Journal: The EMBO Journal

Article Title: The intrinsically disordered CARDs‐Helicase linker in RIG‐I is a molecular gate for RNA proofreading

doi: 10.15252/embj.2021109782

Figure Lengend Snippet: In the autoinhibitory state of RIG‐I, the helicase is in an open conformation and the CARDs are interacting with Hel2i. The missing CHL, in gray, passes near the RNA binding groove of the helicase domain as its spans the 40 Å distance between the C terminus of CARD2 and N terminus of Hel1. PAMP RNA and ATP binding induces a conformational change to close the helicase subdomains around the ligands and release the CHL and CARDs. Models were generated from crystal structures (PDB ID: 4A2W , left; composite model of two crystal structures PDB IDs: 4NQK and 5E3H, right and colored as in (B)). Domain schematic of RIG‐I with the CHL sequence highlighted. Blue arrows, CARD2 residues known to contact the Hel2i domain in the CARD2:Hel2i interface in autoinhibited duck RIG‐I. Negative and positive charged amino acids are labeled red and blue, respectively. Proline residues are in bold. IFN response of mock‐transfected, WT and CHL RIG‐I mutants measured by luciferase reporter assays in HEK293T cells under No RNA or 50 nM 5’ppp ds39 transfected conditions and reported as Relative Luciferase Units (RLU). Each condition was performed three times by two independent workers ( n = 6), and individual trials are plotted. Bars represent mean value, and error bars reflect standard error. Dose‐titration IFN response of mock‐transfected, WT, and constitutively active CHL RIG‐I mutants measured by luciferase reporter assays in HEK293T cells. Cells were transfected with either no RNA or 700 nM, 70 nM, or 7 nM of 5’ ppp ds39 RNA. Each condition was performed two times by two independent workers ( n = 4). Bars represent mean value, and error bars reflect standard error. 5’ppp ds39 stimulated IFN response of mock, WT, and constitutively active CHL RIG‐I mutants. Each total IFN response in (D) was subtracted by each plasmid’s no RNA transfection condition to show the amount of IFN response explicitly due to transfected RNA. Bars represent the average IFN response of each RNA condition minus the average no RNA IFN response, and error bars represent the average standard error of the two. IFN response assay of mock, WT, and Δ190–200 RIG‐I as in (C) except the assay was performed in HEK293T RIG‐I KO cells. Each condition was performed three times by two independent workers ( n = 6), and individual trials are plotted. Bars represent mean value, and error bars reflect standard error.

Article Snippet: ss39 complementary RNA for 5’ppp ds39 with 3nt DNA overhang: 5'‐Desthiobiotin dT(dT‐FAM)dT CCA AAC AAA CCA ACC ACA AAC ACA CAA CAA CCA CCA ACU‐3’ , Trilink , This paper.

Techniques: RNA Binding Assay, Binding Assay, Generated, Sequencing, Labeling, Transfection, Luciferase, Titration

Journal: The EMBO Journal

Article Title: The intrinsically disordered CARDs‐Helicase linker in RIG‐I is a molecular gate for RNA proofreading

doi: 10.15252/embj.2021109782

Figure Lengend Snippet: A AlphaFold predicted structure of RIG‐I shows CHL bound in the RNA binding pocket of the helicase domain. The colors correspond to model per‐residue confidence (pLDDT): Dark blue, very high (pLDDT > 90), light blue, confident (90 > pLDDT > 70), yellow, low (70 > pLDDT > 50), orange, very low (pLDDT < 50). Regions below 50 pLDDT are predicted to be unstructured, like the CHL in orange. B Graph demonstrating per‐residue confidence (pLDDT) of AlphaFold prediction shown in (A), indicating low‐confidence in the prediction of CHL structure. The CHL (186–241) is highlighted in orange. C Western Blot confirms RIG‐I expression in the reporter assays in Fig . In each experiment, pcDNA3.1 myc‐tagged RIG‐I constructs (approximately 108 kDa) were recognized with a primary α‐Myc antibody. β‐actin (approximately 42 kDa) was used as a normalization control. Numbers (left) refer to molecular weight in kDa. Note that HEK293T and HEK293T RIG‐I KO Western blots were performed on separate gels. D–G qRT–PCR assays show the induction of antiviral IFN response genes in the absence and presence of PAMP RNA, 5’ppp ds39. Note the Y‐axis is in log scale. Each bar represents the mean ± SD. Each point represents a mechanical replicate ( n = 2). H Western blot to show induction of pIRF3 in RIG‐I and RIG‐I mutant transfected cells.

Article Snippet: ss39 complementary RNA for 5’ppp ds39 with 3nt DNA overhang: 5'‐Desthiobiotin dT(dT‐FAM)dT CCA AAC AAA CCA ACC ACA AAC ACA CAA CAA CCA CCA ACU‐3’ , Trilink , This paper.

Techniques: RNA Binding Assay, Residue, Western Blot, Expressing, Construct, Control, Molecular Weight, Quantitative RT-PCR, Mutagenesis, Transfection

Journal: The EMBO Journal

Article Title: The intrinsically disordered CARDs‐Helicase linker in RIG‐I is a molecular gate for RNA proofreading

doi: 10.15252/embj.2021109782

Figure Lengend Snippet: Alignment of RIG‐I homolog sequences spanning human to bony fish from residues 180–228 shows the conserved negatively charged amino acids in the CHL of all species. Note, this CHL region 190–220 was identified in Fig as important for regulation. All glutamate and aspartate residues are colored red, and lysines are colored blue. Asterisks indicate when a negative charge, either glutamate or aspartate, is conserved at that position in three out of five homologs tested. pI was calculated for each RIG‐I homolog region aligning with human RIG‐I amino acids 186–241. RIG‐I domain schematic with the sequence of CHL highlighted. The sequences of the charge flip mutants (cf190–200, cf200–210, cf210–220) are shown. IFN response measured by reporter assays in HEK293T cells in Mock‐transfected, WT RIG‐I, a glycine‐serine‐threonine repeat substitution CHL mutant (neu190–200), and CHL charge flip (cf) mutants under No RNA conditions or 50 nM 5’ ppp ds39 RNA transfected conditions and reported as Relative Luciferase Units (RLU). Each condition was performed three times by two independent workers ( n = 6), and individual trials are plotted. Bars represent mean value, and error bars reflect standard error. 5’ppp ds39 stimulated IFN response. No RNA signal from panel C was subtracted from the total amount of signaling after RNA transfection. Bars represent each plasmid’s average 50 nM 5’ ppp ds39 RNA IFN response minus the average no RNA IFN response, while the error bars represent the average standard errors of each construct with and without RNA transfection. IFN response of Mock, WT RIG‐I, and cf200–210 RIG‐I performed as in (C) in HEK293T RIG‐I KO cells. Each condition was performed three times by two independent workers ( n = 6), and individual trails are plotted. Bars represent mean value, and error bars reflect standard error. 5’ppp ds39 stimulated IFN response of Mock, WT RIG‐I, and cf200–210 RIG‐I as in (D) in HEK293T RIG‐I KO cells. Each total IFN response in (E) was subtracted by each plasmid’s no RNA transfection condition to show the amount of IFN response explicitly due to transfected RNA. Bars represent the average IFN response of each RNA condition minus the average no RNA IFN response, and error bars represent the average standard error of the two.

Article Snippet: ss39 complementary RNA for 5’ppp ds39 with 3nt DNA overhang: 5'‐Desthiobiotin dT(dT‐FAM)dT CCA AAC AAA CCA ACC ACA AAC ACA CAA CAA CCA CCA ACU‐3’ , Trilink , This paper.

Techniques: Sequencing, Transfection, Mutagenesis, Luciferase, Construct

Journal: The EMBO Journal

Article Title: The intrinsically disordered CARDs‐Helicase linker in RIG‐I is a molecular gate for RNA proofreading

doi: 10.15252/embj.2021109782

Figure Lengend Snippet: Schematics of the tested RIG‐I constructs. The 190–210 CHL is highlighted by a thicker black bar. The negative amino acids are in red and positive ones in blue. The negatively charged amino acids in the 190–210 region were mutated to lysines to create the CHL‐Hel‐CTD cf190–210 construct. This change resulted in a pI change in the CHL from to 3.7 to 9.5. Fluorimetric titrations show ds26 stem RNA (Cy3 fluorophore at the 13 th position) binding to the various RIG‐I constructs. The RNA binding data of Hel‐CTD and CHL‐Hel‐CTD cf190–210 were fit using a Hill equation (Hel‐CTD, n = 1.8 ± 0.2; CHL‐Hel‐CTD cf190–210 n = 1.9 ± 0.1). Because WT RIG‐I and CHL‐Hel‐CTD did not finish binding within 400 nM, they could not be normalized. Fluorimetric titrations show 5’ppp ds27 (DY547 at 3’ position distal to the 3nt DNA overhang) binding to WT RIG‐I and CHL‐Hel‐CTD. The fluorescence intensity changes were normalized to the highest fluorescence in each RNA. The data were fit using a hyperbola. Fluorimetric titrations show 5’ppp ds27 binding to Hel‐CTD and CHL‐Hel‐CTD cf190–210. Both proteins bind 5’ppp RNA stoichiometrically. Table summarizes the RNA K D values and shows the association and dissociation times of each RIG‐I construct. No binding or n . b. indicates that stem RNA binding was not observed within 400 nM of titrated RIG‐I. Each K D value was repeated twice (biological replicate, n = 2) The average stem RNA association and dissociation times of the various RIG‐I constructs from stopped‐flow experiments (Fig and Appendix Table ).

Article Snippet: ss39 complementary RNA for 5’ppp ds39 with 3nt DNA overhang: 5'‐Desthiobiotin dT(dT‐FAM)dT CCA AAC AAA CCA ACC ACA AAC ACA CAA CAA CCA CCA ACU‐3’ , Trilink , This paper.

Techniques: Construct, Binding Assay, RNA Binding Assay, Fluorescence

Journal: The EMBO Journal

Article Title: The intrinsically disordered CARDs‐Helicase linker in RIG‐I is a molecular gate for RNA proofreading

doi: 10.15252/embj.2021109782

Figure Lengend Snippet:

Article Snippet: ss39 complementary RNA for 5’ppp ds39 with 3nt DNA overhang: 5'‐Desthiobiotin dT(dT‐FAM)dT CCA AAC AAA CCA ACC ACA AAC ACA CAA CAA CCA CCA ACU‐3’ , Trilink , This paper.

Techniques: Recombinant, Reporter Assay, Cloning, In Vitro, Luciferase, Plasmid Preparation, Software