Structure of coxsackievirus cloverleaf RNA and 3C pro dimer establishes the RNA-binding mechanism of enterovirus protease 3C pro

Figures

• 

  Fig. 1.. Structure of the CVB3 cloverleaf RNA and 3Cpro dimer complex. (A) Overall structure of the CVB3 cloverleaf and 3Cpro dimer complex. The structure shows that two copies of 3Cpro interact along the lateral surface of the dsRNA of stem-loop D across from the stem-loop C of cloverleaf. Two 3Cpro monomers are depicted in green (M1) and violet (M2), and the cloverleaf RNA in gold. The 2Fo-Fc electron density map contoured at 1σ is shown as gray cage. (B) Secondary structure of CVB3 cloverleaf RNA. Four stems of cloverleaf RNA (A, B, C, and D) and the A•C-U base triple between stem-loops C and D are indicated. Pyrimidine mismatch region in stem-loop D is boxed. The nucleotide numbering corresponds to the cloverleaf RNA sequence in the genome. (C) Comparison of native and bound cloverleaf RNA structures. The cloverleaf RNA in the 3Cpro complex is colored as in (A), and the native cloverleaf (PDB code 8SP9) is colored in blue. The cloverleaf in the cloverleaf-3Cpro dimer complex can be superposed on the native structure with a root mean square deviation (RMSD) of 2.24 Å across all 70-nt pairs. Interaction with 3Cpro dimer increases the width of major groove between U54 and U67 from 8.8 Å…

• 

  Fig. 2.. Two 3Cpro proteins provide symmetric interaction sites for stem-loop D. (A) Arrangement of 3Cpro dimer on stem-loop D. Two 3Cpro monomers bind stem-loop D in opposite orientations. The 3Cpro residues are colored in green (M1) and violet (M2). The twofold axis of the dimer is indicated. 3Cpro folds into a double–β barrel structure, and the active site (His40, Glu71, and Cys147, pink) is located between the two β barrels on the surface opposite the RNA-binding site. Electrostatic potential of the same surface of the dimer is shown below with positive charge shown in blue and negative charge in red. (B) Bipartite RNA-interaction surface formed by 3Cpro dimer. 3Cpro dimer forms two symmetric RNA-binding sites on stem-loop D. Each site is formed by the N-terminal helix and TGK hairpin of one 3Cpro monomer and FG loop of the other. (C) Sequence alignment of enterovirus 3Cpro. The 3Cpro sequences of CVB3 (AAC60718), PV (NP_740476), HRV14 (AAA45758), and HRV2 (WLV75740) were used for the alignment. The cloverleaf RNA-binding site, protease active site, and predicted 3Dpol-binding site are indicated. Conserved amino acids are shown in bold.

• 

  Fig. 3.. 3Cpro dimer interacts primarily with dsRNA backbone across the major groove of stem-loop D. (A) RNA-binding site in 3Cpro dimer. 3Cpro residues that interact with stem-loop D are shown and colored as in Fig. 1. 3Cpro primarily binds the sugar-phosphate backbone with a few base-specific interactions. The symmetric interactions near the twofold axis of the 3Cpro dimer are shown below. Arg13, Arg84, and Asp85 from each monomer bind across the major groove and maintain the major groove width at 11 Å at both edges. (B) Schematic representation of 3Cpro and stem-loop D interactions. RNA regions that interact with the 3Cpro protein are shown in filled bases. Hydrogen bonds are shown as dotted lines, and van der Waals interactions are shown in boxes. Protein main-chain atoms that are involved in hydrogen bond interactions are indicated in parenthesis. The 3Cpro monomers are colored as in Fig. 1, and the twofold axis is shown.

• 

  Fig. 4.. 3Cpro and 3CD proteins bind cloverleaf as dimers as measured by MP. (A) 3Cpro interaction with cloverleaf RNA. Distribution of molecular mass of 3Cpro was determined in the absence and presence of cloverleaf RNA in a buffer containing either 40 mM (top) or 100 mM KCl (bottom). The oligomerization states are determined using the molecular masses of 3Cpro (21.4 kDa) and cloverleaf RNA (31.8 kDa). 3Cpro exists as a dimer in both 40 and 100 mM KCl conditions and binds cloverleaf RNA as a dimer. (B) 3CD interaction with cloverleaf RNA. Distribution of molecular mass of 3CD was determined as in (A). The oligomerization states are determined on the basis of the molecular mass of 3CD (73.8 kDa) and cloverleaf RNA (31.8 kDa). 3CD exists primarily as a monomer in 40 or 100 mM salt concentrations but binds cloverleaf as a dimer.

• 

  Fig. 5.. 3CD increases the binding affinity to full-length cloverleaf RNA but not stem-loop D. (A) Gel shift assays of 3Cpro and 3CD with cloverleaf stem-loop D. Stem-loop D RNA (SLD; 1 μM) was incubated with increasing concentrations of protein (1 to 10 μM). Both proteins bind stem-loop D similarly. (B) Gel shift assays of 3Cpro and 3CD with cloverleaf RNA. Cloverleaf RNA (CL; 1 μM) was incubated with increasing molar excess of protein (1 to 10 μM). 3CD protein binds cloverleaf RNA with higher affinity than 3Cpro. (C) Model of 3CD and cloverleaf RNA interaction. The 3CD and cloverleaf complex was modeled by superposition of 3Cpro domain of 3CD (PDB code 2IJD) onto each of the 3Cpro domains in the cloverleaf-3Cpro dimer complex. 3Dpol domains (lighter colors) of both 3CD proteins extend toward and flank stem-loop C, likely providing additional interaction with cloverleaf RNA.

• 

  Fig. 6.. 3Cpro interaction with cis-acting RNAs. (A) Sequence and secondary structure comparison of cloverleaf RNA. Cloverleaf RNA from CVB3, PV, HRV2, BEV, and mutants (HRV14InsG and CVB3short) bind CVB3 3Cpro (52). In contrast, the HRV14 cloverleaf and BEV cloverleaf 2 (CL2) RNAs do not bind CVB3 3Cpro (right). The base pairs (BP) that interact with 3Cpro dimer in the cloverleaf-3Cpro dimer complex are boxed, and symmetrically related nucleotides within the 3Cpro interaction sites are shown in bold. (B) The secondary structure of CVB3 and HRV14 CRE. CRE contains a 14-nt apical loop, wherein the AAA sequence positioned at 5 to 7 nt (bold) is used for VPg uridylylation. The predicted bipartite 3Cpro dimer interaction sites are indicated as in (A). (C) Model of the CRE-3Cpro dimer complex. The CRE-3Cpro dimer complex was modeled by superposing CRE (PDB code 1T28) onto stem-loop D of the clovlerleaf-3Cpro dimer complex. The AAA sequence in CRE was colored in orange and labeled. The predicted 3Dpol binding residues (Pro45, Glu96, and Glu99) are depicted on 3Cpro. (D) Model of the CRE-3Cpro dimer-3Dpol complex. 3Dpol was modeled by superposing the template strand in the 3Dpol-template-primer…