Avian reovirus morphogenesis occurs within viral factories and begins with the selective recruitment of sigmaNS and lambdaA to microNS inclusions

We have recently shown that the avian reovirus non-structural protein microNS forms cytoplasmic inclusions in transfected cells and recruits sigmaNS to these structures. In the present study we further demonstrate that microNS mediates the association of the major core protein lambdaA, but not of sigmaA or sigmaC, with inclusions, indicating that the recruitment of viral proteins into avian reovirus factories has specificity. Thus, some proteins appear to be initially recruited to factories by association with microNS, whereas others are recruited subsequently through interaction with as-yet-unknown factors. We next used metabolic pulse-chase radiolabeling combined with cell fractionation and antibody immunoprecipitation to study the recruitment of newly synthesized viral polypeptides into viral factories and virus particles. The results of this combined approach revealed that avian reovirus morphogenesis is a complex and temporally controlled process that takes place exclusively within globular viral factories that are not microtubule-associated. Our findings further suggest that cores are assembled within the first 30 minutes after the synthesis of their polypeptide components, and that reovirion morphogenesis is completed over the next 30 minutes by the subsequent addition of outer capsid proteins.     

Reovirus core protein mu2 determines the filamentous morphology of viral inclusion bodies by interacting with and stabilizing microtubules

Cells infected with mammalian reoviruses often contain large perinuclear inclusion bodies, or "factories," where viral replication and assembly are thought to occur. Here, we report a viral strain difference in the morphology of these inclusions: filamentous inclusions formed in cells infected with reovirus type 1 Lang (T1L), whereas globular inclusions formed in cells infected with our laboratory's isolate of reovirus type 3 Dearing (T3D). Examination by immunofluorescence microscopy revealed the filamentous inclusions to be colinear with microtubules (MTs). The filamentous distribution was dependent on an intact MT network, as depolymerization of MTs early after infection caused globular inclusions to form. The inclusion phenotypes of T1L x T3D reassortant viruses identified the viral M1 genome segment as the primary genetic determinant of the strain difference in inclusion morphology. Filamentous inclusions were seen with 21 of 22 other reovirus strains, including an isolate of T3D obtained from another laboratory. When the mu2 proteins derived from T1L and the other laboratory's T3D isolate were expressed after transfection of their cloned M1 genes, they associated with filamentous structures that colocalized with MTs, whereas the mu2 protein derived from our laboratory's T3D isolate did not. MTs were stabilized in cells infected with the viruses that induced filamentous inclusions and after transfection with the M1 genes derived from those viruses. Evidence for MT stabilization included bundling and hyperacetylation of alpha-tubulin, changes characteristically seen when MT-associated proteins (MAPs) are overexpressed. Sequencing of the M1 segments from the different T1L and T3D isolates revealed that a single-amino-acid difference at position 208 correlated with the inclusion morphology. Two mutant forms of mu2 with the changes Pro-208 to Ser in a background of T1L mu2 and Ser-208 to Pro in a background of T3D mu2 had MT association phenotypes opposite to those of the respective wild-type proteins. We conclude that the mu2 protein of most reovirus strains is a viral MAP and that it plays a key role in the formation and structural organization of reovirus inclusion bodies.     

Increased ubiquitination and other covariant phenotypes attributed to a strain- and temperature-dependent defect of reovirus core protein mu2

Reovirus replication and assembly are thought to occur within cytoplasmic inclusion bodies, which we call viral factories. A strain-dependent difference in the morphology of these structures reflects more effective microtubule association by the mu2 core proteins of some viral strains, which form filamentous factories, than by those of others, which form globular factories. For this report, we identified and characterized another strain-dependent attribute of the factories, namely, the extent to which they colocalized with conjugated ubiquitin (cUb). Among 16 laboratory strains and field isolates, the extent of factory costaining for cUb paralleled factory morphology, with globular strains exhibiting higher levels by far. In reassortant viruses, factory costaining for cUb mapped primarily to the mu2-encoding M1 genome segment, although contributions by the lambda3- and lambda2-encoding L1 and L2 genome segments were also evident. Immunoprecipitations revealed that cells infected with globular strains contained higher levels of ubiquitinated mu2 (Ub-mu2). In M1-transfected cells, cUb commonly colocalized with aggregates formed by mu2 from globular strains but not with microtubules coated by mu2 from filamentous strains, and immunoprecipitations revealed that mu2 from globular strains displayed higher levels of Ub-mu2. Allelic changes at mu2 residue 208 determined these differences. Nocodazole treatment of cells infected with filamentous strains resulted in globular factories that still showed low levels of costaining for cUb, indicating that higher levels of costaining were not a direct result of decreased microtubule association. The factories of globular strains, or their mu2 proteins expressed in transfected cells, were furthermore shown to gain microtubule association and to lose colocalization with cUb when cells were grown at reduced temperature. From the sum of these findings, we propose that mu2 from globular strains is more prone to temperature-dependent misfolding and as a result displays increased aggregation, increased levels of Ub-mu2, and decreased association with microtubules. Because so few of the viral strains formed factories that were regularly associated with ubiquitinated proteins, we conclude that reovirus factories are generally distinct from cellular aggresomes.     

Genome packaging of reovirus is mediated by the scaffolding property of the microtubule network

Reovirus replication occurs in the cytoplasm of the host cell, in virally induced mini‐organelles called virus factories. On the basis of the serotype of the virus, the virus factories can manifest as filamentous (type 1 Lang strain) or globular structures (type 3 Dearing strain). The filamentous factories morphology is dependent on the microtubule cytoskeleton; however, the exact function of the microtubule network in virus replication remains unknown. Using a combination of fluorescent microscopy, electron microscopy, and tomography of high‐pressure frozen and freeze‐substituted cells, we determined the ultrastructural organisation of reovirus factories. Cells infected with the reovirus microtubule‐dependent strain display paracrystalline arrays of progeny virions resulting from their tiered organisation around microtubule filaments. On the contrary, in cells infected with the microtubule‐independent strain, progeny virions lacked organisation. Conversely to the microtubule‐dependent strain, around half of the viral particles present in these viral factories did not contain genomes (genome‐less particles). Complementarily, interference with the microtubule filaments in cells infected with the microtubule‐dependent strain resulted in a significant increase of genome‐less particle number. This decrease of genome packaging efficiency could be rescued by rerouting viral factories on the actin cytoskeleton. These findings demonstrate that the scaffolding properties of the microtubule, and not biochemical nature of tubulin, are critical determinants for reovirus efficient genome packaging. This work establishes, for the first time, a functional correlation between ultrastructural organisation of reovirus factories with genome packaging efficiency and provides novel information on how viruses coordinate assembly of progeny particles.     

Reovirus nonstructural protein mu NS recruits viral core surface proteins and entering core particles to factory-like inclusions

Mammalian reoviruses are thought to assemble and replicate within cytoplasmic, nonmembranous structures called viral factories. The viral nonstructural protein mu NS forms factory-like globular inclusions when expressed in the absence of other viral proteins and binds to the surfaces of the viral core particles in vitro. Given these previous observations, we hypothesized that one or more of the core surface proteins may be recruited to viral factories through specific associations with mu NS. We found that all three of these proteins--lambda 1, lambda 2, and sigma 2--localized to factories in infected cells but were diffusely distributed through the cytoplasm and nucleus when each was separately expressed in the absence of other viral proteins. When separately coexpressed with mu NS, on the other hand, each core surface protein colocalized with mu NS in globular inclusions, supporting the initial hypothesis. We also found that lambda 1, lambda 2, and sigma 2 each localized to filamentous inclusions formed upon the coexpression of mu NS and mu 2, a structurally minor core protein that associates with microtubules. The first 40 residues of mu NS, which are required for association with mu 2 and the RNA-binding nonstructural protein sigma NS, were not required for association with any of the three core surface proteins. When coexpressed with mu 2 in the absence of mu NS, each of the core surface proteins was diffusely distributed and displayed only sporadic, weak associations with mu 2 on filaments. Many of the core particles that entered the cytoplasm of cycloheximide-treated cells following entry and partial uncoating were recruited to inclusions of mu NS that had been preformed in those cells, providing evidence that mu NS can bind to the surfaces of cores in vivo. These findings expand a model for how viral and cellular components are recruited to the viral factories in infected cells and provide further evidence for the central but distinct roles of viral proteins mu NS and mu 2 in this process.     

Reovirus sigma NS protein localizes to inclusions through an association requiring the mu NS amino terminus

Cells infected with mammalian reoviruses contain phase-dense inclusions, called viral factories, in which viral replication and assembly are thought to occur. The major reovirus nonstructural protein mu NS forms morphologically similar phase-dense inclusions when expressed in the absence of other viral proteins, suggesting it is a primary determinant of factory formation. In this study we examined the localization of the other major reovirus nonstructural protein, sigma NS. Although sigma NS colocalized with mu NS in viral factories during infection, it was distributed diffusely throughout the cell when expressed in the absence of mu NS. When coexpressed with mu NS, sigma NS was redistributed and colocalized with mu NS inclusions, indicating that the two proteins associate in the absence of other viral proteins and suggesting that this association may mediate the localization of sigma NS to viral factories in infected cells. We have previously shown that mu NS residues 1 to 40 or 41 are both necessary and sufficient for mu NS association with the viral microtubule-associated protein mu 2. In the present study we found that this same region of micro NS is required for its association with sigma NS. We further dissected this region, identifying residues 1 to 13 of mu NS as necessary for association with sigma NS, but not with mu 2. Deletion of sigma NS residues 1 to 11, which we have previously shown to be required for RNA binding by that protein, resulted in diminished association of sigma NS with mu NS. Furthermore, when treated with RNase, a large portion of sigma NS was released from mu NS coimmunoprecipitates, suggesting that RNA contributes to their association. The results of this study provide further evidence that mu NS plays a key role in forming the reovirus factories and recruiting other components to them.     

Serotype-dependent induction of pulmonary neutrophilia and inflammatory cytokine gene expression by reovirus.

Reovirus type 3 Dearing (T3D) causes a prominent neutrophil influx, substantially greater than seen with reovirus type 1 Lang (T1L) in a rat model of viral pneumonia. We sought to measure reovirus-mediated increases in chemokine mRNA expression in pulmonary cells. We found that the neutrophilia induced by T1L and T3D infection in vivo correlated directly with increased levels of chemokine mRNA expression in T3D-infected compared with those of T1IL-infected lungs. In vitro, reovirus-infected normal alveolar macrophages (AMs) and the rat AM cell line NR8383 expressed greater levels of macrophage inflammatory protein 2, KC, and tumor necrosis factor alpha mRNA. A synergism between reovirus and lipopolysaccharide was also detected for macrophage inflammatory protein 2 and KC mRNA expression. Tumor necrosis factor protein secretion was also increased to a greater extent by T3D than by T1L in primary rat AMs and the NR8383 cells. We conclude that the virus-mediated inflammatory cytokine induction suggests a role for these cytokines in the neutrophil influx observed in the rat reovirus pneumonia model.     

Reovirus sigma NS and mu NS proteins form cytoplasmic inclusion structures in the absence of viral infection

Reovirus replication occurs in the cytoplasm of infected cells and culminates in the formation of crystalline arrays of progeny virions within viral inclusions. Two viral nonstructural proteins, sigma NS and micro NS, and structural protein sigma 3 form protein-RNA complexes early in reovirus infection. To better understand the minimal requirements of viral inclusion formation, we expressed sigma NS, mu NS, and sigma 3 alone and in combination in the absence of viral infection. In contrast to its concentration in inclusion structures during reovirus replication, sigma NS expressed in cells in the absence of infection is distributed diffusely throughout the cytoplasm and does not form structures that resemble viral inclusions. Expressed sigma NS is functional as it complements the defect in temperature-sensitive, sigma NS-mutant virus tsE320. In both transfected and infected cells, mu NS is found in punctate cytoplasmic structures and sigma 3 is distributed diffusely in the cytoplasm and the nucleus. The subcellular localization of mu NS and sigma 3 is not altered when the proteins are expressed together or with sigma NS. However, when expressed with micro NS, sigma NS colocalizes with mu NS to punctate structures similar in morphology to inclusion structures observed early in viral replication. During reovirus infection, both sigma NS and mu NS are detectable 4 h after adsorption and colocalize to punctate structures throughout the viral life cycle. In concordance with these results, sigma NS interacts with mu NS in a yeast two-hybrid assay and by coimmunoprecipitation analysis. These data suggest that sigma NS and mu NS are the minimal viral components required to form inclusions, which then recruit other reovirus proteins and RNA to initiate viral genome replication.     

The proapoptotic Bcl-2 protein Bax plays an important role in the pathogenesis of reovirus encephalitis

Encephalitis induced by reovirus serotype 3 (T3) strains results from the apoptotic death of infected neurons. Extrinsic apoptotic signaling is activated in reovirus-infected neurons in vitro and in vivo, but the role of intrinsic apoptosis signaling during encephalitis is largely unknown. Bax plays a key role in intrinsic apoptotic signaling in neurons by allowing the release of mitochondrial cytochrome c. We found Bax activation and cytochrome c release in neurons following infection of neonatal mice with T3 reoviruses. Bax(-/-) mice infected with T3 Abney (T3A) have reduced central nervous system (CNS) tissue injury and decreased apoptosis, despite viral replication that is similar to that in wild-type (WT) Bax(+/+) mice. In contrast, in the heart, T3A-infected Bax(-/-) mice have viral growth, caspase activation, and injury comparable to those in WT mice, indicating that the role of Bax in pathogenesis is organ specific. Nonmyocarditic T3 Dearing (T3D)-infected Bax(-/-) mice had delayed disease and enhanced survival compared to WT mice. T3D-infected Bax(-/-) mice had significantly lower viral titers and levels of activated caspase 3 in the brain despite unaffected transneuronal spread of virus. Cytochrome c and Smac release occurred in some reovirus-infected neurons in the absence of Bax; however, this was clearly reduced compared to levels seen in Bax(+/+) wild-type mice, indicating that Bax is necessary for efficient activation of proapoptotic mitochondrial signaling in infected neurons. Our studies suggest that Bax is important for reovirus growth and pathogenesis in neurons and that the intrinsic pathway of apoptosis, mediated by Bax, is important for full expression of disease, CNS tissue injury, apoptosis, and viral growth in the CNS of reovirus-infected mice.     

Reovirus sigmaNS protein is required for nucleation of viral assembly complexes and formation of viral inclusions

Progeny virions of mammalian reoviruses are assembled in the cytoplasm of infected cells at discrete sites termed viral inclusions. Studies of temperature-sensitive (ts) mutant viruses indicate that nonstructural protein sigmaNS and core protein mu2 are required for synthesis of double-stranded (ds) RNA, a process that occurs at sites of viral assembly. We used confocal immunofluorescence microscopy and ts mutant reoviruses to define the roles of sigmaNS and mu2 in viral inclusion formation. In cells infected with wild-type (wt) reovirus, sigmaNS and mu2 colocalize to large, perinuclear structures that correspond to viral inclusions. In cells infected at a nonpermissive temperature with sigmaNS-mutant virus tsE320, sigmaNS is distributed diffusely in the cytoplasm and mu2 is contained in small, punctate foci that do not resemble viral inclusions. In cells infected at a nonpermissive temperature with mu2-mutant virus tsH11.2, mu2 is distributed diffusely in the cytoplasm and the nucleus. However, sigmaNS localizes to discrete structures in the cytoplasm that contain other viral proteins and are morphologically indistinguishable from viral inclusions seen in cells infected with wt reovirus. Examination of cells infected with wt reovirus over a time course demonstrates that sigmaNS precedes mu2 in localization to viral inclusions. These findings suggest that viral RNA-protein complexes containing sigmaNS nucleate sites of viral replication to which other viral proteins, including mu2, are recruited to commence dsRNA synthesis.     

Cell receptors for the mammalian reovirus. IV. Reovirus-specific cytolytic T cell lines that have idiotypic receptors recognize anti-idiotypic B cell hybridomas

Cytotoxic T lymphocyte (Tc) cell lines specific for reovirus type 3 lysed an uninfected B cell hybridoma line, 87.92.6, that expresses and secretes an anti-idiotypic antibody that reacts with an anti-viral hemagglutinin monoclonal antibody, 9BG5. Monoclonal anti-idiotype 87.92.6 was shown by fluorescence analysis to specifically bind to reovirus Tc and to block reovirus-specific Tc from killing reovirus-infected target cells or the 87.92.6 hybridoma. An anti-LFA-1 monoclonal antibody, M17, interfered with Tc-mediated lysis of reovirus-infected targets and the 87.92.6 cells, indicating the similarity of cellular interactions mediated by LFA-1 structures when Tc bind to virally infected targets or 87.92.6 targets. Together with studies in which anti-H2 or monoclonal idiotypic antibodies were found to interfere with reovirus-specific Tc recognition of virally infected or 87.92.6 targets, these experiments indicate that some reovirus-specific Tc have conformations in their receptor that can be recognized by anti-idiotype.     

Junctional adhesion molecule a serves as a receptor for prototype and field-isolate strains of mammalian reovirus

Reovirus infections are initiated by the binding of viral attachment protein sigma1 to receptors on the surface of host cells. The sigma1 protein is an elongated fiber comprised of an N-terminal tail that inserts into the virion and a C-terminal head that extends from the virion surface. The prototype reovirus strains type 1 Lang/53 (T1L/53) and type 3 Dearing/55 (T3D/55) use junctional adhesion molecule A (JAM-A) as a receptor. The C-terminal half of the T3D/55 sigma1 protein interacts directly with JAM-A, but the determinants of receptor-binding specificity have not been identified. In this study, we investigated whether JAM-A also mediates the attachment of the prototype reovirus strain type 2 Jones/55 (T2J/55) and a panel of field-isolate strains representing each of the three serotypes. Antibodies specific for JAM-A were capable of inhibiting infections of HeLa cells by T1L/53, T2J/55, and T3D/55, demonstrating that strains of all three serotypes use JAM-A as a receptor. To corroborate these findings, we introduced JAM-A or the structurally related JAM family members JAM-B and JAM-C into Chinese hamster ovary cells, which are poorly permissive for reovirus infection. Both prototype and field-isolate reovirus strains were capable of infecting cells transfected with JAM-A but not those transfected with JAM-B or JAM-C. A sequence analysis of the sigma1-encoding S1 gene segment of the strains chosen for study revealed little conservation in the deduced sigma1 amino acid sequences among the three serotypes. This contrasts markedly with the observed sequence variability within each serotype, which is confined to a small number of amino acids. Mapping of these residues onto the crystal structure of sigma1 identified regions of conservation and variability, suggesting a likely mode of JAM-A binding via a conserved surface at the base of the sigma1 head domain.     

Carboxyl-proximal regions of reovirus nonstructural protein muNS necessary and sufficient for forming factory-like inclusions

Mammalian orthoreoviruses are believed to replicate in distinctive, cytoplasmic inclusion bodies, commonly called viral factories or viroplasms. The viral nonstructural protein muNS has been implicated in forming the matrix of these structures, as well as in recruiting other components to them for putative roles in genome replication and particle assembly. In this study, we sought to identify the regions of muNS that are involved in forming factory-like inclusions in transfected cells in the absence of infection or other viral proteins. Sequences in the carboxyl-terminal one-third of the 721-residue muNS protein were linked to this activity. Deletion of as few as eight residues from the carboxyl terminus of muNS resulted in loss of inclusion formation, suggesting that some portion of these residues is required for the phenotype. A region spanning residues 471 to 721 of muNS was the smallest one shown to be sufficient for forming factory-like inclusions. The region from positions 471 to 721 (471-721 region) includes both of two previously predicted coiled-coil segments in muNS, suggesting that one or both of these segments may also be required for inclusion formation. Deletion of the more amino-terminal one of the two predicted coiled-coil segments from the 471-721 region resulted in loss of the phenotype, although replacement of this segment with Aequorea victoria green fluorescent protein, which is known to weakly dimerize, largely restored inclusion formation. Sequences between the two predicted coiled-coil segments were also required for forming factory-like inclusions, and mutation of either one His residue (His570) or one Cys residue (Cys572) within these sequences disrupted the phenotype. The His and Cys residues are part of a small consensus motif that is conserved across muNS homologs from avian orthoreoviruses and aquareoviruses, suggesting this motif may have a common function in these related viruses. The inclusion-forming 471-721 region of muNS was shown to provide a useful platform for the presentation of peptides for studies of protein-protein association through colocalization to factory-like inclusions in transfected cells.     

Evidence that the sigma 1 protein of reovirus serotype 3 is a multimer.

In this report, we study the reovirus serotype 3 (strain Dearing) sigma 1 protein obtained from various sources: from Escherichia coli expressing sigma 1 protein, from reovirus-infected mouse L cells, and from purified reovirions. We demonstrate that the sigma 1 protein is a multimer in its undisrupted form and present biochemical evidence suggesting that the multimer is made up of four sigma 1 subunits.     

Adaptation of reovirus to growth in the presence of protease inhibitor E64 segregates with a mutation in the carboxy terminus of viral outer-capsid protein sigma3

Reovirus virions are internalized into cells by receptor-mediated endocytosis. Within the endocytic compartment, the viral outer capsid undergoes acid-dependent proteolysis leading to degradation of sigma3 protein and proteolytic cleavage of micro1/micro1C protein. E64 is a specific inhibitor of cysteine-containing proteases that blocks disassembly of reovirus virions. To identify domains in reovirus proteins that influence susceptibility to E64-mediated inhibition of disassembly, we selected variant viruses by serial passage of strain type 3 Dearing (T3D) in murine L929 cells treated with E64. E64-adapted variant viruses (D-EA viruses) produced 7- to 17-fold-greater yields than T3D did after infection of cells treated with 100 microM E64. Viral genes that segregate with growth of D-EA viruses in the presence of E64 were identified by using reassortant viruses isolated from independent crosses of E64-sensitive strain type 1 Lang and two prototype D-EA viruses. Growth of reassortant viruses in the presence of E64 segregated with the S4 gene, which encodes outer-capsid protein sigma3. Sequence analysis of S4 genes of three D-EA viruses isolated from independent passage series revealed a common tyrosine-to-histidine mutation at amino acid 354 in the deduced amino acid sequence of sigma3. Proteolysis of D-EA virions by endocytic protease cathepsin L occurred with faster kinetics than proteolysis of wild-type T3D virions. Treatment of D-EA virions, but not T3D virions, with cathepsin D resulted in proteolysis of sigma3, a property that also was found to segregate with the D-EA S4 gene. These results indicate that a region in sigma3 protein containing amino acid 354 influences susceptibility of sigma3 to proteolysis during reovirus disassembly.     

DNA binding of a nonstructural reovirus protein

The specific early inhibition of DNA synthesis in reovirus-infected cells suggests that the cell nucleus is a target for virus-induced damage. We have now examined the affinity of reovirus proteins for DNA, postulating that such affinity could provide a mechanism for the inhibition. Cytoplasmic and nuclear extracts of cells labeled with [35S] methionine from 6 to 8.5 h after infection at high multiplicity was subjected to chromatography on denatured DNA - cellulose columns. Fractions from both cytoplasm and nucleus eluted with 0.6 N NaCl contained a protein with the same electrophoretic mobility of polyacrylamide slab gels as the nonstructural (NS) reovirus protein of the sigma size class. The protein also exhibited affinity for native DNA - cellulose and denatured DNA - agarose. Electrophoretic analysis is tube gels of cell extracts labeled for 48 h before infection with [14C] leucine and from 6 to 8.5 h after infection with [3H] leucine showed increased 3H label in this protein indicating it is reovirus specific. Small amounts of mu proteins also had DNA affinity. Purified virus did not bind strongly to DNA, suggesting that the binding protein is not a structural protein of the sigma size class on the outer surface of the virus. Our results provide evidence that the sigma NS protein binds to DNA. This affinity could interfere with chromosome function in the infected cell.     

Translational stimulation by reovirus polypeptide sigma 3: substitution for VAI RNA and inhibition of phosphorylation of the alpha subunit of eukaryotic initiation factor 2.

COS cells transfected with plasmids that activate DAI depend on expression of virus-associated I (VAI) RNA to prevent the inhibitory effects of the alpha subunit of eukaryotic initiation factor 2 (eIF-2 alpha) kinase (DAI) and restore the translation of vector-derived dihydrofolate reductase mRNA. This VAI RNA requirement could be completely replaced by reovirus polypeptide sigma 3, consistent with its double-stranded RNA (dsRNA)-binding activity. S4 gene transfection of 293 cells also partially restored adenovirus protein synthesis after infection with the VAI-negative dl331 mutant. In dl331-infected 293 cells, eIF-2 alpha was present mainly in the acidic, phosphorylated form, and trans complementation with polypeptide sigma 3 or VAI RNA decreased the proportion of eIF-2 alpha (P) from approximately 85 to approximately 30%. Activation of DAI by addition of dsRNA to extracts of S4 DNA-transfected COS cells required 10-fold-higher levels of dsRNA than extracts made from cells that were not producing polypeptide sigma 3. In extracts of reovirus-infected mouse L cells, the concentration of dsRNA needed to activate DAI was dependent on the viral serotype used for the infection. Although the proportion of eIF-2 alpha (P) was greater than that in uninfected cells, most of the factor remained in the unphosphorylated form, even at 16 h after infection, consistent with the partial inhibition of host protein synthesis observed with all three viral serotypes. The results indicate that reovirus polypeptide sigma 3 participates in the regulation of protein synthesis by modulating DAI and eIF-2 alpha phosphorylation.