Reovirus nonstructural protein σ1s is required for establishment of viremia and systemic dissemination

Abstract

Serotype-specific patterns of reovirus disease in the CNS of newborn mice segregate with the viral S1 gene segment, which encodes attachment protein σ1 and nonstructural protein σ1s. The importance of receptor recognition in target cell selection by reovirus implicates the σ1 protein as the primary effector of disease outcome. However, the contribution of σ1s to reovirus disease is unknown. To define the function of σ1s in reovirus pathogenesis, we generated a σ1s-deficient virus by altering a single nucleotide to disrupt the σ1s translational start site. Viruses were recovered that contain nine gene segments from strain type 3 Dearing and either the wild-type or σ1s-null S1 gene segment from strain type 1 Lang. Following peroral inoculation of newborn mice, both viruses replicated in the intestine, although the wild-type virus achieved higher yields than the σ1s-null virus. However, unlike the wild-type virus, the σ1s-deficient virus failed to disseminate to sites of secondary viral replication, including the brain, heart, and liver. Within the small intestine, both viruses were detected in Peyer's patches, but only the wild-type virus reached the mesenteric lymph node. Concordantly, wild-type virus, but not σ1s-deficient virus, was detected in the blood of infected animals. Wild-type and σ1s-null viruses produced equivalent titers following intracranial inoculation, indicating that σ1s is dispensable for viral growth in the murine CNS. These results suggest a key role for σ1s in virus spread from intestinal lymphatics to the bloodstream, thereby allowing the establishment of viremia and dissemination to sites of secondary replication within the infected host.

Viruses can disseminate from mucosal sites of inoculation to peripheral tissues to produce organ-specific disease (1). Systemic dissemination requires that the virus effectively navigate diverse intracellular and extracellular environments to infect, replicate, and avoid immune detection and clearance in multiple cell and tissue types (1, 2). However, viral factors that mediate dissemination within the infected host are largely unknown. This gap in knowledge has precluded development of therapeutics designed specifically to interrupt viral spread.

Mammalian orthoreoviruses (reoviruses) are highly tractable models for studies of viral replication and pathogenesis. Reoviruses are nonenveloped, icosahedral viruses that contain 10 segments of double-stranded (ds) RNA (3). In newborn mice, reoviruses are highly virulent, causing injury to a variety of organs, including the central nervous system (CNS), heart, and liver (4). Following oral or intramuscular inoculation, strains of serotype 1 (T1) and serotype 3 (T3) reovirus invade the CNS but use different routes and produce distinct diseases. T1 reovirus spreads hematogenously and infects ependymal cells, causing ependymitis and hydrocephalus (5–7). In contrast, T3 reovirus spreads to the CNS by neural routes and infects neurons, causing lethal encephalitis (7, 8). These differences segregate with the S1 dsRNA gene segment (9, 10), which encodes viral attachment protein σ1 and nonstructural protein σ1s (11, 12). Since receptor recognition dictates target cell selection by many viruses, these studies implicate the σ1 attachment protein as the primary disease determinant. However, the contribution of σ1s to reovirus pathogenesis has not been systematically defined.

The σ1 ORF completely overlaps the σ1s coding sequence; however, the two proteins lie in different reading frames (11, 13, 14). Although little primary sequence identity exists between the σ1s proteins of the three reovirus serotypes, the σ1s ORF is retained by all strains sequenced to date (15), suggesting that σ1s performs an essential function in reovirus replication or transmission. The 14-kDa σ1s protein is expressed at low levels during infection in cell culture and distributes to the nucleus and cytoplasm in infected cells (16). Previous studies suggest that σ1s is involved in reovirus-induced cell cycle arrest (17, 18) and apoptosis (19).

In this study, we used reverse genetics to generate a σ1s-deficient T1 reovirus comprised of nine gene segments from strain type 3 Dearing (T3D) in combination with the S1 allele from strain type 1 Lang (T1L) and compared this virus to wild-type virus following infection of cultured cells and mice. We found that the σ1s-null mutant replicates with equivalent kinetics in cell culture and produces yields that approximate those of wild-type virus, indicating that the T1 σ1s protein is not required for reovirus replication. However, our studies reveal a critical role for σ1s in reovirus pathogenesis. We found that σ1s is required for reovirus spread from the site of inoculation to target organs. Although both wild-type and σ1s-null viruses are taken up by Peyer's patches (PPs) in the intestine, only the wild-type virus replicates in the mesenteric lymph node (MLN) and invades the blood stream. These findings indicate that σ1s is required for the establishment of viremia, which is essential for reovirus to access sites that support secondary viral replication in the host.

Results

Construction and Characterization of a σ1s-Deficient Reovirus.

To define the function of σ1s in reovirus pathogenesis, we generated a recombinant reovirus incapable of expressing σ1s using plasmid-based reverse genetics (20) (SI Text). The start codon for the σ1s open reading frame was disrupted by introducing a single nucleotide change (⁷⁵AUG to ⁷⁵ACG) into the plasmid encoding the cDNA of the S1 gene segment derived from strain T1L (Fig. 1A). Although this mutation alters the σ1s translational start site, the primary amino acid sequence of the overlapping σ1 open reading frame is unchanged. Viable viruses were recovered that contain nine gene segments from strain T3D and either the wild-type or σ1s-null S1 gene from T1L (rsT3D/T1LS1 and rsT3D/T1LS1 σ1s-null, respectively). These viruses were generated to facilitate studies to define the role of σ1s in reovirus pathogenesis following peroral inoculation. The T3D σ1 protein is cleaved and inactivated within the intestinal tract, whereas the T1L σ1 protein is resistant to the intestinal proteases that cleave T3D σ1 (21, 22). Analysis of genomic dsRNA purified from virions of each virus verified that the rescued viruses contain the expected combination of gene segments (Fig. 1B). The identity of the S1 gene segment and the integrity of the σ1s translational start site from each virus were confirmed by direct sequencing of viral RNA. The S1 genes of both strains contained no additional mutations.

Fig. 1.

Construction and characterization of a σ1s-deficient reovirus. (A) Schematic of the reovirus S1 gene segment. The σ1 ORF is shown in gray, the σ1s ORF in black, and the 5′ and 3′ UTRs in white. The wild-type (upper panel) and σ1s-null (lower panel) S1 gene alleles are shown. (B) Electrophoretic mobility of gene segments of parental and recombinant viruses. The S1 gene segments are indicated by arrows. Large (L), medium (M), and small (S) class genes are designated. (C) Expression of reovirus protiens by wild-type and σ1s-null viruses. Whole-cell lysates from infected cells were immunoblotted using antisera specific for T1L σ1s (upper panel), T1L σ1 (middle panel), or reovirus (lower panel). *, nonspecific bands indicate equivalent sample loading.

To confirm that alteration of the σ1s translational start site prevents synthesis of the σ1s protein, we assessed σ1s expression by immunoblotting using T1L σ1s-specific antiserum after infection of murine L929 cells with rsT3D/T1LS1 or rsT3D/T1LS1 σ1s-null (Fig. 1C). At 24 h post-infection, σ1s expression was detected in whole-cell lysates of cells infected with wild-type virus but not the σ1s-null mutant. In parallel, blotting with reovirus-specific polyclonal antiserum showed equivalent steady state levels of viral proteins in cells infected with both viruses, suggesting a comparable level of infection by the wild-type and σ1s-deficient strains. No differences in σ1 protein levels were observed between the wild-type and mutant viruses, indicating that the single nucleotide change in the σ1s-null mutant does not alter synthesis of the σ1 protein. Thus, rsT3D/T1LS1 and rsT3D/T1LS1 σ1s-null are isogenic viruses that vary solely in σ1s expression.

σ1s Is Not Required for Reovirus Replication in Cell Culture.

To determine whether σ1s influences T1 reovirus growth in cell culture, we quantified viral yields following infection of L929 cells (Fig. 2). At three MOIs tested, replication kinetics and yields of infectious progeny for the σ1s-null virus were indistinguishable from those of wild-type virus. Plaque size of the σ1s-null mutant and wild-type virus did not differ (Fig. S1). Following infection of HeLa cells and primary mouse embryo fibroblasts (MEFs), the σ1s-deficient and wild-type viruses did not differ in replication kinetics or viral yields (Fig. S2). These results demonstrate that σ1s is dispensable for T1 reovirus replication in cultured cells.

Fig. 2.

The σ1s protein is dispensable for reovirus growth in cell culture. L929 cells were adsorbed with rsT3D/T1LS1 or rsT3D/T1LS1 σ1s-null at MOIs of 0.01 (A), 0.1 (B), or 1 (C) PFU/cell. Titers of virus in cell lysates were determined by plaque assay at the indicated times post-infection. Results are expressed as viral yields for triplicate samples. Error bars indicate SD.

σ1s Is Required for Systemic Dissemination of Reovirus.

To determine whether σ1s influences reovirus pathogenesis, we inoculated newborn C57/BL6 mice perorally with 10⁴ PFU of rsT3D/T1LS1 or rsT3D/T1LS1 σ1s-null. At days 4, 8, and 12 post-inoculation, we quantified viral titers in selected organs by plaque assay (Fig. 3A). Both viruses replicated in the intestine, achieving peak titers in excess of 10⁴ PFU per intestine. However, yields of wild-type virus were greater than those of the σ1s-null mutant at each time point, suggesting that σ1s enhances viral replication at that site. As anticipated, the wild-type virus produced detectable yields in spleen, liver, heart, and brain. In sharp contrast, the σ1s-deficient virus was not detected at any site of secondary viral replication tested.

Fig. 3.

(A) The σ1s protein is required for systemic reovirus dissemination following peroral inoculation. Newborn C57/BL6 mice were inoculated perorally with 10⁴ PFU of rsT3D/T1LS1 or rsT3D/T1LS1 σ1s-null. At 4, 8, and 12 days post-inoculation, viral titers in the organs shown were determined by plaque assay. (B) The σ1s protein is not required for reovirus replication in the brain. Newborn C57/BL6 mice were inoculated intracranially with 100 PFU of rsT3D/T1LS1 or rsT3D/T1LS1 σ1s-null. At 4, 8, and 12 days post-inoculation, viral titers in the brain were determined by plaque assay. Results are expressed as mean viral titers for six to 10 animals for each time point. Error bars indicate SEM. *, P < 0.05 as determined by Mann-Whitney test in comparison to rsT3D/T1LS1.

We thought it possible that the lower yields of virus in the intestine may be insufficient to allow the σ1s-null virus to achieve a threshold required for systemic spread. To address this possibility, we inoculated newborn C57/BL6 mice perorally with a higher dose of virus, 10⁶ PFU of rsT3D/T1LS1 or rsT3D/T1LS1 σ1s-null, and quantified viral yields in target organs (Fig. S3). Even with a 100-fold increase in the inoculum, only a single animal of seven inoculated with rsT3D/T1LS1 σ1s-null harbored detectable virus in the liver, spleen, heart, and brain. Remarkably, the S1 gene sequence of virus recovered from those organs revealed that the disseminated virus had reverted from ⁷⁵ACG (mutant) to ⁷⁵AUG (wild-type), further underscoring the importance of σ1s to reovirus pathogenesis. The σ1s-null virus also did not produce detectable titer at sites of secondary viral replication following peroral inoculation of Swiss Webster ND4 mice (Fig. S4), suggesting that the function of σ1s is not restricted by the genetic background of the host. Thus, the σ1s-deficient virus has an intrinsic defect that prevents either viral dissemination or viral growth in target tissues.

Although the σ1s-null virus is capable of replication in the intestine (Fig. 3A), we thought it possible that σ1s expression may influence the types of cells infected in the gastrointestinal tract. To determine whether σ1s dictates reovirus tropism in the intestine, we compared histologic sections of intestines from mice inoculated perorally with 10⁷ PFU of rsT3D/T1LS1 or rsT3D/T1LS1 σ1s-null. Following infection with either wild-type or σ1s-null virus, reovirus antigen was detected in columnar epithelial cells lining the intestinal villi (Fig. S5). Reovirus antigen-positive cells localized primarily to the villus tips, with a small number of infected cells found in the intestinal crypts. These observations suggest that σ1s does not alter the site of reovirus replication in the gastrointestinal tract.

σ1s Is Dispensable for Reovirus Growth in the Murine CNS.

Failure to detect the σ1s-null virus in peripheral organs following peroral inoculation suggests that σ1s either facilitates delivery of reovirus to target tissues or is required for viral growth at sites of secondary replication. To distinguish between these possibilities, we inoculated newborn C57/BL6 mice intracranially with 100 PFU of rsT3D/T1LS1 or rsT3D/T1LS1 σ1s-null and determined viral titers in the brain at days 4, 8, and 12 post-inoculation (Fig. 3B). Titers of wild-type and σ1s-null viruses were equivalent at days 4 and 8 post-inoculation, indicating that σ1s is not required for reovirus growth in the brain when virus is inoculated intracranially. Titers of the σ1s-null virus were greater than those of wild-type virus at day 12 post-inoculation, perhaps reflecting enhanced clearance of wild-type virus from that site.

To determine whether σ1s promotes spread of virus from the brain following intracranial infection, we inoculated newborn C57/BL6 mice intracranially with 100 PFU of rsT3D/T1LS1 or rsT3D/T1LS1 σ1s-null and determined viral titers in selected organs at days 4, 8, and 12 post-inoculation (Fig. S6). Similar to findings made in studies of mice inoculated perorally, yields of wild-type virus were markedly higher than those of σ1s-deficient virus in each of the target sites tested. Wild-type virus was detected in the spleen, liver, heart, and intestine at all time points assessed. In contrast, the σ1s-deficient virus was found only in the spleen at day 8 and intestine at day 12 and at titers significantly less than those produced by wild-type virus. Analysis of S1 gene sequences of virus isolates from peripheral organs of mice inoculated intracranially with σ1s-null virus did not reveal reversion mutations, suggesting that σ1s-deficient virus can disseminate from the brain, albeit with markedly diminished efficiency. Therefore, σ1s substantially enhances systemic dissemination of reovirus regardless of the site of primary replication.

σ1s Enhances Virus Spread to the MLN and Is Required for Establishment of Viremia.

Following peroral inoculation, reovirus is transcytosed by intestinal microfold (M) cells and infects underlying PPs. Virus is then thought to traffic via afferent lymphatics to the MLN and bloodstream, where it disseminates systemically. To determine whether σ1s is required for reovirus dissemination within intestinal lymphatics, we inoculated newborn C57/BL6 mice perorally with 10⁷ PFU of rsT3D/T1LS1 or rsT3D/T1LS1 σ1s-null and assessed PP cells for reovirus antigen (Fig. 4 A–D) and quantified reovirus titer in the MLNs (Fig. 4E). Both the wild-type and σ1s-null viruses were detected in mononuclear cells within PPs at days 1 and 3 post-inoculation (Fig. 4 A–D). Although both viruses were capable of reaching the MLN at day 1 post-inoculation (Fig. 4E), titers of wild-type virus were significantly greater than those of σ1s-null virus at days 3 and 5 post-inoculation. These data indicate that σ1s is not required for reovirus transport through intestinal lymphatics, but it may enhance virus growth once it reaches the MLN.

Fig. 4.

(A–D) The σ1s protein is not required for reovirus uptake by PPs. Newborn C57/BL6 mice were inoculated perorally with 10⁷ PFU of rsT3D/T1LS1 or rsT3D/T1LS1 σ1s-null. At 1 and 3 days post-inoculation, PPs were isolated, sectioned, and stained with polyclonal reovirus-specific antiserum. Representative sections are shown for rsT3D/T1LS1 (A and C) and rsT3D/T1LS1 σ1s-null (B and D) at day 1 (A and B) and day 3 (C and D) post-inoculation. (E) The σ1s protein promotes reovirus spread to the MLN. Newborn C57/BL6 mice were inoculated perorally with 10⁷ PFU of rsT3D/T1LS1 or rsT3D/T1LS1 σ1s-null. At 1, 3, and 5 days post-inoculation, viral titers in the MLN were determined by plaque assay. Results are expressed as mean viral titers for four to nine animals for each time point. Error bars indicate SEM. *, P < 0.05 as determined by Mann-Whitney test in comparison to rsT3D/T1LS1 σ1s-null.

We thought it possible that reduced titers of the σ1s-null virus in the MLN might prevent establishment of viremia and hematogenous spread of reovirus to sites of secondary viral replication. To determine whether σ1s is required for hematogenous dissemination, we inoculated newborn C57/BL6 mice perorally with 10⁴ PFU of rsT3D/T1LS1 or rsT3D/T1LS1 σ1s-null and quantified virus titers in the blood at days 4, 8, and 12 post-inoculation (Table 1). Only wild-type virus was detectable in the blood of infected animals. The σ1s-null virus was not found in the blood of any animal tested. Similar results were obtained when viral titers in blood were assessed following IC inoculation with 100 PFU of rsT3D/T1LS1 or rsT3D/T1LS1 σ1s-null (Table S1). Thus, regardless of the route of inoculation, σ1s is required by reovirus to establish viremia.

Table 1.

Viremia following peroral inoculation^*

Virus strain	Time post-infection, days	Number of viremic mice positive/total
rsT3D/T1LS1	4	0/4
	8	3/4
	12	2/8
rsT3D/T1LS1 σ1s-null	4	0/4
	8	0/4
	12	0/4

*Newborn C57/BL6 mice were inoculated perorally with 10⁴ PFU of rsT3D/T1LS1 or rsT3D/T1LS1 σ1s-null. At 4, 8, and 12 days post-inoculation, mice were euthanized, blood was collected, and viral titers in blood were determined by plaque assay.

Discussion

Although nonstructural protein σ1s is expressed by all three reovirus serotypes (13, 16, 23–25), the role of σ1s in reovirus replication and pathogenesis is not well defined. Database searches do not identify proteins with significant primary sequence similarity to σ1s, and structural information about σ1s is not available. Consequently, inferences about σ1s function cannot be made by comparison to other proteins. Previous studies revealed that σ1s is not required for reovirus replication in cell culture (16) and suggest a role for σ1s in reovirus-induced cell cycle arrest (17, 18), apoptosis (19), and neurovirulence (19). However, interpreting previous studies of σ1s function is complicated by the use of a σ1s-null mutant virus that is not isogenic to the parental strain from which the mutant was derived. The recent development of a reverse genetics system for reovirus (20) allowed us to generate a mutant reovirus that does not express σ1s due to the introduction of a single nucleotide change that disrupts the translational initiation codon for the σ1s ORF. The resulting σ1s-null virus is viable, and single-cycle replication kinetics indicate that the σ1s protein is dispensable for reovirus replication in cell culture. However, our studies indicate that σ1s is required for reovirus spread from the point of inoculation to the bloodstream, which allows for viral dissemination to sites of secondary replication.

We envision three possibilities to explain how the σ1s protein promotes reovirus dissemination. First, although our results clearly show that σ1s is dispensable for reovirus growth in the intestine and brain (Fig. 3), σ1s may be required for viral replication in a cell type that is required for reovirus spread. For example, σ1s may be essential for reovirus growth in leukocytes or other cell types required to transport reovirus to the blood. Second, σ1s may antagonize the host response to reovirus infection. Many viruses encode nonstructural proteins that disrupt innate and adaptive defense mechanisms to evade immune detection and viral clearance. For example, rotavirus NSP1 antagonizes innate immune responses by mediating degradation of IFN-regulatory factors (26). Human cytomegalovirus gene product US11 promotes destruction of MHC class I molecules to avoid detection of virus-infected cells by cytotoxic T lymphocytes (27). If σ1s counteracts one or more components of the host antiviral response, failure of the σ1s-null virus to spread from the site of inoculation at early times following infection would suggest that innate immune mechanisms are a likely target, since the block to dissemination occurs before the onset of effector B- and T-cell responses. Third, σ1s may be required for viral egress from infected cells. Although not incorporated into virions, σ1s may be required for efficient release of infectious progeny. Nonstructural protein 3 (NS3) of bluetongue virus (BTV), another member of the Reoviridae, interacts with the host cell ESCRT machinery to facilitate BTV egress via a unique exocytic pathway (28, 29). However, unlike NS3, σ1s is not an integral membrane protein, and reovirus does not use a comparable egress pathway (28–30). Although wild-type and σ1s-null viruses do not differ in egress from cultured cells (SI Text and Fig. S1), it is possible that σ1s is required for virus release from cells in vivo.

Previous studies have suggested a role for σ1s in reovirus-induced cell cycle arrest at the G₂/M boundary (17, 18). However, it is not known whether reovirus induces cell cycle arrest in vivo or whether cell cycle dysregulation contributes to reovirus pathogenesis. In addition, σ1s has been implicated in reovirus-induced apoptosis (19). However, wild-type and σ1s-null viruses do not display differences in apoptotic capacity in cultured cells (SI Text and Fig. S7). Algorithms that predict functional sites within proteins based on amino acid sequence similarities identify several potential interaction motifs within σ1s for cellular proteins and a number of possible phosphorylation sites for cellular kinases (31–33). Several of the potential σ1s interacting partners identified by these searches are proteins associated with cell cycle regulation. Future experiments will focus on determining whether reovirus-induced cell cycle arrest contributes to reovirus pathogenesis.

Understanding mechanisms by which σ1s enhances reovirus dissemination also may have important consequences for applications of reovirus as an oncolytic agent. Transformed cells are much more permissive for reovirus infection (34, 35), and phase II clinical trials are underway to test reovirus as an adjunct to conventional cancer therapy (34, 36, 37). The recently developed reverse genetics system for reovirus permits the design of reovirus-based therapeutics that incorporate mutations to enhance vector safety and efficacy (20). Failure of the σ1s-null virus to disseminate following either peroral or intracranial infection suggests that σ1s-deficient reoviruses will remain localized to the site of inoculation regardless of the route of entry. While it is possible that viruses lacking σ1s could be less effective at combating metastases due to failure to escape the site of primary replication, clinical use of σ1s-deficient viruses could be an important safeguard to limit potential complications from this therapy.

In this study, we identified nonstructural protein σ1s as a key viral determinant of reovirus pathogenesis that is required for the establishment of viremia and systemic dissemination. The capacity to spread systemically correlates with increased pathogenicity for many viruses, including influenza virus (38, 39), severe acute respiratory syndrome coronavirus (40), and poliovirus (41). However, mechanisms that underlie systemic viral dissemination are largely unknown. Understanding how reovirus disseminates may inform studies of other viral pathogens capable of systemic spread and contribute to development of therapeutics designed specifically to interrupt this process.

Materials and Methods

Cells and Viruses.

Murine L929 cells were maintained in Joklik's minimum essential medium (MEM) supplemented to contain 10% FBS, 2 mM L-glutamine (L-Glu), 100 U/mL penicillin, 100 μg/mL streptomycin, and 25 ng/mL amphotericin B (Invitrogen). HeLa cells were maintained in DMEM (Invitrogen) supplemented to contain 10% FBS, 2 mM L-Glu, 100 U/mL penicillin, 100 μg/mL streptomycin, and 25 ng/mL amphotericin B. MEFs were derived from C57/BL6 mice (42) and maintained in DMEM supplemented to contain 10% FBS, 1× MEM nonessential amino acids (Invitrogen), 2 mM L-Glu, 0.1 mM 2-mercaptoethanol (Sigma), 20 mM HEPES, 100 U/mL penicillin, 100 μg/mL streptomycin, and 0.25 μg/mL amphotericin B. Cells at passages 4–5 were used for the experiments in this study.

Recombinant reoviruses were generated as described in the SI Text. Purified reovirus virions were generated using second- or third-passage L-cell lysates stocks of twice-plaque-purified reovirus (43). Viral particles were purified as described (44). Viral titers were determined by plaque assay using L929 cells (45). Electrophoretic mobility of viral dsRNA gene segments was verified using genomic dsRNA purified from CsCl-banded purified virions and 10% polyacrylamide gels (46).

Generation of Reovirus-Specific Antiserum.

Rabbit polyclonal anti-σ1s serum was generated by Antibody Research Corporation. Two New Zealand white rabbits were immunized with a peptide corresponding to the C-terminal region of strain T1L σ1s protein (DEHPLTRQMLEAYGQN) conjugated to keyhole limpet hemocyanin. Rabbits were immunized with peptide four times at 2-week intervals. Purified IgG was obtained by affinity purification with the immunizing peptide.

Rabbit polyclonal anti-σ1 serum was generated by Cocalico Biologicals. A single New Zealand white rabbit was immunized with the T1L σ1 C-terminal head domain. Purification of the T1L σ1 head domain is described in the SI Text. The rabbit was immunized with protein and boosted at 2, 3, and 7 weeks post-immunization.

Immunoblot Assay.

Monolayers of L929 cells in 60-mm dishes (Corning) were either adsorbed with reovirus at an MOI of 100 PFU/cell or mock-infected at room temperature for 1 h in PBS, washed once with PBS, and incubated in serum-containing medium at 37 °C. At 24 h post-adsorption, cells were lysed with 1× RIPA buffer (20 mM Tris-HCl, pH 7.5, 150 mM NaCl, 1 mM EDTA, 1% Igepal, 0.1% SDS, 0.1% deoxycholate). Protein expression was assessed by SDS-PAGE and immunoblotting (47) using σ1s-, σ1-, or reovirus-specific polyclonal antiserum. Blots were visualized using an Odyssey infrared imaging system (LiCor Biosciences).

Virus Replication.

Monolayers of cells in 24-well plates (Corning) were adsorbed in triplicate with reovirus at room temperature for 1 h in PBS, washed twice with PBS, and incubated at 37 °C in serum-containing medium for various intervals. Cells were frozen and thawed twice before determination of viral titer by plaque assay using L929 cells (45). Viral yields were calculated according to the following formula: Log₁₀yield_tx = log₁₀(PFU/mL)_tx - log₁₀(PFU/mL)_t0, where t_x is the time post-infection.

Infection of Mice.

C57BL/6J mice were obtained from Jackson Laboratory, and Swiss-Webster ND4 mice were obtained from Harlan Biosciences. Newborn mice weighing 1.5–2 g were inoculated perorally or intracranially with purified reovirus diluted in PBS as described (42, 48, 49). Blood was collected for analysis of viremia as described (42). Viral titers in organ homogenates were determined by plaque assay (45). For immunohistochemical analysis, mice were euthanized at various intervals post-inoculation, and organs were resected and fixed overnight in 10% formalin, followed by 70% ethanol. Fixed organs were embedded in paraffin, and 6-μm histological sections were prepared. Sections were processed for detection of reovirus protein. Animal husbandry and experimental procedures were performed in accordance with Public Health Service policy and approved by the Vanderbilt University School of Medicine Institutional Animal Care and Use Committee.

Statistical Analysis.

For experiments in which viral titers were determined in an organ or blood, the Mann-Whitney test was used to calculate two-tailed P values. This test is appropriate for experimental data that display a non-Gaussian distribution (50). When all values are less than the limit of detection, a Mann-Whitney test P value cannot be calculated. All statistical analyses were performed using Prism software (GraphPad Software).