Inactivation of the inhibitory kappaB protein kinase/nuclear factor kappaB pathway by Par-4 expression potentiates tumor necrosis factor alpha-induced apoptosis.

Par-4 is a novel protein identified in cells undergoing apoptosis. The ability of Par-4 to promote apoptotic cell death is dependent on the binding and inactivation of the atypical protein kinases C (PKCs). This subfamily of kinases has been reported to control nuclear factor kappaB (NF-kappaB) through the regulation of the IkappaB kinase activity. NF-kappaB activation by tumor necrosis factor alpha (TNFalpha) provides a survival signal that impairs the TNFalpha-induced apoptotic response. We show here that expression of Par-4 inhibits the TNFalpha-induced nuclear translocation of p65 as well as the kappaB-dependent promoter activity. Interestingly, Par-4 expression blocks inhibitory kappaB protein (IkappaB) kinase activity, which leads to the inhibition of IkappaB phosphorylation and degradation, in a manner that is dependent on its ability to inhibit lambda/iotaPKC. Of potential functional relevance, the expression of Par-4 allows TNFalpha to induce apoptosis in NIH-3T3 cells. In addition, the down-regulation of Par-4 levels by oncogenic Ras sensitizes cells to TNFalpha-induced NF-kappaB activation.

Par-4 is a novel protein identified in cells undergoing apoptosis. The ability of Par-4 to promote apoptotic cell death is dependent on the binding and inactivation of the atypical protein kinases C (PKCs). This subfamily of kinases has been reported to control nuclear factor B (NF-B) through the regulation of the IB kinase activity. NF-B activation by tumor necrosis factor ␣ (TNF␣) provides a survival signal that impairs the TNF␣-induced apoptotic response. We show here that expression of Par-4 inhibits the TNF␣-induced nuclear translocation of p65 as well as the B-dependent promoter activity. Interestingly, Par-4 expression blocks inhibitory B protein (IB) kinase activity, which leads to the inhibition of IB phosphorylation and degradation, in a manner that is dependent on its ability to inhibit /PKC.

Of potential functional relevance, the expression of Par-4 allows TNF␣ to induce apoptosis in NIH-3T3 cells. In addition, the down-regulation of Par-4 levels by oncogenic Ras sensitizes cells to TNF␣-induced NF-B activation.
The atypical protein kinase C (PKC) 1 subfamily of isozymes (aPKCs) has recently been the focus of considerable attention. It is composed of two members, PKC and /PKC (1), which appear to be involved in a number of important cellular functions including cell proliferation and survival (2)(3)(4)(5)(6). The mechanisms whereby the aPKCs control these functions most probably involve the ERK cascade (7)(8)(9)(10)(11)(12) and NF-B (4,11,(13)(14)(15)(16)(17), since both signaling pathways are targeted by the atypical PKCs and play critical roles in cell growth and apoptosis (3, 18 -25). In this regard, the aPKCs selectively bind to, and are inhibited by, Par-4, which was initially identified by differential screening in cells that were undergoing apoptosis (3,4,26). Consistently, the ectopic expression of Par-4 in NIH-3T3 cells induces apoptotic cell death in a manner that is dependent on its ability to block the atypical PKCs (3,4). In addition, recent studies show that the expression of Par-4 sensitizes prostate cancer and melanoma cells to apoptotic stimuli (27), as well as that Par-4 may be a mediator of neuronal apoptosis (28). Therefore, the study of the mechanisms of action of Par-4 and the atypical PKCs may be of great importance for the understanding of the signaling events involved in programmed cell death.
Because the atypical PKCs regulate ERK (see above) and this kinase has been shown to be important in cell survival (3, 19 -22), the mechanisms whereby Par-4 induces apoptosis may at least in part involve this pathway. However, this may not be the only way for Par-4 to induce apoptosis. Thus, NF-B that is a target of the atypical PKCs (4,11,(13)(14)(15)(16)(17) is a recognized anti-apoptotic molecule (18,(23)(24)(25). Therefore, it is conceivable that the impairment of the NF-B pathway through the inhibition of the atypical PKCs may contribute to the pro-apoptotic actions of Par-4. TNF␣, although it shares some components of the Fas signaling cascade, is an interesting example of the activation by a single cytokine of two different pathways with opposite effects. Thus, TNF␣ triggers the caspase-8 route to apoptosis and simultaneously activates NF-B, which is an important survival signal (for a recent review, see Ref. 29, and references therein). Therefore, TNF␣ induces apoptosis if the NF-B pathway is inactivated (18,(23)(24)(25). The most classical form of NF-B is a heterodimer of p50 and p65 (Rel A) (30 -32), that is sequestered in the cytosol by IB, which prevents its nuclear translocation and activity (32,33). Upon cell stimulation by inflammatory cytokines such as TNF␣ or IL-1, which are potent activators of the atypical PKCs (17,34,35), IB␣ is phosphorylated in residues 32 and 36, which trigger the ubiquitination and subsequent degradation of IB through the proteosome pathway (33). These events release NF-B, which translocates to the nucleus where it activates several genes involved in cell survival and inflammation (30 -33). Recently, two IB kinases (IKK␣ and IKK␤) have been identified that phosphorylate residues 32 and 36 of IB␣, and whose activity is potently stimulated by TNF and IL-1 (36 -40). Results from this laboratory demonstrated that the atypical PKCs are critical regulators of the IKK activity, which offers a mechanistic explanation to the proposed role of these PKCs on NF-B activation (41).
The studies reported here address the possibility that Par-4 potentiates TNF-induced apoptosis by inhibiting NF-B through the blockade of the aPKC-IKK signaling cascade.

MATERIALS AND METHODS
Plasmids, Cell Culture, and Transfections-The HA-tagged expression plasmids for /PKC, Par-4, Par-4⌬NLS and Par-4⌬NLS⌬, and myc-Par-4 have previously been described (3,4). HA-tagged Ras V12 was generously provided by M. White. The Flag-IKK␤ and IKK␣ constructs were kindly provided by D. Goeddel (Tularik, Inc.). The GST-IB⌬C and GST-IB⌬C A32/36 were transformed into Escherichia coli JM101, and expression of GST fusion proteins and their purification on glutathione-Sepharose were carried out according to the manufacturer's procedures. For generating retroviruses, pBabe-HA-par-4 and pWZL-myc-/PKC-CAAX plasmids (42) and the ecotropic virus packaging line Phoenix were used. Phoenix cells were plated in a 10-cm dish, incubated for 24 h, and then transfected by calcium phosphate with 20 g of a retroviral plasmid (15 h at 37°C). After 48 h, the virus-containing medium was filtered and supplemented with 4 g/ml Polybrene. Target NIH 3T3 cells were plated and incubated overnight. For infections, the culture medium was replaced by the appropriate supernatant, and the culture were incubated at 37°C. NIH 3T3 cells were maintained in high glucose Dulbecco's modified Eagle's medium containing 10% fetal calf serum, penicillin G (100 g/ml), and streptomycin (100 g/ml) (Flow). Subconfluent cells were transfected by the calcium phosphate method (CLON-TECH, Inc.).
Immunofluorescence-NIH 3T3 cells were grown on glass coverslips in growth media. Subconfluent cells were transfected with 5 g of each plasmid. Twenty-hours after transfections, cells were either untreated or stimulated with different amounts of TNF-␣ for 20 min. Cells were rapidly washed twice in ice-cold PBS, and fixed in 4% formaldehyde for 15 min at room temperature. Cells were washed four times with PBS and permeabilized with 0.1% Triton X-100. Free aldehyde groups were quenched with 50 mM NH 4 Cl. Endogenous peroxidase activity was quenched by treatment with 1% H 2 O 2 in PBS for 15 min. The fixed cells were blocked in blocking solution. Cells were incubated with the different antibodies for 1 h at 37°C. Transfected HA-tagged protein was visualized with the monoclonal 12CA5 anti-HA (Roche Molecular Biochemicals) and a fluorescein isothiocyanate-conjugated anti-mouse (Cappel), and the endogenous p65 with a goat polyclonal antibody (Santa Cruz Biotechnology) and the tetramethylrhodamine tyramide TSA-Direct amplification system (NEN Life Science Products), in the double immunofluorescence. For the triple immunofluorescence, transfected HA-tagged protein was detected with the monoclonal 12CA5 anti-HA and a cascade blue-conjugated anti-mouse (Molecular Probes); the myc-tagged constructs were visualized with a polyclonal rabbit anti-Myc (Santa Cruz Biotechnology), followed by a fluorescein isothiocyanate-conjugated anti-rabbit (Cappel); and the endogenous p65 with a goat polyclonal antibody (Santa Cruz Biotechnology) and the tetramethylrhodamine tyramide TSA-Direct amplification system (NEN Life Science Products). Glass coverslips were mounted on Mowiol and were examined with an MRC 1024 Bio-Rad confocal system mounted on a Zeiss Axiovert 135 microscope (Zeiss, Oberkochen, Germany).
Luciferase Reporter Assays-Subconfluent cultures of NIH-3T3 cells were transfected with different amounts of plasmid pCDNA3 -HA-par-4⌬NLS or pCDNA3-HA-par-4⌬NLS⌬ together with 500 ng of either B-Luc or CMV-Luc reporter plasmids, either with or without expression vector for wild type /PKC. After 4 h, the DNA-containing medium was removed and cells were incubated for 16 h, after which cells were stimulated with TNF-␣ for 4 h. Extracts were prepared and luciferase activity determined as described (43).
Preparation of Cytosolic and Nuclear Extracts-Cell extracts fractionation was performed as described (14). Briefly, NIH 3T3 cells were washed with Tris-buffered saline and pellet by centrifugation at 1500 ϫ g for 5 min. The pellet was resuspended in Tris-buffered saline and centrifuged again. The pellet was then resuspended in buffer A (10 mM Hepes, pH 7.9, 10 mM KCl, 0.1 mM EDTA, 0.1 mM EGTA, 1 mM DTT, 0.5 mM PMSF). The cells were allowed to swell on ice for 15 min, after which Nonidet P-40 was added to a final concentration of 0.5% and vigorously vortexed for 10 s. The homogenate was centrifuged for 30 s in a microcentrifuge. The supernatant contained the cytosol. The nuclear pellet was resuspended in ice-cold buffer C (20 mM Hepes, pH 7.9, 0, 4 M NaCl, 1 mM EDTA, 1 mM EGTA, 1 mM DTT, 1 mM PMSF) and the tube vigorously rocked at 4°C for 15 min. The nuclear extract was centrifuged for 5 min at 4°C.

Par-4 Inhibits the Nuclear Translocation of p65 in Response
to TNF␣-Since NF-B has been shown to have anti-apoptotic properties (see above), it was of great interest to determine the effect of Par-4 expression on the nuclear translocation of p65, as a direct measure of NF-B activation. Par-4 was recently reported to bind the nuclear protein WT1 (44), implying that Par-4 can modulate not only the aPKCs, which are located in the cytosol (45), but also a nuclear activity. However, we have recently demonstrated that the possible nuclear function of Par-4 is not required for apoptosis (3). Thus, Par-4⌬NLS, a mutant that lacks the predicted nuclear localization signal located at amino acids 24 -29 of the human Par-4 gene, was always in the cytoplasm of expressing cells and is at least as potent as the wild type protein to induce apoptosis (3). NIH-3T3 cells were then transfected with an HA-tagged version of Par-4⌬NLS, after which they were either untreated or stimulated with TNF␣ for 20 min. Afterward, cells were analyzed by double immunofluorescence microscopy with the monoclonal anti-HA antibody 12CA5 to detect the expressed Par-4 and a goat polyclonal anti-p65 antibody. The upper panels of Fig. 1A show a representative cell expressing Par-4⌬NLS and p65 in the absence of TNF␣; in these panels, Par-4 expression has no effect on the basal localization of p65. The lower panels of Fig.  1A are from an experiment illustrating that the expression of Par-4 inhibits the TNF␣-activated nuclear translocation of p65. Thus, in the field shown in this figure, the cell that does not express the Par-4 construct displays a clear translocation of p65 to the nucleus, whereas the one that expresses Par-4 shows that parameter completely abrogated. When several cell fields were analyzed, it becomes clear that at least more than 70% of the TNF␣-activated cells that express the ectopic Par-4 construct display an impaired nuclear translocation of p65 (Fig.  1B). Similar results were obtained when this experiment was performed with the full-length Par-4 protein (Fig. 1, C and D). It should be noted that no translocation to the nucleus of transfected full-length Par-4 was observed in response to TNF␣ stimulation in these experiments (Fig. 1C). This is further confirmed in the data of Fig. 1E. In that experiment, NIH-3T3 cells were either untrasfected (upper panel) or transfected with the HA-tagged versions of full-length (middle panel) or ⌬NLS (lower panel) Par-4 expression vectors. Afterward, cells were either untreated or stimulated with TNF␣ for different times, and the levels of endogenous Par-4 or of the ectopically expressed Par-4 constructs, were determined in cytosol and nuclear fractions with the corresponding antibodies. No translocation of Par-4 to the nucleus was detected under these conditions (Fig. 1E). Collectively, these results suggest that Par-4 is a cytosolic modulator of the NF-B pathway. To further support this conclusion, NIH-3T3 cell lines (either control or stably overexpressing Par-4) were stimulated with TNF␣ for 30 min, after which cells were fractionated and the transloca-tion of p65 to the nucleus was determined by immunoblot analysis. Results of Fig. 1F confirm that the expression of Par-4 dramatically inhibits the TNF␣-induced nuclear translocation of p65.
Par-4 Inhibits the Activation of a B-dependent Reporter by TNF␣-To further show the implication of Par-4 in NF-B signaling, NIH-3T3 cells were transfected with a B-dependent luciferase reporter plasmid along with increasing concentrations of ⌬NLS (Fig. 2) or full-length (data not shown) Par-4 expression vectors. Afterward, cells were either untreated or stimulated with TNF␣ for 4 h and the luciferase activity was determined. Results shown in the upper panel of Fig. 2 demonstrate that increasing concentrations of the Par-4 construct dramatically inhibits the TNF␣-activated B-dependent reporter activity with little or no effect on a constitutive CMVdependent luciferase reporter. Of note, transfection of a Par-4 deletion mutant that lacks the C-terminal portion responsible for the interaction with the aPKCs, Par-4⌬NLS⌬ (4), shows no effect on the B-dependent reporter activity (Fig. 2, middle   panel). Similar results were obtained when cells were transfected with wild-type Par-4 expression constructs (data not shown). Collectively, these results suggest that Par-4 inhibits NF-B function through the blockade of the atypical PKCs. Consistent with this possibility are the results of the lower panel of Fig. 2, demonstrating that the ability of increasing concentrations of Par-4⌬NLS to inhibit the B-dependent promoter activity is severely abrogated by co-transfection of a /PKC active mutant. Transfection of a kinase-inactive /PKC mutant also inhibits the B-dependent promoter activity, as previously reported (14), and did not block Par-4 effects (data not shown). This clearly indicates that the inhibition of Par-4 effects by /PKC is due to its kinase activity.
Par-4 Expression Inhibits IKK␤ Activation-The fact that the atypical PKCs are critical regulators of IKK␤ activity (41), together with the evidence presented here that Par-4 impairs NF-B signaling, strongly suggest that Par-4 expression could conceivably reduce IKK␤ activation. To address this possibility, cells were transfected with a Flag-IKK␤ expression vector and increasing concentrations of the Par-4 expression plasmid. Afterward, cells were stimulated or not with TNF␣ and the activity of IKK␤ was determined. The expression of Par-4 ( Fig.  3A) or Par-4⌬NLS (Fig. 3B) dramatically inhibits IKK␤ activation by TNF␣. We have previously shown that the atypical PKCs selectively target IKK␤ but not IKK␣ (41). Of note, the expression of Par-4 did not affect TNF␣-stimulated IKK␣ activity (Fig. 3C).
Expression of an Active /PKC Mutant Blocks the Inhibition by Par-4 of IB Phosphorylation and Degradation-To demonstrate that Par-4 inhibits IKK through the atypical PKCs, an activated /PKC mutant (/PKC CAAX ) was introduced into NIH-3T3 cells either alone or together with Par-4, using recombinant replication-deficient retroviruses. Cells were either untreated or stimulated with TNF␣ for different times and the degradation, and Ser 32 phosphorylation of IB␣ was determined as a direct measure of the IKK activity in vivo. Fig. 4 shows that the expression of Par-4 impairs the phosphorylation and degradation of IB in response to TNF␣. In addition, the expression of a /PKC active mutant inhibits to a large extent Twenty hours after transfection, cells were either untreated or stimulated with TNF␣ (30 ng/ml) for 7 min. Afterward, Flag-IKK␤ was immunoprecipitated and its activity was determined as described under "Materials and Methods." Essentially identical results were obtained in another two experiments. B, subconfluent cultures of NIH-3T3 cells in 100-mm plates were transfected with Flag-IKK␤ (10 g) along with 20 g of either HA-Par-4⌬NLS expression vector or control plasmid. Twenty-hours after transfection, cells were either untreated or stimulated with TNF␣ (30 ng/ml) for 7 min. Afterward, Flag-IKK␤ was immunoprecipitated and its activity was determined as above. Essentially identical results were obtained in another two experiments. C, subconfluent cultures of NIH-3T3 cells in 100-mm plates were transfected with Flag-IKK␣ (10 g) along with 20 g of either HA-Par-4 expression vector or control plasmid. Twenty hours after transfection, cells were either untreated or stimulated with TNF␣ (30 ng/ml) for 7 min. Afterward, Flag-IKK␣ was immunoprecipitated and its activity was determined as above. Essentially identical results were obtained in another two experiments. the blockade by Par-4 of this parameter (Fig. 4). Collectively, these findings provide independent evidence, other than the use of dominant negative mutants or antisense oligonucleotides, that the stimulation of the atypical PKCs by TNF␣ accounts for a substantial part of NF-B activation.
Sensitization to the TNF␣ Apoptotic Actions by Expression of Par-4 -The inactivation of the NF-B pathway has been shown to be a key element in the sensitization of cells to the cell killing properties of TNF␣ (see above). Because Par-4 expression provokes the inhibition of NF-B through the aPKC-IKK pathway, we determined in the following experiments if the expression of Par-4 was sufficient to allow the induction of apoptosis by TNF␣. Par-4 was introduced into NIH-3T3 cells either alone or in combination with the /PKC active mutant using recombinant replication-deficient retroviruses as above. Afterward, cells were treated with TNF␣ either in the absence or in the presence of FCS, and cell viability was determined. TNF␣ was unable to induce cell death either in the absence or in the presence of FCS (Fig. 5). The expression of Par-4 induced apoptosis in cells that were incubated in the absence but not in the presence of FCS (Fig. 5). Interestingly, TNF␣ potently induced apoptosis in cells expressing Par-4 irrespective of the presence of FCS (Fig. 5). The co-expression of the /PKC active mutant completely abrogated cell death by TNF␣ in the Par-4-expressing cells (Fig. 5), indicating that Par-4 needs to ablate the activity of the atypical PKCs to promote cell death.
The Down-regulation of Endogenous Par-4 by Oncogenic Ras Sensitizes NIH-3T3 Cells to TNF␣-induced NF-B Activation-We have recently shown that the expression of oncogenic Ras in NIH-3T3 cells completely down-regulates basal levels of endogenous Par-4 by reducing its mRNA content. 2 This is a very interesting observation because it provides us with an useful model system to address the role of changes in endogenous Par-4 levels on NF-B activation, under a determined pathophysiological condition such as the Ras-induced oncogenic transformation. Therefore, a control vector or the HAtagged oncogenic Ras mutant, Ha-Ras V12 , either alone or together with Myc-tagged Par-4, were transfected into NIH-3T3 cells. Afterward, cells were either untreated or stimulated with 10 or 30 ng/ml TNF␣ for 30 min. Cells were analyzed by triple immunofluorescence microscopy with the monoclonal anti-HA antibody 12CA5 to detect the expressed oncogenic Ras, with a rabbit polyclonal anti-Myc antibody to detect the expressed Par-4, and a goat polyclonal anti-p65 antibody to detect the nuclear translocation of p65. Results of Fig. 6 show that the simple expression of oncogenic Ras gives a small although reproducible nuclear translocation of p65 even in the absence of TNF␣. The presence of this cytokine synergistically cooperates with oncogenic Ras to produce that effect. This was not only seen in NIH-3T3 cells (Fig. 6) but also in other cell systems such as HeLa cells, where the synergism between Ras and TNF␣ was even more pronounced (data not shown). Because Ras transformation promotes the down-regulation of Par-4, 2 together with the observations reported here that Par-4 expression inhibits NF-B activation, these results can be interpreted as that the reduction of Par-4 levels promoted by oncogenic Ras sensitizes cells to respond to TNF␣-induced NF-B. Consistent with this model, the ectopic expression of Par-4, which cannot be down-regulated by oncogenic Ras, 2 inhibits the effects of Ras on p65 nuclear translocation (Fig. 6). Collectively, these observations are in good agreement with the notion that changes in the levels of Par-4, either ectopic or endogenous, influence NF-B signaling. DISCUSSION The molecules involved in signaling events controlling cell death and survival are important targets for novel therapies in cancer and inflammatory diseases. The recent discovery that the product of par-4, a gene induced in cells undergoing apoptosis (26), binds to and inhibits the atypical subfamily of PKCs (3,43), suggests that these kinases play important roles in the control of cell survival. Actually, the overexpression of wildtype PKC or /PKC (3,6,43) protects cells from apoptosis induced by different genotoxic and stress insults. Therefore, the investigation of the mechanisms whereby the atypical PKCs regulate cell function is of potential great interest. In this regard, the ability of the atypical PKCs to regulate ERK (3,46) seems important from the point of view of programmed cell death because the basal ERK activity has been shown to be critical for cell survival at least in some cell systems (22,47). Therefore, one of the mechanisms whereby the inhibition of the atypical PKCs by Par-4 induces apoptosis could at least in part involve the reduction of the ERK activity (3,46). This pathway is antagonized by serum survival factors acting through the phosphoinositide 3-kinase/Akt cascade (46), unveiling an interesting cross-talk between different signaling events, which explains why Par-4 induces apoptosis more efficiently in cell cultures maintained under low serum conditions.
There are other apoptotic pathways that are activated in a more direct way such as those of the Fas system (46). Briefly, Fas binds to the adapter molecule FADD, which interacts with FIG. 5. The expression of Par-4 potentiates TNF␣-induced apoptosis. The activated /PKC mutant (/PKC CAAX ) was introduced into NIH-3T3 cells either alone or together with Par-4, using recombinant replication-deficient retroviruses. Afterward, cells were untreated or incubated with TNF␣ (30 ng/ml) for 24 h either in the absence or in the presence of FCS, and cell viability was determined as detailed by Beg and Baltimore (18). Essentially identical results were obtained in another two experiments.
FIG. 6. Oncogenic Ras sensitizes NIH-3T3 cells to TNF␣-induced NF-B activation. NIH-3T3 cells were transfected with 5 g of the HA-tagged oncogenic Ras mutant, Ha-Ras V12 , either alone or together with 5 g of Myc-tagged Par-4. Twenty hours after transfection, cells were either untreated or stimulated with 10 or 30 ng/ml TNF␣ for 30 min. Cells were analyzed by triple immunofluorescence microscopy with the monoclonal anti-HA antibody 12CA5 to detect the expressed oncogenic Ras, with a rabbit polyclonal anti-Myc antibody to detect the expressed Par-4, and a goat polyclonal anti-p65 antibody to detect the nuclear translocation of p65. Results are the mean Ϯ S.D. of several fields (300 cells) from three independent experiments. caspase-8 (also known as FLICE/MACH) triggering the whole caspase cascade (46). Apoptosis induced by TNF␣, although it shares some components of the Fas signaling cascade, is an interesting example of the activation by a single cytokine of two different pathways with opposite effects (30 -32). Thus, TNF␣ triggers the caspase-8 route to apoptosis and simultaneously activates NF-B through independent although highly interconnected elements (29). NF-B is an important survival signal, and TNF␣ induces apoptosis if the NF-B pathway is inactivated (30 -32). Because the atypical PKCs contribute to the activation of NF-B by regulating the IKK enzymatic activity, in the study reported here, we sought to investigate if Par-4 may provide a signal that inactivates NF-B, allowing TNF␣ to induce apoptosis. We show in this study that expression of Par-4 blocks NF-B and IKK activation in a manner that is dependent on the inhibition of the atypical PKCs. These are very interesting observations because they provide independent evidence, other than those obtained with dominant negative mutants (11,13,14,17,43), pseudosubstrate peptide inhibitors (15), or antisense oligonucleotides (16), that the atypical PKCs are critically involved in the control of NF-B and IKK activation (41). Also of potential functional relevance is our finding that expression of Par-4, which by itself is a relatively weak inducer of apoptosis in the presence of high serum concentrations, allows TNF␣ to promote cell death, most probably by inhibiting NF-B. Collectively, these results set an scenario in which the inhibition of the atypical PKCs by Par-4 induces apoptosis in resting cells by reducing ERK activity if the serum/phosphoinositide 3-kinase/Akt system is simultaneously inactivated (46). However, if the induction of apoptosis is an active process, as in the case of TNF␣ signaling, Par-4 contributes to cell death by inactivating the stimulation of the NF-B survival branch. This suggests that potential inhibitors of the atypical PKCs could be used as sensitizers in cancer therapy. How NF-B promotes cell survival is a matter of recent interest. Thus, Baldwin and co-workers (48) have demonstrated that the induction of TRAF1, TRAF2, cIAP1, and cIAP2 may account for the anti-apoptotic effects of TNF␣ signaling. In addition, other molecules such as IEX-1L (49) have also been demonstrated to mediate NF-B-induced protection. Future work in our laboratory will address the impact that Par-4 and the atypical PKCs have on the expression of these survival genes.
A critical question that arises from this study is the pathophysiological implications of these findings. We have evidence that tumor transformation by the Ras oncogene provokes a profound depletion of Par-4 mRNA and protein levels (see above). This may be of potential functional relevance from the point of view of tumor progression because it may constitute a mechanism whereby cancer cells are able to survive and proliferate more efficiently than the normal cells. We demonstrate here that the down-regulation of Par-4 by oncogenic Ras has a measurable impact on TNF␣ signaling toward NF-B activation. Thus, cells expressing oncogenic Ras display a much more efficient nuclear translocation of p65 in response to TNF␣ than control cells. This also seems to be dependent on the downregulation of Par-4, because its ectopic expression dramatically inhibits Ras synergistic actions on NF-B. Together, all these results will be in good agreement with a role of Par-4 during tumorigenic transformation. In this regard, recent data from Rangnekar and co-workers (50) demonstrate the complete depletion of Par-4 in renal cell carcinoma. Of potential interest, these authors presented evidence that the ectopic expression of Par-4 in cell lines derived from those tumors sensitizes them to TNF␣-induced apoptosis. This is clearly in keeping with the data reported in this paper and with the observations that down-regulation of NF-B severely impairs Ras-induced transformation (51). Therefore, Par-4 emerges a novel tumor suppressor gene acting on the aPKCs-IKK signaling axis.