PDBsum entry 2qfb

Hydrolase

PDB id: 2qfb

Contents

Protein chain

(+ 4 more) 121 a.a.

Metals

_ZN ×10

Waters ×49

References listed in PDB file

Key reference

Title

The c-Terminal regulatory domain is the RNA 5'-Triphosphate sensor of rig-I.

Authors

S.Cui, K.Eisenächer, A.Kirchhofer, K.Brzózka, A.Lammens, K.Lammens, T.Fujita, K.K.Conzelmann, A.Krug, K.P.Hopfner.

Ref.

Mol Cell, 2008, 29, 169-179. [DOI no: 10.1016/j.molcel.2007.10.032]

PubMed id

18243112

Abstract

The ATPase RIG-I senses viral RNAs that contain 5'-triphosphates in the cytoplasm. It initiates a signaling cascade that activates innate immune response by interferon and cytokine production, providing essential antiviral protection for the host. The mode of RNA 5'-triphosphate sensing by RIG-I remains elusive. We show that the C-terminal regulatory domain RD of RIG-I binds viral RNA in a 5'-triphosphate-dependent manner and activates the RIG-I ATPase by RNA-dependent dimerization. The crystal structure of RD reveals a zinc-binding domain that is structurally related to GDP/GTP exchange factors of Rab-like GTPases. The zinc coordination site is essential for RIG-I signaling and is also conserved in MDA5 and LGP2, suggesting related RD domains in all three enzymes. Structure-guided mutagenesis identifies a positively charged groove as likely 5'-triphosphate-binding site of RIG-I. This groove is distinct in MDA5 and LGP2, raising the possibility that RD confers ligand specificity.

Figure 1.
Figure 1. Biochemical Analysis of RIG-I Variants
(A) RIG-I and MDA5 variants used in this study.
(B) Catalytic efficiency (k[cat] K[m]^−1) of WT RIG-I and ΔCARD-RIG-I, RIG-I-ΔRD and the DECH domain for pppRVL (black bars), and nonphosphorylated dsRNA (white bars). Error bars represent standard errors of the nonlinear regression analysis (Supplemental Data).

Figure 6.
Figure 6. Localization of the RNA 5′-Triphosphate-Binding Site on RD
(A) Electrostatic surface potential (ranging from blue = 9 kT/e to red = −9 kT/e), displayed in two different views (left, “standard view” used in all other figures; right, 180° rotation around vertical axis). The sites of mutated residues are annotated. A prominent positive groove indicates a likely phosphate-binding site for RNA 5′-triphosphates.
(B) Surface conservation of RIG-I RD in standard view, ranging from dark red (invariant) to white (unconserved). A patch of high sequence conservation colocalizes with the positively charged groove (A, left).
(C) Localization of the mutations, shown in a ribbon model with added side chains. The effect of alanine mutations on pppRVL binding in vitro is highlighted by different colors: red, large effect; orange, medium effect.
(D) Fluorescence anisotropy changes (ΔA) of fluorescently labeled pppRVL in response to titration with WT RD (filled circle, K[d] = 217 ± 11 nM) and mutated RD using the setup of Figure 2A. Two control mutations of conserved residues of the convex side of RD, K807→A (half-filled right-facing triangle, K[d] = 254 ± 16 nM) and D836→A (half-filled square, K[d] = 185 ± 15 nM), did not significantly alter binding affinity of pppRVL. Several mutations in the positively charged groove reduced binding affinity. H830→A (open left-facing triangle, K[d] = 500 ± 30 nM), I875→A (open diamond, K[d] = 1.0 ± 0.1 μM), and K888→A (open down-facing triangle, K[d] = 1.0 ± 0.2 μM) significantly reduced binding affinity. K858→A (open square, K[d] > 5 μM), however, dramatically reduced binding affinity, indicating that this residue is a central recognition site for pppRVL.
(E) HEK293 cells were transfected with IFN-β promoter luciferase reporter constructs and renilla luciferase control vector as well as plasmids expressing WT RIG-I or indicated mutants (10 and 100 ng per transfection). The left panel depicts the more conservative alanine mutants, while the right panel depicts the stronger glutamate charge reversal mutants. Cells were stimulated with transfected pppVSVL or infected with VSV-M51R. IFN-β promoter activity was measured by dual luciferase assay after 18 hr (fold induction compared to mock-treated empty vector control). Mean values and standard deviations (error bars) of three independent experiments are shown.
(F) Proposed model for RNA 5′-triphosphate (gray with red phosphates) activation of RIG-I by ligand-induced dimer formation of RD (yellow with magenta zinc ion). RNA stoichiometry and domain-domain interactions are tentative.

The above figures are reprinted by permission from Cell Press: Mol Cell (2008, 29, 169-179) copyright 2008.