Fig 2: Autophagy factors scaffold TRIM5 interactions with TAK1 and promote TAK1 activation.(A) Immunoblot analysis of the impact of TRIM5 expression on TAK1 activation in control and autophagy factor knockout HEK293T cells. Lysates from cells transiently transfected with GFP-TAK1 and either APEX2-alone or TRIM5-APEX2 were immunoblotted and probed with the indicated antibodies. Right, the abundance of active phospho-TAK1 (phospho-Thr184/187) relative to total GFP-TAK1. N = 3 experiments. (B) Confocal microscopic analysis of GFP-TAK1, TRIM5-HA (rhesus; blue), and endogenous p62 (red) in HeLa cells. Dashed box, region shown in zoomed-in insets. Scale bar, 5 μm. (C) Confocal microscopic analysis of GFP-TAK1, Myc-ULK1 (red) in HeLa cells that stably express TRIM5-HA (blue). Dashed box, indicates regions shown in zoomed-in insets (right). Scale bar, 5 μm. (D, E) Co-immunoprecipitation analysis of interactions between FLAG-TAK1 and GFP-TRIM5 in WT and autophagy factor KO cells. GFP-TRIM5 was immunoprecipitated in (D) and FLAG-TAK1 was immunoprecipitated in (E). Plots show the abundance of the immunoprecipitated protein (FLAG-TAK1 in D and GFP-TRIM5 in E) relative to that protein’s abundance in the input. N = 3 independent experiments. (F) Immunoblot analysis of the impact of WT or LIR2 mutant (WE196AA) TRIM5 on the abundance of active phospho-TAK1 (phospho-Thr184/187). Plot shows the abundance of phospho-TAK1 relative to that of FLAG-TAK1. N = 3 independent experiments. (G) The impact of WT or LIR2 mutant TRIM5 on the activation of an NF-κB luciferase reporter. Each data point represents an independent biological replicate. *, P < 0.05; **, P < 0.01; ****, P < 0.0001 by ANOVA.

Fig 3: Autophagy factors are required for TRIM5-driven inflammatory gene expression in macrophages.(A) Differentiated THP-1 cell lines stably expressing APEX2-V5 alone or APEX2-V5 tagged TRIM5 (human or rhesus) were infected with HIV-GFP and the percentage of infected (GFP-positive) cells was determined by high content imaging. White mask, automatically identified cell boundaries. Scale bar, 50 μm. Plot, the mean above-background GFP intensity per cell (a measure of transduction efficiency) following infection of increasing amounts of HIV-GFP. N = 5 independent biological replicates, >500 cells analyzed per replicate. (B-D) Quantitative PCR analysis of THP-1 macrophage cells stably transduced with lentiviruses expressing rhesus or human TRIM5 fused to APEX2 or APEX2 alone and subjected to the indicated siRNA-mediated knockdowns. The abundance of IFNB1 (B), NLRP1 (C), and IL6 (D) were determined relative to the abundance of 18S ribosomal RNA. Quantitative PCR analysis of knockdown efficiency is shown in S2C and S2D Fig. Data, each data point represents an independent biological replicate; *, P < 0.05; **, P < 0.01; ***, P < 0.001, ****, P < 0.0001; †, not significant by ANOVA.

Fig 4: Multiple components of autophagy/lysosome system are involved in TRIM5 signaling.(A-E) The effect of the indicated knockdowns on TRIM5-mediated NF-κB and AP1 activation in a dual luciferase reporter assay system. Immunoblots demonstrating knockdown efficiency are shown in S1C Fig. (F) HEK293T cells were transfected as above and treated or not with the protease inhibitors e64d and pepstatin A (PepA) for 48 h prior to lysis. Both inhibitors were use at a concentration of 10 μM. Luciferase activity determined as above. Each data point represents an independent biological replicate. *, P < 0.05; **, P < 0.01; ****, P < 0.0001; †, not significant by ANOVA.

Fig 5: Requirement for core autophagy machinery in TRIM5-based activation of inflammatory transcription factors NF-κB and AP1.(A) Schematic of TRIM5-TAK1 signaling axis. (B-D) HEK293T cells were transfected with non-targeting or the indicated siRNA one day prior to transfection with plasmids encoding firefly luciferase under the control of NF-κB- or AP1-responsive promoters, constitutively active Renilla luciferase, and either TRIM5-APEX2 or APEX2 alone. 40–48h after transfection, samples were harvested, and luciferase values determined. Immunoblots to the left indicate knockdown efficiency. Plots show the relative signaling, which is determined by dividing a sample’s firefly luciferase signal by its Renilla luciferase signal. All values are then normalized to the mean of the TRIM5-negative control siRNA-treated sample. Data, each data point represents an independent biological replicate. **, P < 0.01; ****, P < 0.0001; †, not significant by ANOVA.