Inhibiting the interaction between the mitochondrial receptor Tom70 and SARS CoV 2 Orf9b with small molecules

Article activity feed

1. 

   eLife

   Nov 3, 2026

   eLife Assessment

   This important study combines crystallographic fragment-based screening with fluorescence-polarization competition assays to demonstrate that the SARS-CoV-2 accessory protein ORF9b, and its interface with the mitochondrial import receptor TOM70, is accessible to small-molecule binding. This is a result with practical implications beyond coronavirus biology for efforts to restore interferon signalling and the antiviral response. The structural evidence is convincing, resting on an unusually large body of ligand-bound crystal structures whose fragment hotspots are independently corroborated by orthogonal biophysical approaches including fluorescence polarization and SPR. This study will be of interest to the broad readership of eLife, spanning structural biology, drug discovery, and coronavirus communities.

   Read the original source
2. 

   eLife

   Nov 3, 2026

   Reviewer #2 (Public review):

   Summary:

   The authors investigate chemical strategies to disrupt the interaction between the SARS-CoV-2 accessory protein Orf9b and the host mitochondrial receptor Tom70, an interaction implicated in suppression of type-I interferon responses. They employ two discovery approaches: a crystallographic fragment screen against the Orf9b homodimer and a high-throughput fluorescence polarization screen for compounds that compete with Orf9b binding to Tom70. The study identifies fragment-binding hotspots on Orf9b, develops lipidated analogs that stabilize the Orf9b homodimer, and discovers Tom70-binding compounds with low micromolar activity that inhibit Orf9b binding in vitro.

   Strengths:

   Impressive amount of work using a variety of complementary approaches and methods to validate binding (FP, SPR, and computational …

   Reviewer #2 (Public review):

   Summary:

   The authors investigate chemical strategies to disrupt the interaction between the SARS-CoV-2 accessory protein Orf9b and the host mitochondrial receptor Tom70, an interaction implicated in suppression of type-I interferon responses. They employ two discovery approaches: a crystallographic fragment screen against the Orf9b homodimer and a high-throughput fluorescence polarization screen for compounds that compete with Orf9b binding to Tom70. The study identifies fragment-binding hotspots on Orf9b, develops lipidated analogs that stabilize the Orf9b homodimer, and discovers Tom70-binding compounds with low micromolar activity that inhibit Orf9b binding in vitro.

   Strengths:

   Impressive amount of work using a variety of complementary approaches and methods to validate binding (FP, SPR, and computational modelling and SAR). The combination of crystallographic fragment screening on Orfb9 and HTS on Tom70 provides two independent routes for perturbing the Orf9b-Tom70 interaction. The structural work seems to be of very high-quality. The fragment campaign is extensive, yielding a substantial number of fragment-bound structures and identifying biologically meaningful binding hotspots on Orf9b.

   Finally, the screen results in reporting useful chemical starting points. Although potency remains modest, the study provides tractable scaffolds and a clear framework for future optimization.

   Comment on revised version.

   I don't have any further comments. The authors have addressed my suggestion by adding a statement for future work in the discussion.

   Read the original source
3. 

   eLife

   Nov 3, 2026

   Author response:

   The following is the authors’ response to the original reviews.

   Public Reviews:

   Reviewer #1 (Public review):

   Summary:

   The authors used FBDD screening to identify numerous compounds interacting with the ORF9b dimer. They expanded the original fragment hit, soaked the derivatives into the crystals and confirmed their binding poses, and showed that the derivatives bind the target with higher affinity. The authors further targeted the ORF9b binding site on TOM70, and used a fluorescence polarization-based (FP) assay to screen a compound library and obtained several hits. Structure-activity relationship (SAR) optimization yielded hit analogs that have higher binding affinity to TOM70.

   Strengths:

   (1) The study adopted novel drug design strategies, including stabilizing ORF9b homodimer to prevent it from binding …

   Author response:

   The following is the authors’ response to the original reviews.

   Public Reviews:

   Reviewer #1 (Public review):

   Summary:

   The authors used FBDD screening to identify numerous compounds interacting with the ORF9b dimer. They expanded the original fragment hit, soaked the derivatives into the crystals and confirmed their binding poses, and showed that the derivatives bind the target with higher affinity. The authors further targeted the ORF9b binding site on TOM70, and used a fluorescence polarization-based (FP) assay to screen a compound library and obtained several hits. Structure-activity relationship (SAR) optimization yielded hit analogs that have higher binding affinity to TOM70.

   Strengths:

   (1) The study adopted novel drug design strategies, including stabilizing ORF9b homodimer to prevent it from binding TOM70, and blocking ORF9b from binding TOM70 by screening compounds that compete with ORF9b for binding TOM70.

   (2) The work established a feasible high-throughput screening assay. This FP-based assay screened ~50,000 compounds, from which two hit compounds were further optimized to yield analogs with higher binding affinity.

   We thank reviewer #1 for their thoughtful assessment of our work and their constructive feedback.

   Weaknesses:

   (1) The study lacks functional assays to evaluate whether the ORF9b-stabilized compounds or TOM70 binding compounds could affect IFN inhibition caused by ORF9b or virus infection.

   We thank reviewer #1 for their feedback on this subject. We agree that cell-based assays measuring the impact of these compounds on IFN stimulation/suppression would be a valid next step. We have expanded the discussion section to include potential future experiments that could be conducted to probe this:

   “Although we have not assessed the permeability of these compounds and their ability to restore IFN in the context of cell-based assays, one way this could be accomplished is by adapting a previously used reporter cell line. Specifically, this assay uses a reporter cell line that has a Lucia reporter under the control of the IFN-β/ISG56 promoter that is expressed upon mock infection with synthetic dsRNA. When WT Orf9b is co-transfected with the reporter, IFN is suppressed, and, we have shown that phosphomimetic mutants of Orf9b that block its interaction with Tom70 restores IFN signaling. If these compounds are sufficiently permeable, then, for the lipidated-compounds targeting the Orf9b homodimer, we would expect the Orf9b monomer: homodimer equilibrium to shift towards the homodimer due to the stabilizing effect these lipidated-compounds have on the homodimer conformation leading to an increase in IFN signaling. Similarly, for compounds targeting Tom70, we would expect that these compounds block the Orf9b monomer from binding to Tom70 to produce a similar effect.”

   (2) There is a lack of experimental evidence to reveal the binding mode of lipidated-compounds with ORF9b homodimer.

   - We thank reviewer #1 for their feedback. We agree that we were unable to provide direct evidence that our lipidated-compounds are bound to the homodimer through the central lipid-binding channel. Experimental structures of our compounds bound to the Orf9b homodimer would have strengthened our functional studies, however, as we described in the manuscript pertaining to our initial crystallographic fragment screen:

   “Although we have shown that it is possible to refold the Orf9b homodimer to eliminate the lipid from the central channel, we found that apo-Orf9b homodimer crystals were far less tolerant to DMSO and on average diffracted to resolutions worse than 2.5Å. Importantly, we found that microseeding helped increase crystal reproducibility and yielded larger crystals on average. Therefore, for our crystallographic fragment screening campaign we elected to use the lipid-bound protein to identify additional potential binding sites that could be exploited to stabilize the Orf9b homodimer.”

   - We did undertake efforts to co-crystallize our lipidated-compounds with refolded apo-Orf9b homodimers but were unable to generate crystals of sufficient quality to collect X-ray data. Therefore, we relied on both our fluorescence polarization assay and prediction models like Chai1 to investigate if the expected behavior could be recapitulated. We felt this was reasonable for three reasons.

   - First, as we explain in the discussion, none of these compounds or structures were publicly available for models like Chai1 to train on which prevents data leakage:

   “None of these structures were publicly available on the PDB prior to these structural predictions, therefore, they were not available for Chai1 to train on. The placement of the lipidated-analogs in the central channel was in line with the anticipated binding mode, albeit, with discrepancies between the predicted and structurally observed portions of the analog compounds.”

   - Second, as we outlined in the fluorescence polarization assay, we tested these compounds in two control experiments that investigated whether the changes in FP signal over time could also be explained by the compounds binding to Tom70 directly, or, by acting on the obligate Orf9b monomer which was generated by truncating the N and C-termini:

   “We first considered whether the lipidated-analogs had any binding activity for Tom70 on its own and whether changes in FP signal were due to binding to the Orf9b homodimer. We tested the compounds on their own in the presence of the pre-equilibrated Tom70:Orf9b-FITC complex and observed no significant decrease in FP signal indicating that the lipidated-analogs do not directly bind to Tom70 to drive a decrease in FP signal (Figure 2—figure supplement 2A). We next tested whether these compounds acted on the Orf9b homodimer specifically or if they could inhibit the Orf9b monomer from binding to Tom70 by incubating compounds with an Orf9b peptide that acts as an obligate monomer. When chasing with the pre-equilibrated Tom70:Orf9b-FITC complex, we observed no significant differences in kinetics of Orf9b peptide binding compared to the peptide on its own which showed a fast single-exponential kinetic decay in signal suggesting that these compounds do not act on the Orf9b monomer to inhibit binding to Tom70 (Figure 2—figure supplement 2B)”

   (3) There is a lack of experimental evidence to reveal the binding mode of HTS hits or analogs for TOM70.

   We thank reviewer #1 for their feedback. We agree that experimental structures of our HTS compounds bound to Tom70 would provide more insights into how these compounds bind to Tom70 (and their precise location) in ways that the predicted complexes from models like Chai1 were unable to do. At the time of drafting this manuscript, we had not yet identified suitable conditions for structurally resolving Tom70 with our HTS compounds. Therefore, we elected to rely on two orthogonal methods of predicting where our HTS compounds are bound to probe whether the predicted binding sites could plausibly block Orf9b from binding. In the time since releasing this manuscript, one group had been able to crystallize apo-Tom70 which may serve as a suitable platform for future structural studies investigating these compounds and their exact binding mode to Tom70 to enable structure-guided optimisations of these initial hits.

   (4) Overall, none of the compounds shown in the paper have promising potency warranting further development; their binding affinity is limited to the micromolar range.

   We thank reviewer #1 for their feedback. We agree that these compounds should be viewed as starting points for potential future optimization on targets where there were no previous existing ligands. The molecules serve to validate that it is possible to develop small molecules that target Tom70’s C-terminal domain which may be relevant to future studies investigating Tom70’s biology.

   Reviewer #2 (Public review):

   Summary:

   The authors investigate chemical strategies to disrupt the interaction between the SARS-CoV-2 accessory protein Orf9b and the host mitochondrial receptor Tom70, an interaction implicated in suppression of type-I interferon responses. They employ two discovery approaches: a crystallographic fragment screen against the Orf9b homodimer and a high-throughput fluorescence polarization screen for compounds that compete with Orf9b binding to Tom70. The study identifies fragment-binding hotspots on Orf9b, develops lipidated analogs that stabilize the Orf9b homodimer, and discovers Tom70-binding compounds with low micromolar activity that inhibit Orf9b binding in vitro.

   Strengths:

   An impressive amount of work using a variety of complementary approaches and methods to validate binding (FP, SPR, and computational modelling and SAR). The combination of crystallographic fragment screening on Orfb9 and HTS on Tom70 provides two independent routes for perturbing the Orf9b-Tom70 interaction. The structural work seems to be of very high-quality. The fragment campaign is extensive, yielding a substantial number of fragment-bound structures and identifying biologically meaningful binding hotspots on Orf9b.

   Finally, the screen results in reporting useful chemical starting points. Although potency remains modest, the study provides tractable scaffolds and a clear framework for future optimization.

   We thank reviewer #2 for their thoughtful assessment of our work and their constructive feedback.

   Weaknesses:

   General comment:

   (1) Although there is already an incredible amount of data presented, one limitation of this study is the lack of cellular validation - do these drugs enter cells, restore interferon signalling, reduce viral loads, or alter Orfb9 localization?

   We thank reviewer #2 for their feedback. See also the feedback from Reviewer 1. We have not assessed the permeability of these compounds in the context of cell-based assays and IFN signaling and agree that cell-based assays measuring the impact of these compounds on IFN stimulation/suppression would be a valid next step. We have expanded the discussion section to include potential future experiments that could be conducted to probe this:

   “Although we have not assessed the permeability of these compounds and their ability to restore IFN in the context of cell-based assays, one way this could be accomplished is by adapting a previously used reporter cell line. Specifically, this assay uses a reporter cell line that has a Lucia reporter under the control of the IFN-β/ISG56 promoter that is expressed upon mock infection with synthetic dsRNA. When WT Orf9b is co-transfected with the reporter, IFN is suppressed, and, we have shown that phosphomimetic mutants of Orf9b that block its interaction with Tom70 restores IFN signaling. If these compounds are sufficiently permeable, then, for the lipidated-compounds targeting the Orf9b homodimer, we would expect the Orf9b monomer: homodimer equilibrium to shift towards the homodimer due to the stabilizing effect these lipidated-compounds have on the homodimer conformation leading to an increase in IFN signaling. Similarly, for compounds targeting Tom70, we would expect that these compounds block the Orf9b monomer from binding to Tom70 to produce a similar effect.”

   (2) The logic of locking Orfb9 as a dimer is that the monomer binds Tom70 - thus, a more stable dimer means less monomer. In Figure 2, the Orfb9 homerdimer stabilization by compounds should reduce binding affinity to Tom70. A direct binding experiment measuring reduced Tom70 binding with compound treatment would better strengthen this claim.

   We thank reviewer #2 for their observation and agree that compounds targeting the Orf9b homodimer conformation should block binding to Tom70 by preventing Orf9b homodimer dissociation into Orf9b monomers. As schematized in Figure 2D, we described how our fluorescence polarization assay measures the binding of Orf9b to Tom70. It provides insight into how compounds that bind to the Orf9b homodimer may slow the rate of homodimer dissociation. In addition, we have previously demonstrated in a separate publication that there are clear differences in the Orf9b homodimer dissociation kinetics when using either lipid-bound or lipid-free Orf9b homodimers as the competitor due to the stabilizing effect the lipid has on the homodimer conformation.

   Reviewer #3 (Public review):

   Summary:

   This paper attempts to and succeeds in demonstrating that Orf9b is able to bind small molecules using X-ray fragment screening, SPR and FP assays. Exploration of sites from the fragment screening is performed along with fragment linking with inter-dimer lipid moieties.

   Strengths:

   The experimental work looks strong and well performed. The interpretation of the data is appropriate and was often validated through orthogonal methods and follow-up compounds. The use of Tom70 to find binders that might disrupt interactions between Orf9b and Tom70 is elegant.

   Weaknesses:

   The use of Chai-1 to predict co-folded structures with binding molecules was not properly described - no mention of this in the methods. It was not commented on whether the compounds which were found were attempted to be co-crystallised. If they were but negative data was collected (didn't crystallise, didn't diffract or no additional density was found), then this needs to be stated.

   We thank reviewer #3 for their feedback. We have revised the manuscript to clearly point out that beyond our initial attempts to crystallize apo-Orf9b homodimer to enable fragment soaking, we also attempted to both soak our lipidated-compounds into apo-Orf9b homodimer crystals as well as co-crystallize the compounds which did not yield crystals.

   “We attempted two strategies of crystallizing our lipidated-compounds with the apo-Orf9b homodimer to structurally resolve the complex. First, we attempted to soak our compounds into apo-Orf9b homodimer crystals but observed poor tolerance to DMSO which quickly disintegrated the crystals after minutes. Second, we attempted to co-crystallise the lipidated-compounds after refolding the Orf9b homodimer to generate apo-Orf9b homodimers but could not identify conditions that yielded crystals upon re-screening.”

   We have also revised the methods and materials section to better explain the use of Chai1 to co-fold both Orf9b and Tom70 with their respective compounds.

   “Chai1’s web server was used for co-folding either Orf9b or Tom70 with their respective hits. For the Orf9b homodimer, the sequence for a single chain was supplied to represent chains A and B in the homodimer and the ligands were supplied as SMILES strings. In the case where two copies of the ligand were investigated, the number of copies of the ligand was increased to two. Co-folding was performed using MMSeqs2 for the MSA and template searching to ensure the correct folding of the homodimer conformation. Tom70 co-folding was performed by supplying the sequence for hTom70 as a single chain and the compounds as SMILES strings. Co-folding was performed using MMSeqs2 for the MSA with template searching.”

   Recommendations for the authors:

   Reviewer #1 (Recommendations for the authors):

   (1) It is disappointing that the authors screened binders for ORF9b homodimer, identified hits and derivatives binding at sites 1 & 2, measured the binding affinity to the target, but found no effects on stabilizing ORF9b homodimer. The FP competition assay was used to assess the efficacy of the compounds; maybe the authors could use alternative methods like Thermal Shift Assay (TSA) assays to detect some subtle changes in stability.

   We thank the reviewer for their comments and agree that the lack of stabilizing activity observed for the site 1 and 2 analogs was disappointing. However, we do feel that the lipid-binding central channel may be a viable alternative site target that may also yield insights into the role of lipid binding in Orf9b biology. We agree that orthogonal methods such as TSA could be insightful for probing the stabilizing effects of the compounds on the Orf9b homodimer.

   (2) To better illustrate the results of the FP kinetic assay to investigate whether lipidated-compounds binding to the Orf9b homodimer had effects on the stability of the Orf9b homodimer and Orf9b-Tom70 equilibrium, I recommend moving the results shown in Supplemental Figure 4 to a main figure and merging these results with Figure 2CD.

   We think that the eLife publication platform provides appropriate visibility for the supplementary figures and that Figure 2D does the best job of illustrating the effects on kinetics for the initial reader.

   (3) Have the authors tried to co-crystallise their top-rated lipidated-compounds with ORF9b homodimer to see where the lipid part of the compound could extend into the central channel of ORF9b homodimer?

   We have revised the manuscript to clearly point out that beyond our initial attempts to crystallize apo-Orf9b homodimer to enable fragment soaking, we also attempted to both soak our lipidated-compounds into apo-Orf9b homodimer crystals as well as co-crystallize the compounds which did not yield crystals.

   “We attempted two strategies of crystallizing our lipidated-compounds with the apo-Orf9b homodimer to structurally resolve the complex. First, we attempted to soak our compounds into apo-Orf9b homodimer crystals but observed poor tolerance to DMSO which quickly disintegrated the crystals after minutes. Second, we attempted to cocrystallize the lipidated-compounds after refolding the Orf9b homodimer to generate apo-Orf9b homodimers but could not identify conditions that yielded crystals upon re-screening.”

   (4) Have the authors tried to co-crystalize hit compounds or hit analogs resulting from SAR with the TOM70 protein, to really observe the binding mode of these compounds? Only using in silico methods is not satisfactory.

   We initially attempted to develop a crystallization system for Tom70 that would enable us to structurally resolve our bound compounds, but, at the time of writing the manuscript, had not identified a system that yielded crystals. However, another group has published a structure of apo-human Tom70 (https://pmc.ncbi.nlm.nih.gov/articles/PMC13075491/) - we sent them compounds and were unable to identify convincing density by soaking. We agree that a future study in a different crystal system or by pursuing co-crystallization could be fruitful.

   (5) Although the authors improved the binding affinity of HTS hits for TOM70 using SAR, there is a lack of functional assays to evaluate whether the HTS hits or analogs could affect IFN inhibition caused by ORF9b or virus infection (cell-based or animal-based assays).

   We have expanded the discussion section to include potential future experiments that could be conducted to probe this:

   “Although we have not assessed the permeability of these compounds and their ability to restore IFN in the context of cell-based assays, one way this could be accomplished is by adapting a previously used reporter cell line. Specifically, this assay uses a reporter cell line that has a Lucia reporter under the control of the IFN-β/ISG56 promoter that is expressed upon mock infection with synthetic dsRNA. When WT Orf9b is co-transfected with the reporter, IFN is suppressed, and, we have shown that phosphomimetic mutants of Orf9b that block its interaction with Tom70 restores IFN signaling. If these compounds are sufficiently permeable, then, for the lipidated-compounds targeting the Orf9b homodimer, we would expect the Orf9b monomer:homodimer equilibrium to shift towards the homodimer due to the stabilizing effect these lipidated-compounds have on the homodimer conformation leading to an increase in IFN signaling. Similarly, for compounds targeting Tom70, we would expect that these compounds block the Orf9b monomer from binding to Tom70 to produce a similar effect.”

   (6) Figure 4 doesn't have A, B, or C indicated in the figures.

   We have revised Figure 4 to include A, B, and C.

   Reviewer #2 (Recommendations for the authors):

   (1) Page 5 line 23: The authors could comment on the biological relevance of the lipid-bound version of Orfb9 and how it could associate with membranes.

   We thank the reviewer for their comments. Currently, it is unknown what role lipid-binding plays on Orf9b’s function beyond the stabilizing effect it has on the homodimer conformation. While it has been shown that Orf9b co-localizes at the outer mitochondrial membrane through Tom70 association, it has not been shown that Orf9b itself binds to membranes. While this does not exclude the possibility that the Orf9b homodimer could directly bind to membranes, we suspect a more plausible explanation is that Orf9b homodimers can bind to lipid-like metabolites within cells such as palmitate which we had previously shown.

   (2) Figure 3F: These SPR sensorgrams are not all to the same y-axis scale.

   We thank the reviewer for their comment. The sensorgram axis scale is reflective of three separate SPR experiments on three capture points for the ligand. Therefore, differences in ligand capture on each surface will produce slight differences in the total measured response. The axes are scaled to make interpretation easier. We are not quantitatively comparing between treatments, but noting the qualitative behavior of each individually as the concentrations change.

   (3) Discussion: The authors could comment on the next steps, ligands ADME, solubility, permeability as the compounds are at low uM binding affinity, with increased concentrations, comment on selectivity/specificity.

   Given that these are still “hits”, we think a discussion of these properties (which we study intently in the lab!) is premature and beyond the scope of this manuscript.

   Reviewer #3 (Recommendations for the authors):

   (1) The authors should carefully read the manuscript as there are numerous typos and duplicated words across the whole document and supplementary data.

   We have gone through the manuscript and the supplementary material and corrected the typographical errors and duplicated words that we found, including a repeated sentence in the crystallization methods and a repeated word in the purification methods. We have also renamed the supplementary figures to the eLife figure supplement style. We expect copy editing at the VOR stage to catch anything that remains.

   (2) Details of how Chai-1 was run, including constraints, should be given in the Materials and Methods section.

   We have revised the methods section to describe how Chai-1 was used for both Orf9b and Tom70 co-folding predictions. See above.

   (3) There was no comment on whether the compounds which were found were attempted to be co-crystallised. If they were, but negative data were collected (didn't crystallise, didn't diffract, or no additional density found), then this needs to be stated.

   We have revised the manuscript to clearly point out that beyond our initial attempts to crystallize apo-Orf9b homodimer to enable fragment soaking, we also attempted to both soak our lipidated-compounds into apo-Orf9b homodimer crystals as well as co-crystallize the compounds which did not yield crystals.

   “We attempted two strategies of crystallizing our lipidated-compounds with the apo-Orf9b homodimer to structurally resolve the complex. First, we attempted to soak our compounds into apo-Orf9b homodimer crystals but observed poor tolerance to DMSO which quickly disintegrated the crystals after minutes. Second, we attempted to co-crystallize the lipidated-compounds after refolding the Orf9b homodimer to generate apo-Orf9b homodimers but could not identify conditions that yielded crystals upon re-screening.”

   Read the original source
4. 

   Version published to 10.7554/elife.111841.2 on eLife

   Nov 3, 2026
5. 

   Version published to 10.7554/elife.111841 on eLife

   Nov 3, 2026
6. 

   Version published to 10.64898/2026.04.27.721040v2 on bioRxiv

   Sep 21, 2026
7. 

   eLife

   Aug 14, 2026

   eLife Assessment

   This important study combines crystallographic fragment-based screening with fluorescence-polarization competition assays to demonstrate that the SARS-CoV-2 accessory protein ORF9b, and its interface with the mitochondrial import receptor TOM70, is accessible to small-molecule binding. This is a result with practical implications beyond coronavirus biology for efforts to restore interferon signalling and the antiviral response. The structural evidence is convincing, resting on an unusually large body of ligand-bound crystal structures whose fragment hotspots are independently corroborated by orthogonal biophysical approaches including fluorescence polarization and SPR. This work would benefit from additional information regarding the use of CHAI-1 and other attempts to determine the interactions between hit compounds …

   eLife Assessment

   This important study combines crystallographic fragment-based screening with fluorescence-polarization competition assays to demonstrate that the SARS-CoV-2 accessory protein ORF9b, and its interface with the mitochondrial import receptor TOM70, is accessible to small-molecule binding. This is a result with practical implications beyond coronavirus biology for efforts to restore interferon signalling and the antiviral response. The structural evidence is convincing, resting on an unusually large body of ligand-bound crystal structures whose fragment hotspots are independently corroborated by orthogonal biophysical approaches including fluorescence polarization and SPR. This work would benefit from additional information regarding the use of CHAI-1 and other attempts to determine the interactions between hit compounds and Tom70.

   Read the original source
8. 

   eLife

   Aug 14, 2026

   Reviewer #1 (Public review):

   Summary:

   The authors used FBDD screening to identify numerous compounds interacting with the ORF9b dimer. They expanded the original fragment hit, soaked the derivatives into the crystals and confirmed their binding poses, and showed that the derivatives bind the target with higher affinity. The authors further targeted the ORF9b binding site on TOM70, and used a fluorescence polarization-based (FP) assay to screen a compound library and obtained several hits. Structure-activity relationship (SAR) optimization yielded hit analogs that have higher binding affinity to TOM70.

   Strengths:

   (1) The study adopted novel drug design strategies, including stabilizing ORF9b homodimer to prevent it from binding TOM70, and blocking ORF9b from binding TOM70 by screening compounds that compete with ORF9b for binding TOM70.

   (…

   Reviewer #1 (Public review):

   Summary:

   The authors used FBDD screening to identify numerous compounds interacting with the ORF9b dimer. They expanded the original fragment hit, soaked the derivatives into the crystals and confirmed their binding poses, and showed that the derivatives bind the target with higher affinity. The authors further targeted the ORF9b binding site on TOM70, and used a fluorescence polarization-based (FP) assay to screen a compound library and obtained several hits. Structure-activity relationship (SAR) optimization yielded hit analogs that have higher binding affinity to TOM70.

   Strengths:

   (1) The study adopted novel drug design strategies, including stabilizing ORF9b homodimer to prevent it from binding TOM70, and blocking ORF9b from binding TOM70 by screening compounds that compete with ORF9b for binding TOM70.

   (2) The work established a feasible high-throughput screening assay. This FP-based assay screened ~50,000 compounds, from which two hit compounds were further optimized to yield analogs with higher binding affinity.

   Weaknesses:

   (1) The study lacks functional assays to evaluate whether the ORF9b-stabilized compounds or TOM70 binding compounds could affect IFN inhibition caused by ORF9b or virus infection.

   (2) There is a lack of experimental evidence to reveal the binding mode of lipidated-compounds with ORF9b homodimer.

   (3) There is a lack of experimental evidence to reveal the binding mode of HTS hits or analogs for TOM70.

   (4) Overall, none of the compounds shown in the paper have promising potency warranting further development; their binding affinity is limited to the micromolar range.

   Read the original source
9. 

   eLife

   Aug 14, 2026

   Reviewer #2 (Public review):

   Summary:

   The authors investigate chemical strategies to disrupt the interaction between the SARS-CoV-2 accessory protein Orf9b and the host mitochondrial receptor Tom70, an interaction implicated in suppression of type-I interferon responses. They employ two discovery approaches: a crystallographic fragment screen against the Orf9b homodimer and a high-throughput fluorescence polarization screen for compounds that compete with Orf9b binding to Tom70. The study identifies fragment-binding hotspots on Orf9b, develops lipidated analogs that stabilize the Orf9b homodimer, and discovers Tom70-binding compounds with low micromolar activity that inhibit Orf9b binding in vitro.

   Strengths:

   An impressive amount of work using a variety of complementary approaches and methods to validate binding (FP, SPR, and …

   Reviewer #2 (Public review):

   Summary:

   The authors investigate chemical strategies to disrupt the interaction between the SARS-CoV-2 accessory protein Orf9b and the host mitochondrial receptor Tom70, an interaction implicated in suppression of type-I interferon responses. They employ two discovery approaches: a crystallographic fragment screen against the Orf9b homodimer and a high-throughput fluorescence polarization screen for compounds that compete with Orf9b binding to Tom70. The study identifies fragment-binding hotspots on Orf9b, develops lipidated analogs that stabilize the Orf9b homodimer, and discovers Tom70-binding compounds with low micromolar activity that inhibit Orf9b binding in vitro.

   Strengths:

   An impressive amount of work using a variety of complementary approaches and methods to validate binding (FP, SPR, and computational modelling and SAR). The combination of crystallographic fragment screening on Orfb9 and HTS on Tom70 provides two independent routes for perturbing the Orf9b-Tom70 interaction. The structural work seems to be of very high-quality. The fragment campaign is extensive, yielding a substantial number of fragment-bound structures and identifying biologically meaningful binding hotspots on Orf9b.

   Finally, the screen results in reporting useful chemical starting points. Although potency remains modest, the study provides tractable scaffolds and a clear framework for future optimization.

   Weaknesses:

   General comment:

   (1) Although there is already an incredible amount of data presented, one limitation of this study is the lack of cellular validation - do these drugs enter cells, restore interferon signalling, reduce viral loads, or alter Orfb9 localization?

   (2) The logic of locking Orfb9 as a dimer is that the monomer binds Tom70 - thus, a more stable dimer means less monomer. In Figure 2, the Orfb9 homerdimer stabilization by compounds should reduce binding affinity to Tom70. A direct binding experiment measuring reduced Tom70 binding with compound treatment would better strengthen this claim.

   Read the original source
10. 

    eLife

    Aug 14, 2026

    Reviewer #3 (Public review):

    Summary:

    This paper attempts to and succeeds in demonstrating that Orf9b is able to bind small molecules using X-ray fragment screening, SPR and FP assays. Exploration of sites from the fragment screening is performed along with fragment linking with inter-dimer lipid moieties.

    Strengths:

    The experimental work looks strong and well performed. The interpretation of the data is appropriate and was often validated through orthogonal methods and follow-up compounds. The use of Tom70 to find binders that might disrupt interactions between Orf9b and Tom70 is elegant.

    Weaknesses:

    The use of Chai-1 to predict co-folded structures with binding molecules was not properly described - no mention of this in the methods. It was not commented on whether the compounds which were found were attempted to be co-crystallised. …

    Reviewer #3 (Public review):

    Summary:

    This paper attempts to and succeeds in demonstrating that Orf9b is able to bind small molecules using X-ray fragment screening, SPR and FP assays. Exploration of sites from the fragment screening is performed along with fragment linking with inter-dimer lipid moieties.

    Strengths:

    The experimental work looks strong and well performed. The interpretation of the data is appropriate and was often validated through orthogonal methods and follow-up compounds. The use of Tom70 to find binders that might disrupt interactions between Orf9b and Tom70 is elegant.

    Weaknesses:

    The use of Chai-1 to predict co-folded structures with binding molecules was not properly described - no mention of this in the methods. It was not commented on whether the compounds which were found were attempted to be co-crystallised. If they were but negative data was collected (didn't crystallise, didn't diffract or no additional density was found), then this needs to be stated.

    Read the original source
11. 

    Version published to 10.7554/elife.111841.1 on eLife

    Aug 14, 2026
12. Mentioned by @biorxiv-biophys.bsky.social on Bluesky

    Apr 27, 2026

    Inhibiting the interaction between the mitochondrial receptor Tom70 and SARS CoV 2 Orf9b with small molecules https://www.biorxiv.org/content/10.64898/2026.04.27.721040v1

13. Mentioned by @biorxivpreprint.bsky.social on Bluesky

    Apr 27, 2026

    Inhibiting the interaction between the mitochondrial receptor Tom70 and SARS CoV 2 Orf9b with small molecules https://www.biorxiv.org/content/10.64898/2026.04.27.721040v1

14. 

    Version published to 10.64898/2026.04.27.721040v1 on bioRxiv

    Apr 27, 2026