Coupled equilibria of dimerization and lipid binding modulate SARS Cov 2 Orf9b interactions and interferon response

Abstract

Open Reading Frame 9b (Orf9b), an accessory protein of SARS-CoV and -2, is involved in innate immune suppression through its binding to the mitochondrial receptor Translocase of Outer Membrane 70 (Tom70). Previous structural studies of Orf9b in isolation revealed a β-sheet-rich homodimer; however, structures of Orf9b in complex with Tom70 revealed a monomeric helical fold. Here, we developed a biophysical model that quantifies how Orf9b switches between these conformations and binds to Tom70, a requirement for suppressing the type 1 interferon response. We used this model to characterize the effect of lipid binding and mutations in variants of concern to the Orf9b:Tom70 equilibrium. We found that the binding of a lipid to the Orf9b homodimer biases the Orf9b monomer:dimer equilibrium towards the dimer by reducing the dimer dissociation rate ~100 fold. We also found that mutations in variants of concern can alter different microscopic rate constants without significantly affecting binding to Tom70. Together, our results highlight how perturbations to different steps in these coupled equilibria can affect the apparent affinity of Orf9b to Tom70, with potential downstream implications for interferon signaling in coronavirus infection.

Keywords: E. coli; Orf9b; Tom70; biochemistry; chemical biology; coronavirus; human.

Figure 1.. Peptides of Orf9b can be used to model the behavior of monomeric Orf9b to Tom70.

(A) Initial model of the Orf9b-Tom70 equilibrium. (B) (Left) Schematic of SPR binding interaction between Tom70 and Orf9b peptide. (Right) Surface plasmon resonance sensorgram of Orf9b peptide binding to immobilized Tom70 (orange) with model fits (black) showing kinetics and dissociation constant. (C) (Left) Schematic of SPR binding interaction between Tom70 and Orf9b-FITC. (Right) Surface plasmon resonance sensorgram of Orf9b-FITC peptide binding to immobilized Tom70 (green) with model fits (black) showing kinetics and dissociation constant. (D) (Left) Diagram of Orf9b-Tom70 fluorescent polarization assay. (Right) Model ODE (solid lines) overlaid with Orf9b peptide competition kinetics (circles) and model parameters used to generate results.

Figure 1—figure supplement 1.. Determining the KD of Orf9b peptides for Tom70.

(A) Determining the K_D of Orf9b-FITC:Tom70 by titration of Tom70 against a fixed concentration of Orf9b-FITC. 200nM of Orf9b-FITC was kept constant and titrated with increasing concentrations of Tom70. A non-linear regression model using 1:1 binding was used to calculate the K_D. The K_D was 240±6.8Nm. (B) Determining the K_i of unlabeled Orf9b peptide by competition binding. Orf9b peptide was titrated against 200nM Orf9b-FITC and 2.5uM of Tom70. Non-linear regression using a one-site competitive binding model was used to calculate the K_i of the unlabeled Orf9b peptide using the calculated K_D of Orf9b-FITC:Tom70 from (A). The unlabeled Orf9b peptide K_i was 880±5.3nM.

Figure 1—figure supplement 2.. S53E phosphomimetic Orf9b peptides do not bind to Tom70.

(A) Phosphorylated Orf9b at S53 does not bind to Tom70. Zoomed-in view of Orf9b bound to Tom70 showing the location of Orf9b S53 relative to the surrounding Tom70 residues. (B) Illustration of the Orf9b S53E peptide derived from the full-length WT Orf9b sequence. (C) Orf9b S53E peptide FP kinetic assay results showing that the phosphomimetic peptide does not bind to Tom70.

Figure 1—figure supplement 3.. Plots of residuals from Orf9b peptide model showing effect of an increase or decrease by 10% on each model parameter.

All residuals and reporting are with respect to the 100 µM of unlabeled Orf9b peptide condition. Blue dots: reported value. Red dots: 10% increase in reported value. Green dots: 10% decrease in reported value. Table reporting of RMSD values for model fitsafter +/-10% change to model parameter (Left column) and percent change in RMSD relative to reported model RMSD (Right column).

Figure 2.. The Orf9b homodimer can be refolded to eliminate a co-purifying lipid.

(A) Schematic of Orf9b denaturing purification, refolding, and tag-cleavage process. (B) Size exclusion chromatography chromatogram overlay of WT natively folded (black) and WT refolded (blue) Orf9b homodimer. Both proteins purify as a dimer and have identical retention volumes indicating high recovery of refolded Orf9b homodimer. (C) Structural overlay of WT Orf9b homodimer (green-PDB 6Z4U) and WT refolded Orf9b homodimer (magenta/gray-PDB 9MZB). (D) Zoomed-in view of the natively folded Orf9b homodimer central channel. 2Fo-Fc (contoured to 1σ) and Fo-Fc (+/-3σ) density maps show continuous density that supports the placement of a ligand in the dimer central channel. (E) Zoomed-in view of the refolded Orf9b homodimer central channel: 2Fo-Fc (1σ) and Fo-Fc (+/-3σ) density maps show diffuse peaks within the central channel of the refolded Orf9b homodimer. (F) Deconvolved native mass spectra of refolded and natively folded Orf9b homodimers in negative-polarity mode. Lipid-bound homodimers are observed only in the natively folded sample. Both preparations display +60 Da and +257 Da adducts of unknown origin.

Figure 2—figure supplement 1.. Tandem mass spectrometry reveals the identities of possible lipids bound to the Orf9b homodimer in different expression systems.

(A) Mass spectra of lipids extracted from natively folded Orf9b in both (top) E. coli and (bottom) mammalian expression systems. (B) (Top) Tandem mass spectrometry (MS/MS) ionization spectra of 1-Palmitoyl-sn-glycerol and (bottom)1-Strearoyl-sn-glycerol from bacterial expression systems. (C) (Top) Tandem mass spectrometry (MS/MS) ionization spectra of 1-Palmitoyl-sn-glycerol and (bottom) 1-Strearoyl-sn-glycerol from mammalian expression systems.

Figure 2—figure supplement 2.. Polder maps support the absence of lipids bound to the refolded Orf9b homodimer.

(A) Polder map calculated for the lipid-molecule bound to the Orf9b homodimer. Polder map shows strong support for the presence of the lipid with CC1,3 value being 0.9110. (B) Polder map calculated for the refolded Orf9b homodimer with the lipid molecule modeled into the central channel. Polder maps do not support the placement of the lipid with CC1,3 value being 0.4511 indicating that residual Fo-Fc and 2Fo-Fc density peaks are explained by either noise or the bulk solvent mask. (C) Polder maps calculated for the refolded Orf9b homodimer with glycine (from crystallization conditions) modeled into the largest 2Fo-Fc peak within the central channel. Polder maps plausibly support placement of glycine within the central channel with CC1,3 value being 0.9292.

Figure 3.. Serial dilution allows for estimation of Orf9b homodimer dissociation constant.

(A) Size exclusion chromatography overlay of Orf9b homodimer (black) with molecular weight standards (blue). Apo-Orf9b forms two distinct peaks corresponding to the dimer (peak 1) and the monomer (peak 2). (B) Coomassie stain of SDS-PAGE gel from peaks 1 and 2 (from A) showing Orf9b is present in each peak. (C) Size exclusion chromatography chromatogram of serially diluted WT lipid-bound Orf9b homodimer showing no monomer species present. (D) Size exclusion chromatography chromatogram of serially diluted WT apo-Orf9b showing both homodimer and monomer peaks. Running molecular weight standards further supported our observation that the first peak was the Orf9b homodimer due to its close overlap with a 29 kDa standard; however, the monomeric peak eluted after a 6.5 kDa standard rather than between the 13.7 kDa and 6.5 kDA standards as we would expect. We hypothesize that this discrepancy is due to two possibilities: hydrophobic interactions between the monomer and the column could increase the retention volume or monomeric Orf9b has a smaller hydrodynamic radius than expected. (E) Nonlinear regression analysis of the fraction of Orf9b in the homodimer versus the total concentration of Orf9b (from D) injected over the size exclusion column yields a K_D of 1.2±0.1 uM.

Figure 3—figure supplement 1.. Refolded Orf9b binds to Tom70 and co-elutes as a complex by size exclusion chromatography.

(A) Size exclusion chromatogram of Tom70:Orf9b, Tom70, and Orf9b overlaid. Refolded Orf9b and Tom70 show a left shift relative to Tom70 alone, indicating the formation of a higher molecular weight complex. (B) SDS-polyacrylamide gel of the Tom70:Orf9b SEC fractions with molecular weights of all possible species present. The first fractions (14-15) show Tom70 and Orf9b eluting together, which is indicative of the Tom70:Orf9b complex, followed by a fraction that is mostly Tom70 (16) and then two fractions containing mostly Orf9b (17-18).

Figure 4.. Modeling the effect of lipid-binding on Orf9b-Tom70 equilibrium using SPR and FP-based assay.

(A) I. Mathematical model overlay to FP competition kinetic assay using the lipid-bound Orf9b homodimer as the competitor. II. Mathematical model overlay to FP competition kinetic assay using the apo-Orf9b homodimer as the competitor. (B) (Left) Schematic diagram of monomeric Orf9b binding to immobilized monomeric Orf9b to determine dimerization rate constants. (Right) Surface plasmon resonance sensorgram of full-length Orf9b binding to full-length Orf9b monomer and their interaction kinetics. Experimental binding curves (blue) were globally fit (black) using a 1:1 binding model. (C) (Left) Schematic of full-length apo-Orf9b homodimer in exchange with Orf9b monomers. Monomeric Orf9b binds to immobilized Tom70 to determine rate constants of full-length Orf9b binding to Tom70. (Right) Surface plasmon resonance sensorgram of refolded apo-Orf9b binding to Tom70 and their interaction kinetics. Experimental binding curves (dark blue) were globally fit (black) using a 1:1 binding model. (D) Proposed kinetic model of Orf9b-Tom70 equilibrium. Orf9b homodimer dissociates into Orf9b monomers which undergo a conformational change from β-sheet to ɑ-helix. ɑ-helical Orf9b binds to Tom70. (E) Table of model parameters used to model ODEs to both apo and lipid-bound Orf9b homodimer competition binding kinetics.

Figure 4—figure supplement 1.. Orf9b S53E refolded homodimer FP kinetic assay results showing that the phosphomimetic mutation does not bind to Tom70.

Figure 4—figure supplement 2.. Predicted iMTS regions outside of the strucutrally resolved portions of Orf9b bound to Tom70 do not bind to Tom70.

(A) AlphaFold predictions of monomeric Orf9b without structural templates used. The region of Orf9b that is known to bind to Tom70 is highlighted in magenta, and the residues that form the Orf9b homodimer interface are highlighted in pink. A second helix is predicted to form in monomeric Orf9b comprising residues 11–28. The top 5 predicted structures are shown with pLDDT scores less than 50 indicating a low confidence in the predicted monomeric structure of Orf9b. (B) FP kinetic assay using a peptide composed of residues 11-28 shows no binding to Tom70 at the C-terminal domain where the structurally resolved portion of Orf9b binds. (C) SPR using a peptide composed of residues 11–28 against immobilized Tom70 shows no binding indicating that the second predicted helix does not bind to Tom70.

Figure 4—figure supplement 3.. Comparison of kinetic model 1 and 2 in describing experimental results from the kinetic binding assay.

Experimental results using 10uM of refolded Orf9b homodimer are shown as rings with the predicted behavior of model 1 (equilibrium exchange) shown as a dark blue line. The predicted behavior of model 2 (equilibrium exchange with a conformational change between β-sheet and ɑ-helical monomers) is shown as the light blue line. Model parameter values were the same as described in Figure 4D and kept constant in both model comparisons.

Figure 4—figure supplement 4.. Plots of model behavior showing the effect of changes to alpha-beta and beta-alpha monomer interconversion rates compared to experimental values.

Data is modeled with respect to the apo-Orf9b homodimer 5uM condition. The black line represents reported model fit and values used.

Figure 4—figure supplement 5.. Plot of residuals showing the effect of increasing or decreasing individual model parameters 10% compared to the reported values.

All residual plots are with respect to the 5uM apo-Orf9b homodimer condition. Blue dots: reported value. Red dot: 10% increase in reported value. Green dot: 10% decrease in reported value. (Left columns) Table of RMSD values calculated from model fits showing the effect of both +/-10% change to individual model parameters. (Right columns) Percent change in RMSD values subjected to +/-10% change for individual model parameters relative to the RMSD of the reported model.

Figure 5.. Truncated and fused Orf9b constructs can destabilize or stabilize the homodimer to modulate Tom70 binding.

(A) Model of Orf9b homodimer illustrating the ∆91–97 truncation (cyan). Residues 91–97 form 7 hydrogen bonds (7 mediated by backbone backbone interactions) between chains A and B (black dashes) of the homodimer and represent ⅔ of the homodimer binding interface. (B) Berkeley Madonna model (solid lines) overlaid experimental data points (circles) for the FP competition kinetic assay using Orf9b ∆91–97. (C) AlphaFold model of Orf9b homodimer with a 4xSGG linker (green) fusing the C-terminus of chain A (gray) to the N-terminus of chain B (magenta). (D) Fitted model (solid lines) overlaid with experimental data points (circles) for the FP competition kinetic assay using Orf9b 4xSGG fusion construct. (E) Model parameters used for solving ODE in Berkeley Madonna for both ∆91–97 and 4xSGG constructs. (F) Co-immunoprecipitation between Flag-Tom70 and Strep-Orf9b mutants from HEK293T cells and immunoprecipitation was performed using anti-flag magnetic beads. Representative western blots of whole-cell lysates (WCLs) and eluates after IP are shown. Actin was used as a loading control in WCLs. (G) Fold induction of ISRE-reporter activated by 3p-hpRNA upon expression of empty vector, Orf9b WT, Orf9b fused, Orf9b, truncated, and Orf9b S50/53E in HEK293T cells. Fold induction was calculated relative to unstimulated cells.

Figure 5—figure supplement 1.. Mutations in Orf9b variants of concern retain homodimer behavior but do not alter Tom70 binding.

(A) Size exclusion chromatography chromatogram of the Orf9b ∆91-97 truncated construct. The predicted homodimer has a peak centered at a retention volume of 70mL followed by a much larger peak that eluted at the end of the column volume, which we attribute to the monomeric species. (B) (Left) Size exclusion chromatography chromatogram of the Orf9b 4xSGG fusion construct. The predicted homodimer has a retention volume of 70mL. (Right) SDS-PAGE of the homodimer elution peak showing the predicted molecular weight of the Orf9b fusion construct of ~25kDa (red arrow). The minor band between 10kDa and 15kDa we attribute to a minor degradation product.

Figure 6.. Orf9b mutations in variants of concern modulate equilibrium kinetics with Tom70.

(A) An AlphaFold model of the full Orf9b homodimer with mutations found in variants of concern shown as red spheres. Resides that fold into the Tom70-binding ɑ-helix are highlighted in magenta on one protomer and residues that form the dimer interface are highlighted in pink. (B) Berkeley Madonna model (black lines) overlaid on experimental data points (circles) for the FP competition kinetic assay using Lambda, Omicron, and Delta Orf9b variants. (C) Table of model parameters used to model experimental data in B using Berkeley Madonna. (D) Location of point mutations in variants of concern modeled in Pymol: (Left) T60 forms a hydrogen bond with the backbone carbonyl between A68 and F69 of the same chain that is lost in T60A. (Right) Point mutation P10S (red) introduces a serine that forms several hydrogen bonds with R13 and the backbone carbonyl of A11 and L48 within the same chain. Both mutations are mirrored on the opposite protomer. Hydrogen bonds are shown as black dashes. Mutations are modeled in Pymol using point mutation. (E) Co-immunoprecipitation of endogenous Tom70 with Strep-tagged Orf9b from variants of concern in HEK293T cells. (F) Fold induction of ISRE-reporter activated by 3p-hpRNA in HEK293T cells expressing empty vector, Orf9b WT, Orf9b from VOC (Delta, Lambda, and Omicron).

Figure 6—figure supplement 1.. Biological replicates of Orf9b variants with model fits and model parameters used.

(A) Size exclusion chromatogram showing overlaid chromatograms of refolded WT, Delta, Lambda, and Omicron Orf9b homodimers with normalized A280 absorbance. The retention volumes corresponding to the homodimer are centered at ~70mL in all cases, indicating recovery of the homodimer after refolding. (B) (Left) SPR diagram of immobilized Tom70 and Orf9b peptide used as an analyte. (Right) SPR response curves using Orf9b T60A peptide as the analyte. Experimental response curves are shown in green and global 1:1 binding fits are shown in black with corresponding kinetic rate constants and K_D.

Figure 6—figure supplement 2.. Biological replicates of all Orf9b variants and constructs performed in FP kinetic assay with model results (solid black lines) and parameters used.

All concentrations refer to the Orf9b homodimer concentrations unless specified otherwise (see Delta T60A).

Figure 7.. Proposed model of the Orf9b-Tom70 equilibrium.

Orf9b transcripts are translated to produce monomeric Orf9b proteins that can bind to Tom70 leading to a suppression of IFN. Monomeric Orf9b can also undergo a conformational change and bind to a second copy of Orf9b to form the homodimer. The Orf9b homodimer is in equilibrium between free Orf9b monomers and Orf9b monomers bound to Tom70. Upon lipid binding, the homodimer is tightly stabilized and slowly releases monomeric Orf9b.