Roles of Phosphatidylinositol 3-Kinase and Rac in the Nuclear Signaling by Tumor Necrosis Factor-α in Rat-2 Fibroblasts*

We investigated the extent to which phosphatidylinositol 3-kinase (PI 3-kinase) and Rac, a member of the Rho family of small GTPases, are involved in the signaling cascade triggered by tumor necrosis factor (TNF)-α leading to activation ofc-fos serum response element (SRE) and c-Jun amino-terminal kinase (JNK) in Rat-2 fibroblasts. Inhibition of PI 3-kinase by LY294002 or wortmannin, two specific PI 3-kinase antagonists, or co-transfection with a dominant negative mutant of PI 3-kinase dose-dependently blocked stimulation of c-fosSRE by TNF-α. Similarly, LY294002 significantly diminished TNF-α-induced activation of JNK, suggesting that nuclear signaling triggered by TNF-α is dependent on PI 3-kinase-mediated activation of both c-fos SRE and JNK. We also found nuclear signaling by TNF-α to be Rac-dependent, as demonstrated by the inhibitory effect of transient co-transfection with a dominant negative Rac mutant, RacN17. Our findings suggest that Rac is situated downstream of PI 3-kinase in the TNF-α signaling pathway to the nucleus, and we conclude that PI 3-kinase and Rac each plays a pivotal role in the nuclear signaling cascade triggered by TNF-α.

Phosphatidylinositol 3-kinase (PI 3-kinase) 1 is a lipid kinase involved in mitogenic signal transduction and cellular transformation (1). Evidence from intact cells suggests that PI 3-kinase is activated by a variety of growth factors and exerts its cellular effects by elevating of phosphatidylinositol (3,4,5)triphosphate levels (1)(2)(3). In mammalian cells, PI 3-kinase is required for growth factor-induced changes of the actin cytoskeleton that are mediated by Rac, a member of Rho family GTPases (2,4,5). For example, an inhibition of PI 3-kinase was shown to block growth factor induction of membrane ruffling, while activated PI 3-kinase is sufficient to induce membrane ruffling, acting through Rac (2,4). Thus, Rac appears to lie downstream of PI 3-kinase within a signaling pathway that controls actin remodeling.
Rac is also crucially involved in the regulation of signal transduction cascades to the nucleus evoked by environmental stresses and proinflammatory cytokines; elements of such cascades include c-Jun amino-terminal kinase (JNK) (6,7), c-fos serum response element (SRE) (8 -10), p70 S6 kinase (11), and the transcription factor NF-B (12). For instance, in response to exogenous application of H 2 O 2 or ceramide, a second messenger product of sphingomyelin hydrolysis by sphingomyelinase (13), c-fos SRE, was activated via a Rac-dependent signaling pathway, suggesting a role of Rac in stress-induced gene regulation (9,10). Although the role of PI 3-kinase in the regulation of Rac-mediated membrane ruffling has been well studied (2,4,5), almost nothing is known about the potential role of PI 3-kinase in Rac-mediated gene regulation in response to environmental stress or proinflammatory cytokines.
Tumor necrosis factor (TNF)-␣ is one of the most pleiotropic proinflammatory cytokines, signaling a large number of cellular responses, including cytotoxicity, antiviral activity, fibroblast proliferation, and the transcriptional regulation of various genes (14). It is known that a large majority of the pleiotropic activities of TNF are signaled by the TNF receptor-1 (TNFR1; Refs. [15][16][17]. TNF engagement of TNFR1 leads to the recruitment of TNFR1-associated death domain protein, receptor-interacting protein, and TNFR-associated factor-2 (TRAF2) leads to the formation of a receptor complex (18 -20) within which receptor-interacting protein and TRAF2, respectively, transduce signals required for TNF-mediated activation of NF-B (21) and JNK (22)(23)(24). Nonetheless, little is known about the intracellular signaling mediating activation of nuclear transcription factors. In particular, the roles of PI 3-kinase and Rac in the nuclear signaling by TNF-␣ are as yet unclear. In the present study, we investigated the extent to which PI 3-kinase and Rac are involved in the TNF-␣-induced activation of c-fos SRE and JNK. Our findings suggest that both PI 3-kinase and Rac have crucial functions within the intracellular signaling cascade triggered by TNF-␣ in Rat-2 fibroblasts. phosphothioated at both the 5Ј and 3Ј ends (lowercase "s" in sequences). All other chemicals were from standard sources and were molecular biology grade or higher.
Transient transfection was carried out by plating approximately 5ϫ10 5 cells in 100-mm dishes for 24 h and then adding calcium phosphate:DNA precipitates prepared with 20 g of DNA/dish. The quantities of plasmid used were 3 g of reporter gene (pSRE-Luc) and 5 g of small GTPase expression plasmids (e.g. pEXV-RacV12). To control for variations in cell number and transfection efficiency, all clones were co-transfected with 1 g of pCMV-␤GAL, a eukaryotic expression vector in which the Escherichia coli ␤-galactosidase (lacZ) structural gene is under the transcriptional control of the cytomegalovirus promoter. In each transfection, the quantity of DNA used was held constant (20 g) by adding sonicated calf thymus DNA (Sigma). After a 6-h incubation with calcium phosphate:DNA precipitates, cells were rinsed twice with phosphate-buffered saline (PBS) before incubating them in fresh DMEM supplemented with 0.5% FBS for an additional 36 h. Each dish of cells was then rinsed twice with PBS and lysed in 0.2 ml of lysis solution (0.2 M Tris (pH 7.6) ϩ 0.1% Triton X-100), after which lysed cells were scraped and spun for 1 min. Supernatants were assayed for protein concentration as well as luciferase and ␤-galactosidase activities.
Luciferase activity was assayed using 10 l of extract according to the manufacturer's protocol (Promega Luciferase Assay System; Promega, Madison, WI) and counted for 10 s in a Beckman liquid scintillation spectrometer using the tritium channel with the coincidence circuit disconnected. Transfection experiments were performed in triplicate with two independently isolated sets of cells, and the results were averaged. ␤-Galactosidase assays were carried out on 50-l aliquots of extract (diluted with 100 l of H 2 O) using 150 l of 2ϫ reaction buffer (3 mg/ml O-nitrophenyl-␤-galactopyranoside, 2 mM MgCl 2 , 61 mM Na 2 HPO 4 , 39 mM NaH 2 PO 4 , 100 mM 2-mercaptoethanol). Once a faint yellow color appeared, the reactions were stopped by the addition of 350 l of 1 M Na 2 CO 3 . Optical density at 410 nm was then measured in a spectrophotometer and used to normalize luciferase activity to transfection efficiency. Protein concentrations were determined routinely using the Bradford procedure with Bio-Rad dye reagent and bovine serum albumin as a standard.
JNK/Stress-activated Protein Kinase Assays-To assay JNK activity mediated by TNF-␣ or C 2 -ceramide, subconfluent Rat-2 cells were serum-starved for 24 h in DMEM containing 0.5% FBS and then stimulated with TNF-␣ or C 2 -ceramide for 30 min. Each dish of cells was then washed with cold PBS, lysed by incubation for 5 min at 4°C in 0.5 ml of ice-cold lysis buffer (20 mM Tris (pH 7.4) 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% Triton X-100, 2.5 mM sodium pyrophosphate, 1 mM ␤-glycerophosphate, 1 mM Na 3 VO 4 , 1 g/ml leupeptin) with 1 mM phenylmethylsulfonyl fluoride, scraped into Eppendorf tubes, and triturated by 10 passes through a 21.1-gauge needle on ice. The supernatant (cell lysate) was harvested by microcentrifugation at 14,000 rpm for 10 min. Protein concentrations were equalized by normalizing them to the protein levels (assayed by Bradford procedure with Bio-Rad dye reagent) measured before the JNK assay.
JNK activity was determined using a JNK assay kit according to the manufacturer's protocol (New England Biolabs). Briefly, an aminoterminal c-Jun (amino acid residues 1-89) fusion protein bound to glutathione-Sepharose beads was used to pull down JNK from cell lysates. The kinase reaction (50 l) was then carried out using the c-Jun fusion protein as a substrate in the presence of cold ATP. Phosphorylation of the c-Jun fusion protein at Ser-63 was measured by Western blot using an anti-phospho-c-Jun rabbit polyclonal antibody that detects only catalytically activated c-Jun phosphorylated at Ser-63. Protein samples were heated to 95°C for 5 min and subjected to SDSpolyacrylamide gel electrophoresis on 8% acrylamide gels, followed by transfer to polyvinylidene difluoride membranes for 2 h at 100 V using a Novex wet transfer unit. Membranes were then blocked overnight in PBS-T (PBS containing 0.01% (v/v) Tween 20) with 5% (w/v) nonfat dried milk, after which they were incubated for 2 h with primary antibody (anti-phospho-c-Jun) in PBS-T and then for 1 h with horseradish peroxidase-conjugated secondary antibody. The blots were developed using enhanced chemiluminescence kits (ECL, Amersham Pharmacia Biotech). Bands on XAR-5 film (Eastman Kodak Co.) corresponding to phospho-c-Jun were measured by densitometry.

RESULTS
c-fos SRE Is One of the Nuclear Target Sequences of TNF-␣-As an initial approach to understanding the role of PI 3-kinase in the signal transduction pathway between TNF-␣ and the nucleus, we assessed the capacity of TNF-␣ to stimulate c-fos SRE, which is a primary nuclear target for various extracellular signals (9 -11, 30). To accomplish this, Rat-2 cells were transiently transfected with reporter plasmid pSRE-Luc (3 g) containing c-fos SRE oligonucleotides inserted upstream of the c-fos minimal promoter fused to luciferase coding sequences (9). TNF-␣-induced SRE activation was monitored by measuring luciferase activities normalized to co-transfected ␤-galactosidase activity. As shown in Fig. 1, TNF-␣ stimulated c-fos SRE-dependent reporter gene activity in a dose-and timedependent manner. A maximal 5.7-fold increase in the luciferase activity occurred at a TNF-␣ concentration of 10 ng/ml (Fig.  1, left panel) 1 h after its addition (Fig. 1, right panel). No TNF-␣-induced luciferase activity was seen in cells transiently transfected with pO-Luc (vector without SRE; data not shown).
Encouraged by above results, we next directly assayed TNF-␣-evoked PI 3-kinase activity by measuring the levels of the product, phosphatidylinositol phosphate, in serum-starved Rat-2 cells exposed to TNF-␣ for 10 min (Fig. 3). Consistent with the above results, addition of TNF-␣ stimulated PI 3-kinase activity significantly. Interestingly, we observed a similar stimulation of PI 3-kinase activity by TNF-␣ in Rat2-RacN17 cells (28) stably expressing RacN17, a dominant negative Rac1 mutant, which means that in TNF-␣ signaling Rac must act downstream of PI 3-kinase (Fig. 3).
Essential Role of Rac in the Nuclear Signaling by TNF-␣-PI 3-kinase activity induces repertoire of Rac-mediated responses (2,4). Therefore, to investigate the potential role of Rac in the TNF-␣ signaling to c-fos SRE, we tested the effect of transfection with the expression vector encoding RacN17. As shown in Fig. 4A, TNF-␣-induced SRE activation was dramatically in-

FIG. 2. PI 3-kinase activity is required for TNF-␣-induced SRE activation.
A and B, Rat-2 cells were transiently transfected with 3 g of pSRE-Luc reporter plasmid and then serum-starved as described in Fig. 1. Thereafter, cells were incubated for 30 min with selected concentrations of LY294002 (0, 10, and 25 M) (A) or wortmannin (0, 50, and 100 nM) (B) prior to incubation for 1 h with either control buffer, TNF-␣ (10 ng/ml) or C 2 -ceramide (5 M). C, pSRE-Luc (3 g) was transiently co-transfected with selected amounts (1, 3, and 5 g) of pSG5-⌬p85 encoding a dominant negative mutant of PI 3-kinase. DNA sample size was held at 20 g by addition of calf thymus carrier DNA. Transfectants were serum-starved for 36 h, then incubated for 1 h with TNF-␣ (10 ng/ml) or LPA (10 M), after which they were harvested and relative luciferase activity was assayed. Data are expressed as percentage of control. The results shown are representative of at least three independent transfections.

FIG. 3. Stimulation of PI 3-kinase activity by TNF-␣.
Serumstarved Rat-2 and Rat2-RacN17 cells were incubated for 10 min with TNF-␣, after which PI 3-kinase was immunoprecipitated using antiphosphotyrosine agarose beads. PI 3-kinase activity was assayed by measuring levels of phosphatidylinositol phosphate (PIP), the product of PI 3-kinase. hibited by co-transfection with 5 g of pEXV-RacN17 (ϳ65% reduction in luciferase activity), suggesting that Rac activity is crucial for TNF-␣-induced signaling to c-fos SRE. On the other hand, SRE activation induced by 10 M LPA was unaffected by pEXV-RacN17 transfection.
The role of Rac was further investigated by comparing the SRE-luciferase activities in Rat-2 and Rat2-RacN17 cells. Fig.  4B shows TNF-␣-induced SRE activation was inhibited by 50% in serum-starved Rat2-RacN17 cells. In contrast, levels of LPAinduced SRE activation were similar in Rat-2 and Rat2-RacN17 cells (Fig. 4B), while epidermal growth factor-evoked activation of SRE was reduced somewhat in Rat2-RacN17 cells. We, therefore, conclude that TNF-␣ signaling to c-fos SRE is mediated, at least in part, by a Rac-dependent cascade.
Pretreatment with LY294002 Inhibits JNK Activation by TNF-␣-The effect of LY294002 on TNF-␣-induced JNK activation was assessed to determine the extent to which it is dependent on PI 3-kinase and Rac activities. Serum-starved Rat-2 cells were pretreated with LY294002 (ϩ) or control buffer (Ϫ) for 30 min before adding TNF-␣ (10 ng/ml), C 2 -ceramide (5 M), or arachidonic acid (AA; 100 M), a principal product of Rac-activated phospholipase A 2 (33). TNF-␣ and C 2 -ceramide each induced a ϳ5-fold increase of JNK activity as compared with control buffer, an effect that was dramatically inhibited by LY294002 (Fig. 5A). On the other hand, LY294002 had no inhibitory effect on AA-induced JNK activation, which suggests that PI 3-kinase is specifically required for activation of JNK by TNF-␣ or C 2 -ceramide and implies a common, essential role for PI 3-kinase in TNF-␣-evoked activation of both JNK and c-fos SRE.
To determine the function of Rac in TNF-␣ signaling to JNK, levels of JNK activation were compared between control cells and cells stably expressing RacN17. As shown in Fig. 5B, TNF-␣-and C 2 -ceramide-induced JNK activation was dramatically reduced in Rat2-RacN17 cells, indicating the importance of Rac activity in those cases. On the other hand, JNK activation induced by 100 M AA was unaffected by RacN17 expression.
The signaling hierarchy between PI 3-kinase and Rac was investigated further by assessing the LY294002 sensitivity of SRE activation by RacV12, a constitutively activated form of Rac1. LY294002 had no inhibitory effect on SRE activation by RacV12 or RhoV14 (Fig. 6), whereas RasV12-induced SRE activation was significantly and dose-dependently inhibited by LY294002. This is consistent with previous reports showing that PI 3-kinase acts as a downstream mediator of H-Ras within the signaling cascades leading to actin remodeling and transformation (2, 3), and is further evidence that Rac is situated downstream of PI 3-kinase in the nuclear signaling cascade leading to activation of c-fos SRE or JNK. In a separate FIG. 4. Rac is essential for TNF-␣induced SRE activation. A, reporter gene plasmid pSRE-Luc (3 g) was transiently co-transfected with 5 g of pEXV (Vector) or pEXV-RacN17 (Rac1N17). DNA samples were held at 20 g with by addition of calf thymus carrier DNA. The transfectants were serum-starved as described in Fig. 1. TNF-␣ (10 ng/ml) or LPA (10 M) was added 1 h prior to harvest, after which relative luciferase activities was assayed. B, Rat-2 and Rat2-RacN17 cells were transiently transfected with pSRE-Luc (3 g). The transfectants were serum-deprived, and epidermal growth factor (EGF, 50 ng/ml), LPA (10 M), or TNF-␣ (10 ng/ml) was added 1 h prior to harvest, after which relative luciferase activities were assayed. The results shown are representative of at least three independent transfections.

FIG. 5. LY294002 inhibits JNK activation induced by TNF-␣.
A, Rat-2 cells were serum-starved and then incubated for 30 min with C 2 -ceramide (C 2 -Cer, 5 M), TNF-␣ (10 ng/ml), or AA (100 M). Before the addition of agonists, cells were preincubated for 30 min with either LY294002 (20 M) or control buffer. Protein samples of equal size were then assayed for JNK activity using c-Jun fusion protein (1-89) as a substrate. B, Rat-2 and Rat2-RacN17 cells were serumstarved and then incubated with TNF-␣, C 2 -ceramide, or AA as described in A, after which JNK activity was assayed. experiment, we observed that LY294002 had no inhibitory effect on RacV12-induced JNK activation in Rat-2 cells (data not shown).
Role of cPLA 2 in TNF-␣ Signaling to SRE Activation-We previously reported that cytosolic phospholipase A 2 (cPLA 2 ) plays an essential role in mediating Rac signaling to c-fos SRE and thus acts as an important downstream mediator of Rac (34). Considering the linkage between TNF-␣ and Rac signaling, it seems reasonable to hypothesize that cPLA 2 may be involved in TNF-␣ signaling to SRE. To test this possibility, we assessed the extent to which mepacrine, a potent PLA 2 inhibitor, inhibited TNF-␣-induced activation of SRE. Fig. 7A shows that pretreatment with 1 M mepacrine inhibited TNF-␣-induced SRE activation by approximately 50% without affecting LPA-induced activation, suggesting PLA 2 is specifically required for TNF-␣ signaling to c-fos SRE.
To further analyze the role of PLA 2 in TNF-␣ signaling, especially that of cPLA 2 , we examined the effect of transfecting cells with antisense cPLA 2 oligonucleotide on TNF-␣-induced SRE activation. Co-transfection with the antisense oligonucleotide but not the control oligonucleotide significantly inhibited TNF-␣-induced SRE activation (Fig. 7B). For example, cotransfection with 0.5 M cPLA 2 antisense oligomer reduced SRE activation by ϳ45%, which suggests that a Rac-cPLA 2 -linked cascade is involved in TNF-␣ signaling to c-fos SRE. In contrast, LPA-induced SRE activation was unaffected by transfection of the antisense oligonucleotide, suggesting that the involvement of cPLA 2 is specific to TNF-␣-induced signaling to c-fos SRE.

DISCUSSION
In the present study, we provide evidence supporting novel roles for PI 3-kinase and Rac in the nuclear signaling cascade triggered by TNF-␣ in Rat-2 fibroblasts. TNF-␣ was previously reported to rapidly induce protooncogene c-fos in the adipogenic TA1 cell line, although the exact target promoter sequences by which TNF-␣ stimulates c-fos transcription remain unknown (35). Our results clearly indicate that SRE is at least one of the nuclear target sequences by which TNF-␣ stimulates c-fos expression. Consistent with this conclusion, c-fos SRE is also reported to be a nuclear target of ceramide, a putative second messenger for certain stresses (e.g. ultraviolet and xrays) and inflammatory cytokines such as TNF-␣ (9). In addition, our results suggest a role for cPLA 2 that is in good agreement with the earlier report of Haliday et al. (35) showing that AA and its lipoxygenase-generated metabolite are downstream elements in the TNF-␣ signaling pathway to c-fos. The function of AA as a downstream mediator of TNF-␣ signaling was also demonstrated in stromal cells, where AA mediates TNF-␣induced activation of JNK (36).
The involvement of PI 3-kinase in TNF-␣-induced signaling to c-fos SRE was confirmed by the significant inhibitory effects of LT294003 and wortmannin, specific PI 3-kinase antagonists, and of transient transfection with pSG5-⌬p85 encoding a dominant negative PI 3-kinase mutant. Consistent with this conclusion, JNK activation by TNF-␣ was dramatically inhibited by LY294002, implying PI 3-kinase functions broadly as a downstream TNF-␣ mediator in the signaling pathways leading to SRE and JNK activation. That TNF-␣ stimulates PI 3-kinase activity in vitro lends additional support to this idea.
FIG. 6. RacV12-induced SRE activation is not inhibited by LY294002. Reporter gene plasmid pSRE-Luc (3 g) was transiently co-transfected into Rat-2 cells along with 5 g of pEXV (Vector) or vectors expressing RacV12, RasV12, or RhoV14. DNA sample size was held at 20 g by addition of calf thymus carrier DNA. Transfectants were serum-deprived for 12 h prior to incubation for 24 h with selected concentrations of LY294002 (0, 20, and 40 M), after which relative luciferase activity was assayed. Data are expressed as percentage of the control (without LY294002 treatment).
We do not yet know the TNF-␣ target molecule(s) that mediates PI 3-kinase activation; nonetheless, since the mode of action of C 2 -ceramide is quite similar to that of TNF-␣, especially with respect to inhibition by LY294002, we postulate that enhanced production of ceramide might be involved. On the other hand, although further characterization is needed for confirmation, our evidence suggests the role of TRAF2 in the TNF-␣ signaling to SRE or JNK is minimal. For example, a dominant negative mutant of TRAF2 does not inhibit activation of either JNK or SRE in cells exposed to TNF ␣ (data not shown). This finding is in contrast to previous reports (22,24) in which TRAF2 was shown to be essential for TNF-␣-induced JNK activation in lymphocytes, suggesting the function of TRAF2 differs in Rat-2 fibroblasts and lymphocytes. In any event, our present findings make us confident that PI 3-kinase is essential for mediating the nuclear signaling cascades triggered by TNF-␣ or ceramide, which is consistent with increasing evidence indicating that PI 3-kinase is activated by environmental stresses and growth factors (37)(38)(39)(40).
We also found evidence for the role of Rac in TNF-␣ signaling to the nucleus, which is consistent with earlier findings demonstrating an essential role of Rac in the nuclear signaling by C 2 -ceramide, cytokines and environmental stresses (6,7,9). Thus, the present study shows that TNF-␣ stimulates c-fos SRE and JNK via a signaling cascade involving PI 3-kinase and Rac. Although precise determination of the mechanisms of action of PI 3-kinase and Rac will require further study, we postulate a hierarchical relationship among these proteins (TNF-␣ 3 PI 3-kinase 3 Rac), whereby Rac serves as a PI 3-kinase downstream molecule in a TNF-␣-triggered nuclear signaling pathway. Future studies elucidating the linkage between PI 3-kinase and Rac will likely be pivotal to a complete understanding of TNF-␣-evoked intracellular signaling.