A novel mechanism of RNase L inhibition: Theiler's virus L* protein prevents 2-5A from binding to RNase L

Figures

• 

  Fig 1. RNase L ANK repeats 1 and 2 are sufficient for interaction with L*. A. Enhanced Green Fluorescent Protein (eGFP) was fused to the C-terminus of indicated Flag-tagged ankyrin repeats of mouse, rat and human RNase L. results of L* co-immunoprecipitation with eGFP constructs shown in B and C are summarized. NT: not tested. B. Co-immunoprecipitation of indicated Flag-tagged eGFP fusion proteins with HA-tagged L*DA. Right panels show Flag (ANK repeats) and HA (L*) detection after immunoprecipitation of HA. Left panels show detection of Flag and HA in cell lysates (Input). Similar results were obtained in 2 independent experiments. HC: immunoglobulin heavy chain, LC: immunoglobulin light chain. C. Co-immunoprecipitation of Flag-ANK R1-2-eGFP from indicated species with HA-tagged L* of DA or RTV-1. Upper panels show Flag and HA (L*) detection after immunoprecipitation of HA. Lower panels show detection of Flag and HA in cell lysates (Input). Shown are results representative of 2 independent experiments.

• 

  Fig 2. Amino acids 26 to 51 of the mouse RNase L are required for interaction with and inhibition by L*. A. The structure of the ankyrin domain of tested mouse/human RNase L chimeras is presented under a cartoon showing the general organization of RNase L. Blue thick lines (with numbered residues) represent the RNase L segments within the ankyrin domain (to scale with the upper cartoon) that were exchanged between human and mouse RNase L. The table given right of the cartoon summarizes binding and functional data obtained with symmetrical mouse and human RNase L chimeras. B-C. Co-immunoprecipitation of indicated Flag-RNase L (left: mouse RNase L, right: human RNase L) with HA-L*DA. Immunoblots (B) show Flag and HA detection after immunoprecipitation of HA-L* (IP:HA) and in cell lysates (Input). Graphs (C) show the mean and SD of the amount of co-immunoprecipitated RNase L chimera relative to that of WT mouse RNase L (n = 3). *: p<0.05 in a two-way ANOVA followed by Dunnett’s test for multiple comparison. D. Analysis of RNase L-mediated RNA degradation in HeLa-M cells overexpressing indicated Flag-RNase L and L*DA. RNA samples were extracted 7 hours after polyI:C transfection. Arrowheads…

• 

  Fig 3. Amino acids 81 and 82 of the mouse RNase L are also necessary for interaction with and inhibition by L*. A. Segments of decreasing size were swapped between rat and mouse RNase L and the resulting chimeric RNase L were tested for inhibition by and interaction with L*DA (mouse virus) and L*RTV-1 (rat virus). Schematic rat/mouse chimeric RNase L are represented as in Fig 2. The right columns indicate whether L*DA and L*RTV-1 interact with and inhibit the chimeric enzymes. B-C. Immunoblots (B) show Flag and HA detection after immunoprecipitation of HA-L* (IP:HA) and in cell lysates (Input). Graphs (C) show the mean and SD of the amount of co-immunoprecipitated RNase L chimera relative to that of WT RNase L of the corresponding species (n = 3). *: p<0.05 in a two-way ANOVA followed by Dunnett’s test for multiple comparison. D. Analysis of RNase L-mediated RNA degradation in HeLa-M cells overexpressing indicated Flag-RNase L and L*DA or L*RTV-1. RNA samples were extracted 7 hours after polyI:C transfection. Arrowheads point to rRNA cleavage products. Reproducible results were obtained in 2 independent experiments.

• 

  Fig 4. L* footprint on RNase L. Highlighting of residues that were mapped in ANK R1 and R2 as components of the L* footprint on RNase L (in orange). The two RNase L protomers are represented in gray and blue, 2-5A molecules are represented in pink and RNA substrate and ATP mimetics are represented by green dots. (Crystal structure of dimeric human RNase L; [45]; PDB 4OAV). Images were generated using the PyMOL Molecular Graphics System, Version 1.7, Schrödinger, LLC.

• 

  Fig 5. RNase L residues involved in 2-5A binding are not crucial for L* binding to mouse RNase L. A-B. Co-immunoprecipitation of indicated 2-5A binding-defective mouse RNase L mutants with HA-tagged L*DA. A. Immunoblots show Flag (RNase L) and HA (L*) detection after immunoprecipitation of HA (upper panels) and in cell lysates (Input, lower panels). B. Graphs showing the quantification of coimmunoprecipitated RNase L mutants relative to coimmunoprecipitated WT mouse RNase L (n = 3). Differences were non-significant according to one-way ANOVA followed by Tukey's test for multiple comparisons. C-D. Analysis of RNase L-mediated RNA degradation in HeLa-M cells overexpressing indicated Flag-RNase L and HA-L*DA. RNA samples extracted 7 hours after polyI:C transfection were analyzed by RNA chips (C) and quantified (D). Graphs show the quantification of RNA degradation by RNase L mutants in the absence or in the presence of L*DA. Data are normalized to those of WT RNase L in the absence of L*.

• 

  Fig 6. L* inhibits mouse RNase L dimerization and oligomerization. C-terminal Flag-tagged RNase L (RNase L-Flag) and L* were overexpressed in 293T cells by transfection. Cells lysates were divided into samples where RNase L dimerization was induced by adding increasing concentrations of 2-5A. Dimers/oligomers were crosslinked using DSS. A. RNA was extracted from lysates (same samples as in B) and analyzed on agarose gel electrophoresis as a control for RNase L activation. 2-5A concentrations: 0.25, 0.5, 1 and 2μM ATP equivalent. B. The upper panel shows the detection by immunoblot of WT mouse RNase L-Flag monomers, dimers and oligomers using an anti-Flag antibody. Lower panels show detection of L* and GAPDH used as a loading control. 2-5A concentrations: 0.25, 0.5, 1 and 2μM ATP equivalent. C. The upper panel shows the detection of WT human RNase L-Flag monomers, dimers and oligomers using an anti-Flag antibody. Lower panels show detection of L* and GAPDH. 2-5A concentrations: 0, 2, 4 and 8μM ATP equivalent. D. Dimerization-defective (K391R) mouse RNase L was used as a dimerization control. The upper panel shows the detection of RNase L-Flag monomers, dimers and oligomers using an…

• 

  Fig 7. L* interferes with 2-5A binding to mouse RNase L. A. Lysates of 293T cells overexpressing Flag-RNase L and L* were treated with increasing concentrations of 2-5A (0, 2, 4 and 8 μM ATP equivalent of 2-5A) before incubation, RNA extraction and analysis on agarose gel. Arrowheads indicate RNase L-generated rRNA degradation products upon treatment with high concentration of 2-5A, arising despite the presence of L*. B-C. Biotin-labeled 2-5A was immobilized on streptavidin-coated SPR gold chips. (B) Mouse (mu)RNase L (2.5 μM) or (C) human (hu)RNase L (2.5 μM) were injected in the absence or presence of different concentrations of L* as indicated. D. His6-L* binding was immobilized on Ni-NTA chips and mouse RNase L (2.5 μM) was injected in the absence or presence of different concentrations of 2-5A as indicated. RU, response units. E. Kinetic parameters of interactions between mouse RNase L and L* or 2-5A. F-G. Histograms showing mean relative fluorescence units (RFU) resulting from degradation of a FRET RNA probe by RNase L. Experiments were repeated twice in triplicates. F. Mouse RNase L (100 nM) was pre-incubated with mock (circle) or 10μM (triangle) of L* followed by addition of…

• 

  Fig 8. L* compensates ns2 RNase L antagonist activity in bone marrow-derived macrophages (BMM) and in mice. A. Schematic diagram of recombinant MHV. B. L2 fibroblasts were infected (1 PFU/cell). At indicated time points post-infection, virus titers in the cell lysates combined with supernatants were determined by plaque assay (n = 3). C-D. BMM, derived from WT or RNase L−/− mice were infected (1 PFU/cell). At indicated time points post-infection, titers of viruses in the cell lysates combined with supernatants were determined by plaque assay (n = 3). E. Four-week-old WT or RNase L−/− B6 mice were inoculated intrahepatically with WT A59, mutant and chimeric viruses (2000 PFU/mouse). At 5 d.p.i., organs were harvested, homogenized and virus titers determined by plaque assay (n = 4 or 5). Statistics were done using the Mann-Whitney test. Error bars represent standard error of the means.