Fig 1.
Reovirus infectious virus production is delayed relative to core amplification.
(A) Diagram of reovirus whole virion and core indicating respective capsid proteins. In brackets are the genome segments encoding each protein. Non-structural proteins and their genome segments are also listed. Created with Biorender.com. (B) L929 cells were infected with reovirus at an MOI of 3 with or without cycloheximide (100μg/mL) and samples were collected every 3 hours to measure S2 mRNA levels and titers. (B, Left) De novo S2 mRNA levels by RT-qPCR and virus titers at each time point, calculated as fold-increase in cycloheximide-untreated versus cycloheximide-treated samples for each timepoint, n = 4. (B, Right) De novo mRNA and virus titers calculated as % of maximum achieved at 12 hpi for mRNA and 18 hpi for titers.
Fig 2.
Reovirus RNAs and proteins are generated simultaneously.
(A-B) L929 cells were infected with reovirus at an MOI of 3 with or without cycloheximide (100μg/mL) and samples were collected every 3 hours to measure virus mRNA and protein levels. (A, Left) De novo mRNA levels by RT-qPCR at each time point for the indicated reovirus transcripts, calculated as fold-increase in cycloheximide-treated over mock-treated samples for each timepoint, n = 4. (A, Right) Fold increase in de-novo virus mRNA among the time intervals indicated, n = 4. (B, Left) Representative blots show de novo virus protein levels by SDS-PAGE and Western blot analysis using polyclonal antibodies raised against whole virus (detects μ1c and σ3), μ2, σ2, or μNS, or the σ1 N-terminal domain as indicated. (B, Right) Combined protein levels for n = 4. Statistical analysis was performed by two-way ANOVA for A and B, but no significant differences were found. (C-D) H1299 cells were infected with reovirus at an MOI of 3. (C) Similar to B but in H1299 cells. (D) H1299 cells infected with reovirus radio-labelled with S35-methionine/cysteine for the indicated time increments and then immediately lysed. Lysates were subjected to immunoprecipitation with either polyclonal antibodies from rabbits exposed to whole reovirus (pAb whole virus proteins) that predominantly recognize OC μ1c and σ3 but also core proteins λ1/2, or polyclonal antibodies from rabbits exposed to reovirus cores (pAb core proteins) that predominantly recognize core proteins σ2, μ2 and λ1/2. Mock infected cells were used as negative controls. (E-F) H1299 cells infected with reovirus (or mock) were pulse-labelled with S35-methionine/cysteine at the indicated time (label time) for 30 minutes and then chased with normal media until the indicated collection time. (E, Left) Whole lysates show pulse-labelled virus and host proteins. (E, Right) Whole lysates were subjected to immunoprecipitation with polyclonal anti-reovirus antibodies prior to electrophoresis. (F) Whole lysates were subjected to high-speed ultracentrifugation through 1.33g/cc CsCl to pellet reovirus cores and fully-assembled viruses. Minimal host protein contamination is demonstrated by the mock control. (G) Using values from Figs 1 and 2, summary of de-novo mRNA, proteins, progeny cores, and progeny infectious viruses is presented during the course of one reovirus round of replication. Statistical analysis by ANOVA, **** p<0.0001, *** p<0.001, **p<0.05, ns > 0.05.
Fig 3.
Reovirus proteins are spatially compartmentalized.
(A) H1299 cells were infected with reovirus at an MOI of 3 before fixation at 14–20 hpi. Immunofluorescence staining was conducted with antibodies specific to OC proteins indicated in green. The OC proteins were detected with secondary antibodies conjugated to Alexa 488 (pseudo colored green). Co-immunofluorescence in the same cells was conducted using polyclonal rabbit antibodies raised against reovirus cores (α-Core) detected with secondary antibodies conjugated to Alexa 647 (red). In the merged images and their corresponding zoomed-in regions, white arrows show example regions of core-only staining, while cyan arrows indicate core-positive but μ1-negative regions. Similar results were also obtained with monoclonal 10C1 and 4F2 for σ3, 8H6 for μ1, and rabbit σ1-specific polyclonal serum. (B) Represented 3D images created from Z-stacks of reovirus infected H1299 cells probed against σ3 (10C1, red), core (α-Core, green), and nuclei (Hoechst dye, blue). (C) Represented 3D images created from Z-stacks of reovirus infected H1299 cells probed against μ1 (10F6, green), LDs (BODIPY dye, red), and nuclei (Hoechst dye, blue). (bottom) The BODIPY channel showing LDs is toggled off (left) and on (right). (D) Similar to (A) but rabbit polyclonal antibodies to non-structural protein μNS (α-μNS, red) were used instead of α-Core for co-immunofluorescence with antibodies directed towards the OC proteins indicated. Magenta arrows indicate regions that are μNS-positive but OC-negative, and yellow arrows indicated μNS-positive but μ1-negative regions. (E) Represented 3D images created from Z-stacks of reovirus infected H1299 cells probed against σNS (mouse monoclonal 2A9, red), core (α-Core, green), and LDs (BODIPY dye, magenta). (A-E) All images were acquired using spinning disk confocal microscopy and analyzed with Volocity software. Images are representative of at least four images captured for each condition from seven biologically independent experiments. (F, left) The number of core-factory objects co-staining with or without non-structural protein σNS were quantified and plotted as a percent of the total number of core factories. (F, middle) The number of μNS-staining factories were quantified and divided into percentages that overlapped with μ1 or remained at factories independent of μ1. (F, right) The number of outercapsid-σ3 factory objects co-staining with or without non-structural protein μNS were quantified and plotted as a percent of the total number of σ3 factories. Data is representative of a minimum of eight images captured for each condition, plotted as mean +/- 95% CI. Statistical analyses are reported as unpaired t-tests between the mean of each column. **** p<0.0001, *** p<0.001, **p<0.05, ns > 0.05.
Fig 4.
Reovirus outercapsid proteins localize to lipid droplets and perinuclear regions.
(A-B) H1299 cells were transfected with S1pcDNA3 (σ1), S4pcDNA3 (σ3), and M3pcDNA3 (μNS) with or without M2pcDNA3 (μ1). (A, Left) Immunofluorescence staining was conducted with antibodies specific to OC proteins μ1 (monoclonal 10F6) and σ1 (monoclonal G5 directly labelled with AlexaFluor 647) (top) or μ1 (10F6) and σ3 (monoclonal 10C1 directly labelled with AlexaFluor 647) (bottom), BODIPY for LDs, and DAPI staining for nuclei. μ1 was detected with secondary antibodies conjugated to AlexaFluor 488. (A, Right) The number of LDs co-staining with σ1 or σ3 were quantified and plotted as a percentage of the total number of LDs for both cells transfected with or without M2pcDNA3 (+ or–M2). (B) Cell lysates were subjected to SDS-PAGE followed by Western blot analysis with either polyclonal antibodies raised against whole virus (α-Reo, top) or (α-σ1N, bottom) to determine the extent of gene silencing. (C, Left) H1299 cells were transfected with S2pcDNA3 (σ2), S3pcDNA3 (σNS), S4pcDNA3 (σ3), and M3pcDNA3 (μNS) with or without M2pcDNA3 (μ1). Immunofluorescence staining was conducted with antibodies specific to OC protein μ1 (monoclonal 10F6) and polyclonal rabbit antibodies raised against reovirus cores (α-Core), BODIPY for LDs, and DAPI staining for nuclei. μ1 was detected with secondary antibodies conjugated to AlexaFluor 647, and core protein σ2 with secondary antibodies conjugates to AlexaFluor 488. (C, Right) The number of LDs co-staining with σ2 were quantified and plotted as a percentage of the total number of LDs for both cells transfected with or without M2pcDNA3 (+ or–M2). Represented images were created from Z-stacks acquired using immunofluorescent spinning disk confocal microscopy. Images are representative of at least five images captured for each condition from three biologically independent experiments. Data is plotted as mean +/- 95% CI. Statistical analysis is reported as unpaired t-test between the mean of each column. **** p<0.0001, *** p<0.001, **p<0.05, ns > 0.05.
Fig 5.
Core-only factories are spatially segregated from whole virion-containing factories.
(A) Representative SEM image using SEM array tomography sample preparation of an H1299 cell infected with reovirus at an MOI of 3 fixed at 17 hpi. (Green closeup) Distinct region containing cores only. (Blue closeup) Distinct region containing whole viruses. Representative of 8 cells imaged and two independent experiments. (B) H1299 cells were infected with reovirus at an MOI of 3, fixed at 16 hpi and imaged via TEM. Images from various regions around the cell were captured at high magnification and the surface area of viral particles in regions 1–4 were measured using ImageQuant image analysis software. Each point represents a single particle (n = 239). Statistical analysis is reported as a two-tailed student t-test. Data is plotted as mean +/- 95% CI. **** p<0.0001, *** p<0.001, **p<0.05, ns > 0.05. Representative of 18 cells imaged and 4 independent experiments.
Fig 6.
Core-containing factories are transcriptionally active.
H1299 cells were infected with reovirus at an MOI of 3 and at 14 hpi, cells were treated with actinomycin D to reduce cell transcription. (A) Representative image of cells stained for de novo transcribed RNA using an EZ-click RNA Labelling kit (RNA, red in merged image) between 15 hpi and 18 hpi., cells were stained for de novo transcribed RNA using an EZ-click RNA Labelling kit (RNA, red in merged image). Fixed cells were processed for immunofluorescence with rabbit polyclonal α-core antibodies (Alexa Fluor 488, green in merged image) and monoclonal mouse OC protein σ1 antibody G5 (Alexa Fluor 405, blue in merged images). Images were captured by spinning disk confocal microscopy. Scale bar represents 20μm. Highlighted boxes and corresponding close-up images represent example regions positive for core and RNA staining, but negative for σ1. (B) The number of RNA (+) regions containing only core, only σ1 (OC), or both core and σ1 (BOTH) were quantified and graphed as a percentage of total RNA (+) objects within the cell at each timepoint. Graph is plotted as mean +/- 95% CI. Statistical analysis is reported as one-way ANOVA with multiple comparisons between the mean of each column. **** p<0.0001, *** p<0.001, **p<0.05, ns > 0.05. Representative of two independent experiments.
Fig 7.
Temporal and spatial changes to reovirus compartmentalization occur over the course of infection.
(A) Representative images of H1299 cells that were infected with reovirus at an MOI of 3 and fixed at the indicated timepoints. Immunofluorescence spinning disk confocal microscopy was used to capture images of cells stained with monoclonal mouse α-σ3 (10G10, Alexa Fluor 647, red) and polyclonal rabbit α-core (Alexa Fluor 488, green). Cell nuclei were stained with DAPI (blue). White arrows indicate example regions of core-only staining. Images are representative of at least five images for each time point from 3 biological experiments. (B-E) σ3/core and σ1/core data were pooled together. At 8hpi there were no visible core- and OC (shared) factories hence labelled “none detected”. (B) The number of core-alone or core- and OC (shared) viral objects were quantified and graphed as a percentage of total viral objects within the cell at each timepoint. Statistical analysis is reported as a two-way ANOVA with multiple comparisons between the mean of each column. (C) The volume of core-alone or shared factory objects were independently added together for each time point and graphed as a percentage of total viral volume within the cell. Statistical analysis is reported as a two-way ANOVA with multiple comparisons between the mean of each column. (D) The factory volume for core-only objects or shared factory objects was found and plotted as individual points at each timepoint. (E) The edge-to-edge distance for pixels in each factory type was calculated and plotted at each time point by object identity. (F) Representative images of H1299 cells infected with reovirus at an MOI of 3 and fixed at the indicated timepoints. Immunofluorescence confocal microscopy was used to capture cells stained with monoclonal mouse α- μ1 (10F6, Alexa Fluor 647, green). Cell LDs were stained with BODIPY 493/503 dye (red) and nuclei were stained with DAPI (blue). Images representative of at least 5 images for each time point and three biological experiments. (G) The percent of LDs within the cell that co-stain with μ1 over the time course. Statistical analysis is reported as a one-way ANOVA with multiple comparisons, comparing the mean of the 8 hpi group to all the others. (H) The mean distance between cellular LDs at each time point. (I) The edge-to-edge distance of LDs from the nucleus. Statistical analysis was done by one-way ANOVA with multiple comparisons, comparing the mean of the mock group to all the others. Each point represents an individual value from n = 5–6 images for each timepoint. (J) H1299 cells were infected with T3DPL at an MOI of 3 for 6 hours in the presence of 35S-methionine/cysteine. Cell lysates were immunoprecipitated using 5μL of the indicated monoclonal antibodies (σ3: 10C1, 5C3, 10G10, 4F2.) or no-antibody control (No-Ab) prior to resolving by SDS-PAGE. (K) Summary diagram depicting factory/protein and LD localization during reovirus infection at ~8, 12, and 16 hpi. Figure created using Biorender.com. All graphs are plotted as mean +/- 95% CI. Statistical analysis is reported as one-way ANOVA with multiple comparisons between the mean of each column, unless otherwise indicated. **** p<0.0001, *** p<0.001, **p<0.05, ns > 0.05.
Fig 8.
Peripheral viral factories are diminished and diffuse if translation is inhibited during early infection.
(A, B) Representative images of H1299 cells were infected with reovirus at an MOI of 3, and at 10 hours post infection (hpi) cells were either fixed (A) or treated with 100μg/mL cycloheximide or DMSO (B). (A) Cells were immunofluorescently labelled with monoclonal mouse α-σNS (3E10 directly conjugated to AlexaFluor 568 or 2A9 directly conjugated to AlexaFluor 647, red) in combination with DAPI for nuclei staining (blue). (B) At 16hpi, cells were fixed and immunofluorescently labelled with monoclonal mouse α-σ3 (10G10, AlexaFluor 647, red), monoclonal mouse α-σNS (3E10 directly conjugated to AlexaFluor 568, or 2A9 directly conjugated to AlexaFluor 647, cyan) and DAPI for nuclei staining (blue). All images were acquired via immunofluorescence spinning disk confocal microscopy. (C) The total number of factory objects, their volumes, and the edge-to-edge distances from the nucleus were quantified across a minimum of ten cells for each condition. (C, Top) the percentage of σNS factory objects lacking σ3 staining were plotted. (C, middle) The edge-to-edge distance of σNS factory objects lacking σ3. (C, bottom) The volume of σNS factory objects containing σ3. All graphs are plotted as mean +/- 95% CI. Statistical analysis is reported as multiple unpaired t-test between 10hpi and DMSO or CHX. **** p<0.0001, *** p<0.001, **p<0.05, ns > 0.05. (D) Cartoon schematic summarizing the changes in factory distribution upon cycloheximide treatment at 10hpi. Figure created in Biorender.com.
Fig 9.
Outercapsid protein μ1 promotes convergence of peripheral core-only factories into perinuclear core-plus-outercapsid factories.
(A-E) Prior to infecting H1299 cells with reovirus at an MOI of 3, cells were transfected with DsiRNAs: an irrelevant control DsiRNA (IRR), a positive control DsiRNA towards an essential core protein (S2 gene; σ2 reovirus core protein) or test condition (M2 gene; μ1 reovirus outercapsid protein). (A) Cell lysates were subjected to SDS-PAGE followed by Western blot analysis to evaluate the extent of gene silencing. (B) Lysates from cells infected and pre-treated with DsiRNAs were collected at 18 hpi and viral titers for each condition were assessed by plaque assay. Statistical analysis is reported as a one-way ANOVA with multiple comparisons between the mean of each column. **** p<0.0001, *** p<0.001, **p<0.05, ns > 0.05. (C) Close up of dominant particles found in IRR- versus M2/μ1 DsiRNA-treated cells. Violin plot shows size (area) of reovirus particles imaged by TEM of IRR DsiRNA versus M2/μ1 DsiRNA transfected cells. Statistical analysis is reported as a two-tailed student t-test and graph is plotted as mean +/- 95% CI. **** p<0.0001, *** p<0.001, **p<0.05, ns > 0.05. (D) Representative images of DsiRNA-treated infected cells at 12hpi and 16hpi. Cells were immunofluorescently labelled with monoclonal mouse α- μ1 (10F6, Alexa Fluor 647, red) or monoclonal mouse α-σ3 (10G10, Alexa Fluor 647, red) in combination with polyclonal rabbit α-core (Alexa Fluor 488, green), Tye563 for DsiRNA staining (cyan), and DAPI for nuclei staining (blue). (D-E) Data from μ1/core and σ3/core co-staining were pooled. (E) The volume (left) and distance from the nucleus (middle) was measured for each factory at 16hpi in IRR (red) versus M2/μ1 (blue) DsiRNA-treated cells, using the α-core channel to capture both core-only and core+OC shared factories. (Right) Volume versus distance was plotted. Quadrants were established based arbitrarily at 10μm3 volume and 10μm distance from the nucleus to compare ratio of factories between IRR- and M2/μ1 siRNA-treated cells. Data represents eight images per condition and is representative of two independent experiments. Statistical analysis is reported as unpaired t-test between the mean of each column. **** p<0.0001, *** p<0.001, **p<0.05, ns > 0.05.
Fig 10.
Reovirus assembly is spatially and temporally compartmentalized.
A cartoon model summarizing the findings of our study. There exist four segregated areas of reovirus assembly: 1) Core-only peripheral factories, 2) Intermediate regions where residual amplified RNA is translated to proteins and assembled into particles, 3) OC proteins on LDs, and 4) whole virions and core and OC proteins in perinuclear factories. After entry, core particles undergo primary transcription and translation in peripheral core-only factories. Newly synthesized proteins assemble into progeny cores, which then undergo secondary rounds of transcription and translation. As infection progresses, μ1 facilitates OC protein accumulation on LDs, and perinuclear factories form containing predominantly assembled whole virions. Figure created using Biorender.com.