Binding of hnRNP I–vRNA Regulates Sindbis Virus Structural Protein Expression to Promote Particle Infectivity

Figure 1. Protein tethering restores hnRNP I protein binding in the absence of the native interaction site. (A) A nucleotide map of the hnRNP I interaction site in the SINV TE12 3′UTR as defined by prior CLIP-Seq efforts. The specific sequences targeted for deletion in this study are highlighted in red, and sequences belonging to RSE3 (which were included in the original deletion mutant) are highlighted in cyan. (B) A schematic diagram of the viruses used in these studies, including wild-type SINV (SINV.WT) and the hnRNP I interaction-deficient mutants SINV.hnRNP IΔ and SINV.hnRNP ITAR, which incorporated a bovine immunodeficiency virus transactivation response element (BIV-TAR) in lieu of the native interaction site. The SINV repeat sequence elements (RSEs) are denoted by cyan boxes with their relative number labeled inside, similarly the hnRNP I interaction site, the SINV U-rich element and 19-nt 3′ conserved sequence element are indicated with red, purple, and gray boxes labeled with an I, U, or C, respectively. Elements are drawn to scale. (C) Immunoprecipitation of vRNA–hnRNP I complexes derived from mock-transfected HEK293 cells infected with the indicated viruses. (D)…

Figure 2. Restoration of hnRNP I binding results in wild-type-like growth kinetics. The capacity of the hnRNP–vRNA interaction site mutant viruses to replicate in HEK293 cells was assessed using one-step growth curves in (A) mock-transfected or (B) hnRNP ITAT-transfected cells. The titer was quantified using standard plaque assays. Quantitative data shown are the means of at least three minimum biological replicates, with the error bars representing the standard deviation of the means. Statistical significance, as determined by the area under the curve analysis, is shown above.

Figure 3. Specific reconstitution of hnRNP I–vRNA binding restores wild-type growth kinetics in a mutant lacking the native interaction site. (A) Viral titers of wild-type and hnRNP–vRNA interaction site mutant viruses SINV.hnRNP ITAR and SINV.hnRNP IΔ at 24 h post-infection of mock and hnRNP ITAT-transfected HEK293 cells infected at an MOI of 10 PFU per cell. (B) Comparative analysis of the viral titer for each of the aforementioned SINVs in mock-transfected HEK293 cells relative to wild-type SINV. (C) Identical to (B), with the exception that the HEK293 cells were transfected with hnRNP ITAT. Quantitative data shown are the means of at least three minimum biological replicates, with the error bars representing the standard deviation of the means. Statistical significance, as determined by Student’s t-test, are indicated above the specific comparisons (with * ≤ 0.05; ** ≤ 0.01; *** ≤ 0.001).

Figure 4. Restoration of hnRNP I binding abrogates the enhanced structural protein expression observed late during infection. (A) A graphic schematic of the nanoluciferase-based reporter strain derived from SINV TE12 that expresses nanoluciferase in parallel with the SINV Capsid protein during the translation of the subgenomic strand. (B) Mock-transfected or (C) hnRNP ITAT-transfected HEK293 cells were infected with the designated virus and nanoluciferase activity was quantified at the times indicated post-infection. Quantitative data shown are the means of three independent infections, with the error bars representing the standard deviation of the means. Statistical significance as determined by Student’s t-test is indicated above the specific comparisons (with * ≤ 0.05; ** ≤ 0.01).

Figure 5. Viral RNA synthesis is not impacted by the hnRNP–vRNA interaction. Strand-specific quantitative analysis of the three SINV vRNA species in mock-transfected HEK293 cells infected with (A) SINV.WT or (B) SINV.hnRNP ITAR viruses at an MOI of 10 PFU per cell. (C,D) Identical to the previously described panels, with the primary difference being that hnRNP ITAT-transfected HEK293 cells were used. (E,F) Data from the previous panels reconfigured to allow direct comparisons between the viruses in either cell condition. Quantitative data shown are the means of three independent infections, with the error bars representing the standard deviation of the means. Statistical significance as determined by Student’s t-test is indicated above the specific comparisons (with * ≤ 0.05).

Figure 6. Reestablishment of the hnRNP–vRNA interaction restores viral particle infectivity. (A) Virus particles, as defined by the genome equivalents per ml, derived from either mock or hnRNP ITAT-transfected HEK293 cells were quantified via qRT-PCR. (B) Paired viral titer analysis of the samples examined in (A) as measured using standard plaque assays. (C) Quantitative analysis of virus-specific infectivity, as measured by the ratio of particles per infectious unit, for the samples described in the above panels. Quantitative data shown are the means of three independent infections, with the error bars representing the standard deviation of the means. Statistical significance as determined by Student’s t-test is indicated above the specific comparisons (with * ≤ 0.05; **** ≤ 0.001).

Figure 7. Loss of hnRNP I binding does not negatively impact the viral morphology or composition. (A) A graphic model of several working hypotheses as to how increased structural protein expression leads to poor particle infectivity. (B) Representative TEM micrographs of wild-type or SINV.hnRNP ITAR particles purified via low-speed low-temperature centrifugation. Below each micrograph is a histogram of measured particle diameters with the mean and 95% confidence intervals reported inset to each graph. (C) Concentrated SINV.WT and hnRNP I interaction site deletion mutant viral particles were resolved via SDS-PAGE gel and stained with Coomassie blue. Data shown are representative of multiple independent viral preps. (C) The dashed line is indicative of where the gel was cropped and merged to remove intervening lanes for the final presented image.

Figure 8. Loss of hnRNP I binding negatively impacts viral particle attachment. Quantitative analysis of viral attachment via qRT-PCR of total RNAs extracted from HEK293 cells that were incubated with viral particles derived from either (A) mock-transfected or (B) hnRNP ITAT-transfected HEK293 cells. Quantitative data shown are the means of three independent attachment assays, with the error bars representing the standard deviation of the means. Statistical significance as determined by Student’s t-test is indicated above the specific comparisons (with * ≤ 0.05).

Figure 9. The loss of hnRNP I binding negatively impacts the glycosylation of the viral glycoproteins. (A) SINV.WT or SINV.hnRNP ITAR viruses were incubated in the presence or absence of PNGase F overnight at room temperature under nondenaturing conditions. After treatment the viral titer was quantified and the change in viral titer is presented for each pairwise sample. (B) The relative impact of deglycosylation, as determined by the average ratio of treated and untreated samples. Quantitative data shown are the means of three independent PNGase F assays, with the error bars representing the standard deviation of the means. Statistical significance as determined by Student’s t-test is indicated above the specific comparisons (with ** ≤ 0.01).