Novel role for calcium-independent phospholipase A(2) in the macrophage antiviral response of inducible nitric-oxide synthase expression.

The double-stranded (ds) RNA-dependent protein kinase (PKR) is a primary regulator of antiviral responses; however, the ability of dsRNA to activate nuclear factor-kappa B (NF-kappa B) and dsRNA + interferon gamma (IFN-gamma) to stimulate inducible nitric-oxide synthase (iNOS) expression by macrophages isolated from PKR(-/-) mice suggests that signaling pathways in addition to PKR participate in antiviral activities. We have identified a novel phospholipid-signaling cascade that mediates macrophage activation by dsRNA and viral infection. Bromoenol lactone (BEL), a selective inhibitor of the calcium-independent phospholipase A(2) (iPLA(2)), prevents dsRNA- and virus-induced iNOS expression by RAW 264.7 cells and mouse macrophages. BEL does not modulate dsRNA-induced interleukin 1 expression, nor does it affect dsRNA-induced NF-kappa B activation. Protein kinase A (PKA) and the cAMP response element binding protein (CREB) are downstream targets of iPLA(2), because selective PKA inhibition prevents dsRNA-induced iNOS expression, and the inhibitory actions of BEL on dsRNA-induced iNOS expression are overcome by the direct activation of PKA. In addition, BEL inhibits dsRNA-induced CREB phosphorylation and CRE reporter activation. PKR does not participate in iPLA(2) activation or iNOS expression, because dsRNA stimulates iPLA(2) activity and dsRNA + IFN-gamma induces iNOS expression and nitric oxide production to similar levels by macrophages isolated from PKR(+/+) and PKR(-/-) mice. These findings support a PKR-independent signaling role for iPLA(2) in the antiviral response of macrophages.

Double-stranded (ds) 1 RNA, which accumulates at various stages of viral replication, plays a primary role in the activation of the antiviral response in virally infected cells (1). PKR is a 65-to 68-kDa serine-threonine kinase that is activated by binding to dsRNA, an event that results in PKR dimerization and autophosphorylation (2,3). PKR participates in the transcriptional activation of antiviral genes, the inhibition of protein translation mediated by the phosphorylation of the initiation factor eIF2␣, and apoptosis of infected cells (2,4,5). Notwithstanding, recent reports suggest that signaling pathways in addition to PKR participate in the antiviral response of infected cells. We have shown that dsRNA ϩ IFN-␥ stimulates iNOS expression and nitric oxide production to similar levels in macrophages isolated from PKR Ϫ/Ϫ and PKR ϩ/ϩ mice (5). The role of PKR in the activation of NF-B has also been disputed. NF-B, composed of heterodimer or homodimers of the p50/NF-B1 and p65/RelA subunits, is sequestered in the cytoplasm of unstimulated cells in complex with inhibitor protein B (IB) (6). Following phosphorylation, ubiquitination, and proteasome-mediated degradation of IB, NF-B is free to translocate to the nucleus and activate transcription (6,7). Early work supported a role for PKR in dsRNA-induced NF-B activation, either by directly phosphorylating IB or by stimulating IB kinase activity (8,9). Recently, dsRNA treatment and encephalomyocarditis virus (EMCV) infection have been shown to stimulate IB degradation, NF-B nuclear localization, and DNA binding, as well as NF-B promoter activation in embryonic fibroblasts isolated from PKR Ϫ/Ϫ mice (10). These findings support a role for signaling pathways in addition to PKR in the activation of antiviral activities of infected cells.
The antiviral response of macrophages includes the expression of proinflammatory cytokines such as IL-1␣ and IL-1␤ as well as iNOS (11). Nitric oxide appears to play a primary role in host defense against a viral infection. Karupiah et al. (12) first showed that IFN-␥-activated macrophage inhibition of viral replication is mediated by nitric oxide. Viral infection of rodent and human macrophages stimulates the expression of iNOS and production of nitric oxide (13,14), and iNOS-deficient mice have higher mortality rates and reduced viral clearance as compared with wild type mice following virus infection (15).
In this report, we have identified iPLA 2 and the downstream activation of PKA as a novel signaling cascade that regulates the activation of macrophages in response to a viral infection and dsRNA. iPLA 2 , classified as a group VIA PLA 2 , is an 84-kDa protein that catalyzes the hydrolysis of the sn-2 fatty acid from glycerophospholipids to produce a free fatty acid and a 2-lysophospholipid (16,17). The enzymatic activity of iPLA 2 is susceptible to inhibition by the suicide substrate bromoenol lactone (BEL) (18). Previous studies support a role for iPLA 2 in lipid remodeling, supplying lysophospholipid acceptors for in-corporation of arachidonic acid into phospholipids in macrophages (19). In pancreatic ␤-cells the inhibition of iPLA 2 by BEL has been shown to attenuate glucose-induced insulin secretion, suggesting that iPLA 2 or its products participate in signaling events that mediate insulin secretion (20,21). In this study we present evidence supporting a signaling role for iPLA 2 in the activation of iNOS expression by macrophages in response to a viral infection or dsRNA.

MATERIALS AND METHODS
Materials and Animals-PKR Ϫ/Ϫ mice (C57BL/6(J)xSV129 background) were the generous gift of Dr. Randal Kaufman (University of Michigan) and have been previously described (22). C57BL/6(J)xSV129 (PKR ϩ/ϩ ) mice were obtained from The Jackson Laboratories (Bar Harbor, ME). The Saint Louis University Institutional Review Board has approved all animal studies contained in this report. poly IC was purchased from Sigma Chemical Co. (St. Louis, MO) and was prepared as previously described (5). EMC virus was maintained as previously described (14). The PKA inhibitor H-89 was from Calbiochem (San Diego, CA), and 8-Br-cAMP and the Rp-8-bromo-cAMP were purchased from Alexis Biochemicals (San Diego, CA). 1-O-Octadecyl-2-O-methylrac-glycero-3-phosphocholine was purchased from Novabiochem. BEL and PGE 2 enzyme immunosorbant assay kits were obtained from Cayman Chemicals (Ann Arbor, MI). Rabbit anti-IB␣ antiserum was from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA), rabbit anti-CREB and anti-phospho-CREB were from Cell Signaling (Beverly, MA), and rabbit anti-iNOS antiserum was a gift from Dr. Pam Manning (Amersham Biosciences, St. Louis, MO). 3ZD monoclonal mouse anti-IL-1␤ was obtained from the Biological Resources Branch, at the NCI, National Institutes of Health. The iNOS CAT reporter construct was purchased from Oxford Biomedical Research (Oxford, MI). IL-1␣ and IL-1␤ cDNAs and the NF-B CAT, and the CRE CAT reporter constructs were gifts from Drs. Cliff Bellone and John Chrivia (Saint Louis University, St. Louis, MO) and have been previously described (23,24).
Cell Isolation, Culture, and Viral Infection-RAW 264.7 cells were removed from growth flasks by treatment with 0.05% trypsin/0.02% EDTA for 5 min at 37°C. Peritoneal exudate cells (PEC) were isolated from PKR Ϫ/Ϫ and PKR ϩ/ϩ mice by lavage as previously described (25). Following isolation, the cells were washed and then plated at 400,000 cells/400 l of cCMRL-1066 as described previously (5). The cells were allowed to adhere for 2 h prior to the initiation of experiments. RAW 264.7 cells (4 ϫ 10 5 /400 l of Dulbecco's modified Eagle's medium) were infected with the indicated concentrations of EMCV virus for 30 min at 37°C. The infection media was removed, and the cells were culture in 400 l of Dulbecco's modified Eagle's medium for 24 h.
Nitrite, IL-1, cAMP, and iPLA 2 Assays-Nitrite production was determined by Greiss assay (26), and IL-1 release was measured using the RINm5F cell bioassay (27). iPLA 2 activity was determined using a coupled whole cell assay in which arachidonic acid, produced following phospholipase activation, is metabolized by cyclooxygenase to produce a number of prostaglandins, including PGE 2 (28). PGE 2 release and cAMP accumulation were determined by enzyme immunosorbant assay according to the manufacturer's instructions. Membrane and cytosolic fractions were prepared from RAW 264.7 cells treated for 30 min with and without poly IC, and iPLA 2 activity was determined as previously described using [ 3 H]plasmenylcholine as the iPLA 2 substrate (29). BEL sensitivity was examined by addition of the indicated concentrations of this inhibitor to membrane extracts prepared from poly IC-treated RAW 264.7 cells.
CAT, iNOS, and ␤-Galactosidase Reporter Assays-RAW 264.7 cells were transiently transfected with 1 g of NF-B CAT, SS-CRE CAT, or the iNOS-CAT constructs and 1 g of the pCMV-SPORT-␤-galactosidase control plasmid using the Qiagen SuperFect reagent according to the manufacturer's instructions. After an 8-h incubation at 37°C, the experiments were initiated by addition of BEL or poly IC followed by a 24-h incubation. The cells were harvested, lysed in Reporter Lysis Buffer (Promega, Madison, WI), and CAT and ␤-galactosidase assays were performed as previously described (30). The proximal and distal CRE elements in the iNOS promoter ( Ϫ1166 TGAGCCA Ϫ1160 and Ϫ149 TGATGTA Ϫ143 ) were mutated to Ϫ1166 TGTGCCA Ϫ1160 and Ϫ149 TGTGGTA Ϫ143 using the Stratagene (La Jolla, CA) QuikChange PCR-plasmid mutagenesis kit. Transfection efficiency was ϳ70% as determined by ␤-galactosidase activity.

BEL Prevents dsRNA-and Virus-induced iNOS Expression by
Macrophages-Viral infection has been shown to stimulate membrane phospholipid hydrolysis, including the metabolism of plasmalogen substrates (31,32). Because products of iPLA 2catalyzed hydrolysis of membrane phospholipids result in the production of several biologically active metabolites that participate in signaling events and plasmalogens are believed to be a preferred substrate for iPLA 2 (21,(33)(34)(35), the role of iPLA 2 in the regulation of macrophage activation in response to a viral infection was examined using the iPLA 2 suicide substrate inhibitor BEL. In a concentration-related manner BEL inhibits iNOS mRNA accumulation, protein expression, and nitrite production by RAW 264.7 macrophages treated with the synthetic dsRNA molecule poly IC, with complete inhibition at 5 M (Fig.  1, A and B). To confirm that poly IC recapitulates host responses to viral infection, we show that a 24-h infection of RAW 264.7 macrophages with encephalomyocarditis (EMC) virus stimulates iNOS expression and nitrite production ( Fig. 1, C and D). BEL (5 M) attenuates EMCV-induced nitrite production and iNOS protein expression similar to its effects on poly IC-induced iNOS expression by RAW 264.7 cells. Consistent with a role for iPLA 2 in the stimulation of iNOS expression, arachidonyltrifluoromethyl ketone (AACOCF 3 ), a second iPLA 2 inhibitor chemically unrelated to BEL (36), also prevents poly IC-induced iNOS expression and nitrite production by RAW 264.7 cells, with maximal inhibition at 20 M (Fig. 1E).
The inhibitory actions of BEL on the antiviral response of macrophages appears to be selective for iNOS expression, because BEL does not attenuate poly IC-induced IL-1␣ or IL-1␤ mRNA accumulation by RAW 264.7 cells following a 6-h incubation ( Fig. 2A). At 5 M, BEL reduces poly IC-stimulated IL-1 release by ϳ30% following a 24-h incubation (Fig. 2B). The reduction in IL-1 release correlates with the accumulation of pro-IL-1␤ in these macrophages as determined by Western blot analysis (Fig. 2B, inset). Because IL-1␤ requires proteolytic processing by the IL-1␤ converting enzyme (ICE) for activation (37), this result suggests that iPLA 2 may participate in ICE activation, stimulating IL-1␤ proteolytic activation. This finding is consistent with previous studies showing that BEL attenuates IL-1␤ processing in human monocytes (38).
To determine if dsRNA stimulates an increase in iPLA 2 activity, RAW 264.7 cells were treated for 30 min with poly IC, the cells were isolated, and membrane and cytosolic fractions were prepared. As shown in Fig. 3A, iPLA 2 activity is present in both cytosolic and membrane fractions prepared from unstimulated RAW 264.7 cells, and ϳ75% of the total iPLA 2 activity is associated with the membrane fraction. Incubation with poly IC results in a 3-fold increase in the activity of membrane associated iPLA 2 . BEL inhibits poly IC-stimulated membrane-associated iPLA 2 activity in a concentration-related manner with complete inhibition at 5 M. The concentrationdependent inhibition of iPLA 2 activity by BEL is consistent with the concentration of BEL required to inhibit iPLA 2 activity in crude homogenates prepared from P388D1 macrophages (39), and the concentration of BEL required to inhibit poly IC-induced iNOS expression (Fig. 1). Importantly, an 18-h incubation with poly IC does not increase the level of iPLA 2 expressed by RAW 264.7 cells as determined by Western blot analysis (Fig. 3B). These findings indicate that dsRNA-induced iNOS expression is associated with an increase in iPLA 2 activ-ity, and that iPLA 2 activity is primarily associated with RAW 264.7 cell membranes. Although the mechanisms of iPLA 2 activation in response to external stimuli have remained elusive, our findings are consistent with previous reports suggesting membrane translocation as a mechanism for the activation of iPLA 2 following ischemia (40).
BEL Inhibits dsRNA ϩ IFN-␥-induced iNOS Expression by PKR-deficient Macrophages-Peritoneal macrophages (PEC) isolated from PKR ϩ/ϩ and PKR Ϫ/Ϫ mice were used to examine the role of PKR in dsRNA-induced iNOS expression. In contrast to RAW 264.7 macrophages, which express iNOS in response to a single agonist, naive mouse peritoneal macrophages (PEC) require two pro-inflammatory signals for activation (41). Alone neither poly IC nor IFN-␥ induce nitrite formation or iNOS expression following a 24-h exposure; however, in combination, poly IC and IFN-␥ induce iNOS expression and nitrite production to similar levels by PEC isolated from PKR ϩ/ϩ and PKR Ϫ/Ϫ mice (Fig. 4, A and B) (5). BEL prevents poly IC ϩ IFN-␥-induced nitrite production and iNOS expression by PEC isolated from PKR Ϫ/Ϫ and PKR ϩ/ϩ mice. To confirm that PKR is not required for dsRNA-induced iPLA 2 activation, the effects of poly IC on a second measure of iPLA 2 activity, the accumulation of prostaglandin E 2 (PGE 2 ), was examined. In this coupled assay, cyclooxygenase catalyzes the production of PGE 2 using the iPLA 2 product arachidonic acid. Consistent with a role for iPLA 2 in the regulation of iNOS expression, dsRNA stimulates the accumulation of PGE 2 to similar levels by PEC isolated from PKR Ϫ/Ϫ and PKR ϩ/ϩ mice, and BEL (5 M) inhibits PGE 2 accumulation by ϳ90% (Fig. 3C). As expected, IFN-␥ alone does not stimulate nor does it enhance poly IC-induced PGE 2 production by PEC (data not shown). In addition, the inhibitory actions of BEL on iNOS expression and iPLA 2 activity do not appear to be associated with reduced macrophage viability, because BEL does not affect the oxidation of MTT by PEC (data not shown). These findings indicate that PKR is not required for either dsRNAinduced iPLA 2 activation or for dsRNA ϩ IFN-␥-induced iNOS expression by macrophages.
Effects of BEL on dsRNA-induced NF-kB Activation in Macrophages-The effects of BEL on poly IC-induced NF-B iPLA 2 Regulation of Virus-induced iNOS Expression activation were examined because NF-B is required for macrophage expression of iNOS in response to dsRNA (11) and BEL (at ϳ100 M) has been shown to inhibit serine protease activity (42). At a concentration of 5 M, which inhibits dsRNAinduced iNOS expression and nitric oxide production by macrophages (Fig. 1), BEL does not attenuate dsRNA-induced IB␣ degradation, NF-B nuclear translocation, and DNA binding activity following a 30-min incubation or NF-B reporter activity (Fig. 5, A-C). These findings indicate that the inhibitory actions of BEL on dsRNA-induced iNOS expression are not associated with an inhibition of NF-B activation.
Role of PKA in dsRNA-induced iNOS Expression by Macrophages-Recently, Ford and co-workers (43) have shown that lysophospholipids products of iPLA 2 can directly activate PKA in a cAMP-independent manner. Using the selective PKA inhibitors (H-89 and Rp-8-Br-cAMP) the role of PKA in dsRNAinduced iNOS expression was examined. In a concentrationrelated manner H-89 inhibits poly IC-induced nitrite production (Fig. 6A) and iNOS protein expression (Fig. 6B) with maximal inhibition at 10 M. Similar results were obtained using Rp-8-Br-cAMP, a second PKA inhibitor with a different mechanism of action (data not shown). To determine if iPLA 2 mediates PKA activation in response to dsRNA, the effects of directly activating PKA using the non-hydrolyzable analog of cAMP, 8-Br-cAMP, on iNOS expression by macrophages treated with poly IC and BEL were evaluated. In a concentration-related manner, 8-Br-cAMP overcomes the in-hibitory actions of BEL on dsRNA-induced iNOS expression and nitric oxide production with maximal reconstitution at 50 -100 M (Fig. 6C). Alone 8-Br-cAMP (10 -100 M) does not induce nitrite production or iNOS expression by RAW 264.7 macrophages (data not shown). cAMP accumulation was examined in RAW 264.7 cells treated for 0 -60 min with poly IC and the phosphodiesterase inhibitor 3-isobutyl-1-methylxanthine (1 M). Although 8-Br-cAMP reconstitutes poly IC-induced iNOS expression in the presence of BEL (Fig. 6C), cAMP does not accumulate in RAW 264.7 cells following treatment with poly IC (Fig. 6D). These results are consistent with a role for an iPLA 2 lipid-derived signaling molecule in the activation of PKA and in the induction of iNOS expression by poly IC. Similar to RAW 264.7 macrophages, H-89 prevents the stimulatory actions of a 24-h incubation with poly IC ϩ IFN-␥ on nitric oxide production and iNOS expression by PEC isolated from PKR Ϫ/Ϫ and PKR ϩ/ϩ mice with maximal inhibition at 10 M (Fig. 6, D  and E). Similar results were obtained with a second PKA inhibitor Rp-8-Br-cAMP (data not shown). These findings indicate that dsRNA and dsRNA ϩ IFN-␥-induced iNOS expression by RAW 264.7 cells and primary mouse PEC requires the activation of PKA in a cAMP-independent manner.
dsRNA Stimulates CREB Activation and CREB-dependent iNOS Expression-To confirm that iPLA 2 products mediate PKA activation, the effects of poly IC and BEL on CREB phosphorylation and cAMP response element (CRE) reporter gene activity were examined. poly IC induces an increase of ϳ8-fold in CRE reporter gene activation and CREB phosphorylation (Fig. 7, A and B). BEL, at a concentration (5 M) that inhibits iPLA 2 activity and iNOS expression, prevents poly IC-stimulated CRE reporter gene activity and attenuates dsRNA-induced CREB phosphorylation. The inhibitory actions of BEL on CRE reporter activation can be overcome by the direct activa-

FIG. 4. BEL inhibits poly IC ؉ IFN-␥-induced nitrite production and iNOS expression from macrophages isolated from
PKR ؊/؊ and PKR ؉/؉ mice. The combination of poly IC and IFN-␥ stimulate nitrite production (A) and iNOS protein expression (B) to similar levels in PEC isolated from PKR Ϫ/Ϫ and PKR ϩ/ϩ mice following 24-h incubation. BEL (5 M) prevents both iNOS expression and nitrite production by PEC isolated from PKR Ϫ/Ϫ and PKR ϩ/ϩ mice. C, alone poly IC stimulates iPLA 2 activity as determined by PGE 2 release from PEC isolated from PKR Ϫ/Ϫ and PKR ϩ/ϩ mice following a 24-h incubation, and this PGE 2 release is prevented by BEL (5 M). Results for nitrite production and PGE 2 release are the average Ϯ S.E. of three independent experiments, and iNOS protein expression is representative of three independent experiments. tion of PKA using the non-hydrolyzable analog of cAMP, 8-Br-cAMP (Fig. 7B). To determine whether CREB participates in the transcriptional activation of iNOS expression by RAW 264.7 cells, the effects of EMC virus infection on iNOS reporter activity was examined. As shown in Fig. 7C, EMC virus infection stimulates ϳ15-fold increase in iNOS reporter activity. Mutations in either the proximal or distal CRE consensus elements in the iNOS promoter results in a significant attenuate (ϳ40 -60%) of iNOS reporter activity. Mutation of both the proximal and distal CRE consensus elements in the iNOS promoter results in nearly complete inhibition of EMC virusinduced iNOS reporter activity. Taken together, these findings suggest that dsRNA stimulates PKA activation in an iPLA 2dependent manner and that CREB is one downstream target of PKA that participates in the transcriptional activation of the iNOS promoter in response to virus infection.
Reconstitution of BEL-sensitive iNOS Expression by Lysophospholipid-Previous studies have shown that lysoplasmenylcholine can directly activate PKA in a cAMP-independent manner (43,44), and in this study we show that dsRNA stimulates PKA activation (as evidenced by CREB phosphorylation, Fig. 7A) under conditions in which cAMP levels remain unchanged (Fig. 6D). Also, we show that iNOS expression in response to dsRNA is dependent on both iPLA 2 and PKA activation ( Figs. 1 and 6). These findings suggest that lipid-derived products of iPLA 2 may participate in the cellular signal transduction mechanisms by which dsRNA or viral infection stimulates iNOS expression by macrophages. Using a synthetic analog of the iPLA 2 product lysoplasmenylcholine, methyllysoplasmenylcholine (methyl-LPC, 1-O-octadecyl-2-O-methylrac-glycero-3-phosphocholine), it is possible to partially over-come the inhibitory actions of BEL on dsRNA-stimulated iNOS expression by RAW 264.7 cells. As shown in Fig. 8, methyl-LPC, at concentrations of 0.1 and 1 M, overcomes the inhibitory actions of BEL on dsRNA-induced iNOS expression and nitrite production by ϳ50%. Alone methyl-LPC does not stimulate iNOS expression, nor does it enhance dsRNAinduced iNOS expression by RAW 264.7 cells (data not shown). In addition, at concentrations greater than 1 M, methyl-LPC stimulates RAW 264.7 cell death (determined by phase contrast microscopy). The later finding is consistent with the induction of apoptosis of transformed cells by treatment with methyl-LPC at concentrations of 5-10 M (45, 46). Importantly, these findings provide direct evidence in support of a role for iPLA 2 and iPLA 2 -derived products in the regulation of iNOS expression in response to dsRNA. DISCUSSION In this study we provide biochemical evidence to support a novel role for iPLA 2 in the selective regulation of antiviral activities of macrophages stimulated by dsRNA or viral infection ( Fig. 9 for schematic diagram). We show that selective iPLA 2 inhibition using BEL or AACOCF 3 results in an inhibition of dsRNA-and EMC virus-induced iNOS expression. The inhibitory actions of BEL on the antiviral response appear to be selective for iNOS, because this iPLA 2 inhibitor does not mod- ulate IL-1 expression by macrophages in response to dsRNA. BEL does attenuate the amount of IL-1 released by macrophages, and this inhibition is associated with the accumulation of pro-IL-1␤ in macrophages. This finding suggests that BEL may inhibit ICE or that products of iPLA 2 participate in the regulation of ICE activity. The role of iPLA 2 in macrophage activation does not require the presence of PKR, because dsRNA ϩ IFN-␥ stimulates iNOS expression to similar levels in macrophages isolated from PKR Ϫ/Ϫ and PKR ϩ/ϩ mice. Also, PKR does not appear to be required for iPLA 2 activation, because dsRNA stimulates iPLA 2 activity to similar levels in macrophages isolated from PKR Ϫ/Ϫ and PKR ϩ/ϩ mice. These findings suggest that PKR does not participate in either the activation of iPLA 2 or the induction of iNOS expression by macrophages in response to dsRNA.
The intracellular signaling mechanism by which iPLA 2 regulates iNOS expression in response to dsRNA is associated with the activation of PKA. Inhibition of PKA prevents dsRNA and dsRNA ϩ IFN-␥-induced iNOS expression by macrophages, and the inhibitory effects of BEL on dsRNA-induced iNOS expression can be overcome by the direct activation of PKA using non-hydrolyzable cAMP analogs. Although cAMPderived activators of PKA overcome the inhibitory actions of BEL on dsRNA-induced iNOS expression, dsRNA does not stimulate cAMP accumulation in RAW 264.7 cells. These findings suggest that PKA is activated in response to dsRNA in a cAMP-independent manner. Consistent with our findings, in vitro studies have shown that lipid-derived products of iPLA 2 , specifically lysoplasmenylcholine, can activate PKA in a cAMPindependent fashion (43), and that CREB phosphorylation elicited by 5 min of global ischemia of adult rat hearts occurs in the absence of increases in myocardial cAMP content and is attenuated by BEL (35). Although it is possible that local subcellular changes in cAMP levels sufficient to activate PKA occur in response to dsRNA treatment, we favor the hypothesis that lipid-derived products of iPLA 2 are responsible for PKA activation in response to dsRNA in macrophages.
One downstream target of PKA appears to be CREB. Treatment of macrophages with dsRNA results in CREB phosphorylation and CRE reporter activation, effects that are prevented by BEL. The murine iNOS promoter contains two CRE elements, one proximal and one distal to the initiation codon (47). Importantly, the stimulatory actions of EMC virus infection on iNOS reporter activity are attenuated by mutation of either the proximal or distal CRE consensus elements, and mutations of both CRE elements results in a nearly complete inhibition of EMC virus-induced iNOS reporter activity. The role of CREB in the regulation of iNOS expression appears to be stimulus-and cell type-specific. In PEC, iNOS transcription has been reported to be activated by agents that elevate cAMP, whereas in splenic macrophages elevated cAMP levels appear to synergize with tumor necrosis factor and IFN-␥-induced iNOS expression yet inhibit LPS-induced iNOS transcription (48,49). Our findings suggest that transcriptional activation of iNOS in macrophages in response to a viral infection is dependent on CREB.
Although BEL is a selective inhibitor of iPLA 2 , it has been shown to inhibit enzymatic activities in addition to iPLA 2 . At high concentrations, 10-to 20-fold higher than those used in our study (50 -100 M), BEL inhibits serine protease activity (42). This non-selective action does not appear to participate in the inhibitory effects of BEL on dsRNA-induced iNOS expression, because BEL, at a concentration that inhibits iNOS expression (5 M), does not affect the ability of dsRNA to activate NF-B (an event that requires IB degradation by the serineprotease inhibitor-sensitive proteasome). BEL also has been shown to inhibit phosphatidic acid phosphohydrolase (PAP, at ϳ10 -20 M) (50); however, it is not likely that PAP participates in dsRNA-induced iNOS expression, because the PAP-selective inhibitor propranolol (150 M) does not attenuate dsRNAinduced iNOS expression or nitrite production by RAW 264.7 cells (data not shown). Additional evidence in support of a role for iPLA 2 in the regulation of iNOS expression includes the ability of methyl-LPC, an analog of the iPLA 2 product lysoplasmenylcholine, to partially overcome the inhibitory actions of BEL on dsRNA-induced iNOS expression by macrophages (Fig. 8).
The mechanism by which dsRNA or viral infection stimulates iPLA 2 activation has yet to be determined. iPLA 2 contains consensus sites for a number of serine kinases, however, phosphorylation of iPLA 2 or the regulation of iPLA 2 activity by phosphorylation has yet to be demonstrated (21). Recently, the toll-like receptor 3 (TLR3) has been identified as a receptor for dsRNA, and this receptor appears to mediate the activation of NF-B in response to dsRNA (51). Importantly, NF-B activation is required for dsRNA-induced iNOS expression by macrophages (11), an effect that is likely mediated by the TLR3 receptor, and we have recently shown that RAW 264.7 cells 2 and primary mouse macrophages express TLR3 (51). It is possible that TLR3 activation mediates the stimulatory actions of dsRNA on iPLA 2 activity; however, the regulation of iPLA 2 activity by phosphorylation has yet to be demonstrated (21), and there does not appear to be cross-talk between the iPLA 2 and NF-〉 pathways (TLR3) in response to dsRNA (note that BEL does not inhibit dsRNA-induced IB degradation, NF-B nuclear localization, or NF-B reporter activation) (Fig. 5).
In summary, this report has identified a novel signaling role for iPLA 2 in the antiviral response of iNOS expression by macrophages. The role of iPLA 2 in the activation of antiviral activities of macrophages does not require the presence of functional PKR. The mechanism by which iPLA 2 participates in the activation of iNOS expression is associated with PKA activation and the downstream activation of CREB (Fig. 9). Previous studies have shown that NF-B is also required for iNOS expression in response to dsRNA (5,11), and in this report we show that iPLA 2 does not appear to participate in the regulation of NF-B activation in response to dsRNA. These findings suggest, in addition to NF-B, the activation of iPLA 2 and downstream activation of CREB are required for virus and dsRNA-induced iNOS expression by macrophages. These findings highlight the complexity of the antiviral response and suggest that multiple PKR-dependent and -independent pathways participate in the regulation of antiviral activities in infected cells.