanesulphonic acid], 150 mM NaCl and 1% Tween (pH 6.5). Affinity chromatography was performed using SH-SY5Y cell lysates (2–5 mg total protein). Specifically bound proteins were eluted using 100 mM glycine (pH 2) (1 ml), separated by Tris/Glycine PAGE (Invitrogen, Paisley, UK) and visualized using a mass spectrometrycompatible silver stain. Protein bands were excised and digested according to Shevchenko et al. (1996), with some modifications. Mass spectra were collected using a VoyagerDE STR MALDI-TOF (Applied Biosystems, Warrington, UK) and the peptide masses were used to interrogate the SwissProt database using MS-Fit (http://prospector.ucsf.edu/ ucsfhtml4.0/msfit.htm).

Protein expression cDNAs encoding human Bax and nucleophosmin (Invitrogen) were subcloned into pcDNA3.1( þ ) and pCRT7/NT-TOPO (both Invitrogen) and expressed in BL21 cells (Invitrogen) or in vitro (transcribed and translated protein, TNT, Promega, Southampton, UK) using standard molecular procedures. Immunoprecipitation and Western blotting Immunoprecipitation was performed using antibodies against Bax (BaxNT, 6 mg/mg protein; Upstate, Chandlers Ford, UK; and 6A7, 5/200 mg protein; BD Biosciences, Oxford, UK), nucleophosmin (5 mg/mg protein; Cell Signaling Technology, Hitchin, UK) and survival motor neuron protein (SMN, 5 mg/ mg protein; BD Biosciences). Complexes were isolated with protein G-conjugated paramagnetic dynabeads (Dynal Biotech, Paisley, UK) and analysed by Western blotting using Bax (BaxNT, 1/2000; Upstate), nucleophosmin (1/1000; Sigma, Poole, UK) and growth factor receptor-bound protein 2 (GRB2, 1/5000; BD Biosciences) antibodies. Proximity assay AlphaScreen technology (Amplified Luminescent Proximity Homogeneous Assay, Perkin-Elmer, Beaconsfield, UK) was used to measure complex formation between Bax and nucleophosmin following the manufacturer’s instructions. Studies were performed using 384-well white polystyrene proxiplates (Perkin-Elmer). In vitro expressed recombinant Oncogene


2560 nucleophosmin was attached to AlphaScreen acceptor beads via a nucleophosmin monoclonal antibody (Sigma), which was directly conjugated to the acceptor beads using standard amination according to the manufacturer’s protocol. Biotinlabelled S184V Bax C-terminal peptide (CSS-Albachem) was attached to streptavidin donor beads. The AlphaScreen assay contained 20 mg/ml of nucleophosmin antibody-conjugated acceptor beads, 20 mg/ml of streptavidin-coated donor beads and serial dilutions of S184V peptide (starting concentration 1 mg/ml) and nucleophosmin protein (2.5 mg/ml; each prepared in 150 mM NaCl, 10 mM N-2-hydroxyethylpiperazineN0 -2-ethanesulphonic acid-potassium hydroxide (HEPESKOH) (pH 7.4), 0.1% NP-40, 0.01% bovine serum albumin (BSA)). Competition assays were performed using antibodies directed against the C terminus (1D3; Abcam, Cambridge, UK, product number ab16910) or N terminus (1-21aa; Upstate) of Bax and titrations of each antibody ranging from 10 to 0.01 nM were pre-incubated (1 h) with 10 nM Hisnucleophosmin and 75 nM Bax C-terminal peptide before analysis using an Envision Xcite multilabel reader (PerkinElmer). The interaction between full-length Bax and nucleophosmin was examined using recombinant Bax (Expressway in vitro expression system; Invitrogen), which was biotinylated (ProtOn Biotin labelling Kit; Vector Laboratories, Peterborough, UK) to enable attachment to the streptavidin donor beads. The proximity assay was performed as above using 3.7 nM Bax and 5.7 nM nucleophosmin incubated in buffer (150 mM NaCl, 10 mM HEPES-KOH (pH 7.4), 0.01% BSA) containing 0.1% 3-[(3-cholamido propyl)-dimethylammonio]2-hydroxy-1-propanesulphonic acid (CHAPS) or 0.1% NP-40. Cytochrome c ELISA Cells were lysed (1 h) using 3.125 mM digitonin (which selectively lyses the plasma membrane allowing the release of cytoplasmic contents; Kirchhoff et al. (2002) in 40 mM HEPES (pH 7.4), 140 mM KCl, 20 mM NaCl, 5 mM MgCl2, 1 mM ethyleneglycol tetracetate (EGTA), 10 mg/ml each of aprotinin, pepstatin and leupeptin or using the same buffer containing 1% CHAPS (to rupture all cell membranes). Following centrifugation (13 000 r.p.m., 5 min), the cytochrome c contents of the resulting supernatants were measured by enzymelinked immunosorbent assay (ELISA) (Bender MedSystems, Botolph, Clayton, UK). Cytosolic cytochrome c (digitonin buffer) was measured as a percentage of total cellular cytochrome c (CHAPS buffer).

Caspase-3 ELISA Human active caspase-3 was detected by ELISA (BD Biosciences) as described previously (Kerr et al., 2004a). Confocal microscopy SH-SY5Y cells were seeded in glass-bottomed poly-L-lysinecoated Petri dishes (Willco Wells, 500 cells/35/22 mm diameter dish/glass) and allowed to attach overnight. Following induction of apoptosis, cells were fixed and stained with nucleophosmin antibody (B23, 1/200 in block solution; Sigma) and Bax antibody (Bax NT, 1:200 in phosphate-buffered saline; Upstate) according to the manufacturer’s protocols. Bound antibodies were detected using AlexaFluor546- or AlexaFluor488-labelled goat anti-mouse or anti-rabbit immunoglobulin (3 mg/ml, 2 h; both Molecular Probes, Paisley, UK) and visualized using confocal microscopy (Radiance 2000 microscope and LaserSharp 2000 software; both BioRad, Hemel Hempstead, UK). Oncogene

shRNA knockdown Nucleophosmin expression was knocked down using shRNA (pSilencer3.1-H1 hygro, Ambion, Huntingdon, UK) according to the manufacturer’s instructions. Target sequences were nucleotides 182–202 (AAGCAGAGGCAATGAATTACG), 362–382 (AAGATGCAGAGTCAGAAGATG), 470–490 (AACTTGCTGCTGATGAAGATG) and 716–736 (AAGGACCTAGTTCTGTAGAAG) of human nucleophosmin. Cells were plated (1 105 cells/well) in 24-well culture dishes and transfected 24 h later with plasmid DNA (1 mg) using Lipofectamine 2000 (2 ml; Invitrogen). After a further 24 h, apoptosis was induced (500 nM staurosporine, 4 h) before analysis by Western blot or ELISA. Cerebral ischaemia and subcellular fractionation Animal work was performed under licence by the UK Home Office and subject to the Animal (Scientific Procedures) Act of 1986. Monofilament occlusion of the MCA was performed in adult male mice (25–30 g, approximately 10–12 weeks) as described previously (Kerr et al., 2004b). Animals were subjected to 30 min of ischaemia and killed at 3, 6, 12 and 24 h by trans-cardiac perfusion with 4% paraformaldehyde under deep anaesthesia (pentobarbitone, 60 mg/kg intraperitoneally). Brain damage was assessed in thionin-stained coronal brain sections by light microscopy. Samples were maintained at 41C throughout the extraction procedure. For immunoprecipitation studies, the ipsilateral striatum and MCA territory cortex were homogenized in ice-cold buffer (150 mM NaCl, 10 mM HEPES-KOH (pH 7.6), 1% CHAPS, 1 mM phenylmethylsulphonyl fluoride (PMSF), 10 mg/ml each aprotinin, pepstatin A, leupeptin). For subcellular fractionations studies, nuclear, mitochondrial and cytosolic fractions were isolated from mouse brain by differential centrifugation. The ipsilateral MCA territory cortex was dissected out, weighed, minced and placed in 40 ml ice-cold extraction buffer (EB; 10 mM HEPES-KOH (pH 8.0), 0.32 M sucrose, 1 mM EGTA, 25 mM KCl, 5 mM MgCl2, 0.1 mM dithiothreitol (DTT), 1 mM DNase I, 1 mM PMSF, 10 mg/ml each aprotinin, pepstatin A, leupeptin). Homogenates were prepared (six animals, 240 ml/treatment group) then passed through gauze (125 mm clearance). The crude nuclear pellet was isolated by centrifugation (600 g, 10 min), re-suspended in EB and centrifuged (600 g, 5 min) to pellet unbroken cells. The crude nuclear fraction was collected by centrifugation (600 g, 10 min), re-suspended in 3 ml nuclear buffer 1 (NE1; 0.25 M sucrose, 10 mM MgCl2), layered over 3 ml NE2 (0.35 M sucrose, 0.5 mM MgCl2) and centrifuged (1450 g, 5 min). The creamy layer of the pellet was re-suspended in 50 ml EB containing 2% CHAPS and retained as the nuclear fraction. The supernatant from the initial centrifugation step (600 g, 10 min) was homogenized using 15 ‘tight’ up/down strokes and centrifuged (10 000 g, 30 min). The crude mitochondrial (pellet) fraction was washed twice in EB (10 000 g, 30 min), resuspended in 50 ml EB containing 2% CHAPS and centrifuged (600 g, 10 min) to pellet out insoluble material. The crude cytosolic (supernatant) fraction was centrifuged (10 000 g, 30 min) to pellet out residual mitochondria and the supernatant centrifuged (100 000 g, 1 h). The resulting supernatant (B200 ml) was retained as the soluble cytosolic (S100) fraction. The total protein contents of each subcellular fraction were determined (bicinchoninic acid protein assay; Pierce) and analysed by Western blot (20 mg/lane). The marker proteins cjun (nuclear; BD Biosciences), voltage-dependent anion channel (mitochondrial; Oncogene, Nottingham, UK) and cleaved caspase-3 (cytosolic; Cell Signaling Technologies) were used to confirm fraction purity and equal loading. Gel analysis


2561 was performed using FluorChem image analyser software (AlphaInnotech, Braintree, UK). Cytosolic, mitochondrial and nuclear fractions (20 mg protein) from each treatment group (sham and ischaemic) were loaded on each gel. Values obtained for each immunoreactive band were normalized against the loading control for each fraction then used to calculate the total value in six-pooled brains (e.g. 2 ml mitochondrial sample loaded on gel from total 50 ml starting material; multiply value by 25). The fold change in nucleo-

phosmin and Bax expression between different fractions was calculated and relative changes in the subcellular distribution of each protein assessed between treatment groups. Acknowledgements This work was funded by a research grant from Astellas Pharma Inc., Tokyo, Japan.

References Ahn JY, Liu X, Cheng D, Peng J, Chan PK, Wade PA et al. (2005). Nucleophosmin/B23, a nuclear PI(3,4,5)P(3) receptor, mediates the antiapoptotic actions of NGF by inhibiting CAD. Mol Cell 18: 435–445. Akhtar RS, Ness JM, Roth KA. (2004). Bcl-2 family regulation of neuronal development and neurodegeneration. Biochim Biophys Acta 1644: 189–203. Baksh S, Tommasi S, Fenton S, Yu VC, Martins LM, Pfeifer GP et al. (2005). The tumor suppressor RASSF1A and MAP-1 link death receptor signaling to Bax conformational change and cell death. Mol Cell 18: 637–650. Birse-Archbold JL, Kerr LE, Jones PA, McCulloch J, Sharkey J. (2005). Differential profile of Nix upregulation and translocation during hypoxia/ischaemia in vivo versus in vitro. J Cereb Blood Flow Metab 25: 1356–1365. Borer RA, Lehner CF, Eppenberger HM, Nigg EA. (1989). Major nucleolar proteins shuttle between nucleus and cytoplasm. Cell 56: 379–390. Cao G, Minami M, Pei W, Yan C, Chen D, O’Horo C et al. (2001). Intracellular Bax translocation after transient cerebral ischemia: implications for a role of the mitochondrial apoptotic signaling pathway in ischemic neuronal death. J Cereb Blood Flow Metab 21: 321–333. Cartron PF, Priault M, Oliver L, Meflah K, Manon S, Vallette FM. (2003). The N-terminal end of Bax contains a mitochondrial-targeting signal. J Biol Chem 278: 11633– 11641. Chan PK, Bloom DA, Hoang TT. (1999). The N-terminal half of NPM dissociates from nucleoli of HeLa cells after anticancer drug treatments. Biochem Biophys Res Commun 264: 305–309. Chiesa R, Piccardo P, Dossena S, Nowoslawski L, Roth KA, Ghetti B et al. (2005). Bax deletion prevents neuronal loss but not neurological symptoms in a transgenic model of inherited prion disease. Proc Natl Acad Sci USA 102: 238– 243. Colombo E, Marine JC, Danovi D, Falini B, Pelicci PG. (2002). Nucleophosmin regulates the stability and transcriptional activity of p53. Nat Cell Biol 4: 529–533. Cuddeback SM, Yamaguchi H, Komatsu K, Miyashita T, Yamada M, Wu C et al. (2001). Molecular cloning and characterization of Bif-1. A novel Src homology 3 domaincontaining protein that associates with Bax. J Biol Chem 276: 20559–20565. Gao Y, Signore AP, Yin W, Cao G, Yin XM, Sun F et al. (2005). Neuroprotection against focal ischemic brain injury by inhibition of c-Jun N-terminal kinase and attenuation of the mitochondrial apoptosis-signaling pathway. J Cereb Blood Flow Metab 25: 694–712. Gleave M, Jansen B. (2003). Clusterin and IGFBPs as antisense targets in prostate cancer. Ann NY Acad Sci 1002: 95–104.

Guo B, Zhai D, Cabezas E, Welsh K, Nouraini S, Satterthwait AC et al. (2003). Humanin peptide suppresses apoptosis by interfering with Bax activation. Nature 423: 456–461. Gupta S, Knowlton AA. (2002). Cytosolic heat shock protein 60, hypoxia, and apoptosis. Circulation 106: 2727–2733. Gustafsson AB, Tsai JG, Logue SE, Crow MT, Gottlieb RA. (2004). Apoptosis repressor with caspase recruitment domain protects against cell death by interfering with Bax activation. J Biol Chem 279: 21233–21238. Hingorani K, Szebeni A, Olson MO. (2000). Mapping the functional domains of nucleolar protein B23. J Biol Chem 275: 24451–24457. Hsu YT, Youle RJ. (1997). Nonionic detergents induce dimerization among members of the Bcl-2 family. J Biol Chem 272: 13829–13834. Kerr LE, Birse-Archbold JL, Simon A, Logan N, Scott F, Carlson G et al. (2004a). Differential regulation of caspase-3 by pharmacological and developmental stimuli as demonstrated using humanised caspase-3 mice. Apoptosis 9: 739– 747. Kerr LE, McGregor AL, Amet LE, Asada T, Spratt C, Allsopp TE et al. (2004b). Mice overexpressing human caspase 3 appear phenotypically normal but exhibit increased apoptosis and larger lesion volumes in response to transient focal cerebral ischaemia. Cell Death Differ 11: 1102–1111. Kirchhoff SR, Gupta S, Knowlton AA. (2002). Cytosolic heat shock protein 60, apoptosis, and myocardial injury. Circulation 105: 2899–2904. Krummel KA, Lee CJ, Toledo F, Wahl GM. (2005). The C-terminal lysines fine-tune P53 stress responses in a mouse model but are not required for stability control or transactivation. Proc Natl Acad Sci USA 102: 10188–10193. Kurki S, Peltonen K, Latonen L, Kiviharju TM, Ojala PM, Meek D et al. (2004). Nucleolar protein NPM interacts with HDM2 and protects tumor suppressor protein p53 from HDM2-mediated degradation. Cancer Cell 5: 465–475. Li J, Zhang X, Sejas DP, Bagby GC, Pang Q. (2004). Hypoxiainduced nucleophosmin protects cell death through inhibition of p53. J Biol Chem 279: 41275–41279. Li J, Zhang X, Sejas DP, Pang Q. (2005). Negative regulation of p53 by nucleophosmin antagonizes stress-induced apoptosis in human normal and malignant hematopoietic cells. Leukemia Res 29: 1415–1423. Li YP. (1997). Protein B23 is an important human factor for the nucleolar localization of the human immunodeficiency virus protein Tat. J Virol 71: 4098–4102. Lu YY, Lam CY, Yung BY. (1996). Decreased accumulation and dephosphorylation of the mitosis-specific form of nucleophosmin/B23 in staurosporine-induced chromosome decondensation. Biochem J 317: 321–327. Oncogene


2562 Maiguel DA, Jones L, Chakravarty D, Yang C, Carrier F. (2004). Nucleophosmin sets a threshold for p53 response to UV radiation. Mol Cell Biol 24: 3703–3711. Mar Martinez-Senac M, Corbalan-Garcia S, GomezFernandez JC. (2001). Conformation of the C-terminal domain of the pro-apoptotic protein Bax and mutants and its interaction with membranes. Biochemistry 40: 9983–9992. Matsushita K, Matsuyama T, Kitagawa K, Matsumoto M, Yanagihara T, Sugita M. (1998). Alterations of Bcl-2 family proteins precede cytoskeletal proteolysis in the penumbra, but not in infarct centres following focal cerebral ischemia in mice. Neuroscience 83: 439–448. Nam YJ, Mani K, Ashton AW, Peng CF, Krishnamurthy B, Hayakawa Y et al. (2004). Inhibition of both the extrinsic and intrinsic death pathways through nonhomotypic deathfold interactions. Mol Cell 15: 901–912. Narita M, Shimizu S, Ito T, Chittenden T, Lutz RJ, Matsuda H et al. (1998). Bax interacts with the permeability transition pore to induce permeability transition and cytochrome c release in isolated mitochondria. Proc Natl Acad Sci USA 95: 14681–14686. Nechushtan A, Smith CL, Hsu YT, Youle RJ. (1999). Conformation of the Bax C-terminus regulates subcellular location and cell death. EMBO J 18: 2330–2341. Nomura M, Shimizu S, Sugiyama T, Narita M, Ito T, Matsuda H et al. (2003). 14-3-3 interacts directly with and negatively regulates pro-apoptotic Bax. J Biol Chem 278: 2058–2065. Ohtsuka T, Ryu H, Minamishima YA, Macip S, Sagara J, Nakayama KI et al. (2004). ASC is a Bax adaptor and regulates the p53–Bax mitochondrial apoptosis pathway. Nat Cell Biol 6: 121–128. Sawada M, Sun W, Hayes P, Leskov K, Boothman DA, Matsuyama S. (2003). Ku70 suppresses the apoptotic translocation of Bax to mitochondria. Nat Cell Biol 5: 320–329.

Oncogene

Schinzel A, Kaufmann T, Schuler M, Martinalbo J, Grubb D, Borner C. (2004). Conformational control of Bax localization and apoptotic activity by Pro168. J Cell Biol 164: 1021– 1032. Shevchenko A, Wilm M, Vorm O, Mann M. (1996). Mass spectrometric sequencing of proteins silver-stained polyacrylamide gels. Anal Chem 68: 850–858. Suzuki M, Youle RJ, Tjandra N. (2000). Structure of Bax: coregulation of dimer formation and intracellular localization. Cell 103: 645–654. Takahashi Y, Karbowski M, Yamaguchi H, Kazi A, Wu J, Sebti SM et al. (2005). Loss of bif-1 suppresses bax/bak conformational change and mitochondrial apoptosis. Mol Cell Biol 25: 9369–9382. Takemura M, Ohoka F, Perpelescu M, Ogawa M, Matsushita H, Takaba T et al. (2002). Phosphorylation-dependent migration of retinoblastoma protein into the nucleolus triggered by binding to nucleophosmin/B23. Exp Cell Res 276: 233–241. Tan Y, Wu C, De Veyra T, Greer PA. (2006). Ubiquitous calpains promote both apoptosis and survival signals in response to different cell death stimuli. J Biol Chem 281: 17689–17698. Tremblais K, Oliver L, Juin P, Le Cabellec TM, Meflah K, Vallette FM. (1999). The C-terminus of bax is not a membrane addressing/anchoring signal. Biochem Biophys Res Commun 260: 582–591. White FA, Keller-Peck CR, Knudson CM, Korsmeyer SJ, Snider WD. (1998). Widespread elimination of naturally occurring neuronal death in Bax-deficient mice. J Neurosci 18: 1428–1439. Wolter KG, Hsu YT, Smith CL, Nechushtan A, Xi XG, Youle RJ. (1997). Movement of Bax from the cytosol to mitochondria during apoptosis. J Cell Biol 139: 1281–1292. Zhang H, Kim JK, Edwards CA, Xu Z, Taichman R, Wang CY. (2005). Clusterin inhibits apoptosis by interacting with activated Bax. Nat Cell Biol 7: 909–915.