Fig 2: PirB signaling is required for efficient reovirus entry.a–f CHO cells were transfected with the cDNAs shown. a PirB signaling induced by reovirus binding. Cells were treated with Na3VO4 prior to and during T3SA + adsorption. PirB was immunoprecipitated and detected by immunoblotting. Relative pTyr intensity was determined by normalizing to total PirB signal intensity and comparing with the intensity of the pre-adsorption signal (0 min). b Reovirus transcription following infection of WT PirB- or mutant-expressing cells. c Reovirus infection of WT PirB or mutant-expressing cells. Cells were adsorbed with T3SA- (20 or 100 PFU/cell). Infectivity was quantified by IFA at 24 hpa. d Reovirus infection of WT PirB or mutant-expressing cells. Cells were adsorbed with T3SA- (20 PFU/cell). Viral titers in cell lysates were determined by plaque assay. e Effect of SHP-1/2 inhibitor on reovirus internalization into WT or mutant PirB-expressing cells. Cells were incubated with NSC-87877 prior to and during T3SA- adsorption. In b and e, cells were adsorbed with T3SA- and analyzed for s4 RNA levels in cell lysates at the times shown by RT-qPCR. Levels of s4 RNA were normalized to levels of ß-actin RNA. f Reovirus uncoating in cells expressing WT or mutant PirB. CHO cells were adsorbed with T3SA + on ice and incubated at 37 °C for the times shown. Viral proteins were detected by immunoblotting. Levels of µ1 and s3 proteins were normalized to levels of glyceraldehyde 3-phosphate dehydrogenase (GAPDH) and compared with the corresponding levels pre-uncoating (0 min). In a–e and f, experiments were conducted in triplicate and quadruplicate, respectively. Mean values are shown. Error bars indicate SD. P values were calculated using one-way ANOVA with Turkey’s test (a) and two-way ANOVA with Holm-Sidak’s test (b–f). *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.

Fig 3: PirB D3D4 is required for reovirus binding and infectivity.a Schematic of reciprocal exchanges of PirA and PirB ectodomains. PirA (orange) and PirB (blue) have six homologous extracellular Ig-like domains, designated D1 to D6. PirA and PirB extracellular domain sequences were exchanged to yield chimeric receptor constructs. b CHO cells were transfected with the cDNAs shown and scored for reovirus T3SA- binding by flow cytometry. c CHO cells were transfected with the cDNAs shown, absorbed with T3SA- at an MOI of 50 PFU/cell, and scored for infectivity by IFA. Reovirus binding (b) and infectivity (c) assays were conducted in quadruplicate and triplicate, respectively. Mean values are shown. Error bars indicate SD. Statistical analysis was conducted by comparing results of each chimeric receptor with the corresponding parental backbone, which is defined based on the transmembrane and intracellular region. P values were calculated using one-way ANOVA with Turkey’s test. *P < 0.05; **P < 0.01; ****P < 0.0001.

Fig 4: Biophysics of reovirus-PirB interactions.Characterization of reovirus-PirB binding thermodynamics on model surfaces (a–d) and living cells (e–k). a Schematic of quantifying reovirus-PirB interactions using a recombinant PirB-coated model surface. Tips were functionalized with T3SA- virions or recombinant s3. The diagram was prepared using BioRender. b Binding probability on model surfaces with or without PirB-specific mAb treatment. Uncoated surface, nonspecific control. s3-PirB, N = 3; virion-PirB, N = 7; virion-PirB plus mAb, N = 11; uncoating surface, N = 3. c Dynamic force spectroscopy (DFS) plot of the distribution of average rupture forces across eight discrete loading rate ranges. Data corresponding to single and multivalent interactions fit Bell-Evans (solid line) and Williams-Evans models (dotted lines). N = 6605. d Binding probability based on the contact time of reovirus-functionalized tips with a PirB-coated surface. Least-squares fit of the data to a mono-exponential decay model (blue line, r2 of 0.99) provides the average binding kinetic on-rate (kon). N = 5. e Susceptibility of PirB-expressing cells to infection by reovirus virions and ISVPs. Virion, 100 PFU/cell; ISVP, 10,000 ISVPs/cell. N = 3. f Confocal micrograph of PirB-2A-GFP-expressing Lec2 cells. g Representative adhesion map from the boxed area in (f). Reovirus binding is indicated by gray-to-white pixels. In f and g, scale bar, 5 µm. h Reovirus-PirB binding probability using living cells. N = 3. i Adhesion force of reovirus binding to live cells. Virion-PirB, N = 182; virion-PirB plus mAb, N = 140; uncoating surface, N = 93. j DFS plot of prior model surface data (gray) incorporating live-cell data (blue). N = 6605. k Histogram of the force distribution of the live-cell data and a multi-peak Gaussian fit. N = 225. In b–e, h–j, mean values are shown. In b–e, h–j, error bars indicate SD. In b, h, and i, P values were calculated using one-way ANOVA with Turkey’s test. ****P < 0.0001.

Fig 5: PirB is required for efficient T3 reovirus replication and pathogenicity in the murine CNS.a T3 reovirus replication in WT and PirB-/- mice. Mice were inoculated perorally with T3SA- (104 PFU/mouse). N = 11/10/9 at 3/6/9 DPI (WT); N = 9/9/7 at 3/6/9 DPI (PirB-/-). b T3 reovirus replication in the brain of WT and PirB-/- mice. N = 13/13/10 at 2/4/6 DPI (WT); N = 13/10/6 at 2/4/6 DPI (PirB-/-). c T3 reovirus replication in the brain of PirBfl/fl and NspPirB-/- mice. N = 13/19/14 at 2/4/6 DPI (PirBfl/fl); N = 12/9/15 at 2/4/6 DPI (NspPirB-/-). d T3 reovirus virulence in PirBfl/fl and NspPirB-/- mice. e Encephalitis following infection of PirBfl/fl and NspPirB-/- mice. PBS-inoculated PirBfl/fl mice, sham control. Brain inflammation was monitored by MRI at 8 dpi and defined as hyperintensity. Relative level is indicated by hyperintensity voxel percentage. N = 7 (PirBfl/fl); N = 5 (NspPirB-/-); N = 4 (sham). In b–e, mice were inoculated intracranially with T3SA- (25 PFU/mouse). f Reovirus infection of primary cortical neurons is blocked by PirB-specific mAb. Neurons were pre-incubated with mAb or isotype IgG and adsorbed with T3SA+ (100 PFU/cell). g, h T3 reovirus infection of WT and PirB-/- primary neurons. Primary murine cortical neurons were adsorbed with T3SA+ (20 PFU/cell) (g). In f and g, scale bar, 150 µm. Viral titers in tissue (a–c and e) and neuron lysates (h) were determined by plaque assay. In f–h, experiments were conducted in triplicate. WT and PirB-/- mice have comparable tissue weights. In a–c and e, each symbol indicates a single mouse. Mean values are shown. Error bars indicate SD. Statistical analysis: a–c and h, two-way ANOVA with Holm-Sidak’s test; d log-rank (Mantel-Cox) test; e one-way ANOVA with Turkey’s test; f and g two-sided Student’s t-test. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.