Fig 1.
Reovirus protein co-fractionates with EV-enriched fractions released from cells regardless of plasma membrane integrity phenotype.
L cells were adsorbed with three individual clones of T1L or T3D reovirus at an MOI of 1 PFU/cell. (A) Every 24 h, cell lysates were collected, and virus in the supernatant was quantified by FFA. Error bars indicate SD. n = 3. *, P < 0.05; **, P < 0.01; ***, P < 0.001 by two-sample unpaired T test. (B) Cell membrane disruption was quantified for T1L-, T3D-, and mock-infected cells every 24 h for 96 h using trypan blue staining. Error bars indicate SD. n = 3. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001 by one-way ANOVA with Tukey’s multiple comparisons. (C) Schematic showing the EV fraction enrichment protocol described in the text. Created using Biorender.com. (D-G) Infected-cell supernatants were collected every 24 h for 96 h. Mock-infected supernatant was collected at 96 h, but reovirus protein was not detected. Reovirus protein association with large EV, medium EV, and small EV/free virus fractions was quantified following SDS-PAGE and immunoblotting. Representative immunoblots probed using reovirus antiserum for T1L (D) and T3D (F) and graphs showing results quantified from three independent immunoblots for T1L (E) and T3D (G) are shown. Asterisk denotes the reovirus λ3 protein band used for quantitation. Error bars indicate SD. n = 3. *, P < 0.05; ***, P < 0.001 by one-way ANOVA with Tukey’s multiple comparisons prior to normalization. Protein signal was normalized as a percentage of maximum by dividing each adjusted volume value by the highest measured value within the blot. (H) Infected-cell supernatants were harvested at 72 h, and viral infectious units associated with each EV fraction were quantified by plaque assay. Error bars indicate SD. n = 3. **, P < 0.01 by two-way ANOVA with Tukey’s multiple comparisons.
Fig 2.
Extracellular reovirus visually associate with large and medium EVs.
L cells were adsorbed with reovirus strain T1L or T3D at an MOI of 1 PFU/cell for 72 h. Cell supernatants were collected and sequentially centrifuged to enrich for large EVs (A-B) or medium EVs (C-D), which were visualized using negative-stain EM. Arrowheads indicate viral particles. Scale bars = 200 nm.
Fig 3.
Extracellular reovirus particles fail to associate with small EVs.
L cells were adsorbed with three individual clones of reovirus strains T1L or T3D at an MOI of 1 PFU/cell for 72 h. Cell debris, large EVs, and medium EVs were cleared from infected-cell supernatants, as in Fig 1C. (A) The resulting supernatant was centrifuged on a 60% iodixanol cushion to concentrate small EVs and free virus particles. The pellet was loaded onto a 5–40% iodixanol gradient. Twelve 1-ml fractions were collected and analyzed. Created using Biorender.com. (B-E) T1L-infected (B-C) or T3D-infected (D-E) iodixanol gradient fractions were resolved using SDS-PAGE and immunoblotting to detect CD81 (green) and reovirus proteins (red). (B, D). Relative CD81 and reovirus protein signals in fractions 6–12 were quantified in three independent experiments. Asterisk denotes the reovirus λ3 protein band used for quantitation. (C, E). Error bars indicate SD. (F-K) Contents of fractions 7 (F, G), 10 (H, I), and 11 (J, K) were imaged using negative-stain EM. The reovirus strain used to infect the cells from which gradient-separated supernatant fractions were collected is indicated. Scale bars = 200 nm.
Fig 4.
Multiple reovirus particles can be transported by protective, medium-sized EVs and non-protective large EVs.
(A) L cells were adsorbed with three individual clones of reovirus strains T1L or T3D at an MOI of 1 PFU/cell. After 72 h, EV-associated and free reovirus particles were enriched using sequential centrifugation and iodixanol gradient separation, as previously described, then incubated with σ1-specific reovirus antiserum (treated) or with diluent (untreated). Infectious units were quantified by plaque assay. Created using Biorender.com. (B-E) Plaque titer (B, D) and percent relative infectivity, quantified by dividing the treated infectious units by mock-treated infectious units and multiplying by 100 (C, E), for each sample are shown. Error bars indicate SD. “n.d.” = not detected. n = two titers per sample in each of three independent experiments. *, P < 0.05; ****, P < 0.0001 by two-sample unpaired T test. (F) L cells were coinfected with independent inocula of WT or BC T3D reovirus at an MOI of 10 PFU/cell. At 24 h p.i., large EV, medium EV, and small EV/free virus fractions were harvested from the supernatants using sequential centrifugation and subsequently used to inoculate a plaque assay. Plaques representing individual infectious units, which might be EV-associated bundles or free virus particles, were picked and amplified. Viral RNA was genotyped using HRM. Created using Biorender.com. (G) Normalized melt curves for control RNA from WT (red), BC (blue), and 2:1, 1:1, and 1:2 mixtures of WT and BC (green) RNA are shown. (H) Genotype quantitation for the large EV, medium EV, and small EV/free virus fractions as a percentage of total plaques analyzed. Error bars indicate SD. n = 24 plaques represented by each data point in four independent experiments. **, P < 0.01; ***, P < 0.001 by Pearson’s chi-squared analysis with pairwise comparisons.
Fig 5.
EV-mediated reovirus egress occurs in multiple cell types, but protection from neutralization may be strain dependent.
(A-B) Caco-2 cells were infected with three individual clones of T1L (A) or T3D (B) at an MOI of 5 PFU/cell (Caco-2), as quantified in L cells. At the indicated timepoints, cell lysate was collected, and virus titers were determined by FFA. Results for T1L or T3D infection of L cells at an MOI of 1 PFU/cell are duplicated from Fig 1A for comparison. Error bars indicate SD. n = 3. *, P < 0.05; **, P < 0.01; ***, P < 0.001 by two-sample unpaired T test. (C) Caco-2 cells were adsorbed with three individual clones each of reovirus strains T1L or T3D at an MOI of 5 PFU/cell. Cell membrane disruption was quantified for T1L-, T3D-, and mock-infected cells every 24 h for 96 h using trypan blue staining. Error bars indicate SD. n = 3. *, P < 0.05; **, P < 0.01; ****, P < 0.0001 by one-way ANOVA with Tukey’s multiple comparisons. (D—E) Infected Caco-2 cell supernatants were collected every 24 h for 96 h. Mock-infected supernatant was collected at 96 h, but reovirus protein was not detected. Reovirus protein association with large EV, medium EV, and small EV/free virus fractions was quantified following SDS-PAGE and immunoblotting for T1L (D) or T3D (E). Error bars indicate SD. n = 3. *, P < 0.05 by one-way ANOVA with Tukey’s multiple comparisons prior to normalization. Protein signal was normalized as a percentage of maximum by dividing each adjusted volume value by the highest measured value within the blot. (F) Infected-cell supernatants were harvested at 72 h, and viral infectious units associated with each EV fraction were quantified by plaque assay. Error bars indicate SD. n = 3. *, P < 0.05 by two-way ANOVA with Tukey’s multiple comparisons. (G–J) Caco-2 cells were adsorbed with three individual clones of reovirus strains T1L or T3D, as indicated, at an MOI of 1 PFU/cell, and supernatants were fractionated and treated as described in Fig 4A. Infectious units were quantified by plaque assay. Plaque titer (G, I) and relative percent infectivity, quantified by dividing the treated infectious units by mock-treated infectious units and multiplying by 100 (H, J), for each sample are shown. Error bars indicate SD. “n.d.” = not detected. n = two titers per sample in each of three independent experiments. *, P < 0.05; ****, P < 0.0001 by two-sample unpaired T test.
Fig 6.
Reovirus infection enhances EV release compared to uninfected cells.
L cells were adsorbed with media (mock; M) or with three individual clones of T1L or T3D reovirus at an MOI of 1 PFU/cell for 72 h. (A-C) Fractions enriched in large, medium, and small EVs were harvested from supernatants using sequential centrifugation, as previously described, then lysed. Equal lysate volumes were resolved by SDS-PAGE and Coomassie staining (A), three independent experiments were quantified (B), and they were normalized by dividing the average virus-infected value by the average mock-infected value (C). (D-F) Fractions enriched in large, medium, and small EVs were harvested from supernatants using sequential centrifugation, as previously described. Then, EVs were immunoprecipitated using annexin V nanobeads, which bind to phosphatidylserine. Equal volumes of immunoprecipitated material were resolved by SDS-PAGE and Coomassie staining (D), three independent experiments were quantified (E), and normalized by dividing the average virus-infected value by the average mock-infected value (F). (G-H) Fractions enriched in large, medium, and small EVs were harvested from supernatants using sequential centrifugation, as previously described. Each sample was resuspended in an equal volume of salt-balanced buffer, allowed to interact with a fluorescent lipid dye, loaded into the well of a Mattek dish, covered with a sterile glass cover slip, and imaged using confocal microscopy. EVs were counted in 10 random fields of view, each representing an 8 x 8 tile imaging structure (G) and normalized by dividing the average virus-infected value by the average mock-infected value (H). Error bars indicate SD. *, P < 0.05; **, P < 0.01; ***, P < 0.001 by one-way ANOVA with Tukey’s multiple comparisons.
Fig 7.
Model of reovirus release and infection in association with EVs.
Our work indicates that in addition to exiting as free particles, reovirus strains that efficiently or inefficiently disrupt membranes can egress from mouse fibroblast and human colon epithelial cells in association with EVs. Reovirus particles are strain-specifically, cell type-dependently enclosed within and protected from antibody-mediated neutralization by medium EVs. Both large and medium EVs can transport multiple reovirus particles to recipient cells. Furthermore, compared to uninfected cells, reovirus infection enhances cellular release of EVs of all sizes. Created using Biorender.com.