EV-B 3D polymerase remodels viral populations through 5′UTR recombination to subvert cardiac antiviral innate immunity

Figures

• 

  Fig 1. Characterization of genetic recombination events in the 5′ UTR of CV-B leading to infectious recombinants. (A) Recombination sites were mapped from purified infectious recombinants (n = 58) following co-transfection of the 3′ partner (3′P, X-axis) and the 5′ partner (5′P, Y-axis) in a 19:1 ratio in human cardiomyocytes (HCM). The locations of recombination events in the 5′ UTR are depicted, with spacers 1 and 2 identified as recombination hotspots, termed HSA and HSB, respectively. (B) Hotspots HSA and HSB are highlighted, with recombination types depicted. (C) Representative recombinants of each type homologous (H), non-homologous with deletion (NH−), and non-homologous with insertion (NH+) are shown in HSA and HSB. Sequences from the 5′ partner CV-B6 (green) and 3′ partner CV-B3 D50 (blue) are indicated. (D) The proportions and numbers of recombinant viruses are presented by recombination type and their position within the viral genome. (E) Distribution of NH recombinants according to the length (in bases) of the duplication or insertion. (F) Virus titers by recombinant type after 72 h of amplification in HCM. Data represent the mean ± SD. ****: P < 0.0001 by Mann–Whitney U…

• 

  Fig 2. Fitness of CV-B recombinant strains compared to parental viruses. Human cardiomyocytes (HCM) were infected with a selected recombinant (H, NH−, or NH+) or with CV-B3 or CV-B6 parental viruses at an MOI of 10. (A) Viral titers of recombinant and parental viruses at 24 h post-infection. (B) Total viral RNA (intracellular and extracellular) from CV-B3 or CV-B6 parental viruses and homologous recombinants was measured by RT-qPCR at the indicated time points post-infection. (C) Total viral RNA (intracellular and extracellular) from CV-B3 or CV-B6 parental viruses and non-homologous recombinants was measured by RT-qPCR at the indicated time points post-infection. Data represent the mean ± SD for three independent experiments. *: P < 0.05; **: P < 0.01; The Mann–Whitney test was used for two groups comparison. CV-B: Coxsackievirus, H: homologous, HS: Hotspot, MOI: multiplicity of infection, NH: non-homologous, PFU: plage forming unit.

• 

  Fig 3. Assessment of 3D polymerase action on recombination mechanisms between 5′ and 3′ RNA partners. (A) Ribbon representation of CV-B3 3D polymerase, showing fingers subdomain (light green), thumb (dark blue), and palm (gray). Mutations impairing copy-choice recombination D79H, Y276H, and L421A are indicated. (B) Co-transfection of 5′ partner with either wild-type or mutant 3′ partners (3′P 3Dpol D79H, 3′P 3Dpol Y276H, or 3′P 3Dpol L421A) in a 19:1 ratio at 24 h post-transfection in HCM; infectious recombinant production was analyzed. (C) Co-transfection of 5′ partner with wild-type or mutant 3′ partner (3Dpol Y276H) at various time points post-transfection in HCM, with infectious recombinant production analysis. (D, E) Transfection of CV-B3 or CV-B3 Y276H mutant: (D) assessment of viral RNA replication activity and (E) analysis of infectious recombinant production. Data represent the mean ± SD for three independent experiments. **: P < 0.01; ***: P < 0.001; ****: P < 0.0001 by Mann Whitney U test. HCM: human cardiac myocytes, P: partner, 3Dpol: 3D polymerase, PFU: plaque forming unit.

• 

  Fig 4. Impact of interaction between full-length and deleted CV-B3 RNA forms on viral translation activities. (A) Schematic representation of full-length (FL) and 5′ terminally deleted (TD50, D50) CV-B3 nanoluciferase-tagged (NL) RNAs. (B) Relative luciferase activity in human cardiomyocytes (HCM) following transfection with CV-B3 NL-tagged RNA, with or without the Y276H mutation in 3Dpol. (C) Relative luciferase activity in HCM after transfection with TD50 CV-B3 NL-tagged RNA, with or without Y276H mutation in 3Dpol. (D) Relative luciferase activity, normalized to total protein, following transfection of 5% FL CV-B3 NL-tagged RNA (with or without Y276H mutation in 3Dpol), alone or co-transfected with 95% TD50 CV-B3 RNA (with or without Y276H mutation). (E) Relative luciferase activity, normalized to total protein, following transfection of 95% TD50 CV-B3 NL-tagged RNA (with or without Y276H mutation in 3Dpol), alone or co-transfected with 5% FL CV-B3 RNA (with or without Y276H mutation). Data represent the mean ± SD for three independent experiments *: P < 0.05; ****: P < 0.0001 by Mann–Whitney U test. 3Dpol: 3D polymerase, CV-B: Coxsackievirus, IRES: Internal Ribosome Entry Site, NL:…

• 

  Fig 5. Characterization of cardiovirulence and genomic RNA replication of CV-B3 recombinants, 3Dpol mutants, and parental CV-B3 in DBA/2J mice hearts. (A) Weight loss in DBA/2J mice infected with 106 PFU/mice of parental CV-B3 (n = 15), homologous (HSA, HSB, n = 5–10) or non-homologous (NH+ (HSA, and HSB) or NH−, n = 5–10) recombinants, or the 3Dpol Y276H-mutant CV-B3 strain (n = 15), and in uninfected controls (n = 3), expressed as mean ± s.e.m. for the indicated number of mice. (B) Histological scores of cardiac lesions in mice infected with the indicated viruses (n = 4–6) or in uninfected controls (n = 3) at 10 days post-infection. Data shown as mean ± SD; **P < 0.01 by Mann–Whitney U test. (C) Immunofluorescence staining of viral capsid protein VP1 (red) and dystrophin (green) in heart sections (nuclei in blue) from mice infected with the indicated viruses or left uninfected at 10 days post-infection: magnification ×20. The images are representative of three mice per group. (D) Viral RNA loads in heart tissue, quantified by qRT-PCR, in mice infected with recombinant viruses, the CV-B3 Y276H mutant, or parental CV-B3 (n = 4–6) at the indicated time points post-infection (3, 7, and 10…

• 

  Fig 6. Emergence and maintenance of 5′TD CV-B3 populations in heart tissues of DBA/2J mice. (A–C) Proportions of full-length (FL) and 5′-terminally deleted (5′TD) CV-B3 RNA populations were measured by RACE-PCR and micro-electrophoresis (Agilent) in mice heart tissues at 3-, 7-, and 10-days post-infection (n = 3-6). Data represent mean ± SD. Statistical significance: **P < 0.01, ***P < 0.001 by two-way ANOVA with multiple comparisons. (A) [Left] Comparison of CV-B RNA population proportions in mice infected with 3Dpol Y276H mutant or parental CV-B3 strains. [Right] Schematic cloverleaf structure of CV-B showing positions of 8, 36, and 50 nt deletions: FL populations (gray, 0–8 nt deletions), 5′TD 9–36 nt populations (blue, 9–36 nt deletions), and 5′TD 37–50 nt populations (pink, 37–50 nt deletions). (B) Comparison of RNA population proportions in mice infected with homologous recombinants and parental CV-B3 strain. (C) Comparison of RNA population proportions in mice infected with non-homologous recombinants and 3Dpol Y276H mutant CV-B3 strain. (D and E) IFN-β and ISG15 mRNA expression reported as fold change relative to uninfected mice, normalized to housekeeping gene expression, in…

• 

  Fig 7. Impact of 5′TD CV-B3 populations on IFN-β pathway activation in human cardiomyocytes. (A) Experimental design: Full-length CV-B3, 5′TD with 15 nt deletions (TD15), and 5′TD with 50 nt deletions (TD50) synthetic RNAs were transfected into human cardiomyocytes (HCM), mimicking the proportions of FL and 5′TD RNA populations observed in mouse hearts infected with parental CV-B3 (Profile A) or 3Dpol Y276H-mutant CV-B3 (Profile B) at 7 days post-infection. (B) Viral RNA replication activity in HCM at 24 h post-transfection, assessed by RT-qPCR. (C, D) IFN-β and ISG15 protein levels in HCM supernatants at 24 h post-transfection, quantified by ELISA. (E) HCM cells were transfected with synthetic RNA and incubated up to 24 h for western blot analysis of TBK1/NAK phosphorylation and total TBK1/NAK. Data represent the mean ± SD for three independent experiments. *: P < 0.05; **: P < 0.01; ***: P < 0.001 by Mann–Whitney U test. The Poly(I:C) high molecular weight (HMW) was used as the positive control in these experiments.3Dpol: 3D polymerase, CV-B: Coxsackievirus, FL: full-length, MW: molecular weight, TD: 5′terminally deleted. RNA extraction: Created in BioRender. GLENET, M. (2026)…