Fig 2: CVB3 3Cpro cleavage of ninein-like (NINL) prevents rapalog-induced dynein-dependent transport of intracellular cargoes.(A) Schematic of the peroxisomal trafficking assay. The peroxisomal targeting signal (PTS1) of human PEX3 (amino acids 1–42) was fused to mEmerald and FKBP and a truncated NINL (amino acids 1–1062) was fused to FRB. Dynein-dependent accumulation of peroxisomes at the centrosome, where most minus-ends are located, is initiated by the rapalog-mediated heterodimerization of FKBP and FRB. Blue arrow indicates dynein motility. (B) Indicated FRB and FKBP constructs transiently expressed with (+) or without (-) the transient co-expression of HA-tagged CVB3 3Cpro in HEK293T cells. Immunoblots were probed with anti-FLAG, anti-FKBP, anti-GAPDH, and anti-HA antibodies. Protein molecular weight markers are shown in kilodaltons (kDa) to the left of each immunoblot. Representative images from three biological replicates are shown. (C) Confocal micrographs are displayed as maximum intensity projections of U-2 OS cells, transfected with Pex3-mEmerald-FKBP and the indicated cleavable or uncleavable NINL-FRB fusion constructs with or without the co-expression of CVB3 3Cpro. Where indicated, cells were treated for 1 hr with ethanol (EtOH) as a control or 1 µM rapalog in EtOH prior to fixation. Centrosomes were immunostained with anti-pericentrin and nuclei were visualized with DAPI. 15 µm scale bars indicated in lower-left corner of merged micrographs. Yellow rectangles denote region of cropped inset. Dashed white lines denote cellular boundaries. Representative micrographs from three biological replicates are shown. (D) Schematic of the analysis pipeline. (E) Quantification of peroxisomal trafficking assays from three biological replicates. The fluorescence intensity of Pex3-mEmerald-FKBP at the centrosome was normalized to the whole-cell fluorescence, and to the areas of the regions of interest used to quantify centrosome versus whole-cell fluorescence. Each datapoint corresponds to an individual cell. The dark gray, large, outlined circles correspond to the mean for each biological replicate. For each condition, n = ~80. The mean across all replicates is denoted by the bold line. Data were analyzed using Kruskal–Wallis with Dunn’s post-hoc test for multiple comparisons. ****p<0.0001, ns, not significant. Figure 6—source data 1.Full raw unedited images for Figure 6B. Figure 6—source data 2.Individual data values for Figure 6E and Figure 6—figure supplement 1C.

Fig 3: Ninein-like (NINL) knockout (KO) cells have reduced pSTAT1 nuclear localization.(A) Confocal micrographs displayed as maximum intensity projections of wild-type (WT), NINL KO, and NIN KO A549 cells. Where indicated, cells were treated for 1 hr with 1000U interferon alpha (IFNa). pSTAT1 was immunostained with anti-phosphorylated STAT1 (Y701) and nuclei were visualized with DAPI. Dashed white lines denote cell boundaries. Representative micrographs from three biological replicates are shown. 15 µm scale bars are also shown. (B) Quantification of the mean nuclear to mean cytoplasmic fluorescence intensity of pSTAT1 from three biological replicates. Light gray datapoints correspond to the mean ratio of all cells in a field of view. Dark gray, large, outlined circles correspond to the mean for each of the three biological replicates. For each condition, n = 30. The mean across all replicates is denoted by the bold line. Data were analyzed using Kruskal–Wallis with Dunn’s post-hoc test for multiple comparisons. ****p<0.0001, ns, not significant. (C) Schematic depicting where in the IFN signaling pathway NINL acts. Figure 4—source data 1.Individual data values for Figure 4B.

Fig 4: Ninein-like (NINL) knockout (KO) cells fail to mount an effective interferon (IFN) response.(A) Immunoblot of extracts from wild-type (WT), NINL KO, and NIN KO A549 cells untreated (-) or treated (+) with IFNa. Immunoblots were probed with anti-STAT2, anti-phospho-STAT2 (Tyr690), anti-STAT1, anti-phospho-STAT1 (Tyr701), anti-Mx1, anti-IFIT3, anti-OAS1, anti-ISG15, and anti-ß-actin antibodies. Predicted protein molecular weights are shown in kilodaltons (kDa) to the left of each immunoblot. Representative images from three biological replicates are shown. (B) Differential interferon-stimulated gene (ISG) expression in WT, NINL KO, and NIN KO cells induced with IFNa. ISGs were identified as the 88 genes whose expression was upregulated in WT cells after IFNa pretreatment (Figure 3—figure supplement 2). Data are displayed as a violin plot of ISG expression in NIN KO or NINL KO cells relative to WT cells. ****p<0.0001 based on paired t-test. Dotted line indicates mean. Individual datapoints for ISGs shown in panel (A) are indicated. (C) A549 WT cells were treated with 1000U IFNa for 24 hr, then infected with wild-type vaccinia virus (WT VacV) or J3 mutant vaccinia virus (J3 VacV) (50,000 PFU/mL, MOI ˜ 0.1). Cell-associated virus was collected 24 hr post-infection and viral titers (y-axis, PFU/mL) were determined by plaque assay. (D) A549 NINL KO cells were treated, infected, harvested, and quantified as described in (C). (C, D) Data are presented as mean ± standard deviation of three experiments, with individual points shown. Data were analyzed by two-way ANOVA with Tukey’s method adjustment for multiple comparisons for IFNa treatment within each cell line, two-way ANOVA interaction comparison for IFNa interaction between cell lines. ****p<0.0001, ns, not significant. Figure 3—source data 1.Full raw unedited images for Figure 3A. Figure 3—source data 2.Individual data values for Figure 3C.

Fig 5: Cleavage of ninein-like (NINL) during viral infection prevents dynein-dependent transport of an intracellular cargo.(A) Confocal micrographs displayed as maximum intensity projections of uninfected or CVB3 infected U-2 OS cells. Cells were transfected with Pex3-mEmerald-FKBP and the indicated cleavable or uncleavable NINL-FRB fusion constructs, and infected (or mock infected) with CVB3 (500,000 PFU/ml, MOI = ~2) for 5 hr. Cells were then treated for 1 hr with ethanol (EtOH) or 1 µM rapalog prior to fixation. Centrosomes were immunostained with anti-pericentrin and nuclei were visualized with DAPI. 15 µm scale bars are indicated in the lower-left corner of the merged micrographs. Yellow rectangles denote region of cropped inset. Dashed white lines denote cellular boundaries. Representative micrographs from three biological replicates are shown. (B) Quantification of peroxisomal trafficking assays from three biological replicates. The fluorescence intensity of Pex3-mEmerald-FKBP at the centrosome was normalized to the whole-cell fluorescence, and to the areas of the regions of interest used to quantify centrosome versus whole-cell fluorescence. Each datapoint corresponds to an individual cell and biological replicates can be distinguished by shade. For each condition, n = ~80. The mean across all replicates is denoted by the bold line. Bold circles correspond to the mean for each biological replicate. Data were analyzed using Kruskal–Wallis with Dunn’s post-hoc test for multiple comparisons. ****p<0.0001, ns, not significant. (C) Schematic of the rapalog-induced pericentrosomal accumulation of peroxisomes and loss of accumulation upon viral infection. Figure 7—source data 1.Individual data values for Figure 7B and Figure 7—figure supplement 1B.