Alphaviral Capsid Proteins Inhibit Stress Granule Assembly via Competitive RNA Binding With G3BP1

FIGURE 1. SFV capsid inhibits stress granule formation. (A) Diagram of SFV genome and encoding proteins. (B) Immunofluorescent analysis of mCherry‐tagged SFV encoding proteins and endogenous G3BP1 in 0.5 mM SA‐treated U2OS cells. Scale bar, 10 µm. (C) Quantification of SG area and number in (B). Data are plotted as minimum to maximum; lines indicate the first quartile (lower), median, and the third quartile (upper) with n ≥ 50 for each sample (one‐way ANOVA with Mixed‐effects analysis). *p < 0.05, **p < 0.01, ****p < 0.0001, ns is not significant. Data are from one representative experiment of two independent experiments. (D) Immunofluorescence of mCherry‐SFV capsid and endogenous G3BP1 in heat‐stressed U2OS cells (left), with corresponding quantification of SG area and number (right). Scale bar, 10 µm. Data are plotted as minimum to maximum; lines indicate the first quartile (lower), median, and the third quartile (upper) with n ≥ 50 for each sample (Student's t‐test, unpaired, two‐tailed). ****p < 0.0001, ns is not significant. Data are from one representative experiment of two independent experiments. (E) Immunofluorescence of mCherry‐SFV capsid and endogenous eIF3η in 0.4 M…

FIGURE 2. Identification of a minimal fragment of SFV capsid required for SG inhibition. (A) Schematic structures of SFV capsid N‐terminal variants. (B) Immunofluorescent analysis of SA‐induced SG formation in U2OS cells expressing the N‐terminal variants in (A). Scale bar, 10 µm. (C, D) Quantification of SG area and number in (B). Data are plotted as minimum to maximum; lines indicate the first quartile (lower), median, and the third quartile (upper) with n ≥ 50 for each sample (one‐way ANOVA with Mixed‐effects analysis). **p < 0.01, ***p < 0.001, ****p < 0.0001, ns is not significant. Data are from one representative experiment of at least two independent experiments.

FIGURE 3. Oligomerization mediated by a helix within SFV capsid is required for SG inhibition. (A) Schematic showing the sequence of the α‐helix between the two positively charged blocks in the N‐terminal of SFV capsid. (B) Schematic of the SFV capsid 22–110 aa full‐length and shorter variants. (C) Immunofluorescent analysis of SA‐induced SG formation in U2OS cells expressing mCherry‐tagged α‐helix variants in (B). Scale bar, 10 µm. (D, E) Quantification of SG area and number related to (C). Data are plotted as minimum to maximum; lines indicate the first quartile (lower), median, and the third quartile (upper) with n ≥ 50 for each sample (one‐way ANOVA with Mixed‐effects analysis). *p < 0.05, **p < 0.01, ****p < 0.0001, ns is not significant. Data are from one representative experiment of two independent experiments. (F) The amino acid sequences and the CD spectra of two peptides spanning the α‐helix 35–64 aa WT and the 2LP mutant. The mutation sites were highlighted in blue and red. (G) AlphaFold3 prediction of the potential SFV capsid α‐helix WT conformation as monomer and multimer (Top) and the conformational prediction of its 2LP mutant (Bottom). pLDDT, predicted local distance…

FIGURE 4. Disruption of α‐helix of capsid dampens SFV replication and SG inhibition activity in cells. (A) Schematic of visualizing SFV capsid during viral infection with a recombinant SFV containing the GFP11 tag in‐frame with the capsid ORF. Infection of the GFP1‐10 stable U2OS cells with the GFP11‐SFV enabled the reassembly of a functional GFP and a fluorescent readout. The diagram was created with BioRender.com. (B) Confocal imaging of SFV capsid in GFP‐11 SFV‐infected U2OS cells co‐stained with SFV capsid antibody. Representative patterns of diffuse and puncta were shown. Scale bar, 10 µm. (C) Confocal imaging of the capsid expression pattern in WT SFV‐infected U2OS cells at the indicated time points. Percentage of the patterns was quantified by examining ≥50 cells for each time point from one of the two independent experiments. Scale bar, 10 µm. (D) Confocal imaging of SFV capsid in GFP‐11 SFV‐infected U2OS cells co‐stained with SFV nsP3 antibody. Three representative patterns are detected: full co‐localization of capsid with nsP3 foci (top); partial co‐localization of capsid with nsP3 foci (middle); diffuse capsid without substantial co‐localization with nsP3 foci (bottom). Scale…

FIGURE 5. The valency and density of the positive charges are required for SG modulation. (A) Schematic of SFV capsid with positively charged block 2 and 3 scramble mutants. Lysine and arginine residues in each sequence were highlighted in green. (B) Immunofluorescent analysis of SA‐induced SG formation in U2OS cells expressing the mCherry‐tagged block 2 scrambles in the context of full‐length and N‐terminal truncation. Scale bar, 10 µm. (C) Quantification of SG area and number in (B). Data are plotted as minimum to maximum; lines indicate the first quartile (lower), median, and the third quartile (upper) with n ≥ 50 for each sample (one‐way ANOVA with Mixed‐effects analysis). ****p < 0.0001, ns is not significant. Data are from one representative experiment for at least two independent experiments. (D) Schematic of lysine pattern rearrangement of SFV capsid block 2. (E) Immunofluorescent analysis of SA‐induced SG formation in U2OS cells expressing the indicated lysine‐rearranged 40–110 aa constructs. Scale bar, 10 µm. (F) Quantification of SG area and number related to (E). Data are plotted as minimum to maximum; lines indicate the first quartile (lower), median, and the third quartile…

FIGURE 6. SFV Capsid inhibits G3BP1‐RNA phase separation via RNA competition. (A, B) LLPS of purified G3BP1 with in vitro‐transcribed SFV nsP3 RNA at the indicated concentrations. (B) The phase diagram of (A). (C) LLPS of purified G3BP1 with in vitro‐transcribed SFV nsP3 RNA at the indicated concentrations in the presence of 25 µM SFV capsid. Scale bar, 10 µm. (D) The phase diagram of (C). (E) LLPS of 25 µM G3BP1 and 100 ng/µL polyA RNA in the presence of increasing concentrations of MBP, MBP‐fused N‐terminal, and C‐terminal SFV capsid. Scale bar, 10 µm. (F) Fluorescence polarization (FP) assay to measure the binding of purified G3BP1 or SFV capsid to FAM‐labeled host (n = 2) and viral (n = 2) RNA probes. Data are mean ± SEM with n = 3 technical replicates in one representative experiment. (G) Excess RNA reverses capsid inhibition of G3BP1‐RNA LLPS. Increasing concentrations of polyA RNA, G3BP1, and BSA were individually added into a specified capsid‐G3BP1‐RNA system, and the reoccurrence of LLPS was captured via DIC imaging. Scale bar, 10 µm. (H) IP‐WB to map the SFV capsid domains required for G3BP1 interaction. Note: the double band for mCherry fusion is likely due to the proteolysis of…

FIGURE 7. A shared feature of alphavirus capsids on SG inhibition. (A) Immunofluorescent analysis of SA‐induced SG formation in U2OS cells expressing the indicated mCherry‐tagged alphaviral capsids. Scale bar, 10 µm. (B) Quantification of SG area and number in (A). Data are plotted as minimum to maximum; lines indicate the first quartile (lower), median, and the third quartile (upper) with n ≥ 50 for each sample (one‐way ANOVA with Mixed‐effects analysis). ****p < 0.0001. Data are from one representative experiment of two independent experiments. (C) Immunofluorescent analysis of SA‐induced SG formation in U2OS cells expressing the mCherry‐fused α‐helix‐block 2 region of the indicated alphaviral capsids. Scale bar, 10 µm. (D) Quantification of SG area and number in (C). Data are plotted as minimum to maximum; lines indicate the first quartile (lower), median, and the third quartile (upper) with n ≥ 50 for each sample (one‐way ANOVA with Mixed‐effects analysis). ****p < 0.0001. Data are from one representative experiment of at least two independent experiments. (E) Schematic architecture of VEEV capsid protein. The block 2 contained a nuclear localization signal (NLS). (F) Immunofluorescent…

FIGURE 8. SARS‐CoV‐2_N inhibits SG via direct G3BP1 interaction. (A) Schematic illustration of SARS‐CoV‐2 nucleocapsid (N). A motif of ITFG and its alanine mutations was highlighted. (B) Immunofluorescent analysis of SA‐induced SG formation in U2OS cells expressing the indicated mCherry‐tagged mutants of SARS‐CoV‐2_N protein. Scale bar, 10 µm. (C) Quantification of SG area and number in (B). Data are plotted as minimum to maximum; lines indicate the first quartile (lower), median, and the third quartile (upper) with n ≥ 50 for each sample (one‐way ANOVA with Mixed‐effects analysis). **p < 0.01, ***p < 0.001, ****p < 0.0001. Data are from one representative experiment of two independent experiments. (D, E) Heatmap of representative TurboID‐G3BP1‐enriched proteins in untreated (D) or SA‐treated (E) U2OS cells co‐transfected with SARS‐CoV‐2_N WT or F17A.

FIGURE 9. A working model. The alphavirus capsid harbors a conserved molecular feature, a helix‐coupled lysine‐rich region, that can efficiently inhibit SG formation by decoupling RNA from G3BP1.