B-Myb-Dependent Regulation of c-Myc Expression by Cytosolic Phospholipase A 2 *

Cytosolic phospholipase A 2 (cPLA 2 ) cleaves mem- brane phospholipids to release arachidonic acid, initiat-ing lipoxygenase and cyclooxygenase pathways. Mice lacking a gene for cPLA 2 suggested important roles of the protein in allergic responses, fertility, and neural cell death. Here we show that cPLA 2 negatively regu- lates c-Myc expression in a B-Myb-dependent manner. Overexpression of cPLA 2 protein but not a mutant cPLA 2 protein that lacks in vitro binding ability with B-Myb inhibits B-Myb-dependent c- myc gene expression. The inhibition was associated with physical interaction of B-Myb protein with cPLA 2 both in the cytoplasm and the nucleus. Binding site analysis demonstrated that both the N and C termini of cPLA 2 interact with B-Myb. Macrophage colony stimulating factor (MCSF) stimu-lated cPLA 2 redistribution into the nucleus and also association with B-Myb in human monocytes. Impor-tantly, macrophages from mice with a disrupted cPLA 2 gene demonstrated significantly increased levels of c-Myc protein in the nucleus compared with cells from the wild-type mice, whereas B-Myb levels were similar in the cells from the cPLA 2 (cid:1) / (cid:1) and FBS. macrophages, cells harvested the with phosphate-buffered saline (PBS). Lung fibroblasts were isolated from cPLA 2 (cid:1) (cid:1) and cPLA 2 (cid:3) (cid:3) mice digesting and 5 (cid:5) -CA- GAAGGTGATCCAGACTCTG, spanning exon 2 and exon 3 of the human c- myc genome. Real-time PCR was performed with a LightCycler instrument (Roche Diagnostics) using a commercially available kit (LightCycler RNA amplification kit SYBR Green I, Roche Diagnostics). The level of c- myc mRNA was expressed as a relative value to glyceraldehydes 3-phosphate dehydrogenase mRNA. cPLA 2 Ablation by RNA-mediated Interference and Antisense Oligo-nucleotide— The RNA-mediated ablation of endogenous cPLA 2 in MCSF-activated human monocytes was performed essentially described as previously (19). A 21-nucleotide small interfering RNA du- plex with 3 (cid:5) -dTdT overhangs corresponding to human cPLA 2 mRNA (5 (cid:5) -CCUGACGUUUCAGAGCUGATT and 5 (cid:5) -TTGGACUGCAAAGU- CUCGACU) was synthesized (Japan Bio Services). RNA-mediated interference transfection was performed using Oligofectamine reagent (Invitrogen). Harvested human peripheral blood monocytes were stim-ulated with MCSF for 24 h and then transfected once using the man- ufacturer’s protocol (Invitrogen). The ablation of endogenous cPLA 2 in human monocytes was also performed using antisense oligonucleotide, which was complementary to nucleotide 219–238 of human cPLA 2 (5 (cid:5) -GTGCTGGTAAGGATCTAT) (20). The mis-sense oligomer was used for the control (5 (cid:5) -GTGCTCCTAAGTTTCTAT). The phosphorothioate-modified oligonucleotides were synthesized and purified by high per- formance liquid chromatography (Sigma).

cPLA 2 1 is activated by cytokines, submicromolar concentrations of Ca 2ϩ ions, and mitogen-activated protein kinasemediated phosphorylation of serine residues in the protein (1,2). Activated cPLA 2 protein distributes preferentially to the perinuclear region of the cell (3)(4)(5), where the enzyme is thought to participate in arachidonic acid release. Consistent with this notion, 5-lipoxygenase and cyclooxygenase, two major enzymes downstream of cPLA 2 , reside in the nuclear envelope and oxidize arachidonic acid (6 -10). However, the entry of cPLA 2 into the nucleus does not occur in these conditions (5). cPLA 2 localizes in the nucleus of subconfluent endothelial cells (11). cPLA 2 can regulate the NF-B-dependent transcription (12). These findings along with the observations that more downstream enzymes such as lipoxygenase and cyclooxygenase reside in the nuclear envelope prompted us to test a hypothesis that cPLA 2 can enter the nucleus under some cellular conditions. Because the cPLA 2 protein does not harbor any apparent classical nuclear localization signal (NLS), cPLA 2 might be directed toward the nucleus by a partner protein(s).
Recent studies using mice lacking a gene for cPLA 2 showed that these mice were associated with impairment of allergic responses and fertility and also suggested important contributions of cPLA 2 to the pathophysiology of neural cell death (13,14). These studies imply that cPLA 2 may play a more important role than previously thought.
B-Myb is a nuclear transcriptional factor belonging to the Myb family that is ubiquitously expressed in tissues and is involved in cell growth control, differentiation, and cancer (15). Furthermore, B-Myb interacts with other transcriptional factors and modulates their biological functions. Here we demonstrate so-far undefined physical associations of cPLA 2 and B-Myb transcriptional factor, which facilitates redistribution of cPLA 2 -B-Myb complexes into the nucleus. Importantly, we also show that redistribution of cytoplasmic cPLA 2 after its specific binding to B-Myb transcriptional factor can regulate B-Mybdependent c-myc gene expression.

EXPERIMENTAL PROCEDURES
Cell Culture-African Green monkey kidney cells (CV-1) and human kidney cells (293T) were cultured in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum (FBS), 4 mM L-glutamine, 100 units/ml penicillin, and 100 g/ml streptomycin. Human monocytes were isolated from the peripheral blood of healthy volunteers by Ficoll-Paque separation followed by adherence for 1 h and removal of non-adherent cells. The adherent cells were collected and washed 3 times with culture medium. The resultant cell population consisted of Ͼ90% monocytes as judged by morphological examination and ␣-naphthyl acetate esterase staining (16). Monocytes were cultured in RPMI1640 medium supplemented with 10% FBS, 4 mM L-glutamine, 100 units/ml penicillin, and 100 g/ml streptomycin. The monocytes were treated with 10 3 units/ml MCSF (Leukoprol; Morinaga Co., Tokyo, Japan). Thioglycolate-elicited peritoneal cells were collected from cPLA 2 ϩ/ϩ and cPLA 2 Ϫ/Ϫ mice and seeded at 4 ϫ 10 5 cells/ml in RPMI1640 supplemented with 10% FBS. To obtain mature macrophages, the collected cells were allowed to adhere to the plastic dish bottom for 2 h in culture medium, and adherent cells were harvested after washing the dish with phosphate-buffered saline (PBS). Lung fibroblasts were isolated from cPLA 2 ϩ/ϩ and cPLA 2 Ϫ/Ϫ mice by digesting * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Plasmid Construction and Transfection-The FLAG-tagged cPLA 2  fragments 1-749 (full-length), 1-524, 1-294, 1-202, and 202-749 were  constructed by subcloning restriction fragments of cPLA 2 into a pact  vector. A FLAG-tagged C-terminal truncated mutant cPLA 2 -(493-749) and FLAG-tagged cPLA 2 mutated in nuclear export signal (NES) were created by PCR and were also subcloned into the same vector. We carried out DNA transfection by LipofectAMINE reagents (Invitrogen) or by calcium phosphate precipitation method. We used 3 and 65 g of plasmids, respectively, for LipofectAMINE and calcium phosphate precipitation methods. Transfection of mouse peritoneal macrophages from cPLA 2 Ϫ/Ϫ mice was carried out according to a commercially available protocol using a Nucleofector transfection system and Mouse Neuron Nucleofector Solution (Amaxa Biosystems). We transfected 4 ϫ 10 6 peritoneal macrophages with 0.5 g of pact plasmid containing fulllength cPLA 2 .
Immunofluorescence Staining and Confocal Microscopy-Cells were fixed for 10 min by 4% paraformaldehyde in PBS at room temperature, permeabilized by 0.1% Triton X-100 in PBS, and blocked with 1.5% skimmed milk and 1.5% bovine serum albumin in PBS. Primary antibodies (10 g/ml) were prepared in 2% bovine serum albumin in PBS. Cells were then incubated with a fluorescein isothiocyanate-conjugated goat anti-mouse antibody (1:20; Zymed Laboratories Inc.) and Cy3conjugated goat anti-rabbit antibody (1:100; Amersham Biosciences). Confocal imaging was performed using a Leica TCS SP2 AOBS laserscanning microscope.
Cell Fractionation-Cell were suspended in ice-cold hypotonic buffer (10 mM Hepes, pH 7.4, 1.5 mM MgCl 2 ,10 mM KCl) supplemented with 0.25% Nonidet P-40, 0.5 mM dithiothreitol, and 0.5 mM phenylmethylsulfonyl fluoride and incubated for 5 min on ice. After centrifugation at 16,000 ϫ g for 10 min at 4°C, the supernatant (cytoplasmic fraction) was collected. The pellets were washed twice with ice-cold buffer. The pellets were resuspended in an appropriate volume of high salt buffer (20 mM Hepes, 25% glycerol, 0.42 M KCl, 1.5 mM MgCl 2 , 0.2 mM EDTA, 0.5 mM dithiothreitol, and 0.5 mM phenylmethylsulfonyl fluoride) and incubated for 15 min on ice. After centrifugation at 16,000 ϫ g for 10 min at 4°C, the supernatants were collected (nuclear fraction).
Immunoprecipitation and Western Blotting-For the immunoprecipitation experiments we used monoclonal antibodies specific for cPLA 2 (Santa Cruz) or FLAG (Eastman Kodak Co.) or polyclonal antibodies for B-Myb (Santa Cruz). Immunoprecipitates were then electrophoresed in an SDS-PAGE gel and transferred to a polyvinylidene difluoride filter (Millipore), and the protein-protein interaction was confirmed by Western blotting using the above antibodies.
Reporter Gene Activity Assays-CV-1 cells were co-transfected with empty vector or a pact plasmid containing B-Myb-(1-700) or pact plasmid containing cPLA 2 -(1-749) or both. The reporter gene activity was measured as described previously (17). The chloramphenicol acetyltransferase (CAT) reporter plasmid pmycCAT was used for the c-myc transcription assays in CV-1 cells expressing B-Myb and/or wild-type or mutant cPLA 2 . The amount of cell extracts used for chloramphenicol acetyltransferase assays were normalized with respect to cotransfected pRL-SV40 plasmid (Promega). The radioactivity of either [ 14 C]chloramphenicol or its acetylated form was measured using a BAS 1500 imager (Fuji Film). Data were shown as averages of three experiments. PLA 2 Assay-In vitro PLA 2 assay was performed on cell lysates after subcellular fractionation into the nuclear and cytoplasmic components using the same protocol for Western blotting as described above. The PLA 2 assay was performed as described in the previous publication (16). PLA 2 activity was assessed in the presence of 4 mM Ca 2ϩ by measuring labeled arachidonic acid release from the sn-2 position of 1-palmitoyl-2-arachidonoyl phosphatidylcholine. The radioactive spots on TLC plates, each corresponding to released arachidonic acid, were visualized by a BAS 1500 imager (Fuji Film) and were quantitated by densitometric scanning of the spots.
Generation of cPLA 2 Ϫ/Ϫ Mice-cPLA 2 Ϫ/Ϫ mice were generated by using homologous recombination as described previously (14). The mouse strains used (C57BL/6J and 129/Ola) are congenitally defective in type II PLA 2 , which is a secretory PLA 2 participating in the release of arachidonate in the inflammatory lesions (18).
Reverse Transcription-PCR-c-myc expression in MCSF-activated human monocytes was assessed by reverse transcription-PCR using a commercially available kit (RT-PCR high-Plus, Toyobo). Primers used were human c-myc 5Ј-GCCAAGCTCGTCTCAGAGAAG and 5Ј-CA-GAAGGTGATCCAGACTCTG, spanning exon 2 and exon 3 of the hu-man c-myc genome. Real-time PCR was performed with a LightCycler instrument (Roche Diagnostics) using a commercially available kit (LightCycler RNA amplification kit SYBR Green I, Roche Diagnostics). The level of c-myc mRNA was expressed as a relative value to glyceraldehydes 3-phosphate dehydrogenase mRNA. cPLA 2 Ablation by RNA-mediated Interference and Antisense Oligonucleotide-The RNA-mediated ablation of endogenous cPLA 2 in MCSF-activated human monocytes was performed essentially described as previously (19). A 21-nucleotide small interfering RNA duplex with 3Ј-dTdT overhangs corresponding to human cPLA 2 mRNA (5Ј-CCUGACGUUUCAGAGCUGATT and 5Ј-TTGGACUGCAAAGU-CUCGACU) was synthesized (Japan Bio Services). RNA-mediated interference transfection was performed using Oligofectamine reagent (Invitrogen). Harvested human peripheral blood monocytes were stimulated with MCSF for 24 h and then transfected once using the manufacturer's protocol (Invitrogen). The ablation of endogenous cPLA 2 in human monocytes was also performed using antisense oligonucleotide, which was complementary to nucleotide 219 -238 of human cPLA 2 (5Ј-GTGCTGGTAAGGATCTAT) (20). The mis-sense oligomer was used for the control (5Ј-GTGCTCCTAAGTTTCTAT). The phosphorothioatemodified oligonucleotides were synthesized and purified by high performance liquid chromatography (Sigma).

Subcellular
Localization of cPLA 2 -To identify the sequence(s) regulating subcellular distribution of cPLA 2 , we created serial deletion mutants and assayed their localization by indirect immunostaining and confocal microscopy. Deletions of the C-terminal amino acids up to but not including the putative NES (PLLLLTP), which resides in the N-terminal region of the PH-like domain, did not affect the distribution of cPLA 2 (Fig. 1,  A-D). Further deletion of the C-terminal 547 amino acids, cPLA 2 -(1-202), resulted in nuclear localization of the protein (Fig. 1, A and E). cPLA 2 distributed diffusely in the nucleus and cytoplasm when CV-1 cells were transfected with a C-terminaltruncated cPLA 2 (amino acids 493-749) (Fig. 1, A and F). Extension of the truncation to the amino acid 202, cPLA 2 -(202-749), however, resulted in exclusively cytoplasmic localization (Fig. 1, A and G). We obtained similar results with 293T cells.
Replacement of all four leucines with alanines in amino acids 263-269 in a truncated (1-294) (Fig. 1, A and H) and a full-length (1-794) (Fig. 1, A and I) cPLA 2 or complete deletion of this sequence (not shown) resulted in predominant nuclear localization of the proteins in CV-1 cells. Collectively, these findings suggest that the mid-portion of cPLA 2 protein directs the protein to the cytoplasm, probably via the nuclear export signal (PLLLLTP).
In Vivo Interaction of cPLA 2 and B-Myb-Because the cPLA 2 protein does not harbor any apparent classical NLS, cPLA 2 might be directed toward the nucleus by a partner protein(s). We, therefore, investigated a possible adapter protein(s) that binds to cPLA 2 and drives the protein into the nucleus. Physical or functional interactions have been suggested between cPLA 2 and proteins involved in transcriptional regulation (12,21). Therefore, we tested the possibility that B-Myb, another protein that regulates transcription of several genes, may play a role as an adopter protein for nuclear transport of cPLA 2 . B-Myb is a potential candidate for this adapter protein, since it is ubiquitously expressed in tissues and is involved in cell growth control, differentiation, and cancer (15). Furthermore, B-Myb interacts with other transcriptional factors and modulates their biological functions (22).
To test this hypothesis we first determined the subcellular distribution of cPLA 2 after co-transfection with vectors containing either wild-type or mutant B-Myb protein. As described above, the wild-type full-length cPLA 2 -(1-749) predominantly localizes in the cytoplasm (Fig. 1B). Surprisingly, cPLA 2 redistributes into the nucleus and colocalizes with B-Myb in the nucleus when the wild-type B-Myb is co-expressed ( Fig. 2A). In contrast, the wild-type cPLA 2 remains in the cytoplasm of cells expressing NLS-deleted B-Myb (B-Myb NLS) (Fig. 2B), suggesting that the NLS of B-Myb is required for the nuclear entry of cPLA 2 -B-Myb complexes. Similarly, truncated cPLA 2 that is cytoplasmic when expressed alone translocates into the nucleus after co-expression with the wild-type B-Myb but not with NLS-mutated B-Myb (not shown). In contrast, co-expression of c-Myb, another member of the Myb family functioning as a transcriptional transactivator, does not direct cPLA 2 into the nucleus (Fig. 2C), indicating that interaction between cPLA 2 and B-Myb is specific. We next examined in vivo interaction between cPLA 2 and B-Myb by immunoprecipitating cPLA 2 followed by Western blotting with monoclonal antibody specific for B-Myb after transfection of 293T cells with the wild-type cPLA 2 tagged with FLAG (Fig. 2F). In this condition, cPLA 2 coimmunoprecipitated with endogenous B-Myb. A reciprocal experiment using FLAGspecific antibodies for immunoprecipitation of the full-length cPLA 2 followed by Western blotting for B-Myb confirmed in vivo interaction of cPLA 2 and B-Myb (Fig. 2G). We detected B-Myb protein in cPLA 2 immunoprecipitates from 293T cells overexpressed with the N-or C-terminal-truncated cPLA 2 (Fig.  2H), suggesting that the B-Myb protein can interact with both the N-and C-terminal regions of cPLA 2 .
Nuclear Colocalization of cPLA 2 and B-Myb in Human Macrophages-We ask next whether the specific interaction between cPLA 2 and B-Myb and subsequent entry of cPLA 2 into the nucleus might have any biological relevance. To address this question, we examined cPLA 2 -B-Myb interaction and the subcellular distributions of these proteins in human monocytes. MCSF up-regulates cPLA 2 protein levels and intrinsic enzymatic activity in human monocytes (16). In resting human monocytes, cPLA 2 was exclusively cytoplasmic, and the endogenous B-Myb protein level was very low in the nucleus (Fig. 3, A and C). MCSF treatment resulted in a significant nuclear redistribution of cPLA 2 , which colocalizes with B-Myb (Fig. 3, B and C). The interaction was confirmed by Western blotting using antibodies specific for cPLA 2 of immunoprecipitates isolated by antibodies specific for B-Myb (Fig. 3D). A reciprocal experiment using antibodies specific for cPLA 2 with immunoprecipitation and antibodies specific for B-Myb with Western blotting also showed an in vivo interaction of these proteins (not shown). In vitro cPLA 2 enzyme assay showed that the nuclear fraction of resting monocytes contained a greater activity than the cytoplasmic fraction (Fig. 3E), indicating that a significant part of cPLA 2 resides in the cytoplasmic fraction. The nuclear redistribution of cPLA 2 was associated with a 2.6-fold increase in cPLA 2 enzymatic activity in the nuclear fraction from MCSF-activated monocytes. In contrast, MCSF treatment resulted in only a 40% increase in cPLA 2 activity in the cytoplasmic fraction, consistent with the notion that MCSF-induced terminal differentiation of human monocytes is associated with selective induction of cPLA 2 activity in the nucleus.
Regulation of B-Myb-dependent Gene Expression by cPLA 2 -B-Myb participates in regulation of several genes, each of which has an important role in cell growth. For example, the c-myc gene bears a putative B-Myb-responsive element on its promoter (23). Therefore, we tested whether nuclear cPLA 2 could affect the B-Myb-dependent c-myc gene expression. To this end we measured the B-Myb-dependent reporter gene activity in CV-1 cells transiently expressing the pmycCAT reporter in the presence or absence of coexpressed wild-type cPLA 2 (Fig. 4A). As expected, transient expression of B-Myb activated the reporter gene activity. Interestingly, coexpression of the full-length cPLA 2 -(1-749) greatly inhibits the B-Myb-dependent reporter gene activity to the control level.
These findings suggest that cPLA 2 negatively regulates cmyc gene expression in a B-Myb-dependent manner. However, it is unclear how the interaction between cPLA 2 and B-Myb would be effective. To explore this, we next asked whether the action of B-Myb would be influenced by changes in levels of cPLA 2 . To test this, we transfected CV-1 cells with varying doses of plasmid containing full-length cPLA 2 and then monitored the B-Myb-dependent activation of c-myc reporters. We found that cPLA 2 inhibited B-Myb-dependent c-myc expression in a dose-dependent manner (Fig. 4B). These findings suggest that cPLA 2 can inhibit the transcriptional activity of M-Myb.
Up-regulation of c-Myc in Macrophages from cPLA 2 -null Mice-To investigate further the biological relevance of cPLA 2 -B-Myb interaction we examined c-Myc protein levels in cells that lack cPLA 2 activity. To this end we used mice with disrupted cPLA 2 gene (cPLA 2 Ϫ/Ϫ mice) (14). Peritoneal macrophages were isolated from these mice and were then monitored for the c-Myc protein levels in the nucleus and cytoplasm. Peritoneal macrophages from cPLA 2 ϩ/ϩ mice showed substantial amounts of cPLA 2 in the cytoplasm and much smaller but detectable levels of cPLA 2 in the nuclei (Fig. 5A). The absence of immunoreactive cPLA 2 protein and cPLA 2 transcripts (not shown) of peritoneal macrophages confirmed the disruption of the cPLA 2 gene in the cells from cPLA 2 Ϫ/Ϫ mice. Peritoneal macrophages from cPLA 2 ϩ/ϩ and cPLA 2 Ϫ/Ϫ mice contained equivalent quantities of B-Myb protein, suggesting that disruption of the cPLA 2 gene does not affect the B-Myb protein levels in these cells. However, peritoneal macrophages from cPLA 2 Ϫ/Ϫ mice showed a substantial level of c-Myc protein in the nucleus, whereas c-Myc protein was barely detectable in macrophages from cPLA 2 ϩ/ϩ mice. Next, we tested whether the restoration of cPLA 2 expression could affect the c-Myc protein level in cPLA 2 Ϫ/Ϫ mouse peritoneal macrophages. We obtained low levels of cPLA 2 after transfection in these primary cells, whereas the c-Myc expression was inhibited to 69% of the levels in non-transfected cPLA 2 Ϫ/Ϫ cells (Fig. 5B). We further examined the effects of cPLA 2 inhibition in normal human peripheral blood monocytes. When the cPLA 2 expression was specifically inhibited by using antisense oligonucleotide or small interfering RNA in MCSF-activated human monocytes, the c-myc expression was up-regulated in these cells (Fig. 5, C-E). Collectively, these results suggest that cPLA 2 interacts directly with B-Myb to enter into the nucleus and thereby functions as an inhibitory factor for the c-Myc activity.
A causative relationship was found between cPLA 2 activation and UV irradiation and apoptosis (24,25). Therefore, we assessed the lack of cPLA 2 on cell survival after DNA-damaging agents using lung fibroblasts isolated from cPLA 2 ϩ/ϩ and cPLA 2 Ϫ/Ϫ mice. The growth rate was similar between cPLA 2 ϩ/ϩ and cPLA 2 Ϫ/Ϫ cells without treatment (data not shown). However, compared with cPLA 2 ϩ/ϩ cells, cPLA 2 Ϫ/Ϫ cells showed decreased numbers of cells 48 h after treatment with UV irradiation or H 2 O 2 (Table I). In contrast, cPLA 2 Ϫ/Ϫ cells were more resistant to ionizing radiation than cPLA 2 ϩ/ϩ cells, as assessed 96 h after irradiation. Taken together these results suggest that the lack of cPLA 2 may affect cell growth and/or apoptosis after DNA damage, depending on the types of cells and damaging.

DISCUSSION
Our report has demonstrated thus far undefined physical associations of cPLA 2 and B-Myb transcriptional factor that facilitates redistribution of cPLA 2 -B-Myb complexes into the nucleus. cPLA 2 protein does not harbor any apparent intrinsic NLS, whereas B-Myb protein has multiple functional NLSs (26). Furthermore, a deletion of the NLS from the B-Myb protein results in a mutant protein that cannot direct cPLA 2 protein into the nucleus. Therefore, B-Myb protein may serve as an adapter protein for the nuclear entry of cPLA 2 . We also found that the physical interaction of cPLA 2 and B-Myb proteins is specific and c-Myb, which is another Myb family transcriptional factor, does not direct cPLA 2 into the nucleus. More importantly, our results suggest that, after binding to B-Myb protein, nuclear cPLA 2 may participate in the transcriptional regulation of B-Myb-dependent expression of c-myc gene.
It is surprising that cPLA 2 is not confined to the cytoplasm but is redistributed into the nucleus to serve as a regulator for B-Mybdependent gene expression. Originally, cPLA 2 was considered to translocate onto the cytoplasmic membrane after appropriate stimuli to release arachidonic acid from membrane phospholipids (3). However, accumulating evidence suggests that cPLA 2 can translocate from the cytosolic compartment to the nuclear membrane (2,4). The ATP depletion may contribute to the nuclear redistribution of cPLA 2 (27). Although the localization of downstream enzyme 5-lipoxygenase and cyclooxygenase in the nuclear membrane (6 -10) may explain in part the physiologic relevance of the redistribution of cPLA 2 to the nuclear membrane, the definite role of the "nuclear" cPLA 2 is not well understood. Recent findings that an acetyltransferase Tip60 interacted and colocal- ized with cPLA 2 in the nucleus and that the introduction of protein complexes into the nucleus was associated with stimulation of apoptosis (28) strongly support the notion that cPLA 2 plays additional roles in the nucleus other than its enzymatic activity at the nuclear membrane.
Our results suggest that the c-myc gene is one of the major targets of cPLA 2 -B-Myb complexes in the nucleus. We also foundthatcPLA 2 -B-MybcomplexesinthenucleusinhibitB-Myb-dependent gene expressions of cdc25C, but the effect was minimal (data not shown). cPLA 2 was reported to be involved in the induction of apoptosis caused by tumor necrosis factor (29). Tumor necrosis factor is considered to transduce apoptotic signals from the cytoplasmic membrane to mitochondria. During these processes the 100-kDa cPLA 2 is cleaved in a caspase 3-dependent fashion to a 70-kDa proteolytic fragment that loses its catalytic activity but contributes to apoptosis (30,31). Although we have shown in the present study that the nuclear cPLA 2 exhibited in vitro enzyme activity, the enzyme activity may be inhibited or masked in the nucleus by an adaptor protein such as B-Myb and Tip60.
The c-myc protooncogene encodes a transcriptional factor that participates in the regulation of cellular proliferation and apoptosis (32,33). In particular, the Myc protein can induce S-phase entry and apoptosis by independent mechanisms (34).  regulate peroxisome proliferator activated receptor-mediated gene transcription, implying further a role in the regulation of genes that participate in differentiation and apoptosis. However, cPLA 2 may exert its regulatory function for peroxisome proliferator-activated receptor-dependent gene transcription in the cytoplasm, since this property was apparent only after stimulation by calcium ionophore A23187, where cPLA 2 did not enter the nucleus. In this report we showed that lack of cPLA 2 affects cell survival after DNA damages (Table I). In this context it is noteworthy that Haq et al. (35) show that deletion of cPLA 2 sustained activation of insulin-like growth factor signaling and promoted striated muscle growth. Therefore, we postulate the possibility that cPLA 2 has separate roles for cell proliferation and apoptosis in the different compartment of the cell; the protein participates in stimulation of cell death in the cytoplasm, whereas the protein plays a role in the inhibition of proliferation and cell death in the nucleus.