当前位置: 首页 > 医学版 > 期刊论文 > 内科学 > 内分泌学杂志 > 2005年 > 第7期 > 正文
编号:11168484
Enhancement of Tumor Necrosis Factor--Induced Growth Inhibition by Insulin-Like Growth Factor-Binding Protein-5 (IGFBP-5), But Not IGFBP-3 i
     Kolling Institute of Medical Research, University of Sydney, Royal North Shore Hospital, St. Leonards, New South Wales 2065, Australia

    Address all correspondence and requests for reprints to: Dr. Alison J. Butt, Cancer Research Program, Garvan Institute of Medical Research, 384 Victoria Street, Darlinghurst, New South Wales 2010, Australia. E-mail: abutt@garvan.org.au.

    Abstract

    Expression of IGF-binding protein-3 (IGFBP-3) and IGFBP-5 in human breast cancer cells induces apoptosis and is associated with modulations in Bcl-2 proteins, suggesting that these IGFBPs induce an intrinsic apoptotic pathway. In this study we demonstrate that although both IGFBPs induced the activation of caspase-8 and caspase-9, the expression of IGFBP-5, but not IGFBP-3, sensitized MDA-MB-231 breast cancer cells to the inhibitory effects of TNF. This sensitivity to TNF was associated with a block in nuclear factor-B-mediated cell survival signals. IGFBP-5 expression was also associated with a caspase-8-independent activation of Bid, increased levels of cytosolic second mitochondria-derived activator of caspase (Smac)/direct inhibitor of apoptosis proteins (IAP) binding protein with low pI (DIABLO), and an enhanced phosphorylation of c-Jun N-terminal kinase, both basally and in response to TNF. These results suggest that IGFBP-5 expression may influence extrinsic apoptotic pathways via a differential modulation of downstream cell survival and cell death pathways. Furthermore, although IGFBP-3 and IGFBP-5 share much structural and functional homology, they can modulate distinct apoptotic pathways in human breast cancer cells.

    Introduction

    IGF-BINDING PROTEIN-3 (IGFBP-3) and IGFBP-5 are members of a family of high-affinity binding proteins that modulate the mitogenic and antiapoptotic effects of IGFs. However, it is now becoming clear that both have direct effects on the growth of cancer cells that are independent of their ability to bind IGFs (1). For example, we and others have demonstrated that IGFBP-5 and IGFBP-3 have IGF-independent, cytostatic and cytotoxic effects on the growth of human breast (2, 3, 4, 5) and prostate (6) cancer cells.

    How these intrinsic growth effects of IGFBPs are mediated is largely unclear. Receptors for both IGFBP-3 and IGFBP-5 have been postulated (7, 8, 9); however, these remain to be fully characterized. IGFBP-3 and -5 translocate to the nucleus in breast cancer cells (10, 11), and we have also shown that both can transcriptionally modulate the expression of apoptotic genes such as bax and bcl-2 (2, 3, 4). However, neither interaction with the cell surface nor nuclear translocation of IGFBP-3 is required for its growth inhibitory and proapoptotic effects in breast cancer cells, suggesting that these effects may be initiated in the cytoplasm (4). Exogenous IGFBP-5 has variable effects on the growth and survival of human breast cancer cells (2, 12), but we have shown that it is not internalized by MDA-MB-231 cells, suggesting that it also does not elicit its growth effects through a cell surface receptor and may act via an intracrine mechanism in these cells (2).

    Recent studies have emphasized the important role of the mitochondria in the regulation of apoptosis (13). Apoptotic stimuli, such as growth factor deprivation, glucocorticoids, and cytotoxic agents, lead to an increase in the permeability of the outer mitochondrial membrane and promote the release of cytochrome c into the cytosol. Cytochrome c binds to apoptotic protease-activating factor-1 and caspase-9, promoting activation of the latter that subsequently activates the effector caspase-3 (14). Although the exact mechanism of mitochondrial release of cytochrome c is unclear, members of the Bcl-2 family have been shown to influence this process (15). Generally, antiapoptotic members of this family (e.g. Bcl-2 and Bcl-xL) inhibit the release of mitochondrial proteins, whereas proapoptotic members (e.g. Bax, Bid, and Bad) are translocated from the cytoplasm to the mitochondria during the induction of apoptosis and induce the release of cytochrome c.

    Apoptosis can also be induced via an extrinsic pathway involving ligand-mediated activation of death receptors, such as TNF receptor 1 (TNFR1). Upon binding of TNF to the TNFR1, an intracellular death effector complex is formed, consisting of adaptor molecules such as Fas-associated death domain protein, and an inactive precursor form of caspase-8 (16). Formation of this complex leads to cleavage of caspase-8 into active subunits and the subsequent proteolysis of downstream substrates. Activation of the transcription factor nuclear factor-B (NFB) by nuclear translocation elicits a potent survival signal and blocks this death receptor-mediated apoptotic pathway (17).

    Although distinct intrinsic and extrinsic pathways to apoptosis have been described, recent studies have suggested that there may be much cross-talk between the two. Apoptotic signals from death receptors can be amplified at the level of the mitochondria through caspase-8-mediated cleavage of the proapoptotic protein Bid, producing a truncated fragment that translocates to the mitochondria and induces cytochrome c release (18). There is evidence that Bid-mediated release of cytochrome c is critical for execution of an extrinsic apoptotic pathway in some cells types (19). Recently, Deng et al. (20) reported that TNF-mediated activation of c-Jun N-terminal kinase (JNK) leads to a caspase-8-independent cleavage of Bid and the release of second mitochondria-derived activator of caspase [second mitochondria-derived activator of caspase (Smac) or direct inhibitor of apoptosis proteins (IAP) binding protein with low pI (DIABLO)] (21, 22) from mitochondria. This provides additional evidence of an important functional link between extrinsic and intrinsic apoptotic pathways.

    We have previously demonstrated that IGFBP-3- and IGFBP-5-induced apoptosis in breast carcinoma cells is associated with modulation of Bcl-2 family members (2, 3, 4). Furthermore, both IGFBPs can enhance the apoptotic effects of ionizing radiation in these cells, suggesting that they may influence apoptotic pathways initiated in the mitochondria. However, despite these functional similarities, recent studies have suggested some divergence in their intrinsic, IGF-independent roles, particularly in relation to breast cancer cell growth. Both IGFBP-5 and IGFBP-3 are expressed in breast cancer tissue (23) and cell lines (24), but their expression differs in relation to estrogen receptor (ER) status, with IGFBP-5 expression higher in ER-positive tumors and IGFBP-3 expression higher in ER-negative specimens (24). Furthermore, high breast tumor expression of IGFBP-3 is associated with poor prognosis (25), with some evidence that this may be due to altered epidermal growth factor signaling during tumorigenic progression (26, 27). These differences emphasize the clinical relevance of further elucidating the mechanisms governing the growth effects of IGFBP-3 and -5 in human breast cancer. In this present study we have delineated at which point IGFBPs engage the apoptotic machinery and have demonstrated that IGFBP-3 and IGFBP-5 initiate distinct apoptotic pathways in human breast cancer cells.

    Materials and Methods

    Cell culture of breast cancer cells

    The human breast cancer cell line MDA-MB-231 was routinely maintained in RPMI 1640 supplemented with 10% fetal calf serum (FCS), 10 μg/ml insulin, and 2.92 mg/ml glutamine under standard conditions.

    Stable transfection

    The stable transfection of MDA-MB-231 cells with human IGFBP-5 cDNA is previously described (2, 4). To allow for a direct comparison between IGFBPs, MDA-MB-231 cells were stably transfected with human IGFBP-3 cDNA, essentially as previously described (2, 4). The corresponding vector controls for each transfection were used in subsequent experiments. To overcome the potential variation in individual clones, mixed populations of transfectants were grown up for subsequent experiments.

    Adenoviral-mediated expression of IGFBPs

    IGFBP-3 or IGFBP-5 cDNAs were subcloned into the adenoviral vector, pAd-Track-cytomegalovirus, and replication-deficient adenoviruses were produced as previously described (28). Proliferating cultures of MDA-MB-231 cells were incubated with adenovirus stock (4.5 plaque-forming units/μl) for 6 h in serum-free/insulin-free (SF) medium essentially as previously described (2, 4). The adenoviral constructs also code for a green fluorescent protein marker that was used to determine equal infection efficiency.

    Analysis of conditioned medium

    Concentrations of IGFBPs in 24-h conditioned medium (CM) from stable transfectants were assayed using specific RIA for IGFBP-5 (29) or IGFBP-3 (30) as previously described.

    Cell proliferation assays

    Cells were plated in 12-well plates at 5 x 104/well for 24 h, then treated with 10 ng/ml recombinant human TNF (Sigma-Aldrich Corp., St. Louis, MO) or vehicle in 10% FCS medium. On d 3, 5, and 7 of treatment, cells were trypsinized, and the viable cell number was determined by trypan blue exclusion and cell counting. At each time point, cells were treated with fresh 10% FCS with or without TNF.

    [3H]Thymidine incorporation

    Stable transfectants were plated in 48-well plates at 5 x 104/well for 24 h before being treated with or without 10 ng/ml TNF in SF medium for an additional 24 h. DNA synthesis was determined using incorporation of [3H]thymidine essentially as previously described (31).

    Survival assays

    The long-term survival of MDA-MB-231 transfectants after treatment with TNF was assessed by survival assay. Cells (2 x 103) were seeded into a six-well plate in triplicate for 24 h, then exposed to various doses of TNF (1, 5, or 10 ng/ml) with or without the proteasome inhibitor N-acetyl-L-leucinyl-L-norleucinal (LLnL; 5 μM) or the JNK-specific inhibitor SP600125 (20 μM; after a 30-min pretreatment) in SF medium for 2 h. Medium was replaced with 10% FCS medium, and cells were incubated for 14 d. At this time, individual cells were counted, and the percent survival was determined by the proportion of attached cells surviving (assessed by trypan blue exclusion) relative to untreated controls. Because MDA-MB-231 cells tend to grow quite diffusely over the culture dish and do not form distinct colonies, we evaluated individual cells rather than counted colonies. Typically, approximately 6–7 x 105 cells were counted in the vector controls. We have previously used this method to demonstrate that the expression of either IGFBP-3 (3, 4) or IGFBP-5 (2) alone significantly reduces the survival of breast cancer cells. However, in this study, which measured the additional effects of TNF, the survival of untreated IGFBP transfectants was expressed as 100%.

    Measurement of DNA fragmentation by flow cytometry

    Adenoviral-infected cells were incubated for 48 h in SF medium with or without the caspase-8-specific inhibitor z-Ile-Glu-Thr-Asp-fluoromethyl ketone (z-IETD-fmk) or the caspase-9-specific inhibitor, z-Leu-Glu-His-Asp-fluoromethyl ketone (z-LEHD-fmk) (both 50 μM; BD Pharmingen, San Diego, CA). Floating and attached cell populations were then combined, and aliquots of 1 x 106 cells were fixed in 70% ethanol at –20 C for at least 24 h. Fixed cells were washed and suspended in 1 ml fluorochrome solution (50 μg/ml propidium iodide and 1 mg/ml ribonuclease A) for at least 1 h in the dark at 4 C. Flow cytometric analysis was performed using an ELITE flow cytometer (Coulter, Hialeah, FL). Twenty thousand cells were analyzed for each sample, and the percent apoptosis was defined as the percentage of the population in the pre-G1 fraction.

    Immunoblot analysis

    Proteins from whole cell lysates were resolved under reducing conditions on 12% sodium dodecyl sulfate-polyacrylamide gels using standard methods. For analysis of RelA levels, 20 μg nuclear extracts (prepared using NE-PER nuclear and cytoplasmic extraction reagents, Pierce Chemical Co., Rockford, IL) were electrophoresed under reducing conditions. Resolved proteins were transferred to nitrocellulose membranes and probed with anti-Bid polyclonal (Santa Cruz Biotechnology, Inc., Santa Cruz, CA), anticaspase-8 monoclonal (551234; BD Pharmingen), anticaspase-9 precursor polyclonal (550437; BD Pharmingen), anti-RelA polyclonal (Santa Cruz Biotechnology, Inc.), anticytochrome c monoclonal (7H8.2C12, BD Pharmingen), anti-TNFR1 monoclonal (Santa Cruz Biotechnology, Inc.), anti-Smac polyclonal (Santa Cruz Biotechnology, Inc.), anticytochrome c oxidase (subunit IV, Molecular Probes, Eugene, OR), and anti-JNK, phospho-specific (sc-6254; Santa Cruz Biotechnology, Inc.) and total (sc-571; Santa Cruz Biotechnology, Inc.), antibodies overnight at 4 C or, for RelA and TNFR1 immunoblots, for 2 h at room temperature. Immunoreactive protein bands were detected by the relevant anti-IgG antibodies conjugated with horseradish peroxidase, followed by enhanced chemiluminescence (Pierce Chemical Co.). Blots were checked for equal loading by reprobing with anti--tubulin (Sigma-Aldrich Corp.) or antiproliferating cell nuclear antigen (Santa Cruz Biotechnology, Inc.) antibodies.

    Analysis of cytoplasmic cytochrome c and Smac/DIABLO levels

    Levels of cytochrome c and Smac/DIABLO were determined by Western blotting of cytoplasmic extracts prepared essentially as previously described (32). Briefly, cells were trypsinized, washed with PBS, and resuspended in digitonin lysis buffer (75 mM NaCl, 1 mM NaH2PO4, 8 mM Na2HPO4, 250 mM sucrose, and 190 μg/ml digitonin) at 4 C for 10 min. Cells were microfuged for 10 min at 14,000 rpm at 4 C, then supernatants were removed and stored at –80 C until use.

    Immunofluorescence

    Cells were seeded on sterile coverslips, then allowed to attach and grow for 24 h. Cell monolayers were washed with SF medium and incubated for an additional 24 h. After this time, cells were gently washed in PBS, then fixed for 10 min in ice-cold 100% methanol. Fixed cells were blocked with 3% BSA in PBS for 1 h, then exposed to anticytochrome c monoclonal (6H2.B4; BD Pharmingen) antibody for 1 h at room temperature. After washing, cells were exposed to Alexa Fluor 594 (1 μg/ml; Molecular Probes) antirabbit IgG for 1 h at room temperature, then washed and mounted in Histochoice (Amresco Inc., Solon, OH). Samples were observed using a fluorescence microscope.

    Isolation of IGFBPs from CM

    IGFBP-3 and IGFBP-5 fractions were isolated from CM of adenoviral-infected MDA-MB-231 cells by affinity chromatography and reverse phase HPLC essentially as previously described (2).

    Isolation and analysis of mitochondria

    The isolation procedure was carried out at 4 C. Approximately 4 x 107 MDA-MB-231 cells were washed in PBS, then resuspended in 5 ml mitochondrial isolation buffer [250 mM sucrose, 10 mM HEPES (pH 7.9), 10 mM KCl, 1.5 mM MgCl2, 0.5 mM dithiothreitol, and 50 μl/ml protease inhibitor cocktail; Sigma-Aldrich Corp.]. Cell suspensions were then homogenized in a 15-ml glass homogenizer with a Teflon pestle by 20 strokes, a 1-min incubation on ice, and then another 20 strokes. Cell lysates were centrifuged at 600 x g for 10 min, then the supernatants were centrifuged at 10,000 x g for 15 min. Mitochondrial pellets were enriched by repeating the centrifugation steps. Mitochondria were resuspended in 500 μl PT buffer [50 mM sucrose, 10 mM succinate, 10 mM HEPES-KOH (pH 7.4), 0.1% BSA, 5 mM potassium phosphate, pH 7.4]. To examine cytochrome c release, freshly isolated mitochondria (2 μg protein) were incubated at 37 C for 30 min with 1 μg IGFBP-3 or -5 fractions (described above). Mitochondria were then pelleted by centrifugation for 10 min at 13,000 x g, supernatants were removed, and the mitochondrial pellet was analyzed for cytochrome c content by immunoblotting with an anticytochrome c monoclonal antibody (BD Pharmingen). Expression of cytochrome c oxidase (subunit IV; Molecular Probes), determined by immunoblot, was used as a loading control.

    Statistical analysis

    Statistical analysis was carried out using StatView 4.02 (Abacus Concepts, Inc., Berkeley, CA). Differences between groups were evaluated by Fisher’s protected least significant difference test after ANOVA, using repeated measures or factorial analysis where appropriate.

    Results

    Activation of caspase-8 and caspase-9 in IGFBP transfectants

    Both IGFBP-5 and -3 induce a caspase-dependent apoptotic pathway in human breast cancer cells (2, 4). To further delineate the specific apoptotic pathway initiated by these proteins, activation of both caspase-8 and caspase-9 was examined by Western immunoblot in cells infected with IGFBP-3, IGFBP-5, or control-adenovirus after 24 h in SF medium. Figure 1A demonstrates that adenovirus-mediated expression of either IGFBP-5 or -3 was associated with activation of caspase-8, visualized by the appearance of the active subunits of 40 and 23 kDa. The inactive, precursor form predominated in control lysates, although some basal activation was seen. Similarly, activation of caspase-9 was detected in lysates from IGFBP-5- and IGFBP-3-expressing cells by a decrease in the precursor form (46 kDa). The antibody used only detects the inactive form of caspase-9.

    FIG. 1. IGFBP-mediated apoptosis is associated with activation of caspase-8 and caspase-9. A, MDA-MB-231 cells were infected with IGFBP-3-, IGFBP-5-, or vector-adenovirus; cell lysates were prepared 24 h after infection and immunoblotted for expression of caspase-8-inactive precursor and active subunits and caspase-9-inactive precursor, as indicated. Expression of -tubulin was used as a loading control. Representative blots from at least three independent experiments are shown. B, MDA-MB-231 cells were infected with IGFBP-3-, IGFBP-5-, or vector-adenovirus and incubated in the presence () or absence () of the caspase-8-specific inhibitor, z-IETD-fmk, or the caspase-9-specific inhibitor, z-LEHD-fmk, for 48 h before analysis of apoptosis by flow cytometry. Values shown are the mean of triplicate wells ± SE from three independent experiments. *, P < 0.05; **, P < 0.02 (for inhibitor-treated cells vs. untreated controls).

    We also examined the effects of the caspase-8 and caspase-9-specific inhibitors, z-IETD-fmk and z-LEHD-fmk, on IGFBP-induced apoptosis by flow cytometric analysis of the hypodiploid (pre-G1) population. MDA-MB-231 cells were infected with IGFBP-3, IGFBP-5, or control adenovirus, then incubated with or without z-IETD-fmk or z-LEHD-fmk for 48 h in SF medium (Fig. 1B). Incubation with either z-IETD-fmk or z-LEHD-fmk resulted in a significant decrease in IGFBP-mediated apoptosis (P < 0.01, respectively) to levels that were not significantly different from vector control values. Levels of apoptosis in vector-infected cells were not significantly affected by the inhibitor (data not shown).

    Activation of an intrinsic apoptotic pathway by IGFBPs

    Activation of caspase-9 during IGFBP-mediated apoptosis suggests the induction of an intrinsic apoptotic pathway. To investigate this, we examined the release of mitochondrial cytochrome c in MDA-MB-231 cells infected with IGFBP-5, IGFBP-3, and control-adenovirus. Cytochrome c localization was visualized by immunocytochemical staining and fluorescence microscopy in cells 24 h after adenoviral infection. Figure 2A demonstrates the classical, punctate staining of mitochondrial cytochrome c observed in vector-infected cells. However, in cells expressing either IGFBP-5 or IGFBP-3, a diffuse, cytoplasmic staining pattern was seen, consistent with a release of cytochrome c from mitochondria. This was confirmed by immunoblotting of cytoplasmic extracts of adenovirus-infected cells, where cytochrome c was present in the cytosol of IGFBP-expressing cells, but not in vector controls (Fig. 2B).

    FIG. 2. IGFBPs induce an intrinsic apoptotic pathway. A, Localization of cytochrome c was determined by immunocytochemistry and visualized by fluorescence microscopy 24 h after adenoviral infection. B, Immunoblot analysis of cytochrome c expression in cytosolic extracts from IGFBP-expressing cells 24 h after adenoviral infection. C, Effects of IGFBPs on the release of cytochrome c from isolated mitochondria. Mitochondria were incubated with CM-derived IGFBPs, then analyzed for cytochrome c expression by immunoblotting. Lane 1, Untreated control; lane 2, 30-kDa intact IGFBP-5; lane 3, 20-kDa IGFBP-5 fragment; lane 4, 20- and 14-kDa IGFBP-5 fragments; lane 5, IGFBP-3; lane 6, 1 mM CaCl2 (positive control). Expression of cytochrome c oxidase (subunit IV) was used as a loading control (Cox IV). Representative blots from three independent experiments are shown.

    To determine whether IGFBP-3 and -5 could directly induce mitochondrial cytochrome c release, isolated mitochondria were incubated with 1 μg CM-derived IGFBP-3 or IGFBP-5 (including naturally occurring IGFBP-5 fragments), then examined for cytochrome c expression by immunoblotting. Figure 2C demonstrates that similar levels of expression of cytochrome c were observed in mitochondria incubated with IGFBPs compared with untreated mitochondria.

    IGFBP-5, but not IGFBP-3, sensitizes human breast cancer cells to TNF

    MDA-MB-231 human breast cancer cells are relatively resistant to the inhibitory effects of TNF (33). We investigated the effects of stable expression of IGFBP-3 and IGFBP-5 on the response of MDA-MB-231 cells to TNF. Although MDA-MB-231 secreted low levels of endogenous IGFBP-3, stable transfectants secreted comparable levels of either IGFBP-3 or IGFBP-5 (Fig. 3A). Levels of IGFBPs in vector cells were below the limit of detection of the assays (<5 ng/ml for IGFBP-5 and <2 ng/ml for IGFBP-3). Cell growth was assessed over a 7-d period in the presence or absence of TNF. Figure 3B demonstrates that although IGFBP-3 and IGFBP-5 expression alone was growth inhibitory compared with vector controls, as previously described (2, 3), IGFBP-5 expression sensitized MDA-MB-2321 cells to the inhibitory effects of TNF (Fig. 3B, left graph; P < 0.0001, by repeated measures ANOVA). Thus, although mean cell counts on d 7 in vector controls (6.13 ± 0.48 x 105) were unaffected by the presence of TNF (6.29 ± 0.48 x 105), cell counts in IGFBP-5-expressing cells (1.79 ± 0.43 x 105) were decreased by approximately 75% (0.45 ± 0.04 x 105) by TNF treatment. This effect was not observed in IGFBP-3 transfectants, in which treatment had no effect on cell numbers (Fig. 3B, right graph). We previously demonstrated that stable expression of either IGFBP-3 or IGFBP-5 alone significantly reduces DNA synthesis by approximately 15% and 20%, respectively (2, 4). To examine the additional effect of TNF treatment, levels of DNA synthesis (assessed by incorporation of [3H]thymidine) in TNF-treated cells were expressed as a percentage of untreated cells. Figure 3C demonstrates that TNF treatment for 24 h significantly reduced DNA synthesis in MDA/BP-5 cells compared with MDA/VEC (Fig. 3C, left graph; P < 0.0001), an effect that was not observed in MDA/BP-3 cells (Fig. 3C, right graph).

    FIG. 3. Expression of IGFBP-5 sensitizes human breast cancer cells to TNF. A, Levels of IGFBP-3 and IGFBP-5 secreted by MDA/VEC and MDA/BP stable transfectants. CM was collected after 24-h incubation under SF conditions and assayed by specific RIA. The shaded areas show the limit of detection of the assay. B, MDA/VEC and MDA/BP stable transfectants were grown in 10% FCS, either with 10 ng/ml TNF ( and , respectively) or without TNF ( and , respectively). At the indicated time points after seeding, viable cell numbers were determined by trypan blue exclusion and cell counting, and fresh medium with or without TNF was added. Values shown are the mean of triplicate wells ± SE from three independent experiments. *, P < 0.005; **, P < 0.0001 (for IGFBP transfectants vs. vector controls). C, DNA synthesis was determined by [3H]thymidine incorporation after 24-h incubation in SF medium in the presence or absence of 10 ng/ml TNF. Data shown are for TNF-treated cells as a percentage of untreated cells. Values shown are the mean of quadruple wells from two independent experiments ± SE. D, MDA/VEC () and MDA/BP () were treated with various doses of TNF as indicated for 2 h, and survival was measured 14 d after treatment by clonogenic assays. Values shown are the mean of triplicate wells from two independent experiments ± SE. *, P < 0.05; **, P < 0.01; ***, P < 0.0001 (for IGFBP-expressing cells vs. vector controls).

    The effects of IGFBP-3 and IGFBP-5 on the long-term survival of MDA-MB-231 cells in response to TNF was examined by clonogenic survival assays in stable transfectants. We previously demonstrated using this assay that stable expression of either IGFBP-3 (3) or IGFBP-5 (2) alone significantly reduces breast cancer cell survival (by 50%). However, to examine the additional effects of TNF, the survival of untreated IGFBP transfectants was expressed as 100%. Figure 3D demonstrates that in response to TNF, IGFBP-5 transfectants had significantly reduced survival over 14 d compared with vector controls (Fig. 3D, left graph; P < 0.01), an effect that was not observed in IGFBP-3 transfectants (Fig. 3D, right graph).

    IGFBP-5 expression blocks NFB-mediated survival signals in response to TNF

    We examined whether the enhanced sensitivity to TNF observed in IGFBP-5-expressing cells was due to changes in expression of the TNFR1. TNFR1 expression was examined both basally and in response to TNF in stable IGFBP-3 and IGFBP-5 transfectants by immunoblotting. Figure 4A demonstrates that no significant difference in TNFR1 expression was observed after treatment, suggesting that the differential response to TNF is mediated downstream of the TNFR1.

    FIG. 4. IGFBP-5 expression blocks NFB-mediated survival signals in response to TNF. A, TNFR1 expression levels in IGFBP transfectants and vector controls. Cells were treated with or without TNF (10 ng/ml) for 24 h, then cell lysates were immunoblotted for TNFR1 expression. Actin expression was used as a loading control. B, Upper panel, Levels of RelA (the active subunit of NFB) in total cell extracts from untreated IGFBP transfectants and vector controls. Lower panel, Expression of RelA in nuclear extracts of IGFBP transfectants and vector controls. Cells were treated with or without TNF (10 ng/ml) for 24 h, then nuclear extracts were prepared and immunoblotted for RelA expression. The expression of actin was used as a loading control. Representative blots from two independent experiments are shown.

    We also examined at which point IGFBP-5 acts to sensitize MDA-MB-231 cells to the cytotoxic effects of TNF. After binding of TNF to the TNFR1, survival signals are mediated by the nuclear translocation of the RelA (p65) subunit of the transcription factor NFB (17). Total RelA levels were not significantly different in the IGFBP transfectants and vector controls (Fig. 4B, upper panel). RelA levels were examined in nuclear extracts of stable transfectants treated with or without TNF for 24 h (Fig. 4B). Proliferating cell nuclear antigen expression was used as a loading control. Densitometric analysis of the bands demonstrated that nuclear RelA abundance was increased approximately 3-fold in response to TNF in IGFBP-3-expressing cells (P < 0.02) and vector controls (P < 0.05), but not in MDA/BP-5 cells. This suggests that the sensitivity of MDA/BP-5 cells to the cytotoxic effects of TNF may be due to a block in the survival signal initiated by the nuclear translocation of RelA.

    To determine whether inhibition of nuclear RelA translocation would be sufficient to render IGFBP-3-expressing cells similarly sensitive to TNF, we used the proteasome inhibitor LLnL, which inhibits the activation of NFB by blocking its nuclear translocation. LLnL has been shown to sensitize melanoma cells to death receptor-mediated apoptosis (34). IGFBP transfectants were treated with TNF in the presence or absence of LLnL (5 μM) for 24 h, then nuclear extracts were examined for RelA abundance. Figure 5A shows that the nuclear expression of RelA induced by TNF (shown in Fig. 4B) is equally blocked by treatment with LLnL in IGFBP-3- and IGFBP-5-expressing cells. To determine the importance of NFB-mediated survival signals in the resistance of IGFBP-3 cells to TNF, we determined the ability of LLnL to alter the sensitivity of IGFBP transfectants to TNF. Cells were incubated with LLnL (5 μM) in the presence or absence of TNF (10 ng/ml) for 2 h, then 14-d survival was measured by clonogenic assay. Figure 5B demonstrates that treatment with LLnL did not alter the response of IGFBP transfectants to TNF, i.e., blocking NFB-mediated survival signals was not sufficient to sensitize IGFBP-3-expressing cells to TNF. This suggests that there may be additional mechanisms mediating the differential responsiveness of IGFBP-3- vs. IGFBP-5-expressing cells to TNF-.

    FIG. 5. The proteasome inhibitor, LLnL, blocks NFB activation, but does not restore TNF sensitivity to IGFBP-3-expressing cells. A, Cells were incubated with TNF (10 ng/ml) in the presence or absence of LLnL (5 μM) for 24 h, then nuclear extracts were prepared and immunoblotted for RelA expression. The expression of actin was used as a loading control. B, Cells were treated with LLnL (5 μM) in the presence () or absence () of TNF (10 ng/ml) for 2 h. Survival was measured 14 d after treatment by clonogenic assays and expressed as a percentage of cells not treated with TNF. Values shown are the mean of triplicate wells from two independent experiments ± SE. *, P < 0.001 (for TNF-treated MDA/BP-5 cells vs. untreated MDA/BP-5 cells).

    Caspase-8-independent activation of Bid in IGFBP-5-expressing, but not IGFBP-3-expressing, cells

    To examine possible additional mechanisms mediating the differential responsiveness to TNF, activation of Bid was examined both basally and in response to TNF treatment. Bid cleavage was examined by immunoblotting 24 h after infection with IGFBP-5-, IGFBP-3-, or vector-adenovirus. The antibody used only detects full-length Bid protein. Figure 6A demonstrates that there was a significant decrease in inactive Bid in IGFBP-5-expressing cells compared with IGFBP-3-expressing cells or vector controls, suggesting that Bid is activated in MDA/BP-5 cells. Bid expression was also examined in stable transfectants after a 24-h incubation in the presence or absence of TNF (10 ng/ml). No activation of Bid was observed in MDA/BP-3 or MDA/VEC cells, either basally or in the presence of TNF (Fig. 6B). However, a decrease in full-length, inactive Bid was observed in the IGFBP-5 transfectants, both basally and in the presence of TNF, although the additional effect of TNF was relatively small compared with the decrease in Bid induced by IGFBP-5 alone. We then determined whether the activation of Bid in IGFBP-5-expressing cells was induced by caspase-8. Figure 6C shows expression of Bid in cells treated with TNF in the presence or absence of the caspase-8-inhibitor, IETD-fmk (50 μM). The activation of Bid observed in MDA/BP-5 cells was not ablated in the presence of IETD-fmk, suggesting that it is caspase-8 independent. Figure 6C (right panel) demonstrates that IETD-fmk at this concentration (50 μM) is able to effectively inhibit caspase-8 activation in TNF-treated MDA/BP-5 cells.

    FIG. 6. Differential activation of Bid by IGFBP-5 and IGFBP-3. A, MDA-MB-231 cells were infected with IGFBP-3-, IGFBP-5-, or vector-adenovirus, then cell lysates were prepared 24 h after infection and immunoblotted for expression of the inactive, precursor form of Bid. The expression of -tubulin was used as a loading control. B, Immunoblot analysis of Bid expression in cell lysates from stable transfectants 24 h after treatment with TNF (10 ng/ml) or in untreated controls. The expression of -tubulin was used as a loading control. C, Left panel, Immunoblot analysis of Bid expression in cell lysates from TNF-treated (24 h; 10 ng/ml) stable transfectants in the presence or absence of the caspase-8 inhibitor, IETD-fmk (50 μM). The expression of -tubulin was used as a loading control. Right panel, IETD-fmk (50 μM) blocks caspase-8 activation in TNF-treated (24 h; 10 ng/ml) MDA-BP-5 cells. D, Immunoblot analysis of Smac/DIABLO expression in cytoplasmic extracts from IGFBP-expressing cells after 24-h incubation in the presence or absence of TNF (10 ng/ml). The expression of -tubulin was used as a loading control. Representative blots from two independent experiments are shown.

    Deng et al. (20) recently reported caspase-8-independent activation of Bid after JNK activation and mitochondrial release of Smac/DIABLO. We investigated the mitochondrial release of Smac/DIABLO in IGFBP-5- and IGFBP-3-expressing cells. Figure 6D shows Smac/DIABLO expression in cytoplasmic extracts from IGFBP-expressing cells after 24-h incubation in the presence or absence of TNF. Densitometric analysis of the detected bands showed a significant increase in levels of Smac/DIABLO in cytoplasmic extracts from IGFBP-5-expressing cells, both basally (P < 0.01) and in response to TNF (3-fold; P < 0.04) compared with the respective vector controls. This increase in release of Smac/DIABLO from mitochondria into the cytosol in response to TNF was not observed in IGFBP-3-expressing cells or vector controls (Fig. 6D).

    Differential phosphorylation of JNK in IGFBP-expressing cells

    To investigate the involvement of JNK in the differential responsiveness to TNF, we examined the activation of JNK using a phospho-specific antibody. Stable transfectants were incubated in the presence or absence of TNF for 24 h, then immunoblotted for phospho-JNK expression. Figure 7A demonstrates that treatment with TNF resulted in an increase in the levels of phospho-JNK in the stable transfectants. However, levels of phospho-JNK in IGFBP-5-expressing cells were higher than in the other cell populations, both basally and in response to TNF-.

    FIG. 7. Enhanced activation of JNK in IGFBP-5-expressing cells. A, Immunoblot analysis of phospho-JNK1 expression in cell lysates from stable transfectants 24 h after treatment with TNF (10 ng/ml) or in untreated controls. The expression of total JNK (JNK1 and JNK2) was used as a loading control. B, Expression of full-length Bid in lysates from 24-h TNF-treated cells in the presence or absence of JNK-specific inhibitor (SP600125; 20 μM). Also shown is the effect of SP600125 on the expression of Bid in MDA/BP-5 not treated with TNF. C, Stable transfectants were treated with SP600125 (20 μM) in the presence or absence of TNF (10 ng/ml) as indicated for 2 h as described in Materials and Methods. Survival was measured 14 d after treatment by clonogenic assays and expressed as a percentage of cells not treated with TNF. Values shown are the mean of triplicate wells from two independent experiments ± SE. *, P < 0.01; **, P < 0.001 (for inhibitor-treated MDA/BP-5 cells vs. untreated MDA/BP-5 cells).

    Next we investigated the ability of the JNK-specific inhibitor, SP600125, to ablate the inhibitory effects of TNF in IGFBP-5-expressing cells. Previous studies have shown that SP600125 (20 μM) can effectively inhibit stress-induced activation of JNK in MDA-MB-231 cells (35). Although SP600125 had no significant effect on Bid cleavage in vector cells, it inhibited the cleavage of Bid observed in MDA/BP-5 cells, both basally and in the presence of TNF (Fig. 7B), suggesting that the Bid cleavage observed under these conditions is mediated via the activation of JNK. We then determined the effects of JNK inhibition on the differential response of transfectants to TNF. Stable transfectants were treated with or without SP600125 (20 μM) in the presence or absence of TNF (10 ng/ml) for 2 h, then clonogenic survival was measured as previously. As discussed in Fig. 3, survival in IGFBP transfectants was reduced compared with their vector controls, but normalized to 100%. Figure 7C shows that specifically blocking activation of JNK inhibits the ability of IGFBP-5-expressing cells to respond to TNF.

    Discussion

    Both IGFBP-3 and IGFBP-5 have potent, caspase-dependent, proapoptotic effects in human cancer cells (2, 4, 6) and modulate the effects of other apoptotic stimuli, such as ionizing radiation (3, 36). Two caspase-mediated apoptotic pathways have been characterized in mammalian cells: the intrinsic pathway, involving mitochondrial release of cytochrome c and activation of caspase-9, and the extrinsic or death receptor-mediated pathway, resulting in the activation of caspase-8. Cross-talk between the two pathways centers upon the caspase-8-mediated cleavage of Bid, which then amplifies the apoptotic cascade at the level of the mitochondria. In this study we have delineated the specific apoptotic pathways initiated by IGFBP-3 and IGFBP-5 in MDA-MB-231 human breast cancer cells and have demonstrated a differential modulation of TNF-mediated apoptotic pathways.

    The apical caspases, caspase-8 and -9, are activated in response to apoptotic stimuli by proteolytic cleavage of the precursor forms into the active subunits of the enzyme (37). Adenovirus-mediated expression of either IGFBP-3 or -5 induced apoptosis to a comparable level in MDA-MB-231 cells and was associated with activation of both caspase-8 and caspase-9. The activation of caspase-9 may be initiated by modulation of proapoptotic Bcl-2-like proteins, which we have previously observed in association with the expression of IGFBP-3 and -5 (2, 3). Our results, demonstrating the activation of caspase-8 during IGFBP-mediated apoptosis, correlate with a recent study by Kim et al. (38), who demonstrated cleavage of caspase-8 in IGFBP-3-expressing breast cancer cells. However, this is the first report of caspase-8 activation in IGFBP-5-induced apoptosis. This basal activation of caspase-8 may occur in a death receptor-independent manner downstream of mitochondrial cytochrome c release and caspase-9 activation (39), as has been observed in the breast cancer cell line MCF-7 in response to staurosporine (20), and may act to amplify the apoptotic signal initiated by IGFBPs in these cells. Interestingly, the use of z-IETD-fmk and z-LEHD-fmk at concentrations shown to inhibit caspase activation ablated IGFBP-mediated apoptosis. This may suggest that these apical caspases are activated along a common pathway by IGFBP-3 and -5; thus, blocking either would ablate cell death. Alternatively, caspase-9 may be the initiating caspase that subsequently activates caspase-8, as described above, and z-IETD-fmk at these concentrations may delay IGFBP-mediated apoptosis rather than prevent it.

    Adenovirus-mediated expression of either IGFBP-3 or IGFBP-5 was associated with a release of cytochrome c from the mitochondria into the cytoplasmic compartment. However, in a cell-free system, IGFBPs did not directly induce mitochondrial cytochrome c release, suggesting that this effect may be initiated intracellularly via IGFBP-mediated modulation of Bcl-2 proteins, namely, an induction of proapoptotic Bax and Bad, and down-regulation of antiapoptotic Bcl-2 and Bcl-xL proteins, as we have previously demonstrated (2, 3). These data suggest that IGFBPs can engage an intrinsic apoptotic pathway in breast cancer cells and are consistent with the observations that both IGFBP-3 and IGFBP-5 can enhance apoptosis induced by DNA-damaging agents such as ionizing radiation (2, 3, 36), which act via this pathway.

    Upon binding of its ligand, TNF, the TNFR1 can mediate either cell death or cell survival signals. The former is initiated via the formation of an intracellular death-effector complex and subsequent activation of caspase-8 (40). Cell survival is mediated via the nuclear translocation of the RelA/p50 heterodimer of the transcription factor, NFB, which leads to the transcription of prosurvival genes. MDA-MB-231 cells are relatively resistant to the inhibitory effects of TNF (33). However, the expression of IGFBP-5 was associated with increased cytostatic and cytotoxic effects in the presence of TNF, an effect that was not observed in IGFBP-3-expressing cells. Similar to previous reports in MDA-MB-231 cells, this differential sensitivity to TNF was not related to changes in TNFR1 expression levels (41). The resistance of MDA-MB-231 cells to growth inhibition in the presence of TNF is associated with an increase in nuclear RelA. This fails to occur in TNF-sensitive, IGFBP-5-expressing cells, suggesting that the latter influences TNF sensitivity upstream of NFB-mediated survival signaling. NFB activation is prevented by its interaction with the inhibitor, IB, which masks NFB’s nuclear translocation signal and holds it in the cytosol (42), and it is possible that IGFBP-5 may act to prevent NFB activation by modulating levels of IB in these cells. The proteasome inhibitor, LLnL can block NFB activation by preventing the degradation of IB (34). However, treatment of IGFBP-3-expressing cells with LLnL did not alter their sensitivity to TNF-induced growth inhibition. This suggests that blockade of nuclear RelA alone is not sufficient to sensitize the cells to TNF, and that additional, NFB-independent mechanisms may govern the differential responsiveness of IGFBP-3- vs. IGFBP-5-expressing cells.

    Activation of the proapoptotic protein Bid has been shown to influence sensitivity to death receptor apoptotic pathways (19). We observed a correlation between Bid activation and responsiveness to TNF in IGFBP-expressing cells; namely, Bid was activated both basally and in response to TNF in sensitive MDA/BP-5 cells, but not in resistant MDA/BP-3 cells. Interestingly, IGFBP-5 expression alone caused a significant activation of Bid, with smaller, additional effects seen in the presence of TNF. Bid can be activated by a caspase-8-dependent pathway to produce truncated Bid, which translocates to the mitochondria and induces cytochrome c release (18). However, Bid activation in IGFBP-5-expressing cells was not blocked by the caspase-8-specific inhibitor, z-IETD-fmk, suggesting that it may occur via a caspase-8-independent pathway in these cells. This correlates with a recent study by Deng et al. (20), who demonstrated that activation of JNK leads to a caspase-8-independent activation of Bid and the subsequent release of Smac/DIABLO from mitochondria. Additional support for the involvement of this mechanism can be found in our observation of increased levels of Smac/DIABLO in the cytosol of IGFBP-5-expressing cells, both basally and in response to TNF, compared with vector controls. Upon additional investigation, we demonstrated enhanced phosphorylation of JNK in IGFBP-5-expressing cells. Furthermore, MDA/BP-5 cells had a greatly reduced responsiveness to TNF in the presence of the specific JNK inhibitor, SP600125, suggesting that the enhanced responsiveness in IGFBP-5-expressing cells may be mediated via the activation of JNK. The mechanism of this enhanced activation of JNK in IGFBP-5-expressing cells is unclear. However, our previous studies have provided evidence that the inhibitory and proapoptotic effects of IGFBP-5 are mediated intracellularly (2), suggesting that IGFBP-5 may interact with intracellular signaling pathways.

    Interestingly, in TNF-sensitive breast (43) and prostate (44) cell lines, IGFBP-3 expression is up-regulated in response to TNF, with some evidence that it may mediate the inhibitory effects of TNF in these cells. This suggests that IGFBP-3 may also play a role in mediating extrinsic apoptotic pathways in cells that are sensitive to the inhibitory effects of TNF-.

    Our data suggest the following model by which basal apoptosis initiated by IGFBP-3 and IGFBP-5 involves activation of an intrinsic apoptotic pathway via modulation of proapoptotic Bcl-2 proteins. This results in the activation of caspase-9, which may then lead to activation of caspase-8 via a previously described, death receptor-independent, amplification step (39). In IGFBP-5-expressing cells, JNK is highly phosphorylated, which may act to make these cells more permissive for cell sensitization toward TNF. Thus, in response to TNF, a cell death pathway is initiated that may involve blocking NFB-mediated survival signals, cleavage of the proapoptotic protein Bid, and release of mitochondrial Smac/DIABLO, the effect being reversed by JNK inhibition. This proposal is consistent with the previous observation that JNK activation is necessary, but not sufficient, for TNF-induced apoptosis in breast cancer cells (45). Because in this and other systems, JNK activation appears to be a critical mediator of the response to TNF, it will be important in future studies to determine the mechanism of this activation by IGFBP-5. IGFBP-5 is reported to activate a serine-threonine kinase cell surface receptor (8), which may initiate these effects.

    In conclusion, we have demonstrated that although IGFBP-3 and IGFBP-5 share much structural and functional homology and are both potently proapoptotic in human breast cancer cells, they may act via distinct inhibitory pathways. Additional evidence to support this can be found in our observations that IGFBP-3 and IGFBP-5 induce a differential cell cycle arrest in breast cancer cells; IGFBP-3 induces an arrest in G1/S phase (4), whereas IGFBP-5 induces predominantly a G2/M arrest (2). In addition, IGFBP-3 and IGFBP-5 expressions are associated with distinct effects on breast tumor growth in vivo; IGFBP-5 potently inhibits breast tumor growth (2), but expression of IGFBP-3 is associated with growth stimulation (27). This may have important implications for the potential use of IGFBPs as therapeutic agents or adjuncts to current therapies for breast cancer treatment.

    References

    Firth SM, Baxter RC 2002 Cellular actions of the insulin-like growth factor binding proteins. Endocr Rev 23:824–854

    Butt AJ, Dickson KA, McDougall F, Baxter RC 2003 Insulin-like growth factor binding protein-5 inhibits the growth of human breast cancer cells in vitro and in vivo. J Biol Chem 278:29676–29685

    Butt AJ, Firth SM, King MA, Baxter RC 2000 Insulin-like growth factor binding protein-3 modulates expression of Bax and Bcl-2 and potentiates p53-independent radiation-induced apoptosis in human breast cancer cells. J Biol Chem 275:39174–39181

    Butt AJ, Fraley KA, Firth SM, Baxter RC 2002 Insulin-like growth factor binding protein-3-induced growth inhibition and apoptosis do not require cell-surface binding and nuclear translocation in human breast cancer cells. Endocrinology 143:2693–2699

    Gill ZP, Perks CM, Newcomb PV, Holly JMP 1997 Insulin-like growth factor-binding protein (IGFBP)-3 predisposes breast cancer cells to programmed cell death in a non-IGF-dependent manner. J Biol Chem 272:25602–25607

    Rajah R, Valentinis B, Cohen P 1997 Insulin-like growth factor (IGF)-binding protein-3 induces apoptosis and mediates the effects of transforming growth factor-?1 on programmed cell death through a p53- and IGF-independent mechanism. J Biol Chem 272:12181–12188

    Andress DL 1995 Heparin modulates the binding of insulin-like growth factor (IGF) binding protein-5 to a membrane protein in osteoblastic cells. J Biol Chem 270:28289–28296

    Andress DL 1998 Insulin-like growth factor-binding protein-5 (IGFBP-5) stimulates phosphorylation of the IGFBP-5 receptor. Am J Physiol 274:E744–E750

    Oh Y, Muller HL, Pham H, Rosenfeld RG 1993 Demonstration of receptors for insulin-like growth factor binding protein-3 on Hs578T human breast cancer cells. J Biol Chem 268:26045–26048

    Schedlich LJ, Le Page SL, Firth S, Briggs LJ, Jans DA, Baxter RC 2000 Nuclear import of insulin-like growth factor binding protein-3 (IGFBP-3) and IGFBP-5 is mediated by the importin ? subunit. J Biol Chem 275:23462–23470

    Schedlich LJ, Young TY, Firth SM, Baxter RC 1998 Insulin-like growth factor-binding protein (IGFBP)-3 and IGFBP-5 share a common nuclear transport pathway in T47D human breast carcinoma cells. J Biol Chem 273:18347–18352

    McCaig C, Perks CM, Holly JM 2002 Intrinsic actions of IGFBP-3 and IGFBP-5 on Hs578T breast cancer epithelial cells: inhibition or accentuation of attachment and survival is dependent upon the presence of fibronectin. J Cell Sci 115:4293–4303

    Bernardi P, Petronilli V, Di Lisa F, Forte M 2001 A mitochondrial perspective on cell death. Trends Biochem Sci 26:112–117

    Li P, Nijhawan D, Budihardjo I, Srinivasula SM, Ahmad M, Alnemri ES, Wang X 1997 Cytochrome c and dATP-dependent formation of Apaf-1/caspase-9 complex initiates an apoptotic protease cascade. Cell 91:479–489

    Gross A, McDonnell JM, Korsmeyer SJ 1999 BCL-2 family members and the mitochondria in apoptosis. Genes, Dev 13:1899–1911[Free Full Text]

    Ashkenazi A, Dixit VM 1998 Death receptors: signaling and modulation. Science 281:1305–1308

    Beg AA, Baltimore D 1996 An essential role for NFB in preventing TNF--induced cell death. Science 274:782–787

    Luo X, Budihardjo I, Zou H, Slaughter C, Wang X 1998 Bid, a Bcl2 interacting protein, mediates cytochrome c release from mitochondria in response to activation of cell surface death receptors. Cell 94:481–490

    Kandasamy K, Srinivasula SM, Alnemri ES, Thompson CB, Korsmeyer SJ, Bryant JL, Srivastava RK 2003 Involvement of proapoptotic molecules Bax and Bak in tumor necrosis factor-related apoptosis-inducing ligand (TRAIL)-induced mitochondrial disruption and apoptosis: differential regulation of cytochrome c and Smac/DIABLO release. Cancer Res 63:1712–1721

    Deng Y, Ren X, Yang L, Lin Y, Wu X 2003 A JNK-dependent pathway is required for TNF-induced apoptosis. Cell 115:61–70

    Du C, Fang B, Li Y, Li L, Wang X 2000 Smac, a mitochondrial protein that promotes cytochrome c-dependent caspase activation by eliminating IAP inhibition. Cell 102:33–42

    Verhagen AM, Ekert PG, Pakusch M, Silke J, Connolly LM, Reid GE, Moritz ML, Simpson RJ, Vaux DL 2000 Identification of DIABLO, a mammalian protein that promotes apoptosis by binding to and antagonizing IAP proteins. Cell 102:43–53

    Yee D, Sharma J, Hilsenbeck SG 1994 Prognostic significance of insulin-like growth factor-binding protein expression in axillary lymph node-negative breast cancer. J Natl Cancer Inst 86:1785–1789

    Figueroa JA, Jackson JG, McGuire WL, Krywicki RF, Yee D 1993 Expression of insulin-like growth factor binding proteins in human breast cancer correlates with estrogen receptor status. J Cell Biochem 52:196–205

    Yu H, Levesque MA, Khosravi MJ, Papanastasiou-Diamandi A, Clark GM, Diamandis EP 1996 Associations between insulin-like growth factors and their binding proteins and other prognostic indicators in breast cancer. Br J Cancer 74:1242–1247

    Martin JL, Weenink S, Baxter RC 2003 Insulin-like growth factor binding protein-3 potentiates EGF action in MCF-10A mammary epithelial cells: Involvement of p44/42 and p38 MAP kinases. J Biol Chem 278:2969–2976

    Butt AJ, Martin JL, Dickson KA, McDougall F, Firth S, Baxter RC 2004 Insulin-like growth factor binding protein-3 expression is associated with growth stimulation in T47D human breast cancer cells: the role of altered epidermal growth factor signalling. J Clin Endocrinol Metab 89:1950–1956

    Firth SM, Ganeshprasad U, Poronnik P, Cook DI, Baxter RC 1999 Adenoviral expression of human insulin-like growth factor binding protein-3. Protein Expression Purif 16:202–211

    Baxter RC, Meka S, Firth SM 2002 Molecular distribution of IGF binding protein-5 in human serum. J Clin Endocrinol Metab 87:271–276

    Baxter RC, Martin JL 1986 Radioimmunoassay of growth hormone-dependent insulin-like growth factor binding protein in human plasma. J Clin Invest 78:1504–1512

    Martin JL, Baxter RC 1999 Oncogenic ras causes resistance to the growth inhibitor insulin-like growth factor binding protein-3 (IGFBP-3) in breast cancer cells. J Biol Chem 274:16407–16411

    Heibein JA, Barry M, Motyka B, Bleackley RC 1999 Granzyme B-induced loss of mitochondrial inner membrane potential and cytochrome c release are caspase independent. J Immunol 163:4683–4693

    Basu A, Mohanty S, Sun B 2001 Differential sensitivity of breast cancer cells to tumor necrosis factor-: involvement of protein kinase C. Biochem Biophys Res Commun 280:883–891

    Franco AV, Zhang XD, Van Berkel E, Sanders JE, Zhang XY, Thomas WD, Nguyen T, Hersey P 2001 The role of NF-B in TNF-related apoptosis-inducing ligand (TRAIL)-induced apoptosis of melanoma cells. J Immunol 166:5337–5345

    Mingo-Sion AM, Marietta PM, Koller E, Wolf DM, Van den Berg CL 2004 Inhibition of JNK reduces G2/M transit independent of p53, leading to endoreduplication, decreased proliferation, and apoptosis in breast cancer cells. Oncogene 23:596–604

    Williams AC, Collard TJ, Perks CM, Newcomb P, Moorghen M, Holly JM, Paraskeva C 2000 Increased p53-dependent apoptosis by the insulin-like growth factor binding protein IGFBP-3 in human colonic adenoma-derived cells. Cancer Res 60:22–27

    Butt AJ, Kumar S 1999 Caspases: the molecular effectors of apoptosis. In: Lavin M, Watters D, eds. Signalling pathways in apoptosis. Bristol: Harwood Academic; 247–262

    Kim H-S, Ingermann AR, Tsubaki J, Twigg SM, Walker GE, Oh Y 2004 Insulin-like growth factor-binding protein 3 induces caspase-dependent apoptosis through a death receptor-mediated pathway in MCF-7 human breast cancer cells. Cancer Res 64:2229–2237

    Slee EA, Harte MT, Kluck RM, Wolf BB, Casiano CA, Newmeyer DD, Wang HG, Reed JC, Nicholson DW, Alnemri ES, Green DR, Martin SJ 1999 Ordering the cytochrome c-initiated caspase cascade: hierarchical activation of caspases-2, -3, -6, -7, -8, and -10 in a caspase-9-dependent manner. J Cell Biol 144:281–292

    Sartorius U, Schmitz I, Krammer PH 2001 Molecular mechanisms of death-receptor-mediated apoptosis. Chembiochemistry 2:20–29

    Mueller H, Flury N, Liu R, Scheidegger S, Eppenberger U 1996 Tumour necrosis factor and interferon are selectively cytostatic in vitro for hormone-dependent and hormone-independent human breast cancer cells. Eur J Cancer 32A:2312–2318

    Wajant H, Pfizenmaier K, Scheurich P 2003 Tumor necrosis factor signaling. Cell Death Differ 10:45–65

    Rozen F, Zhang J, Pollak M 1998 Antiproliferative action of tumor necrosis factor- on MCF-7 breast cancer cells is associated with increased insulin-like growth factor binding protein-3 accumulation. Int J Oncol 13:865–869

    Rajah R, Lee K-W, Cohen P 2002 Insulin-like growth factor binding protein-3 mediates tumor necrosis factor--induced apoptosis: role of Bcl-2 phosphorylation. Cell Growth Differ 13:163–171

    Tang F, Tang G, Xiang J, Dai Q, Rosner MR, Lin A 2002 The absence of NFB-mediated inhibition of c-Jun N-terminal kinase activation contributes to tumor necrosis factor -induced apoptosis. Mol Cell Biol 22:8571–8579(Alison J. Butt, Kristie A)