Introductory Review Cellscience Reviews Vol 2 No.2 ISSN 1742-8130

The Cellular Response to Virus Infection: The Role of PKR and Novel PKR-independent Pathways

Jason M. Moran, Robert Mark L. Buller & John A. Corbett

Abstract

Virus infection induces numerous characteristic responses in cells that prevent productive infection, viral replication and virion production. The expression of proinflammatory and antiviral genes by infected cells contributes to the inhibition of virus replication, although the mechanisms of transcriptional regulation are incompletely understood. Cellular antiviral responses are believed to be initiated primarily through the recognition of accumulating viral dsRNA, a consequence of the replication of both RNA and DNA viruses. The dsRNA-dependent protein kinase (PKR) is one dsRNA-sensitive protein thought to mediate the activation of antiviral activities based on extensive genetic studies. Notwithstanding, the activation of dsRNA-induced gene expression in the absence of functional PKR suggests additional PKR-independent pathways participate in the antiviral response. The recent identification and characterization of PKR-independent pathways that regulate gene expression in virus-infected macrophages will be reviewed. Specific attention will be placed on the regulation of iNOS, IL-1 and COX-2. In addition, we will briefly discuss other proteins (TLR3 and RIG-I) capable of interacting with dsRNA as potential regulators of PKR-independent gene expression in virus-infected cells.

Introduction

The capacity of cells to not only recognize an invading virus but to mount appropriate protective responses is critical to the health and survival of organisms. Cellular responses to virus infection have been the subject of intensive investigation for several decades and the well recognized paradigm that has emerged from these studies is that antiviral activities in cells are believed to be activated, in part, by the intracellular accumulation of double-stranded (ds) RNA, a consequence of the replication of both RNA and DNA viruses.

An extensive body of literature has shown that synthetic as well as naturally occurring dsRNAs from a variety of sources can induce the expression and release of interferons (IFNs) and confer protection from subsequent virus challenge. The IFN response is critical to host defense against viral infection as evidenced by reports that type I IFN expression limits viral replication and enhances survival of virus-infected mice (Harle et al., 2001; Noisakran et al., 1999), administration of IFN neutralizing antibodies following virus infection increases viral titers 1000-fold (Hendricks et al., 1991), and mice deficient in the common IFN receptor subunit are highly sensitive to virus infection (Muller et al., 1994).

Initial studies of the regulation of dsRNA-induced cellular responses, in particular IFN production, indicated a requirement for the dsRNA-dependent protein kinase (PKR) and thus cultivated the hypothesis that host responses to dsRNA were mediated by PKR. Notwithstanding, recent experimental evidence from mouse embryonic fibroblasts (MEFs) and macrophages is consistent with the presence of additional cellular antiviral responses independent of PKR and novel dsRNA-induced signaling pathways that participate in gene expression are beginning to be identified.

Macrophages represent an integral part of innate immunity based on their ability to generate robust inflammatory and antiviral responses. The production and release of cytokines and inflammatory mediators such as interleukin-1 (IL-1) and cytotoxic molecules such as nitric oxide contribute to the suppression of virus replication. This is particularly evident given the observation that the IFN-mediated inhibition of virus replication is dependent upon nitric oxide (Karupiah et al., 1993). Despite the implication from previous historical findings that PKR mediates cellular responses to dsRNA, the transcriptional activation of several genes studied in macrophages does not support a PKR-dependent mechanism of regulation, thus suggesting that additional pathways contribute to antiviral responses in cells that function independently of PKR. As the purpose of this manuscript is to review the current knowledge of the PKR-independent regulation of proinflammatory and antiviral gene expression in macrophages activated by dsRNA or virus infection, a brief perspective of the historical findings describing cellular responses to dsRNA may be appropriate. Therefore, we begin with a discussion of the initial experimental observations that cultivated the notion that dsRNA, as an active component of virus infection, stimulates antiviral responses in cells.

Cellular Responses to dsRNA: A Historical Perspective

For more than four decades work has diligently continued to unravel the mechanisms by which viral infection stimulates protective host responses. It is now generally regarded that the intracellular accumulation of viral dsRNA, a consequence of the replication of both RNA and DNA viruses, induces cellular antiviral responses through the activation of proteins that are exquisitely sensitive to dsRNA. However, there was a time when our understanding of the effects of dsRNA on cells was not always so complete.

The inherent ability of dsRNA to induce antiviral responses in cells was perhaps first recognized following the observation that a substance, termed helenine, derived from the mold Penicillium funiculosum conferred resistance in mice to virus infection (Shope, 1966). Although the active component of helenine responsible for its antiviral properties remained a matter of conjecture for several years, subsequent fractionation and purification revealed that helenine-derived dsRNA was a potent inducer of IFN release and host resistance to virus infection in vivo (Lampson et al., 1967; Lewis, 1960). Hilleman and colleagues further demonstrated that in addition to naturally occurring dsRNA, synthetic dsRNA injected intravenously into rabbits in low microgram quantities was also capable of stimulating the release of IFN. This activity appeared not only dependent upon ribonucleotide content but also upon multi-stranded character. Double-stranded polynucleotides derived from the interaction of polyinosinic acid and polycytidylic acid (designated as I:C) induced IFN and afforded protection against virus infection in vivo and in vitro, while single-stranded homopolymers of inosine or cytidine as well as other single-stranded and double-stranded ribonucleotide polymers (comprised of adenine or uracil) were unable to induce IFN responses (Field et al., 1967). To our knowledge, these studies were the first to suggest that synthetic dsRNA possessed the capacity to induce protective antiviral responses, results that were subsequently confirmed by several groups (Finkelstein et al., 1968; Vilcek et al., 1968; Youngner and Hallum, 1968). As a testament to the significance of this observation, polyinosinic:polycytidylic acid (polyIC) has been used for nearly 40 years as a viral dsRNA mimetic to study general cellular responses to viral infection. The gravity of these experimental findings, however, was not fully appreciated until the concurrent and independent recognition by a number of laboratories that virus-specific dsRNA could be isolated from cells infected with both RNA and DNA viruses (Baltimore et al., 1964; Burdon et al., 1964; Colby and Duesberg, 1969; Gomatos and Tamm, 1963; Kaerner and Hoffmann-Berling, 1964; Montagnier and Sanders, 1963; Shipp and Haselkorn, 1964; Tytell et al., 1967).

The formation of dsRNA within virus infected cells, as well as the ability of synthetic dsRNA to induce IFN release and afford protection from virus infection in vivo, led to the speculation that activation of cellular responses was due to the formation and intracellular accumulation of dsRNA during viral infection. However, it had yet to be demonstrated experimentally that virus-specific dsRNA possessed the capacity to induce similar antiviral responses as had previously been documented for polyIC. Non-infectious, non-replicative forms of single-stranded ribonucleic acids from Newcastle disease virus, influenza A virus and tobacco mosaic virus failed to stimulate cellular IFN release, again confirming a requirement for the double-stranded nature of RNA for the activation of antiviral responses (Field et al., 1967; Lampson et al., 1967). Direct evidence to suggest that viral dsRNA could activate cellular antiviral responses was eventually obtained when genomic dsRNA isolated from virions of type 3 reovirus actively induced IFN release in vivo and rendered cells resistant to virus infection in vitro (Tytell et al., 1967). The culmination of these studies provided a compelling correlation between the accumulation of viral dsRNA in infected cells and the activation of protective host responses, yet the molecular mechanism by which dsRNA activated cells remained enigmatic.

In the years following these observations, extensive effort has been invested to identify cellular proteins sensitive to dsRNA and to elucidate the mechanisms of host responsiveness to dsRNA or viral infection. Among the proteins now known to bind dsRNA, perhaps the most comprehensively studied is the protein kinase PKR. Below we discuss the experimental findings that have suggested a critical and primary role for PKR in the regulation of various cellular responses to virus infection.

PKR and the Antiviral Response

The double-stranded RNA-dependent protein kinase (PKR) is a 65-68 kDa serine/threonine protein kinase ubiquitously expressed in nearly all mammalian cells that participates in host responses to viral infection. While cells express modest quantities of latent enzyme, PKR expression is dramatically induced in many cell types by type I IFNs (IFNα and IFNβ) (Samuel et al., 1997; Stark et al., 1998). The amino terminal portion of PKR is characterized by two tandem copies of a dsRNA binding motif that are both necessary and sufficient for interaction with dsRNA in a sequence-independent manner (Feng et al., 1992; McCormack et al., 1994; Patel and Sen, 1992). Binding to dsRNA promotes the dimerization and trans-autophosphorylation of PKR, which renders the catalytic kinase domain active (Galabru and Hovanessian, 1987; Galabru et al., 1989; Ortega et al., 1996; Thomis and Samuel, 1993). As the detailed mechanism of PKR activation is beyond the scope of this review and has been eloquently described elsewhere, we kindly refer the reader to the following reviews (Kaufman, 2000; Williams, 1999).

Abrogation of PKR function, either by the overexpression of dominant-negative mutants, introduction of antisense oligonucleotides or through gene deletion have enabled the study of PKR function in various dsRNA-induced cellular responses. The biological significance of PKR in host protection from virus infection is readily apparent from the observations that constitutive expression of PKR renders cultured cells resistant to EMCV or vaccinia virus infection (Lee and Esteban, 1993; Meurs et al., 1992) and that multiple viruses (adenovirus, Epstein-Barr virus, human immunodeficiency virus, vaccinia, influenza, reovirus, hepatitis C, poliovirus) have developed strategies to inhibit PKR activity (Gale and Katze, 1998). Studies over the past two decades have attempted to elucidate the mechanisms by which PKR manifests cellular antiviral responses and the culmination of these experimental findings has subsequently revealed a role for PKR in the inhibition of translation, activation of gene transcription and induction of apoptosis (Figure 1).

Figure 1: The activation and function of PKR. The intracellular accumulation of RNA with secondary structure promotes the dimerization, transphosphorylation and activation of PKR. Activated PKR has been implicated in the regulation 0f the cellular antiviral response by several discrete pathways. PKR can prevent protein synthesis by phosphorylation of the translation initiation factor eIF2α, upregulate gene transcription by activating transcription factors such as NF-κB or participate in cellular apoptotic responses.

PKR & Translation: In 1971, Hunt and Ehrenfeld demonstrated that dsRNA extracted from poliovirus-infected HeLa cells inhibited protein synthesis in rabbit reticulocytes (Ehrenfeld and Hunt, 1971; Hunt and Ehrenfeld, 1971). This inhibition of protein synthesis was confirmed for a variety of synthetic and naturally occurring dsRNAs and in cell-free translation systems prepared from virus-infected cells (Hunter et al., 1975; Kerr, 1974). These observations, along with subsequent work, led to the intriguing consideration that one cellular response to dsRNA or virus infection is the translational repression of cellular mRNAs. A functional consequence of dsRNA on the cellular translational machinery seemed a logical host defense against infection given that virus replication is dependent upon cell-mediated synthesis of virus proteins. It is now well established that the mechanism by which dsRNA attenuates translation involves the direct phosphorylation of the eukaryotic translation initiation factor eIF2α on serine-51 by PKR (Hovanessian, 1993; Kaufman, 2000; Levin and London, 1978). A preliminary but essential step in the synthesis of polypeptides is the recruitment of ribosomes and the initiator methionyl-tRNA to mRNAs, a process mediated, in part, by eIF2α (Pain, 1996). Phosphorylation of eIF2α effectively reduces the levels of functional eIF2 by increasing the stability of its interaction with the guanine nucleotide exchange factor eIF2B, thereby limiting translation initiation on all mRNAs within the cell (Safer, 1983). This rapid and reversible phosphorylation of eIF2α in response to dsRNA, therefore, provides a fundamental mechanism by which PKR mediates a protective response to virus infection.

PKR & Gene Expression: An early indication that PKR participates in the activation of dsRNA-sensitive genes emerged from indirect evidence that the PKR inhibitor 2-aminopurine could inhibit the expression of the IFN-β gene in MG63 osteosarcoma cells treated with dsRNA (Zinn et al., 1988). Although the mechanism by which PKR contributes to dsRNA-induced gene expression was unclear, independent examination of the regulation and responsiveness of the IFN-β promoter to dsRNA was consistent with an essential role for nuclear factor-κB (NF-κB) (Lenardo et al., 1989; Visvanathan and Goodbourn, 1989). A transcription factor required for the expression of many genes involved in immunity, NF-κB is a dimer comprised of p50 and p65 subunits and is sequestered in the cytoplasm of unstimulated cells by its association with the inhibitory protein IκB (Baeuerle and Henkel, 1994; Haskill et al., 1991). Stimuli that activate NF-κB induce the phosphorylation, polyubiquitination and subsequent proteasome-mediated degradation of IκB, which allows the nuclear translocation and DNA binding of NF-κB to responsive promoters (Liou and Baltimore, 1993). Given that PKR function appeared to be associated with the expression of dsRNA-induced genes, the hypothesis arose that PKR mediates gene expression in response to dsRNA via the activation of transcription factors such as NF-κB. Direct evidence to suggest that PKR could modulate the activity of NF-κB was obtained following the recognition that 1) PKR could phosphorylate IκB and induce NF-κB DNA binding activity in vitro using purified recombinant proteins (Kumar et al., 1994) and 2) dsRNA-induced IκB degradation and NF-κB DNA binding was prevented in mouse embryonic fibroblasts (MEFs) derived from PKR-null mice (Yang et al., 1995; Zamanian-Daryoush et al., 2000). The mechanism by which PKR participates in the activation of NF-κB has been proposed to be either by the direct phosphorylation of IκB (Kumar et al., 1994) or indirectly through the stimulation of IκB kinase (Chu et al., 1999). The activation of NF-κB in cells exposed to dsRNA or infected with a virus provides a mechanism by which PKR participates in the transcriptional activation of genes critical for cellular innate immunity.

As experimental data consistent with a requirement for PKR in dsRNA- or virus-induced inhibition of translation and activation of transcription factors began to accumulate in the late 1990’s, it fostered the hypothesis that PKR may be indispensable for cellular antiviral responses to dsRNA. Perhaps these ideas were manifest at the time because our knowledge of additional proteins activated by dsRNA was somewhat limited or perhaps it was due to the focused study of a limited number of cell culture models. Notwithstanding, more recent published reports have identified novel dsRNA-induced antiviral responses that function independent of PKR, suggesting that additional innate immune responses sensitive to accumulating viral dsRNA are present in cells.

The Emergence of PKR-independent Signaling Pathways

The working model that emerged from several independent investigations of the mechanism of NF-κB activation in response to dsRNA or virus infection proposed that PKR was required for NF-κB DNA binding activity and NF-κB-dependent gene expression. Published reports demonstrated that PKR is essential for dsRNA-induced NF-κB activation based on the observations that overexpression of dominant-negative PKR mutants prevent poly IC-induced NF-κB reporter activity (Kumar et al., 1994) and poly IC- or virus-induced NF-κB DNA binding and NF-κB-dependent gene expression is severely reduced in PKR^-/- MEFs (Chu et al., 1999; Kumar et al., 1997). While these findings were once believed to clearly demonstrate a requirement for PKR in dsRNA- or virus-induced activation of NF-?B, the reproducibility of these results by other groups as well as the recognition that these findings do not universally extend to other cell types beyond MEFs has recently caused the re-evaluation of the role of PKR in NF-κB activation in virus-infected cells.

PKR-independent NF-κB Activation: Perhaps the first inclination that dsRNA could activate NF-κB in a PKR-independent manner came from the demonstration by Yang et al. (Yang et al., 1995) that the deficiency in dsRNA-induced NF-κB activation in PKR^-/- MEFs is not absolute and can be overcome if the cells are first primed with type I IFNs. These results suggest that additional signaling pathways independent of PKR can compensate for or participate in cellular responses to dsRNA. Subsequent studies by Magun and coworkers (Iordanov et al., 2001) provided conflicting observations as those previously documented in MEFs and suggest that poly IC-induced IκBα degradation, NF-κB nuclear translocation, NF-κB DNA binding activity and NF-κB dependent gene expression occur in both PKR^+/+ and PKR^-/- MEFs in an indistinguishable manner. Although the dramatic differences in experimental observations between groups are unclear and remain to be reconciled, functional characterization of the PKR gene products in the two PKR^-/- mouse models suggests that both are incomplete knock-outs and that the expression of differing truncated PKR variants may account, in part, for the signaling differences reported (Baltzis et al., 2002). Notwithstanding, the question that arises from these studies is whether PKR is required for the dsRNA-induced activation of NF-κB in other cell types or whether these inconsistencies in PKR-dependence are simply peculiar to MEFs. In vivo evidence does not support a role for PKR in dsRNA-induced NF-κB activation or gene expression as has been suggested in MEFs, since Newcastle disease virus (NDV) infection or peritoneal injection of poly IC results in similar levels of type I IFN expression and NF-κB activation in spleen, liver and lung of PKR^+/+ and PKR^-/- mice (Yang et al., 1995). Since PKR may not mediate innate immunity to viruses at the level of the organism, the physiological relevance of PKR-mediated NF-κB activation in MEFs is at present uncertain.

Given that PKR does not appear to be required for NF-κB activation or antiviral gene expression in response to dsRNA or virus infection in vivo, the contribution of PKR to dsRNA-induced responses in cell types that possess a decisive role in innate immunity are of particular interest. Experimental observations in macrophages are consistent with a PKR-independent mechanism of NF-κB activation in response to virus infection as poly IC or encephalomyocarditis virus (EMCV) induce the complete degradation of IκB in RAW264.7 cells despite the overexpression of dominant negative (dn) mutants of PKR (Maggi et al., 2000; Moran et al., 2005b). Furthermore, poly IC induces comparable levels of NF-κB nuclear localization and DNA binding activity in primary macrophages isolated from either PKR^+/+ or PKR^-/- mice (Maggi et al., 2000), suggesting that PKR is not required for NF-κB activation in macrophages in response to dsRNA or virus infection. Importantly, these results in macrophages are consistent with in vivo data suggesting a PKR-independent mechanism of NF-κB activation by dsRNA and further supports the hypothesis that additional dsRNA-sensitive signaling pathways participate in the regulation of NF-κB activity and antiviral gene expression in virus-infected cells.

PKR & MAPK Activation: The PKR-independent activation of NF-κB in response to dsRNA was perhaps the first indication that pathways in addition to PKR participate in innate immunity to virus infection. Although the role of PKR in dsRNA-induced NF-κB activation remains controversial, multiple signaling pathways have subsequently been identified in macrophages that are activated by dsRNA or virus infection in a manner independent of PKR. The mitogen-activated protein kinases (MAPK) represent a family of stress-responsive kinases whose enzymatic activity is stimulated in cells by a number of conditions including osmotic stress (Galcheva-Gargova et al., 1994), UV radiation (Iordanov et al., 1998), RNA damaging agents (Iordanov et al., 1997), cytokines (Raingeaud et al., 1995) and endotoxin (Hambleton et al., 1996). In addition, the phosphorylation and activation of MAPK have been observed in various cell types either exposed to dsRNA or following virus infection (Alexopoulou et al., 2001; Goh et al., 2000; Iordanov et al., 2000; Li et al., 2004; Meusel and Imani, 2003; Tamanini et al., 2003). Recently, the treatment of RAW264.7 cells or primary macrophages with poly IC or EMCV was documented to result in the rapid and robust activation of ERK, JNK and p38 (Alexopoulou et al., 2001; Maggi et al., 2003; Moran et al., 2005b), suggesting that MAPK signaling pathways participate in the macrophage response to virus infection. Direct evidence to support a PKR-independent mechanism of MAPK activation by dsRNA or virus infection emerged from the recognition that poly IC or EMCV induces ERK, JNK and p38 phosphorylation in macrophages despite the overexpression of dnPKR mutants or the genetic absence of PKR (Maggi et al., 2003; Moran et al., 2005b; S.A. Steer and J.A. Corbett, unpublished observation). Additional evidence to support a PKR-independent mechanism of MAPK activation in virus-infected macrophages has been obtained following the observation that ERK, JNK and p38 phosphorylation does not temporally correlate with PKR activation in EMCV-infected macrophages, but rather precedes it by greater than 6 hours (Moran et al., 2005b). Taken together, these data suggest that dsRNA or virus infection induce MAPK signaling cascades in macrophages in a manner that does not require PKR.

PKR & eIF2α Phosphorylation: Perhaps even more compelling than the PKR-independent activation of NF-κB or MAPK in macrophages by dsRNA or virus infection, recent experimental evidence is consistent with a PKR-independent mechanism of eIF2α phosphorylation in virus-infected macrophages. Based on extensive time-course studies, EMCV infection of macrophages appears to induce a biphasic phosphorylation of eIF2α that is characterized by an early induction at 30 minutes post-infection and a late induction at 18-24 hours post-infection. Interestingly, while the latter phase of eIF2α phosphorylation correlates temporally with PKR activation, the immediate early macrophage response of eIF2α phosphorylation is more robust and is not associated with detectable levels of PKR activation. The role of PKR in EMCV-induced eIF2α phosphorylation was confirmed using macrophages isolated from PKR^-/- mice and although EMCV-induced eIF2α phosphorylation at 18-24 hours post-infection is prevented in PKR^-/- macrophages, phosphorylation of eIF2α following a 30 minute exposure to EMCV occurs to similar levels in both PKR^+/+ and PKR^-/- macrophages (Moran et al., 2005b). At present, the functional significance of this early eIF2α phosphorylation is unclear and it remains to be determined whether other kinases can compensate for the loss of PKR in PKR^-/- macrophages. However, phosphorylation of eIF2α has been observed in MEFs isolated from mice with a targeted disruption of the C-terminal kinase domain of PKR (PKR^-/-) and this result is consistent with the hypothesis that eIF2α phosphorylation in the absence of PKR enzymatic function must result from a previously unappreciated kinase activity (Abraham et al., 1999). With the exception of PKR, to our knowledge none of the other eIF2α kinases (GCN2, PERK and HRI) are known to be directly or immediately activated by dsRNA or virus infection, suggesting an as-of-yet unidentified kinase or novel signaling pathway is responsible for eIF2α phosphorylation early during EMCV infection of macrophages.

As we have provided a brief overview of the experimental observations that have provided strong support that cellular antiviral responses to dsRNA or virus infection can occur by mechanisms independent of PKR, we will now discuss the identification and characterization of PKR-independent signaling pathways in macrophages that participate in the regulation of dsRNA- and virus-induced proinflammatory and antiviral gene expression.

PKR-independent Antiviral Gene Expression: The Transcriptional Regulation of iNOS

Viral infection of both rodent and human macrophages or exposure of macrophages to dsRNA results in the expression of the inducible isoform of nitric oxide synthase (iNOS) (Bukrinsky et al., 1995; Heitmeier et al., 1998; Hirasawa et al., 1999), suggesting that iNOS may participate in the macrophage antiviral response. iNOS catalyzes the five electron oxidation of one of the chemically equivalent guanidino nitrogens of L-arginine to generate L-citrulline and liberate the free radical nitric oxide (Nathan, 1995). Direct evidence to indicate the importance of iNOS expression and nitric oxide production in the antiviral response has been obtained from the observations that IFN-γ-induced inhibition of ectromelia, vaccinia and herpesvirus replication is mediated by the macrophage production of nitric oxide (Karupiah et al., 1993). In addition, the abrogation of iNOS activity, either by administration of iNOS inhibitors or due to genetic absence (iNOS^-/- mice), increases viral titers and enhances mortality of virus-infected mice (Flodstrom et al., 2001; Lowenstein et al., 1996).

The role of NF-κB: Although the mechanism by which viral infection stimulates the expression of iNOS by macrophages is incompletely defined, NF-κB appears to be required for the transcriptional activation of iNOS in response to dsRNA based on experimental data indicating that poly IC-induced iNOS expression and nitrite production by RAW264.7 cells are sensitive to pharmacological inhibitors of NF-κB activation, including the antioxidant pyrrolidinedithiocarbamate (PDTC) or the proteasome inhibitor MG-132 (Heitmeier et al., 1998, Figure 2). The activation of NF-κB, while necessary for dsRNA-induced iNOS expression, alone it is not sufficient since stimuli that induce IκBα degradation and NF-κB activation in macrophages do not universally activate iNOS expression. Thus the selectivity of gene expression in macrophages activated by dsRNA or virus infection is thought to be afforded by the activation of multiple independent signaling pathways that converge at the promoters of genes whose expression is critical to macrophage proinflammatory and antiviral responses.

Figure 2: NF-κB is required for dsRNA- or virus-induced iNOS expression and nitric oxide (NO) production by macrophages. However, additional PKR-independent pathways also participate in the transcriptional regulation of iNOS.

The role of PKR: In an attempt to identify additional signaling pathways that participate in the regulation of iNOS expression by virus-infected macrophages, a critical contribution of PKR proved to be an attractive hypothesis given the experimental observations that iNOS expression is induced in macrophages by dsRNA and that NF-κB, a potential downstream target of PKR, is required for the transcriptional activation of iNOS (Heitmeier et al., 1998). iNOS expression in RAW264.7 cells treated with poly IC is inhibited in cells transiently transfected with dominant-negative mutants of PKR, suggesting that PKR participates in the regulation of iNOS expression by macrophages in response to dsRNA (Maggi et al., 2000). However, this requirement for PKR is not absolute as the inhibitory effects of dnPKR expression can be overcome by the addition of IFN-γ (Maggi et al., 2000). Interestingly, these results are reminiscent of those by Yang et al. (Yang et al., 1995) in which the PKR dependence of IFNα/β expression and NF-κB activation in poly IC-treated MEFs could be overcome by priming cells with IFNs. Although the mechanism by which IFN-γ overcomes the PKR dependence of iNOS expression is unclear, these results can be interpreted to suggest that IFN-γ activates signaling pathways that supplement or compensate for those inhibited by dnPKR, thereby restoring the transcriptional activation of iNOS. Further experimental support to suggest a PKR-independent regulation of iNOS expression was obtained from macrophages isolated from PKR^-/- mice. Neither poly IC nor IFN-γ alone induces iNOS expression or nitrite production by primary macrophages, which is consistent with a requirement for two proinflammatory stimuli for the induction of iNOS expression by mouse macrophages. However, poly IC + IFN-γ induces comparable levels of iNOS expression and nitrite production in PKR^+/+ and PKR^-/- macrophages (Maggi et al., 2000). These data provided some of the first evidence to suggest that the expression of a dsRNA-sensitive antiviral gene in macrophages was not entirely dependent upon PKR.

The question that emerged and remained unanswered from the conclusion of these studies was whether these results were peculiar to dsRNA or if macrophage expression of iNOS in response to virus infection was also independent of PKR. Confirmation that macrophage responses to virus infection were faithfully recapitulated by poly IC came from the examination of the PKR-dependence of iNOS expression in macrophages infected with a non-enveloped ssRNA virus of the Picornaviridae family known to accumulate viral dsRNA in infected cells. Consistent with the PKR-independent activation of iNOS expression by polyIC, EMCV-induced iNOS expression is unaffected by the presence of dnPKR (Moran et al., 2005b). Further evidence to support a PKR-independent regulation of iNOS expression was obtained from the observation that EMCV-induced iNOS mRNA accumulation precedes PKR activation by several hours (Moran et al., 2005b). These findings provide strong evidence that poly IC faithfully recapitulates macrophage responses to virus infection and suggests that both dsRNA and virus infection can activate antiviral gene expression in macrophages in a PKR-independent manner.

The role of iPLA2: Since PKR did not appear to participate in the regulation of iNOS, the identity of virus-induced signaling pathways that activate iNOS expression were of considerable interest. Virus infection has been associated with alterations in cellular membrane phospholipid metabolism (Horrocks et al., 1978; Malewicz et al., 1981) and specifically, increases in arachidonic acid release have been documented in cells infected with poxviruses (Palumbo et al., 1993). As alterations in phospholipid metabolism correlate with the infection and replication of several viruses and products of membrane catabolism contribute to inflammatory responses, one intriguing hypothesis is that phospholipases such as the calcium-independent phospholipase A2 (iPLA2) participate in the regulation of cellular defenses against viral infection. Using the iPLA2-selective irreversible mechanism-based inhibitor bromoenol lactone (BEL), which inhibits both mammalian isoforms of iPLA2 (iPLA2β and iPLA2γ) at low micromolar concentrations, the potential role of iPLA2 in the regulation of proinflammatory and antiviral gene expression by macrophages exposed to dsRNA or virus infection was examined. The possibility that phospholipid metabolism participates in innate immunity was realized when RAW264.7 cells were treated with poly IC or EMCV and a concentration-dependent decrease in the expression of iNOS and production of nitrite was observed in the presence of BEL (Maggi et al., 2002). But what was the basis for the BEL-sensitivity of dsRNA- or virus-induced iNOS expression, particularly since there was no precedence for the participation of iPLA2 in the regulation of dsRNA-induced gene expression and the literature was replete with examples of iPLA2 function in membrane phospholipid remodeling and maintenance of membrane homeostasis (Baburina and Jackowski, 1999; Balsinde et al., 1997; Balsinde et al., 1995; Balsinde and Dennis, 1997)? The first obvious explanation lay in the possibility that BEL is not so selective after all, but rather inhibits another enzyme required for dsRNA-induced iNOS expression. This possibility is potentially consistent with published reports that BEL exhibits inhibitory properties toward another enzyme, phosphotidate phosphohydrolase-1 (PAP-1) (Balsinde and Dennis, 1996). This explanation did not turn out to be true, however, since 1) the concentrations of BEL required to prevent virus-induced iNOS expression ware 2- to 4-fold lower than those necessary to inhibit PAP-1 activity in vitro (Balsinde and Dennis, 1996) and 2) PAP-1-selective inhibitors have no effect on the expression of iNOS in virus-infected macrophages (L.B. Maggi, Jr. and J.A. Corbett, unpublished observation).

In time, both pharmacologic evidence (using individual BEL enantiomers) and genetic evidence (using macrophages isolated from iPLA2β^-/- mice) were obtained to implicate a requirement for the beta isoform of iPLA2 in the transcriptional activation of iNOS during virus infection of macrophages (Moran et al., 2005a). The expression of iNOS in EMCV-infected macrophages is sensitive to the S-enantiomer of BEL, which has previously been characterized to be 10-fold more selective for iPLA2β than the other mammalian isoform iPLA2γ, while the same concentration of the R-enantiomer has little inhibitory effect on virus-induced iNOS expression (Moran et al., 2005a). Consistent with this pharmacological approach, virus-induced iNOS mRNA accumulation and protein expression has been documented to be completely prevented in iPLA2β^-/- macrophages, suggesting that the beta isoform of iPLA2 is selectively and absolutely required for transcriptional activation of iNOS in macrophages during virus infection (Moran et al., 2005a). Importantly, the role of iPLA2 in innate immune responses to virus infection appears to be selective for iNOS as the expression of other inflammatory genes in macrophages induced by dsRNA or virus infection, including interleukin-1 and cyclooxygenase-2, are not sensitive to iPLA2 inhibition or iPLA2β gene deletion (Maggi et al., 2002; Moran et al., 2005a; Steer et al., 2003).

The downstream targets of iPLA2: Given the experimental observations consistent with a requirement for iPLA2β in the regulation of virus-induced iNOS expression, the possibility was envisaged that iPLA2 enzymatic activity in virus-infected macrophages liberates phospholipid-derived products that participate in the induction of the iNOS gene. When RAW264.7 cells or primary mouse macrophages are incubated with either poly IC or EMCV, a robust increase in membrane-associated iPLA2 enzymatic activity is readily apparent (Maggi et al., 2002). Importantly, this increase in iPLA2 activity is sensitive to BEL at the same concentrations that inhibit dsRNA- or virus-induced iNOS expression and nitrite production. While it can be demonstrated that virus infection stimulates iPLA2 enzymatic activity and that iNOS expression is absolutely dependent upon functional iPLA2, there is no precedence for iPLA2 DNA binding activity or phospholipids directly inducing alterations in transcription, thus a direct effect of iPLA2 at the iNOS promoter is difficult to reconcile. The possibility that iPLA2 activates gene transcription indirectly through the liberation of phospholipid mediators capable of activating proteins involved in iNOS expression must be subsequently considered. One protein previously identified to be sensitive to modulation by phospholipid-derived products is the cAMP-dependent protein kinase (PKA), as studies by Williams and Ford (Williams and Ford, 1997) have revealed that lysophospholipids derived from iPLA2 enzymatic activity can activate PKA in vitro. Using a variety of pharmacological and molecular biology approaches, experimental evidence consistent with a role for PKA and the transcription factor CREB in mediating iPLA2-dependent iNOS expression has been obtained (Maggi et al., 2002; Figure 3). What these experiments lack, however, is direct evidence that iPLA2-derived phospholipids, such as those defined to be capable of activating PKA in vitro, can stimulate iNOS transcription. To address these issues, the inhibitory effects of BEL on virus-induced iNOS expression have been demonstrated to be overcome by the exogenous addition of lysophospholipids capable of activating PKA (Maggi et al., 2002).

Figure 3: Novel iPLA2-mediated regulation of iNOS expression by dsRNA or virus infection. iPLA2 participates in the transcriptional activation of iNOS by liberating phospholipid products capable of directly activating PKA. Although the transcription factors CREB and NF-κB are required for iNOS expression, the transcriptional regulation is independent of PKR.

Although these studies do not define the product of iPLA2 activity that regulates iNOS expression, they do provide compelling circumstantial evidence that lysophospholipids can contribute to the transcriptional activation of iNOS. These studies not only define a new physiological role for iPLA2 but they also establish a novel PKR-independent signaling pathway induced by dsRNA or viral infection that regulates the expression of the antiviral gene iNOS.

PKR-independent Proinflammatory Gene Expression: The Transcriptional Regulation of COX-2 and IL-1

The response of macrophages to viral pathogens is characterized, in part, by the robust expression of a battery of proinflammatory genes that participate in the inhibition of virus replication and promote virus clearance. Prostaglandins are abundant, ubiquitous lipid mediators produced during inflammatory reactions that contribute to numerous physiological responses. Importantly, proinflammatory lipids and the enzymes that are responsible for their metabolism contribute critically to immune responses. Cyclooxygenase-2 (COX-2) participates in the production of prostaglandins by catalyzing the oxidation of arachidonic acid to PGH2, which is subsequently isomerized to various protanoids including PGE2. A role for COX-2 and PGE2 in host responses to virus infection can be appreciated from the observation that COX-2 is rapidly induced by viral infection (Foley et al., 1992; Janelle et al., 2002; Steer and Corbett, 2003; Steer et al., 2003; Zhu et al., 2002). Consistent with virus-induced increases in COX-2 expression, with a few exceptions PGE2 production enhances the replication of most viruses that have been examined to date, although the mechanisms responsible for the augmentation in replication remain unclear (Chen et al., 2002; Chen et al., 2000; Khyatti and Menezes, 1990; Rott et al., 2003). Activation of proinflammatory cellular responses may be host immune strategies against an invading viral pathogen or likewise may represent a direct virus-mediated process requisite for successful replication, although these possibilities are not mutually exclusive.

Although the cumulative results from a number of studies that have examined the expression of COX-2 in response to a variety of RNA and DNA viruses suggest that products of COX-2 enzymatic activity correlate with virus infection and replication, the regulatory mechanisms of COX-2 expression in virus-infected macrophages have not been completely defined. Treatment of RAW264.7 cells with dsRNA or infection with EMCV results in COX-2 mRNA accumulation, protein expression and robust PGE2 release (Steer et al., 2003). While the expression of COX-2 is clearly stimulated by dsRNA, the transcriptional activation of this gene does not appear to be dependent upon PKR as overexpression of dnPKR mutants or the genetic absence of PKR has no inhibitory effect on dsRNA or EMCV-induced COX-2 expression (Steer et al., 2003). These results argue that functional PKR is not essential for the regulation of COX-2 expression in virus-infected macrophages.

The transcriptional activation of many genes that participate in cellular immune defenses requires the transcription factor NF-κB. Multiple lines of evidence are consistent with a requirement for NF-κB in the transcriptional activation of COX-2 in macrophages treated with dsRNA or infected with virus. Conditions that prevent the activation of NF-κB, either pharmacologically using the proteasome inhibitors PDTC or MG132 or by adenovirus-mediated overexpression of an IκB super-repressor, prevent dsRNA or virus-induced COX-2 expression by macrophages, suggesting that NF-κB participates in the transcriptional regulation of COX2 (Steer et al., 2003). In addition to NF-κB, preliminary observations are consistent with a role for p38 and JNK in the regulation of dsRNA- or virus-induced COX-2 expression in both RAW264.7 cells and primary mouse macrophages (Figure 4). JNK and p38 appears to selectively participate in the regulation of COX-2 expression as inhibitors of these MAPK do not attenuate virus-induced IL-1 or iNOS expression (Steer and Corbett, 2003; S.A. Steer and J.A. Corbett, unpublished observation).

Figure 4: The transcriptional regulation of COX-2 in virus-infected macrophages. The activation of COX-2 expression by macrophages in response to dsRNA or virus infection requires NF-κB as well as the MAPK p38 and JNK. Importantly, neither the activation of these signaling pathways nor the expression of COX-2 are dependent upon PKR.

In addition to the production of prostaglandins, the expression and release of cytokines also contribute to proinflammatory cellular responses to virus infection. The expression of the cytokine interleukin-1 (IL-1) is robustly induced by dsRNA or viral infection, suggesting that it may participate in host protection to viral pathogens. Despite the implication that IL-1 contributes to protective host responses, the mechanism by which virus infection activates IL-1 expression in macrophages has not been extensively characterized. Recently, one mechanism by which viruses stimulate the expression of IL-1 has been reported to involve MAPK signaling. Treatment of RAW264.7 cells or primary macrophages with dsRNA or EMCV results in IL-1 mRNA accumulation, protein expression and IL-1 release as well as the rapid induction of ERK phosphorylation and enzymatic activity (Maggi et al., 2003; Moran et al., 2005b). Experimental data is consistent with a requirement for ERK in virus-induced IL-1 expression based on the observation that the transcriptional activation of the IL-1 gene is prevented in the presence of the selective ERK inhibitors U0126 and PD98059 or by overexpression of a catalytically inactive dnERK (Maggi et al., 2003). Importantly, ERK appears to selectively participate in the transcriptional regulation of IL-1 in virus-infected macrophages given that ERK inhibitors fail to attenuate dsRNA-induced expression of iNOS or COX-2 (Maggi et al., 2002; S.A. Steer and J.A. Corbett, unpublished observation). This is in contrast to the regulation of the expression of iNOS in response to endotoxin as reports support a direct requirement for ERK in LPS-induced iNOS expression (Ajizian et al., 1999; Chan and Riches, 2001; Chan et al., 1999). These results are consistent with the hypothesis that the activation of multiple, independent signaling pathways that converge at the promoters of individual genes affords the ability of cells to generate appropriate yet differing responses to many diverse stimuli.

Figure 5: The regulation of IL-1 expression in virus-infected macrophages. In addition to NF-κB, ERK and PU.1 are required for dsRNA- or virus-induced IL-1 expression. PU.1 may be one downstream target of ERK that mediates ERK-dependent IL-1 expression. PKR is not required for dsRNA-induced ERK activation or IL-1 expression in macrophages.

The mechanism by which ERK participates in virus-induced IL-1 expression by macrophages has been proposed to be, at least in part, through the activation of PU.1, a transcription factor implicated in the regulation of LPS-induced IL-1 expression (Buras et al., 1995; Lodie et al., 1997). A potential requirement for PU.1 in the regulation of IL-1 expression was first envisaged following the observations that 1) dsRNA stimulates PU.1 DNA binding activity and 2) mutation or deletion of the PU.1 binding site within the IL-1β promoter completely prevents poly IC-induced reporter activity (L.B. Maggi, Jr., J.M. Moran and J.A. Corbett, unpublished observations). Although these results are consistent with the hypothesis that PU.1 is required for dsRNA-induced IL-1 expression by macrophages, they do not address whether ERK is involved in PU.1 activation. Experimental evidence to suggest a contribution of ERK in the activation of PU.1 has emerged from gel shift analysis and chromatin immunoprecipitation studies, which have revealed that dsRNA induces PU.1 DNA binding activity in a manner sensitive to ERK inhibitors and the in vivo accumulation of PU.1 at the IL-1β promoter temporally correlates with ERK activation (L.B. Maggi, Jr., L. McClemore, J.M. Moran and J.A. Corbett, unpublished observation). While these results do not provide definitive evidence to suggest that PU.1 is one downstream target of ERK in macrophages exposed to dsRNA or virus infection, they provide strong circumstantial evidence in support of PU.1 mediating ERK-dependent IL-1 expression (Figure 5). Importantly, PU.1 does not appear to be the sole transcription factor responsible for the transcriptional activation of IL-1 expression since inhibition of NF-κB activity abrogates poly IC-induced IL-1 expression (Heitmeier et al., 1998). Moreover, both the activation of ERK and the expression of IL-1 in macrophages in response to poly IC or EMCV infection appear to be independent of PKR since ERK activation and IL-1 protein expression occur to similar levels in PKR^+/+ and PKR^-/- macrophages (Maggi et al., 2000; Maggi et al., 2003; Moran et al., 2005b).

Taken together, these experimental findings provide evidence to suggest that the transcriptional activation of COX-2 and IL-1 is regulated by unique signaling pathways in cooperation with NF-κB.

New DsRNA Binding Proteins: PKR, TLR3 and RIG-I

The recognition that eIF2α phosphorylation, NF-κB activation and proinflammatory and antiviral gene expression induced by dsRNA can occur by PKR-independent mechanisms has in many ways provided more questions than answers. Although a direct role for PKR has been excluded from several antiviral responses using molecular and genetic techniques to compromise PKR function, with few exceptions has there been a systematic characterization of dsRNA-induced signaling pathways that regulate gene expression in a PKR-independent manner. Likewise, the identity of cellular proteins capable of mediating PKR-independent responses to dsRNA remains to be definitively established. These works have been greatly enhanced in recent years, however, by the discovery of several proteins that possess the inherent ability to interact with dsRNA and activate responses that participate in innate immunity. We will discuss two such proteins, TLR3 and RIG-I, and comment on their potential role in PKR-independent antiviral signaling pathways activated by dsRNA.

dsRNA and TLR3: The toll-like receptors (TLRs) represent a family of proteins that have been shown to participate in innate immune responses through the recognition of distinct molecular patters characteristic of microbial pathogens (Akira, 2001). Originally identified in Drosophila, the Toll proteins were characterized as important regulators of immunity and the structural and functional similarities in these proteins from flies to humans has been proposed to suggest that TLRs represent the ancient arm of the mammalian innate immune system (Brightbill and Modlin, 2000). To date, more than 10 mammalian TLRs have been described, each of which recognize a different pathogenic stimulus such as lipopolysaccharide (LPS), CpG DNA or dsRNA. Of all the TLRs studied, TLR3 has been identified to be particularly responsive to dsRNA (both of synthetic and viral origin) and activation of TLR3 induces NF-κB activity and type I IFN gene expression. Importantly, TLR3 appears to participate critically to cellular responses to dsRNA based on the observation that TLR3^-/- macrophages have impaired type I IFN expression, cytokine production, and NF-κB activation in response to poly IC when compared to wild-type macrophages (Alexopoulou et al., 2001). Furthermore, in vivo studies have shown that TLR3^-/- mice are more resistant to poly IC-induced shock and death than their TLR3^+/+ counterparts (Alexopoulou et al., 2001). These data have contributed to the hypothesis that TLR3 likely plays an important role in host defense against virus infection by recognizing the accumulation of viral dsRNA. However, experimental results from TLR3^-/- mice challenged with several RNA or DNA viruses does not support this hypothesis as TLR3-deficient mice were equally susceptible to reovirus, lymphocytic choriomeningitis virus, vesicular stomatitis virus and murine cytomegalovirus (MCMV) infection and generated similar CD4 and CD8 T-cell immune responses to these viruses as TLR3-sufficient mice (Edelmann et al., 2004). Notwithstanding, mice deficient in downstream components of TLR3 signaling have been reported to have enhanced susceptibility to MCMV at higher doses (Hoebe et al., 2003). Despite the discrepancies in these findings, taken together these results may suggest that TLR3 does not universally participate in innate immune responses to all virus infections. Alternatively, it is also possible that TLR3 is not capable of recognizing viral dsRNA intermediates from some viruses (due to viral RNA sequestration, etc) or that endogenous levels of dsRNA during viral infection are insufficient to activate TLR3 signaling.

dsRNA and RIG-I: The retinoic acid inducible gene-I (RIG-I) has been implicated as an essential component of innate immune responses to dsRNA or virus infection. RIG-I is an intracellular RNA helicase belonging to the DExD/H box family of helicases that are found in almost all organisms from viruses to mammals. Cloned and characterized by Fujita and colleagues (Yoneyama et al., 2004), RIG-I has two N-terminal caspase recruitment domain (CARD) motifs that are required for IRF-3 and NF-κB activation and a C-terminal helicase domain that appears to confer dsRNA binding activity. The observation that overexpression of RIG-I dramatically enhances IFN-β reporter activity in cells treated with dsRNA or infected with Newcastle disease virus (NDV) was consistent with the hypothesis that RIG-I participates in dsRNA-induced activation of gene transcription. Accordingly, polyIC or NDV infection enhances RIG-I-mediated IRF-3 and NF-κB reporter activity and DNA binding activity. Importantly, RIG-I is indispensable for the expression of IFN-β and several IFN-inducible chemokine genes as the transcriptional activation of these genes is prevented in cells expressing siRNA to RIG-I or in RIG-I^-/- MEFs (Kato et al., 2005; Yoneyama et al., 2004). Not only does RIG-I participate in host antiviral responses by inducing IFN expression, but it also appears to have a direct effect on virus replication. Consistent with a role for RIG-I in the inhibition of virus replication, the stable overexpression of RIG-I protects EMCV- or VSV-infected cells from cytopathic effect and reduces progeny virion production by two to three logs while viral yield in VSV-infected RIG-I^-/- cells is augmented approximately 100-fold (Kato et al., 2005; Yoneyama et al., 2004). Importantly, the mechanism by which RIG-I inhibits virus replication does not appear to be due to RIG-I-mediated expression of antiviral IFNs since RIG-I suppresses virus replication in the presence of neutralizing antibodies to type IFN.

While it remains to be experimentally determined whether the contribution of RIG-I to dsRNA-induced gene expression is intact and functional in macrophages, it is intriguing to consider a primary role for RIG-I in the activation of dsRNA-induced events that have previously been determined to be independent of PKR. Results in dendritic cells support a role for RIG-I in IFN gene expression as bone marrow-derived dendritic cells from RIG-I^-/- mice exhibit severely impaired IFN-β and cytokine induction after NDV infection (Kato et al., 2005).

Provided that PKR, TLR3 and RIG-I are all activated by dsRNA and participate in the activation of NF-κB and the expression of IFNs and cytokines, one intriguing consideration is the significance of having three functionally redundant mechanisms present in cells to respond to viral dsRNA. The existence of multiple dsRNA-sensitive proteins would be expected to facilitate robust host responses to virus infection and ensure that the health of the organism was maintained. However, the assumption that PKR, TLR3 and RIG-I are all activated by virus infection to mediate proinflammatory and antiviral gene expression by redundant pathways is likely not completely accurate. It is perhaps more appropriate to envision that these proteins, given their subcellular localization, selectively recognize dsRNA in very discrete and separate locations. Therefore, the activation of PKR, TLR3 or RIG-I during virus infection would be largely dictated by the location of dsRNA. The extracellular accumulation of viral dsRNA, perhaps due to the release of viral dsRNA intermediates from neighboring infected cells undergoing lytic virus replication or necrosis, would be expected to exclusively activate cell surface receptors such as TLR3. In contrast, the intracellular accumulation of dsRNA during viral replication would be predicted to activate cytosolic proteins such as PKR and RIG-I but not TLR3. These interpretations are consistent with the experimental observations that 1) TLR3-induced antiviral responses are unaffected by the expression of dnRIG-I (Yoneyama et al., 2004), and 2) virus-induced host responses and virus replication are unaffected by the absence of TLR3 (Edelmann et al., 2004).

The presence of multiple, independent mechanisms of dsRNA recognition, therefore, confers the advantage of sensing virus infection from a variety of cellular locations. Moreover, due to the recent characterization of dsRNA-sensitive proteins in addition to PKR, the identification of PKR-independent antiviral responses described herein does not challenge the importance of PKR in host responses to virus infection but rather provides a potential explanation to reconcile the PKR-independence of host responses to dsRNA. With time, it is expected that the contribution of TLR3 and RIG-I to PKR-independent antiviral responses will be further elucidated and will provide a more comprehensive understanding of cellular responses to dsRNA and virus infection.

dsRNA-independent antiviral responses: Several studies that have examined the activation of proinflammatory and antiviral gene expression in virus infected cells have supported the hypothesis that host responses can be independent of PKR. As PKR is the most well characterized dsRNA-sensitive protein, the possibility must be considered that PKR-independent responses to virus infection are either mediated by other dsRNA-sensitive proteins or that the activation of these responses are not due to intracellular accumulation of dsRNA. Recent experimental evidence is consistent with the latter. Interestingly, heating EMCV at 54°C, conditions which have been previously shown to liberate the ssRNA genome from virions while maintaining intact capsid structures (Smirnov et al., 1983), does not prevent the ability of EMCV to activate NF-κB, CREB or ERK, nor does it attenuate EMCV-induced iNOS expression (Moran et al., 2005b). Furthermore, the activation of these pathways was confirmed to be independent of viral dsRNA accumulation as activation of ERK, NF-κB and CREB precedes EMCV RNA accumulation by several hours and heat-treated EMCV does not introduce detectable levels of genomic RNA into infected macrophages (Moran et al., 2005b). Taken together, these results suggest that the mechanism by which EMCV activates macrophages to express antiviral genes is not associated with the intracellular accumulation of viral dsRNA. The mechanism by which EMCV activates macrophages has been postulated to be due to an interaction between the EMCV virion and a macrophage cell surface receptor early during infection based on the observation that ERK, NF-κB and CREB are all activated within 15 minutes post-infection (Moran et al., 2005b). These experimental findings are consistent with reports showing that the association of virus particles with the cell surface can activate various signaling pathways as the binding of HCMV to TLR2 activates NF-κB (Compton et al., 2003) and replication-defective adenovirus stimulates MAPK signaling in A549 cells within 10 minutes post-infection (Terenzi et al., 1999). Not only do these results support the hypothesis that virus-induced macrophage activation can occur following specific molecular interactions between virions and host cells, they also provide a potential explanation for the PKR-independent regulation of gene expression in EMCV-infected macrophages.

Conclusions: A Circle Closed

As we began this review with a discussion of the historical observations that led to the conclusion that dsRNA could activate IFN production, it is perhaps appropriate that we conclude with a perspective on the current knowledge of the regulation of IFN expression in response to dsRNA or virus infection. The experimental investigation into the regulation of IFN expression in dsRNA-treated cells represents a 40 year journey in which the importance of several cellular proteins sensitive to dsRNA were identified. It was studies of IFN expression that provided the first indication that dsRNA-induced gene expression might be dependent upon PKR. Although the expression of type I IFNs are clearly induced in numerous cell types by dsRNA or viral infection and participate critically in host protective responses, in vivo evidence from PKR^-/- mice argues against a PKR-dependent mechanism of gene regulation. These results are consistent with the presence of additional dsRNA-sensitive, PKR-independent pathways that regulate IFN expression and novel cellular proteins capable of binding dsRNA and activating gene expression are beginning to be identified. Recently, TLR3 and RIG-I have been implicated in the regulation of dsRNA-induced IFN expression from various overexpression or knock-out models. The presence of multiple yet independent dsRNA-induced pathways that induce IFN expression may suggest the evolution of functional redundancy given the importance of antiviral responses for host survival during virus infection. Additionally, multiple pathways may offer a selective advantage to allow cells to recognize an invading virus from a variety of locations.

Importantly, the PKR-independent regulation of gene expression in virus-infected cells is not limited to the IFN genes but has been extended to proinflammmatory and antiviral genes including iNOS, COX-2 and IL-1. The complete complement of PKR-independent genes that participate in innate immunity expressed in virus-infected cells will undoubtedly continue to grow as will the number of proteins that are known to be activated by dsRNA. Advancements in the current understanding of the molecular mechanisms of cellular antiviral responses are expected to enhance our knowledge of viral pathogenesis and may provide novel targets for antiviral therapeutics.

Acknowledgements

We would like to thank Drs. Abdul Waheed and Michael M. Moxley for helpful scientific discussions and Colleen Bratcher and Kari Chambers for assistance in preparing this review. This work was supported by National Institutes of Health Grants DK-52194 and AI-44458 (J.A.C). Correspondence should be addressed to Dr. Moran (current address: Northwestern University, Department of Biochemistry, 2205 Tech Drive, Hogan 2-100, Evanston, IL 60208; Ph: (847-491-3714 ) or Dr. Corbett (Department of Biochemistry and Molecular Biology, Saint Louis University Sch. of Med., 1402 S. Grand Blvd., St. Louis, MO 63104; email: corbettj@slu.edu).

References

Abraham, N., Stojdl, D. F., Duncan, P. I., Methot, N., Ishii, T., Dube, M., Vanderhyden, B. C., Atkins, H. L., Gray, D. A., McBurney, M. W., et al. (1999). Characterization of transgenic mice with targeted disruption of the catalytic domain of the double-stranded RNA-dependent protein kinase, PKR. J Biol Chem 274, 5953-5962.

Ajizian, S. J., English, B. K., and Meals, E. A. (1999). Specific inhibitors of p38 and extracellular signal-regulated kinase mitogen-activated protein kinase pathways block inducible nitric oxide synthase and tumor necrosis factor accumulation in murine macrophages stimulated with lipopolysaccharide and interferon-gamma. Journal of Infectious Diseases 179, 939-944.

Akira, S. (2001). Toll-like receptors and innate immunity. Adv Immunol 78, 1-56.

Alexopoulou, L., Holt, A. C., Medzhitov, R., and Flavell, R. A. (2001). Recognition of double-stranded RNA and activation of NF-kappaB by Toll-like receptor 3. Nature 413, 732-738.

Baburina, I., and Jackowski, S. (1999). Cellular responses to excess phospholipid. J Biol Chem 274, 9400-9408.

Baeuerle, P. A., and Henkel, T. (1994). Function and activation of NF-kappa B in the immune system. Annu Rev Immunol 12, 141-179.

Balsinde, J., Balboa, M. A., and Dennis, E. A. (1997). Antisense inhibition of group VI Ca2+-independent phospholipase A2 blocks phospholipid fatty acid remodeling in murine P388D1 macrophages. J Biol Chem 272, 29317-29321.

Balsinde, J., Bianco, I. D., Ackermann, E. J., Conde-Frieboes, K., and Dennis, E. A. (1995). Inhibition of calcium-independent phospholipase A2 prevents arachidonic acid incorporation and phospholipid remodeling in P388D1 macrophages. Proc Natl Acad Sci U S A 92, 8527-8531.

Balsinde, J., and Dennis, E. A. (1996). Bromoenol lactone inhibits magnesium-dependent phosphatidate phosphohydrolase and blocks triacylglycerol biosynthesis in mouse P388D1 macrophages. J Biol Chem 271, 31937-31941.

Balsinde, J., and Dennis, E. A. (1997). Function of calcium-independent phospholipase A2 in arachidonic acid metabolism in P388D1 macrophages. Adv Exp Med Biol 407, 99-103.

Baltimore, D., Becker, Y., and Darnell, J. E. (1964). Virus-Specific Double-Stranded Rna in Poliovirus-Infected Cells. Science 143, 1034-1036.

Baltzis, D., Li, S., and Koromilas, A. E. (2002). Functional characterization of pkr gene products expressed in cells from mice with a targeted deletion of the N terminus or C terminus domain of PKR. J Biol Chem 277, 38364-38372.

Brightbill, H. D., and Modlin, R. L. (2000). Toll-like receptors: molecular mechanisms of the mammalian immune response. Immunology 101, 1-10.

Bukrinsky, M. I., Nottet, H. S., Schmidtmayerova, H., Dubrovsky, L., Flanagan, C. R., Mullins, M. E., Lipton, S. A., and Gendelman, H. E. (1995). Regulation of nitric oxide synthase activity in human immunodeficiency virus type 1 (HIV-1)-infected monocytes: implications for HIV-associated neurological disease. J Exp Med 181, 735-745.

Buras, J. A., Reenstra, W. R., and Fenton, M. J. (1995). NF beta A, a factor required for maximal interleukin-1 beta gene expression is identical to the ets family member PU.1. Evidence for structural alteration following LPS activation. Mol Immunol 32, 541-554.

Burdon, R. H., Billeter, M. A., Weissmann, C., Warner, R. C., Ochoa, S., and Knight, C. A. (1964). Replication of Viral Rna, V. Presence of a Virus-Specific Double-Stranded Rna in Leaves Infected with Tobacco Mosaic Virus. Proc Natl Acad Sci U S A 52, 768-775.

Chan, E. D., and Riches, D. W. (2001). IFN-gamma + LPS induction of iNOS is modulated by ERK, JNK/SAPK, and p38(mapk) in a mouse macrophage cell line. American Journal of Physiology - Cell Physiology 280, C441-450.

Chan, E. D., Winston, B. W., Uh, S. T., Wynes, M. W., Rose, D. M., and Riches, D. W. (1999). Evaluation of the role of mitogen-activated protein kinases in the expression of inducible nitric oxide synthase by IFN-gamma and TNF-alpha in mouse macrophages. Journal of Immunology 162, 415-422.

Chen, N., Restivo, A., and Reiss, C. S. (2002). Selective inhibition of COX-2 is beneficial to mice infected intranasally with VSV. Prostaglandins Other Lipid Mediat 67, 143-155.

Chen, N., Warner, J. L., and Reiss, C. S. (2000). NSAID treatment suppresses VSV propagation in mouse CNS. Virology 276, 44-51.

Chu, W. M., Ostertag, D., Li, Z. W., Chang, L., Chen, Y., Hu, Y., Williams, B., Perrault, J., and Karin, M. (1999). JNK2 and IKKbeta are required for activating the innate response to viral infection. Immunity 11, 721-731.

Colby, C., and Duesberg, P. H. (1969). Double-stranded RNA in vaccinia virus infected cells. Nature 222, 940-944.

Compton, T., Kurt-Jones, E. A., Boehme, K. W., Belko, J., Latz, E., Golenbock, D. T., and Finberg, R. W. (2003). Human cytomegalovirus activates inflammatory cytokine responses via CD14 and Toll-like receptor 2. J Virol 77, 4588-4596.

Edelmann, K. H., Richardson-Burns, S., Alexopoulou, L., Tyler, K. L., Flavell, R. A., and Oldstone, M. B. (2004). Does Toll-like receptor 3 play a biological role in virus infections? Virology 322, 231-238.

Ehrenfeld, E., and Hunt, T. (1971). Double-stranded poliovirus RNA inhibits initiation of protein synthesis by reticulocyte lysates. Proc Natl Acad Sci U S A 68, 1075-1078.

Feng, G. S., Chong, K., Kumar, A., and Williams, B. R. (1992). Identification of double-stranded RNA-binding domains in the interferon-induced double-stranded RNA-activated p68 kinase. Proc Natl Acad Sci U S A 89, 5447-5451.

Field, A. K., Tytell, A. A., Lampson, G. P., and Hilleman, M. R. (1967). Inducers of interferon and host resistance. II. Multistranded synthetic polynucleotide complexes. Proc Natl Acad Sci U S A 58, 1004-1010.

Finkelstein, M. S., Bausek, G. H., and Merigan, T. C. (1968). Interferon inducers in vitro: difference in sensitivity to inhbitiros of RNA and protein synthesis. Science 161, 465-468.

Flodstrom, M., Horwitz, M. S., Maday, A., Balakrishna, D., Rodriguez, E., and Sarvetnick, N. (2001). A critical role for inducible nitric oxide synthase in host survival following coxsackievirus B4 infection. Virology 281, 205-215.

Foley, P., Kazazi, F., Biti, R., Sorrell, T. C., and Cunningham, A. L. (1992). HIV infection of monocytes inhibits the T-lymphocyte proliferative response to recall antigens, via production of eicosanoids. Immunology 75, 391-397.

Galabru, J., and Hovanessian, A. (1987). Autophosphorylation of the protein kinase dependent on double-stranded RNA. J Biol Chem 262, 15538-15544.

Galabru, J., Katze, M. G., Robert, N., and Hovanessian, A. G. (1989). The binding of double-stranded RNA and adenovirus VAI RNA to the interferon-induced protein kinase. Eur J Biochem 178, 581-589.

Galcheva-Gargova, Z., Derijard, B., Wu, I. H., and Davis, R. J. (1994). An osmosensing signal transduction pathway in mammalian cells. Science 265, 806-808.

Gale, M., Jr., and Katze, M. G. (1998). Molecular mechanisms of interferon resistance mediated by viral-directed inhibition of PKR, the interferon-induced protein kinase. Pharmacol Ther 78, 29-46.

Goh, K. C., deVeer, M. J., and Williams, B. R. (2000). The protein kinase PKR is required for p38 MAPK activation and the innate immune response to bacterial endotoxin. Embo J 19, 4292-4297.

Gomatos, P. J., and Tamm, I. (1963). Animal and Plant Viruses with Double-Helical Rna. Proc Natl Acad Sci U S A 50, 878-885.

Hambleton, J., Weinstein, S. L., Lem, L., and DeFranco, A. L. (1996). Activation of c-Jun N-terminal kinase in bacterial lipopolysaccharide-stimulated macrophages. Proc Natl Acad Sci U S A 93, 2774-2778.

Harle, P., Noisakran, S., and Carr, D. J. (2001). The application of a plasmid DNA encoding IFN-alpha 1 postinfection enhances cumulative survival of herpes simplex virus type 2 vaginally infected mice. J Immunol 166, 1803-1812.

Haskill, S., Beg, A. A., Tompkins, S. M., Morris, J. S., Yurochko, A. D., Sampson-Johannes, A., Mondal, K., Ralph, P., and Baldwin, A. S., Jr. (1991). Characterization of an immediate-early gene induced in adherent monocytes that encodes I kappa B-like activity. Cell 65, 1281-1289.

Heitmeier, M. R., Scarim, A. L., and Corbett, J. A. (1998). Double-stranded RNA-induced inducible nitric-oxide synthase expression and interleukin-1 release by murine macrophages requires NF-kappaB activation. J Biol Chem 273, 15301-15307.

Hendricks, R. L., Weber, P. C., Taylor, J. L., Koumbis, A., Tumpey, T. M., and Glorioso, J. C. (1991). Endogenously produced interferon alpha protects mice from herpes simplex virus type 1 corneal disease. J Gen Virol 72 ( Pt 7), 1601-1610.

Hirasawa, K., Jun, H. S., Han, H. S., Zhang, M. L., Hollenberg, M. D., and Yoon, J. W. (1999). Prevention of encephalomyocarditis virus-induced diabetes in mice by inhibition of the tyrosine kinase signalling pathway and subsequent suppression of nitric oxide production in macrophages. J Virol 73, 8541-8548.

Hoebe, K., Du, X., Georgel, P., Janssen, E., Tabeta, K., Kim, S. O., Goode, J., Lin, P., Mann, N., Mudd, S., et al. (2003). Identification of Lps2 as a key transducer of MyD88-independent TIR signalling. Nature 424, 743-748.

Horrocks, L. A., Spanner, S., Mozzi, R., Fu, S. C., D'Amato, R. A., and Krakowka, S. (1978). Plasmalogenase is elevated in early demyelinating lesions. Adv Exp Med Biol 100, 423-438.

Hovanessian, A. (1993). Interferon-induced and double-stranded RNA-activated proteins as key enzymes regulating protein synthesis. In Translational Regulation of Gene Expression 2, J. Ilan, ed. (New York, Plenum Press), pp. 163-185.

Hunt, T., and Ehrenfeld, E. (1971). Cytoplasm from poliovirus-infected HeLa cells inhibits cell-free haemoglobin synthesis. Nat New Biol 230, 91-94.

Hunter, T., Hunt, T., Jackson, R. J., and Robertson, H. D. (1975). The characteristics of inhibition of protein synthesis by double-stranded ribonucleic acid in reticulocyte lysates. J Biol Chem 250, 409-417.

Iordanov, M. S., Paranjape, J. M., Zhou, A., Wong, J., Williams, B. R., Meurs, E. F., Silverman, R. H., and Magun, B. E. (2000). Activation of p38 mitogen-activated protein kinase and c-Jun NH(2)-terminal kinase by double-stranded RNA and encephalomyocarditis virus: involvement of RNase L, protein kinase R, and alternative pathways. Mol Cell Biol 20, 617-627.

Iordanov, M. S., Pribnow, D., Magun, J. L., Dinh, T. H., Pearson, J. A., Chen, S. L., and Magun, B. E. (1997). Ribotoxic stress response: activation of the stress-activated protein kinase JNK1 by inhibitors of the peptidyl transferase reaction and by sequence-specific RNA damage to the alpha-sarcin/ricin loop in the 28S rRNA. Mol Cell Biol 17, 3373-3381.

Iordanov, M. S., Pribnow, D., Magun, J. L., Dinh, T. H., Pearson, J. A., and Magun, B. E. (1998). Ultraviolet radiation triggers the ribotoxic stress response in mammalian cells. J Biol Chem 273, 15794-15803.

Iordanov, M. S., Wong, J., Bell, J. C., and Magun, B. E. (2001). Activation of NF-kappaB by double-stranded RNA (dsRNA) in the absence of protein kinase R and RNase L demonstrates the existence of two separate dsRNA-triggered antiviral programs. Mol Cell Biol 21, 61-72.

Janelle, M. E., Gravel, A., Gosselin, J., Tremblay, M. J., and Flamand, L. (2002). Activation of monocyte cyclooxygenase-2 gene expression by human herpesvirus 6. Role for cyclic AMP-responsive element-binding protein and activator protein-1. J Biol Chem 277, 30665-30674.

Kaerner, H. C., and Hoffmann-Berling, H. (1964). Synthesis of Double-Stranded Rna in Rna-Phage Infected E. Coli Cells. Nature 202, 1012-1013.

Karupiah, G., Xie, Q. W., Buller, R. M., Nathan, C., Duarte, C., and MacMicking, J. D. (1993). Inhibition of viral replication by interferon-gamma-induced nitric oxide synthase. Science 261, 1445-1448.

Kato, H., Sato, S., Yoneyama, M., Yamamoto, M., Uematsu, S., Matsui, K., Tsujimura, T., Takeda, K., Fujita, T., Takeuchi, O., and Akira, S. (2005). Cell type-specific involvement of RIG-I in antiviral response. Immunity 23, 19-28.

Kaufman, R. J. (2000). The Double-stranded RNA-activated Protein Kinase PKR. In Translational Control of Gene Expression (Cold Spring Harbor Laboratory Press), pp. 503-527.

Kerr, I. M., Friedman, R.E., Brown, R.E., Ball, A.L., and Brown, J.C. (1974). Inhibition of protein synthesis in cell-free systems from interferon-treated infected cells: Further characterization and effect of formylmethionyl-tRNA. J Virol 13, 9-21.

Khyatti, M., and Menezes, J. (1990). The effect of indomethacin, prostaglandin E2 and interferon on the multiplication of herpes simplex virus type 1 in human lymphoid cells. Antiviral Res 14, 161-172.

Kumar, A., Haque, J., Lacoste, J., Hiscott, J., and Williams, B. R. (1994). Double-stranded RNA-dependent protein kinase activates transcription factor NF-kappa B by phosphorylating I kappa B. Proc Natl Acad Sci U S A 91, 6288-6292.

Kumar, A., Yang, Y. L., Flati, V., Der, S., Kadereit, S., Deb, A., Haque, J., Reis, L., Weissmann, C., and Williams, B. R. (1997). Deficient cytokine signaling in mouse embryo fibroblasts with a targeted deletion in the PKR gene: role of IRF-1 and NF-kappaB. Embo J 16, 406-416.

Lampson, G. P., Tytell, A. A., Field, A. K., Nemes, M. M., and Hilleman, M. R. (1967). Inducers of interferon and host resistance. I. Double-stranded RNA from extracts of Penicillium funiculosum. Proc Natl Acad Sci U S A 58, 782-789.

Lee, S. B., and Esteban, M. (1993). The interferon-induced double-stranded RNA-activated human p68 protein kinase inhibits the replication of vaccinia virus. Virology 193, 1037-1041.

Lenardo, M. J., Fan, C. M., Maniatis, T., and Baltimore, D. (1989). The involvement of NF-kappa B in beta-interferon gene regulation reveals its role as widely inducible mediator of signal transduction. Cell 57, 287-294.

Levin, D., and London, I. M. (1978). Regulation of protein synthesis: activation by double-stranded RNA of a protein kinase that phosphorylates eukaryotic initiation factor 2. Proc Natl Acad Sci U S A 75, 1121-1125.

Lewis, U. J., Rickes, E.L., Williams, D.E., McClelland, L., and Brink, N.G. (1960). J Am Chem Soc 82, 5178.

Li, G., Xiang, Y., Sabapathy, K., and Silverman, R. H. (2004). An apoptotic signaling pathway in the interferon antiviral response mediated by RNase L and c-Jun NH2-terminal kinase. J Biol Chem 279, 1123-1131.

Liou, H. C., and Baltimore, D. (1993). Regulation of the NF-kappa B/rel transcription factor and I kappa B inhibitor system. Curr Opin Cell Biol 5, 477-487.

Lodie, T. A., Savedra, R., Jr., Golenbock, D. T., Van Beveren, C. P., Maki, R. A., and Fenton, M. J. (1997). Stimulation of macrophages by lipopolysaccharide alters the phosphorylation state, conformation, and function of PU.1 via activation of casein kinase II. J Immunol 158, 1848-1856.

Lowenstein, C. J., Hill, S. L., Lafond-Walker, A., Wu, J., Allen, G., Landavere, M., Rose, N. R., and Herskowitz, A. (1996). Nitric oxide inhibits viral replication in murine myocarditis. J Clin Invest 97, 1837-1843.

Maggi, L. B., Jr., Heitmeier, M. R., Scheuner, D., Kaufman, R. J., Buller, R. M., and Corbett, J. A. (2000). Potential role of PKR in double-stranded RNA-induced macrophage activation. Embo J 19, 3630-3638.

Maggi, L. B., Jr., Moran, J. M., Buller, R. M., and Corbett, J. A. (2003). ERK activation is required for double-stranded RNA- and virus-induced interleukin-1 expression by macrophages. J Biol Chem 278, 16683-16689.

Maggi, L. B., Jr., Moran, J. M., Scarim, A. L., Ford, D. A., Yoon, J. W., McHowat, J., Buller, R. M., and Corbett, J. A. (2002). Novel role for calcium-independent phospholipase A(2) in the macrophage antiviral response of inducible nitric-oxide synthase expression. J Biol Chem 277, 38449-38455.

Malewicz, B., Parthasarathy, S., Jenkin, H. M., and Baumann, W. J. (1981). Rapid phospholipase A2 stimulation and diacylglycerol cholinephosphotransferase inhibition in baby hamster kidney cells during initiation of dengue virus infection. Biochem Biophys Res Commun 101, 404-410.

McCormack, S. J., Ortega, L. G., Doohan, J. P., and Samuel, C. E. (1994). Mechanism of interferon action motif I of the interferon-induced, RNA-dependent protein kinase (PKR) is sufficient to mediate RNA-binding activity. Virology 198, 92-99.

Meurs, E. F., Watanabe, Y., Kadereit, S., Barber, G. N., Katze, M. G., Chong, K., Williams, B. R., and Hovanessian, A. G. (1992). Constitutive expression of human double-stranded RNA-activated p68 kinase in murine cells mediates phosphorylation of eukaryotic initiation factor 2 and partial resistance to encephalomyocarditis virus growth. J Virol 66, 5805-5814.

Meusel, T. R., and Imani, F. (2003). Viral induction of inflammatory cytokines in human epithelial cells follows a p38 mitogen-activated protein kinase-dependent but NF-kappa B-independent pathway. J Immunol 171, 3768-3774.

vMontagnier, L., and Sanders, F. K. (1963). Replicative Form of Encephalomyocarditis Virus Ribonucleic Acid. Nature 199, 664-667.

Moran, J. M., Buller, R. M., McHowat, J., Turk, J., Wohltmann, M., Gross, R. W., and Corbett, J. A. (2005a). Genetic and pharmacologic evidence that calcium-independent phospholipase A2beta regulates virus-induced inducible nitric-oxide synthase expression by macrophages. J Biol Chem 280, 28162-28168.

Moran, J. M., Moxley, M. A., Buller, R. M., and Corbett, J. A. (2005b). Encephalomyocarditis virus induces PKR-independent mitogen-activated protein kinase activation in macrophages. J Virol 79, 10226-10236.

Muller, U., Steinhoff, U., Reis, L. F., Hemmi, S., Pavlovic, J., Zinkernagel, R. M., and Aguet, M. (1994). Functional role of type I and type II interferons in antiviral defense. Science 264, 1918-1921.

Nathan, C. (1995). Inducible nitric oxide synthase: regulation subserves function. Curr Top Microbiol Immunol 196, 1-4.

Noisakran, S., Campbell, I. L., and Carr, D. J. (1999). Ectopic expression of DNA encoding IFN-alpha 1 in the cornea protects mice from herpes simplex virus type 1-induced encephalitis. J Immunol 162, 4184-4190.

Ortega, L. G., McCotter, M. D., Henry, G. L., McCormack, S. J., Thomis, D. C., and Samuel, C. E. (1996). Mechanism of interferon action. Biochemical and genetic evidence for the intermolecular association of the RNA-dependent protein kinase PKR from human cells. Virology 215, 31-39.

Pain, V. M. (1996). Initiation of protein synthesis in eukaryotic cells. Eur J Biochem 236, 747-771.

Palumbo, G. J., Glasgow, W. C., and Buller, R. M. (1993). Poxvirus-induced alteration of arachidonate metabolism. Proc Natl Acad Sci U S A 90, 2020-2024.

Patel, R. C., and Sen, G. C. (1992). Identification of the double-stranded RNA-binding domain of the human interferon-inducible protein kinase. J Biol Chem 267, 7671-7676.

Raingeaud, J., Gupta, S., Rogers, J. S., Dickens, M., Han, J., Ulevitch, R. J., and Davis, R. J. (1995). Pro-inflammatory cytokines and environmental stress cause p38 mitogen-activated protein kinase activation by dual phosphorylation on tyrosine and threonine. J Biol Chem 270, 7420-7426.

Rott, D., Zhu, J., Burnett, M. S., Zhou, Y. F., Zalles-Ganley, A., Ogunmakinwa, J., and Epstein, S. E. (2003). Effects of MF-tricyclic, a selective cyclooxygenase-2 inhibitor, on atherosclerosis progression and susceptibility to cytomegalovirus replication in apolipoprotein-E knockout mice. J Am Coll Cardiol 41, 1812-1819.

Safer, B. (1983). 2B or not 2B: regulation of the catalytic utilization of eIF-2. Cell 33, 7-8.

Samuel, C. E., Kuhen, K. L., George, C. X., Ortega, L. G., Rende-Fournier, R., and Tanaka, H. (1997). The PKR protein kinase--an interferon-inducible regulator of cell growth and differentiation. Int J Hematol 65, 227-237.

Shipp, W., and Haselkorn, R. (1964). Double-Stranded Rna from Tobacco Leaves Infected with Tmv. Proc Natl Acad Sci U S A 52, 401-408.

Shope, R. E. (1966). An antiviral substance from Penicillium funiculosum. IV. Inquiry into the mechanism by which helenine exerts its antiviral effect. J Exp Med 123, 213-227.

Smirnov, Y. A., Rodrigues-Molto, M. P., and Famadas, M. T. (1983). Protein-RNA interaction in encephalomyocarditis virus as revealed by UV light-induced covalent linkages. J Virol 45, 1048-1055.

Stark, G. R., Kerr, I. M., Williams, B. R., Silverman, R. H., and Schreiber, R. D. (1998). How cells respond to interferons. Annu Rev Biochem 67, 227-264.

Steer, S. A., and Corbett, J. A. (2003). The role and regulation of COX-2 during viral infection. Viral Immunol 16, 447-460.

Steer, S. A., Moran, J. M., Maggi, L. B., Jr., Buller, R. M., Perlman, H., and Corbett, J. A. (2003). Regulation of cyclooxygenase-2 expression by macrophages in response to double-stranded RNA and viral infection. J Immunol 170, 1070-1076.

Tamanini, A., Rolfini, R., Nicolis, E., Melotti, P., and Cabrini, G. (2003). MAP kinases and NF-kappaB collaborate to induce ICAM-1 gene expression in the early phase of adenovirus infection. Virology 307, 228-242.

Terenzi, F., deVeer, M. J., Ying, H., Restifo, N. P., Williams, B. R., and Silverman, R. H. (1999). The antiviral enzymes PKR and RNase L suppress gene expression from viral and non-viral based vectors. Nucleic Acids Res 27, 4369-4375.

Thomis, D. C., and Samuel, C. E. (1993). Mechanism of interferon action: evidence for intermolecular autophosphorylation and autoactivation of the interferon-induced, RNA-dependent protein kinase PKR. J Virol 67, 7695-7700.

Tytell, A. A., Lampson, G. P., Field, A. K., and Hilleman, M. R. (1967). Inducers of interferon and host resistance. 3. Double-stranded RNA from reovirus type 3 virions (reo 3-RNA). Proc Natl Acad Sci U S A 58, 1719-1722.

Vilcek, J., Ng, M. H., Friedman-Kien, A. E., and Krawciw, T. (1968). Induction of interferon synthesis by synthetic double-stranded polynucleotides. J Virol 2, 648-650.

Visvanathan, K. V., and Goodbourn, S. (1989). Double-stranded RNA activates binding of NF-kappa B to an inducible element in the human beta-interferon promoter. Embo J 8, 1129-1138.

Williams, B. R. (1999). PKR; a sentinel kinase for cellular stress. Oncogene 18, 6112-6120.

Williams, S. D., and Ford, D. A. (1997). Activation of myocardial cAMP-dependent protein kinase by lysoplasmenylcholine. FEBS Lett 420, 33-38.

Yang, Y. L., Reis, L. F., Pavlovic, J., Aguzzi, A., Schafer, R., Kumar, A., Williams, B. R., Aguet, M., and Weissmann, C. (1995). Deficient signaling in mice devoid of double-stranded RNA-dependent protein kinase. Embo J 14, 6095-6106.

Yoneyama, M., Kikuchi, M., Natsukawa, T., Shinobu, N., Imaizumi, T., Miyagishi, M., Taira, K., Akira, S., and Fujita, T. (2004). The RNA helicase RIG-I has an essential function in double-stranded RNA-induced innate antiviral responses. Nat Immunol 5, 730-737.

Youngner, J. S., and Hallum, J. V. (1968). Interferon production in mice by double-stranded synthetic polynucleotides: induction or release? Virology 35, 177-179.

Zamanian-Daryoush, M., Mogensen, T. H., DiDonato, J. A., and Williams, B. R. (2000). NF-kappaB activation by double-stranded-RNA-activated protein kinase (PKR) is mediated through NF-kappaB-inducing kinase and IkappaB kinase. Mol Cell Biol 20, 1278-1290.

Zhu, H., Cong, J. P., Yu, D., Bresnahan, W. A., and Shenk, T. E. (2002). Inhibition of cyclooxygenase 2 blocks human cytomegalovirus replication. Proc Natl Acad Sci U S A 99, 3932-3937.

Zinn, K., Keller, A., Whittemore, L. A., and Maniatis, T. (1988). 2-Aminopurine selectively inhibits the induction of beta-interferon, c-fos, and c-myc gene expression. Science 240, 210-213.