|Home | About | Journals | Submit | Contact Us | Français|
The phosphatidylinositol 3-kinase (PI3-K)/Akt pathway is well known for the regulation of cell survival, proliferation, and some metabolic routes.
In this study, we document a novel role for the PI3-K/Akt pathway during cell death induced by apoptin, a tumour-selective inducer of apoptosis.
We show for the first time that apoptin interacts with the p85 regulatory subunit, leading to constitutive activation of PI3-K. The inhibition of PI3-K activation either by chemical inhibitors or by genetic approaches severely impairs cell death induced by apoptin. Downstream of PI3-K, Akt is activated and translocated to the nucleus together with apoptin. Direct interaction between apoptin and Akt is documented. Co-expression of nuclear Akt significantly potentiates cell death induced by apoptin. Thus, apoptin-facilitated nuclear Akt, in contrast to when in its cytoplasmic pool, appears to be a positive regulator, rather than repressor of apoptosis.
Our observations indicate that PI3-K/Akt pathways have a dual role in both survival and cell death processes depending on the stimulus. Nuclear Akt acts as apoptosis stimulator rather than as a repressor, as it likely gains access to a new set of substrates in the nucleus. The implicated link between survival and cell death pathways during apoptosis opens new pharmacological opportunities to modulate apoptosis in cancer, for example through the manipulation of Akt’s cellular localization.
The phosphatidylinositol 3-kinase (PI3-K) is a lipid kinase that catalyses phosphorylation of the inositol ring of phosphoinositides [PI, PI(4)P and phosphatidylinositol (4,5)-bisphosphate (PI(4,5)P2)] at the D3 position (Fruman et al. 1999; Rameh & Cantley 1999). Three classes of PI3-K have been identified which differ in their primary structure, regulation and substrate specificity (reviewed in Vanhaesebroeck & Waterfield 1999). Class I PI3-Ks have been the major focus of PI3-K studies and are heterodimers composed of a catalytic subunit (p110) and a regulatory subunit (p85). Interaction of the nSH2 domain with the tyrosine-phosphorylated proteins is universally accepted to be responsible for PI3-K activation, but there are other studies that report activation of PI3-K via alternative mechanisms. For example, the conformational switch within the p85–p110 holoenzyme can also occur via interactions of SH3 domain/proline rich sequences, BCR-homology domain/GTP-loaded adaptor proteins and others (Liu et al. 1993; Prasad et al. 1993; Pleiman et al. 1994; Zheng et al. 1994).
Akt/PKB, a serine/threonine kinase, is a crucial kinase downstream of the PDK1 recruitment to the phosphatidylinositol (3,4,5)-trisphosphate [PtdIns(3,4,5)P3] lipid messenger, produced via PI3-K activation (Marte & Downward 1997). Activated Akt modulates the function of numerous substrates related to the regulation of cell proliferation, such as glycogen synthase kinase-3 (GSK-3), cyclin-dependent kinase inhibitors, p21Cip1/Waf1, p27kip1, and mammalian target of rapamycin (mTOR) (Fruman et al. 1998; Song et al. 2005). Another important function of activated PI3-K/Akt in cells is maintaining cell survival by inhibition of apoptosis. The targets of Akt during this process include Bad phosphorylation (a pro-apoptotic Bcl-2 family member), FKHRL1 inactivation (a transcription factor for FasL and Bim) and nuclear factor kappa B activation (Coffer et al. 1998; Vanhaesebroeck & Alessi 2000).
Apoptin is a 14-kDa protein derived from the chicken anaemia virus. Apoptin selectively induces apoptosis in cancer cells but not in non-malignant cells (reviewed in Alvisi et al. 2006; Maddika et al. 2006); in these, apoptin remains in the cytoplasm, while in transformed cells it migrates into the nucleus and ultimately kills the cell by activation of the mitochondrial death pathway, in a Nur77-dependent manner, independently of death receptors (Los et al. 1995b; Danen-Van Oorschot et al. 2003; Maddika et al. 2005; Poon et al. 2005b). However, targeted translocation of apoptin into the nuclei of normal cells is not sufficient for apoptin’s toxicity. Thus, additional interaction partners or specific activation of other signalling pathways in cancer cells, preceding nuclear accumulation, might be necessary for apoptin’s tumour specific toxicity. Nuclear accumulation is linked to phosphorylation of apoptin at the threonine-108 (Thr-108) residue by an as yet unknown kinase (Rohn et al. 2002; Poon et al. 2005a) specifically in cancer cells but not in normal cells. Thus, identifying the pathways responsible for phosphorylation of apoptin might provide clues about its tumour-specific toxicity. These pathways may also represent attractive targets for the development of new highly selective anticancer drugs.
In this study, we have investigated the role of the PI3-K/Akt pathway in apoptin-induced cell death. Our results intriguingly demonstrate that apoptin interacts with the p85 regulatory subunit, which leads to the constitutive activation of PI3-K; inhibition of this activation process severely impairs apoptin-induced cell death. We also demonstrate nuclear translocation of Akt during apoptin-induced cell death. Nuclear translocation of Akt alone does not have a toxic effect, but it strongly potentiates the toxicity of apoptin.
PC-3 and MCF-7 cells were grown in RPMI-1640 medium supplemented with 10% foetal bovine serum (Hyclone, Logan, UT, USA), 100 μg/mL penicillin and 0.1 μg/mL streptomycin (Gibco BRL, Grand Island, NY, USA). The cells were grown at 37 °C with 5% CO2 in a humidified incubator. Peripheral blood lymphocytes were isolated from chronic lymphocytic leukaemia (CLL) patients or from normal healthy individuals, by Ficoll gradient fractionation, as described previously (Los et al. 1995a), and were maintained in RPMI medium. The following antibodies were used: murine anti-PI3-K (p85), antimouse immunoglobulin G (IgG) horseradish peroxidase (HRP), antirabbit IgG HRP (all from Upstate Cell Signalling, Beverly, MA, USA), goat anti-Akt, murine antitubulin, rabbit anti-GFP, antigoat IgG HRP (all from Santa Cruz Biotechnologies, Santa Cruz, CA, USA), murine antiphospho-Akt Ser-473 (Cell Signalling), antihuman CD5-FITC, CD19-PerCP (BD Biosciences, San Jose, CA, USA), antigoat Cy3, antirabbit Cy3 and antimurine Cy3 (all from Sigma, St. Louis, MO, USA). The following inhibitors were used: wortmannin (IC50 5 nM), LY294002 (IC50 1.4 μM) (all from Calbiochem, San Diego, CA, USA).
Peripheral blood lymphocytes from normal individuals and CLL patients either left untreated or TAT-apoptin treated for the indicated times were washed twice with ice-cold phosphate-buffered saline (PBS), and then were incubated (individually) with both CD5-FITC and CD19-PerCP antibodies (each 0.5 μg per 106 cells) for 30 min at 4 °C in dark. The cells were then washed twice with ice-cold PBS and were re-suspended in 300 μL of PBS. Samples were analysed by flow cytometry by using both FL1 (FITC) and FL2 (PerCP) channels, and percentage of B cells obtained by gating the double positive cells compared to unstained controls.
The TAT-GFP and TAT-apoptin proteins were purified as previously described (Maddika et al. 2005). Glutathione S-transferase (GST) and GST-apoptin were purified by using glutathione-Sepharose high-performance beads (Amersham Biosciences, Piscataway, NJ, USA) according to the manufacturer’s protocol. The GST pull-down assay was performed to detect apoptin’s interacting partners. Briefly, either purified GST or GST-apoptin along with total PC-3 cell lysate was immobilized on glutathione-Sepharose beads overnight at 4 °C. The beads were washed three times with ice-cold lysis buffer and bound proteins were isolated on sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Proteins specific for apoptin were subjected to in-gel digestion and were further identified by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry, at the proteomics centre of the University of Manitoba, Manitoba, Canada. Finally, proteins from the GST pull-down assay were identified by immunoblotting.
The following plasmid were used: GFP-apoptin (apoptin cloned into pEGFP-C1 vector, clonetech), GST-apoptin (apoptin cloned into PGEX-2T vector, Amersham Biosciences), PI3-K dominant-negative vector, Akt wild-type vector [J. Downward, UK (Kauffmann-Zeh et al. 1997)], PDK1 dominant-negative and constitutively active vectors (A. Halayko, Winnipeg, , Manitoba, Canada), phosphatase and Tensin homolog (PTEN) wild-type and PTEN C124S phosphatase-dead mutant (D. H. Anderson, Saskatchewan Cancer Agency, Saskatchewan, Canada), Ad-NLS-Akt [M. A. Sussman, San Diego, CA, USA (Shiraishi et al. 2004), and Ad-Akt dominant-negative vector K. Walsh (Luo et al. 2000)]. Transfection was performed using Lipofectamine (Invitrogen, Burlington, ON, Canada) according to the manufacturer’s recommendations. Adenoviral transfections with Akt dominant-negative mutant vector and nuclear-targeted Akt were performed as previously described (Fujio et al. 1999).
A non-radioactive competitive ELISA-based assay was used to assess the PI3-K activity, according to the manufacturer’s protocol (Echleon Biosciences, Salt Lake City, UT, USA). Briefly, equal amounts of PI3-K from the PC-3 cell lysates were immuno-precipitated with anti-p85 antibodies overnight at 4 °C and then were incubated with protein A-Sepharose beads for 1 h at 4 °C. Bead-bound enzymes were incubated with 100 pM of PI(4,5)P2 substrate in kinase reaction buffer (4 mM MgCl2, 20 mM Tris, pH 7.4, 10 mM NaCl and 25 μM ATP) for 2 h at room temperature. The mixtures were then incubated with PI(3,4,5)P3 detector for 1 h at room temperature in the dark, subsequently were added to PI(3,4,5)P3-coated microplate wells and were incubated for 30 min at room temperature in the dark. After thorough washing, peroxidase-linked secondary detection reagent was added, and PI(3,4,5)P3 detector protein binding to the plate was assessed by measuring absorbance at 450 nm. Data for the kinase activity are expressed as fold induction in transfected cells compared to the activity in untreated cells.
Cells were washed twice with ice-cold PBS, lysed with ice-cold lysis buffer (50 mM Tris, pH 7.5, 1% Nonidet P-40, 150 mM NaCl, 1 mM Na3VO4, 2 mM EGTA, Protease inhibitor cocktail), were incubated for 10 min on ice, and then were centrifuged for 10 min at 4 °C. Immuno-precipitations were performed with the indicated antibodies (Fig. 1c) and the immuno-complexes were captured with protein A-agarose beads (Amersham Pharmacia Biotech). After three washing steps with cell lysis buffer, bead-bound proteins were subjected to Western blot analysis as described previously (Maddika et al. 2005). When appropriate, quantification of Western blot data was performed using the ImageQuant 5.2 software (Molecular Dynamics, Sunnyvale, CA, USA).
PC-3 cells were transfected to express apoptin and were fixed 24 h later in 4% paraformaldehyde in PBS, permeabilized in 0.2% Triton X-100 and were stained with either anti-p85 or anti-Akt antibodies followed by their respective appropriate secondary antibodies, conjugated to Cy3. The fluorescent images were analysed by confocal microscopy.
Statistical significance of the data was confirmed using Student’s t-test.
The measurement of apoptosis was performed using the Nicoletti method followed by flow cytometry, as previously described (Maddika et al. 2005). Cell death was quantitatively confirmed by the microtiter tetrazolium (MTT) assay as previously described (Ghavami et al. 2005).
First, we re-assessed the cancer cell-selective toxicity of apoptin, a controversial question in apoptin biology, using peripheral blood lymphocytes (PBLs) from CLL patients compared to peripheral blood lymphocytes from a normal healthy individual. B cells in CLL and normal PBLs were detected by double staining for the specific surface markers CD5 and CD19. In normal PBLs, there was no significant difference in the B-cell population before (6.1%) or after (5.2%) 48 h of treatment with TAT-apoptin. In contrast, there was a significant decrease in the CLL PBLs upon TAT-apoptin treatment (11.2%) compared to the control (40.1%) (Fig. 1a). Thus, the data indicate that apoptin effectively kills malignant B cells from CLL patients, but not the normal counterparts, therefore confirming apoptin’s selective toxicity (Fig. 1a). TAT-GFP, used as a negative control for this toxicity, had no significant effect on the survival of either normal or malignant cells.
As a first step to identify cellular targets of apoptin action in the cell, we generated an expression construct encoding GST fused to the apoptin N-terminus. GST-apoptin and control GST were used in a pull-down assay with cell extracts derived from the PC-3 prostate cancer cells (Fig. 1b). Mass spectrometric analysis of proteins specifically bound to apoptin identified the two major components of the PI3-K/Akt pathway the p85 regulatory subunit of PI3-K, and the serine/threonine kinase Akt, a kinase downstream of PI3-K. We also found that apoptin interacts with proteins such as chaperones, actin and tubulin family members (data not shown), as observed previously (Teodoro et al. 2004). Interaction of apoptin with PI3-K and Akt was also examined in vivo, whereby PC-3 cells were transfected with GFP-apoptin, and total cell extracts immuno-precipitated with anti-GFP antibody at different time points, post-transfection, and composition of the immune complexes was analysed for PI3-K and Akt by immunoblotting. In a separate series of experiments, PI3-K (p85) and Akt were immuno-precipitated, and apoptin detected by Western blot analysis. Figure 1c shows that PI3-K and Akt both interact with apoptin confirming the data from Fig. 1b. Interaction of apoptin with PI3-K could be detected 6 h after apoptin transfection, thus preceding apoptin-induced cell death, which was initiated at least 18–24 h later. The above experiment indicates that PI3-K interaction with apoptin is a very early event in apoptin-induced cell death. As shown in Fig. 1c, apoptin strongly interacts with PI3-K within 6–24 h post-transfection. In some further experiments, the interaction could be observed much longer – for over 48 h. On the contrary, the interaction of apoptin with Akt seems weaker and is seen only at certain time points after apoptin treatment.
To determine the functional significance of apoptin’s interaction with the p85 regulatory subunit, we measured PI3-K activity using a non-radioactive ELISA-based method. Surprisingly, MCF-7 and PC-3 cells transfected to express GFP-apoptin revealed constitutive activation of PI3-K in apoptin-transfected cells (Fig. 2a). PI3-K activity was increased nearly 4-fold in apoptin-treated MCF-7 cells and up to 6-fold in PC-3 cells, compared to the GFP-transfected control. PI3-K activation was seen around 6 h after transfection, consistent with the interaction data (Fig. 1b,c), with activation retained at a similar level for up to a further 40 h. To support the positive role of PI3-K/Akt pathway and to investigate the direct effect of PI3-K inhibition on apoptin-induced cell death, we pretreated the cells with the PI3-K inhibitors, wortmannin and LY294002, 30 min before transfection to express apoptin, and cell death was assayed 24 and 48 h later. To our surprise, both inhibitors afforded significant protection against apoptin-induced cell death, despite generally being known to enhance the cell death process through inhibition of PI3-K activity (Fig. 2b). This effect was further confirmed by co-transfection of cells to express apoptin together with a dominant-negative derivative of PI3-K, which significantly protected cells from apoptin-induced cell death (Fig. 2b).
We next investigated the effect of apoptin on downstream targets of the PI3-K pathway by transfecting the PC-3 cells to express apoptin, and assessing Akt activation by Western blotting at different time points post-transfection. In agreement with the results for PI3-K activation, increased levels of activated Akt were seen around 6–12 h post-transfection, with pronounced levels of phosphorylated Akt detectable even up to 24 h post-transfection (Fig. 3a). Activation of Akt is downstream of, and is dependent on PI3-K activation; pre-treatment of cells with the inhibitors wortmannin or LY294002 (Fig. 3b), or cotransfection of apoptin-expressing cells with constructs encoding PI3-K or PDK1 dominant-negative mutants, impaired Akt activation. Furthermore, overexpression of wild-type PTEN, a negative regulator of Akt activation, severely reduced apoptin-induced Akt activation, while overexpression of a phosphatase-deficient PTEN mutant had no effect (Fig. 3c). No effect was seen on total Akt levels by any of these treatments (data not shown). To further confirm the role of PI3-K in apoptin-induced cell death, and to validate the role of Akt in this process, we transiently overexpressed a dominant-negative mutant of PDK1 (PDK1-DN), and/or the dominant-negative Akt (Akt-DN) using adenoviral vectors. Both dominant-negative kinase mutants as well as overexpression of wild-type PTEN significantly protected against apoptin-induced cell death, thus confirming the key role of the PI3-K/Akt pathway in this process (Fig. 3d).
Akt is generally regarded as a survival or proliferation-promoting kinase and not as a pro-apoptotic factor. However, in the presence of apoptin, Akt is clearly acting as a pro-cell death molecule as its inhibition severely inhibits cell death pathways triggered by apoptin. Next, we asked if apoptin redirects Akt to different cellular targets by modulating its subcellular localization. As shown in Fig. 4a, Akt is mainly localized in the cytoplasm of untransfected PC-3 cells in the absence of apoptin expression, but is clearly nuclear in its presence. Analysis of the kinetics of Akt nuclear translocation revealed that most of the cytoplasmic Akt translocates to the nucleus within 12 h of transfection to express apoptin (data not shown). To determine whether nuclear Akt alone is sufficient to induce cell death, even in the absence of an apoptotic stimulus, we infected PC-3 cells with a nuclear localization signal (NLS)-Akt adenoviral vector in the absence and in the presence of apoptin expression. Quantitative analysis using flow cytometry, to determine the extent of cell death (Fig. 4b), indicated that NLS-Akt after 48 h of transduction, exhibits slightly increased toxicity (13%) compared to the negative (wild-type) control (7%). Importantly, co-expression of NLS-Akt together with apoptin sensitized the cells towards apoptin-induced cell death at 24 and 48 h after apoptin transfection. Together, the data indicate that nuclear Akt alone may not be sufficient for cell death induction, but can clearly facilitate cell death in the presence of certain apoptotic stimuli.
In this study, we have investigated for the first time the role of the PI3-K/Akt pathway in cancer cell-selective death, triggered by apoptin. Interestingly, our results show that the PI3-K/Akt signalling pathway, universally accepted as promoti ng cell survival and proliferation, can also have a critical role in promoting cell death in response to certain stimuli. Moreover, we demonstrate that the inhibition of PI3-K activation by either pharmacological inhibitors or by genetic methods can abrogate cell death induced by the viral protein apoptin. The apoptin-inhibitory effect of PI3-K inhibitors could be observed, despite their inherent death promoting activity, if used at higher concentrations. Several molecules in the literature have been assigned a dual role in both cell survival and cell death mechanisms. For instance, the oncogene c-myc (Henriksson & Luscher 1996; Hueber el al. 1997), nuclear factor kappa B (Barkett & Gilmore 1999), molecules of Ras/Map kinase pathway (Downward 2003; Brown & Benchimol 2006), Bcl-2 (Cory & Adams 2002; Subramanian & Chinnadurai 2003), caspases (Los el al. 2001), cyclins A, B, D, E (Maddika et al. 2007) and even an orphan nuclear receptor Nur77 (Lin et al. 2004) are all involved in promoting either cell proliferation or cell death, depending on the context and the stimulus. So far, the PI3-K/Akt pathway has been implicated only in the preferment of cell survival, proliferation, growth, transcription and translation (Cantley 2002). The specific role of the PI3-K/Akt pathway in a pro-cell death pathway has thus far not been clarified.
Although PI3-K is known to be involved in cell survival, several publications have hinted at the fact that active PI3-K may contribute to apoptosis under certain conditions (Aki et al. 2003; Nimbalkar et al. 2003; Maddika et al. 2007). Intriguingly in this context, Akt inhibitors have proved to be only moderately successful in experimental cancer therapy (Stein 2001; Workman 2004). We document here for the first time that activated Akt, if translocated to the nucleus, can stimulate rather than protect against apoptosis induced by cytotoxic stimuli such as administration of apoptin. We hypothesize that in the presence of apoptin, Akt is aberrantly activated and targets cellular substrates and/or pathways in the nucleus that are different from its targets during cell survival mechanisms. Thus, the net outcome of Akt activation could vary according to signalling context, type of stimuli and temporal characteristics of signals that they trigger (e.g. transient versus constitutive signalling). There are well-established examples of such context-dependent, dramatic changes in the final outcome of activation of certain signalling pathways. For example, the proto-oncogene c-myc stimulates cell proliferation in the presence of appropriate survival stimuli (including activated PI3-K/Akt pathway) and triggers apoptosis in their absence (Pelengaris et al. 2002a). This dual capacity ensures that cell growth is restricted to the correct paracrine environment, co-activation of a pro-survival signalling pathway, and/or co-expression of anti-apoptotic molecules, and is thereby strictly controlled by multiple mechanisms (Kauffmann-Zeh et al. 1997; Pelengaris et al. 2002b; Baudino et al. 2003).
We have demonstrated that interaction of apoptin with the p85 regulatory subunit constitutively activates PI3-K. In addition to apoptin’s interaction with PI3-K, we observed transient interaction with Akt. Furthermore, interaction of Akt with apoptin appears to facilitate Akt nuclear localization. Apoptin thus effects Akt’s nuclear translocation, presumably enabling access to pro-apoptotic phosphorylation targets that do not normally come into contact with cytoplasmic Akt. Nuclear access of Akt alone is not sufficient to induce apoptosis, because NLS-Akt alone does not induce cell death, but can potentiate apoptin-induced cell death. Recently, Trotman et al. (2006) reported that the promyelocytic leukaemia tumour suppressor prevents cancer by dephosphorylating and inactivating Akt inside the nucleus. Other workers have reported that phosphorylated nuclear, but not cytoplasmic Akt, interacts with Ebp1 (an inhibitor of caspase-activated DNase-dependent apoptotic DNA fragmentation), and enhances its anti-apoptotic action independently of Akt kinase activity, in a cell-free experimental system (Ahn et al. 2006). Thus, nuclear Akt contributes to cell death pathways only in the presence of certain apoptotic stimuli, implying that nuclear Akt may have pro-survival and proliferation-promoting function, depending on the experimental conditions.
There are several reported nuclear targets of Akt, including FOXO3a (Brunet et al. 1999), Nur77 (Pekarsky et al. 2001) and p21cip/waf (Li et al. 2002). We have previously shown that Nur77 translocates from the nucleus to mitochondria during apoptin-induced cell death (Maddika et al. 2005); we still have to define the connection between nuclear Akt and this translocation. We have also observed down-regulation of p21cip/waf (unpublished data) 24 h after cell death induction by apoptin, although this occurs independently of Akt activation, because Akt inhibition by wortmannin did not affect p21cip/waf levels. Down-regulation of p21cip/waf may be the result of cleavage by caspases, as a secondary event, downstream of mitochondrial death pathway activation. Further studies in our laboratory are focused on identifying different nuclear targets of Akt during apoptin-induced cell death.
Different components of the PI3-K/Akt pathway are involved in tumourigenesis and are highly active in various types of cancer cell compared to normal cells (reviewed in Vivanco & Sawyers 2002; Luo et al. 2003). Furthermore, PTEN, a phosphatase that counteracts PI3-K’s action, is the second most commonly mutated tumour suppressor gene after p53 (Vivanco & Sawyers 2002). Hyper-activation of the PI3-K/Akt pathway indicates poor clinical prognosis and also contributes to drug resistance during cancer treatment. Thus, apoptin’s targeting of these very pathways that are hyperactive in cancer, may explain its unique properties of tumour-specific toxicity. Our data strongly indicate that apoptin hijacks these survival pathways and redirects them from their normal survival/proliferatory action towards the activation of the cell death.
T. J. Kroczak has been supported by the Manitoba Institute of Child Health. M. Los thankfully acknowledges the support by the Canada Foundation for Innovation-Canada Research Chairs program, the Prostate Cancer Research Foundation of Canada-, the Canadian Institutes of Health Research (CIHR)-, CIHR Intellectual Property Mobilization-, CCMF- and MHRC-financed programs. We acknowledge the support of CancerCare Manitoba Translational Research program and Manitoba CLL tissue bank.