CD81 is an Ebola virus inhibiting factor that is antagonized by GP and VP40

Article activity feed

1. 

   Review Commons

   Mar 30, 2026

   Note: This response was posted by the corresponding author to Review Commons. The content has not been altered except for formatting.

   Learn more at Review Commons

   ---

   Reply to the reviewers

   The authors do not wish to provide a response at this time.

   Read the original source
2. 

   Review Commons

   Mar 30, 2026

   Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

   Learn more at Review Commons

   ---

   Referee #3

   Evidence, reproducibility and clarity

   Summary:

   This study identifies CD81 as an anti-viral host factor that restricts EBOV infection. In-turn, EBOV encodes CD81 antagonism activity in GP and VP40. The two major lines of inquiry of this study are how CD81 restricts EBOV infection and how EBOV antagonizes CD81. These are evaluated using BSL-2 level models and authentic EBOV infection in different cell types, including human primary macrophages. This study underscores the intricate interplay between mechanisms of host innate immunity and mechanisms of virus immune evasion and antagonism.

   CD81 restriction of EBOV

   A previous study published by this group demonstrated that …

   Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

   Learn more at Review Commons

   ---

   Referee #3

   Evidence, reproducibility and clarity

   Summary:

   This study identifies CD81 as an anti-viral host factor that restricts EBOV infection. In-turn, EBOV encodes CD81 antagonism activity in GP and VP40. The two major lines of inquiry of this study are how CD81 restricts EBOV infection and how EBOV antagonizes CD81. These are evaluated using BSL-2 level models and authentic EBOV infection in different cell types, including human primary macrophages. This study underscores the intricate interplay between mechanisms of host innate immunity and mechanisms of virus immune evasion and antagonism.

   CD81 restriction of EBOV

   A previous study published by this group demonstrated that CD81 antagonizes NF-κB signaling. In this study and a very recently published study by another group, VP40 is demonstrated to activate NF-κB signaling. This study shows that VP40 activates NF-κB through a TLR4-independnet mechanism that is antagonized by CD81. This study also demonstrated that CD81 suppresses authentic EBOV infection in HEK293T cells and human primary blood monocyte derived macrophages. Using a transcription and replication-competent tetracistronic minigenome system (trVLPs), this study found that CD81 reduces the abundance of all viral RNA species (mRNA, vRNA genome, cRNA genome complement) but does not interfere with the NP-driven formation of inclusion bodies where viral RNA synthesis occurs. CD81 was also found to suppress cellular micropinocytosis activity and restrict EBOV entry. The CD81-stimulating antibody 5A6 suppressed both trVLP and authentic EBOV infection.

   EBOV antagonism of CD81

   CD81 (as well as the other tetraspanins CD63 and CD9) was identified to be downregulated on the cell surface of HeLa and HEK293 cells transfected with an EBOV GP expression plasmid as a part of a larger screen for EBOV GP's impact on cell surface receptor expression. Using a combination of bimolecular fluorescent complementation (BiFC), fluorescence resonance energy transfer (FRET), and proximity ligation assays, EBOV GP was shown to directly interact with CD81. Imaging analysis showed that EBOV GP colocalizes with CD81 at the cell surface. Here, GP blocks accessibility of CD81 through a glycan shielding mechanism that is reversed with PNGase F treatment. EBOV GP also reduces the total cellular abundance of CD81 protein by inducing CD81 degradation through both lysosomal and proteasome mediated degradation pathways without altering CD81 mRNA transcript levels. The GPs of other relative filoviruses (Marburg virus, Sudan virus, Reston virus, and Taï Forest virus) also downregulated accessible cell surface CD81, CD63, and CD9 in HEK293T cells. This CD81 downregulating activity appears to be both conserved amongst and specific to filovirus GPs, as none of the glycoproteins from 14 other RNA viruses tested (including arenaviruses, rhabdoviruses, influenza viruses, coronaviruses, and retroviruses) significantly altered cell surface CD81, CD63, or CD9 in HEK293T cells. In human primary blood monocyte derived macrophages, authentic EBOV infection was found to reduce the abundance of CD81 and CD9 accessible at the cell surface. EBOV VP40 was the only other EBOV structural protein that downregulated total CD81, though its effect was mild. In contrast to the mechanisms utilized by GP, VP40 was found to induce CD81 degradation mainly through the proteasome mediated degradation pathway.

   Major Comments:

   The claims made by the authors are appropriate and are supported by their data and their use of appropriate controls which yielded the expected results based on references from the literature. There are no new experiments that must be introduced to support the claims made by the authors. The methods section is excellent and provides extensive detail for techniques and organized lists of plasmid and antibody reagents used and their original sources. Graphed data shows individual replicates, representative flow scatter plots and images are show, and appropriate statistical analyses were used and reported. Excitingly, this study opens many new lines of question which can be addressed in future studies.

   The only major comment:

   • It would be valuable to add more discussion around something the data presented in Figure 7 b-d suggest, and that is that EBOV entry appears to be targeted by CD81 by multiple mechanisms. Figure 7d demonstrates CD81 suppresses cellular macropinocytosis activity, which would yield less uptake of EBOV which utilizes PS receptors to be internalized through the macropinocytosis pathway. Since PS receptors recognize the PS in the viral envelope, and not the viral GP, it makes sense that trVLPs pseudotyped with VSV-G were restricted like those with EBOV GP (Fig 7C). However, in the pseudotyped lentivirus system, EBOV GP-mediated entry was significantly suppressed by CD81 while VSV-G mediated entry was not (Fig 7B). Together this data shows that CD81 restricts EBOV entry in both viral envelope-targeted and GP-targeted mechanisms, demonstrating the vast innate immune mechanisms of CD81 against EBOV. I think this is of great impact and should be discussed more.

   Minor Comments:

   Some modest experimental and analysis suggestions that are not required to support the claims of the paper but would add additional depth are:

   • The experiments in Figure 8E would have benefited from collection and titering of supernatants from infected cells.
   • While bafilomycin treatment or MG132 treatment can partially rescue CD81 from the degradation induced by EBOV GP or VP40, neither drug is able to fully rescue this for either viral protein. It would be insightful to assess if co-treatment of bafilomycin and MG132 would yield a full rescue of total CD81. Based on the presented results, I would expect co-treatment to fully restore total CD81 in VP40-expressing cells, but I would expect co-treated GP-expressing cells to still have increased CD81 surface downregulation equivalent to the strength of which the glycans of EBOV GP shield CD81 from recognition. These experiments would give valuable insight into the relative strengths of the glycan shielding effect and induced degradation effects in EBOV GP's antagonism of CD81.
   • The impact of the imaging experiments shown in Fig. 5D would be strengthened with a quantitative colocalization analysis such as Pearson's Coefficient.
   • Another publication (citation 132 in this study) shows cooperativity of GP in VP40's ability to activate NF-ΚB. This study shows in Fig. 6E that CD81's suppressive activity can overcome VP40's activation of NF-ΚB. It would be valuable to assess (in the same format as the experiment done in Fig 6E) if this would remain true in cells co-expressing EBOV GP and VP40; ie, would CD81 still be able to overcome, and at a similar rate, the NF-ΚB-activating activity of VP40 in the presence of the CD81-antagonizing activity of GP.
   • Understandably, the VP40 gene was used as a probe in the Fig. 6 trVLP experiments because it is encoded within the tetracistronic minigenome. However, this became mildly confusing when reading Figure 6 because other parts of the manuscript discuss and measure the effects and activities of VP40. One thought is to probe the VP24 gene. However, a simple way to reduce the initial confusion in interpreting the data in the figure is to just remove "(VP40)" from Fig.6 A-C. The methodology of using VP40 primers to probe for the viral RNAs is adequately detailed in the figure caption and methods section.

   Significance

   This paper gives more insight into another mechanism of innate immune responses of the host and mechanisms of evasion/antagonism by EBOV VP40 and GP. This is the first report of CD81 as an anti-viral host factor of EBOV. This study shows that CD81 interferes with EBOV viral entry and reduces the abundance of EBOV RNAs using different mechanism. Additionally, CD81 was found to have a previously unrecognized negative regulatory role in cellular micropinocytosis. In terms of how EBOV antagonizes CD81, this study demonstrates multiple mechanisms of antagonism and more broadly demonstrate the diverse mechanisms of innate immune evasion encoded by EBOV. Building on previous reports of the N-linked glycans of EBOV having a steric shielding effect that blocks accessibility of cell surface human leukocyte antigen class-1 (HLA-I) and MHC class I polypeptide-related sequence A (MICA), this study found that these glycans also sterically shield the accessibility of cell surface CD81. This opens important new lines of inquiry, such as: what else do the EBOV glycans sterically interfere with at the cell surface?

   This group previously demonstrated that CD81 suppresses NF-κB activation and this study builds on those observations by demonstrating that despite the CD81-antagonizing activity of VP40, VP40's ability to activate NF-κB can be overcome by CD81. This is in line with some of the most impactful findings in this paper, namely that despite encoding CD81 antagonism activity in at least GP and VP40, EBOV remains susceptible to CD81 activation, as demonstrated by inhibition of both the trVLP system and authentic EBOV infection by the CD81-stimulating antibody 5A6.

   Read the original source
3. 

   Review Commons

   Mar 30, 2026

   Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

   Learn more at Review Commons

   ---

   Referee #2

   Evidence, reproducibility and clarity

   The manuscript entitled "CD81 is an Ebola virus inhibiting factor that is antagonized by GP and VP40" by Hu et al. investigates the role of the tetraspanins CD81, CD63, and CD9 during Ebola virus infection. They found that CD81, among the three tetraspanins, plays a major role as a cellular antiviral factor by interfering with EBOV glycoprotein and VP40. CD81 suppressed NFkB signaling and was found to restrict EBOV replication and VLP uptake. Overall, the study design, choice of experimental approaches, and presentation of the data meet an excellent scientific standard. The figures and their legends are comprehensive and very …

   Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

   Learn more at Review Commons

   ---

   Referee #2

   Evidence, reproducibility and clarity

   The manuscript entitled "CD81 is an Ebola virus inhibiting factor that is antagonized by GP and VP40" by Hu et al. investigates the role of the tetraspanins CD81, CD63, and CD9 during Ebola virus infection. They found that CD81, among the three tetraspanins, plays a major role as a cellular antiviral factor by interfering with EBOV glycoprotein and VP40. CD81 suppressed NFkB signaling and was found to restrict EBOV replication and VLP uptake. Overall, the study design, choice of experimental approaches, and presentation of the data meet an excellent scientific standard. The figures and their legends are comprehensive and very clearly written. However, I recommend that the authors make the following clarifications:

   Major:

   1. I recommend including a figure that depicts the domain organization of the tetraspanins, as this would help readers better appreciate how these three tetraspanins differ from one another. Did the authors determine the minimal region of CD81 required for interaction with EBOV GP or VP40? How are tetraspanins trafficked from the plasma membrane to intracellular compartments or into extracellular vesicles, and are these trafficking pathways also altered during EBOV infection?
   2. Given the clear role of VP40 in this CD81-dependent mechanism, it is important to demonstrate whether VP40 and CD81 interact directly. As the BiFC assay did not resolve this question, I recommend using a complementary approach, such as co-immunoprecipitation followed by western blotting, to address it.

   Minor:

   1. Please clarify how the '+' and '++' GFP categories are quantitatively defined.
   2. In Figure 4b (right panels), what explains the different effects of the DMSO control on surface versus total CD81?
   3. For clarity, I suggest defining the exact numerical boundaries of the individual domains shown in Figure S2C.
   4. In Figure S3C, the data presentation could be improved by using different colors for the control and KO groups, or by increasing the size of the symbols representing the data points.

   Significance

   This article is well written, and the study underscores the critical role of the tetraspanin CD81 as a cellular antiviral factor during EBOV infection and defines its role in filoviral immune response regulation. This article can be accepted for publication after the minor revision.

   Read the original source
4. 

   Review Commons

   Mar 30, 2026

   Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

   Learn more at Review Commons

   ---

   Referee #1

   Evidence, reproducibility and clarity

   To identify cell surface receptors modulated by GP, the authors performed a flow cytometry-based screen using the LEGENDScreen Human Cell PE Kit, which targets 332 host cell surface markers. Among the validated hits in two human cell lines, the authors focused on the tetraspanin CD81 as its expression was selectively reduced by EBOV GP and other filovirus GPs, but not by GPs from unrelated viruses, suggesting a filovirus GP-specific mechanism. This is analogous to the downregulation of CD81 expression by viral proteins such as Vpu and Gag from HIV, and NS5A from HCV. The authors also found that VP40 overexpression reduced CD81 …

   Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

   Learn more at Review Commons

   ---

   Referee #1

   Evidence, reproducibility and clarity

   To identify cell surface receptors modulated by GP, the authors performed a flow cytometry-based screen using the LEGENDScreen Human Cell PE Kit, which targets 332 host cell surface markers. Among the validated hits in two human cell lines, the authors focused on the tetraspanin CD81 as its expression was selectively reduced by EBOV GP and other filovirus GPs, but not by GPs from unrelated viruses, suggesting a filovirus GP-specific mechanism. This is analogous to the downregulation of CD81 expression by viral proteins such as Vpu and Gag from HIV, and NS5A from HCV. The authors also found that VP40 overexpression reduced CD81 levels, potentially enhancing VP40-mediated NF-κB activation. The authors suggest that CD81 reduction may result from degradation led by GP-CD81 interaction. CD81 downregulation was also observed in infected monocyte-derived macrophages (MDMs). Detailed analysis using CD81-KO cells and the transcription- and replication-competent VLP (trVLP) system demonstrated that CD81 is involved in EBOV entry and replication steps. While these data provide key insights, concerns remain regarding their statistical significance and interpretation.

   1. Have the authors investigated the functional consequences of CD81 downregulation by GP, VP40, or viral infection? In particular, could this enable superinfection? This can be examined using the approaches used in the manuscript.
   2. "Fold of modification (GP-/GP+)" in Figure 1a does not appear to match the results presented in Figure S2 and Table S2.
   3. Where appropriate, please indicate 'n.s.' for comparisons that are not statistically significant. With n = 3, the results may be unreliable; increasing the number of replicates to five would be recommended, as this is critical for supporting the manuscript's conclusions.
   4. Although the authors conclude that GP suppresses or counteracts CD81-mediated inhibition of viral replication (e.g., VP40 protein expression) based on experiments using trVLPs with or without GP, the presented data do not support this conclusion. In fact, higher VP40 expression was detected in trVLPΔGP-infected cells compared to trVLP-infected control cells (Fig. 3a and 3b), or no statistically significant differences was provided. These results seem inconsistent with the authors' interpretation and require clarification.
   5. Increased p65 expression does not necessarily indicate activation of p65 or NF-κB signaling. Indeed, VP40-induced or infection-induced increase in p65 expression level was not significantly different between wild-type and CD81-KO cells (Fig. 6c and 6d). To properly assess the NF-κB activation, the phosphorylation status of p65 and/or nuclear translocation should be examined.
   6. The authors suggest that CD81 is involved in macropinocytosis based on experiments using CD81-KO cells (Fig. 7) and anti-CD81 antibody (5A6 clone) (Fig. 8). Have the authors examined whether CD81 regulates macropinocytosis-associated signaling pathway (e.g., the PI3K/AKT1 pathway)? It is possible that AKT1 is constitutively activated in CD81-KO cells, given the increased dextran uptake. Such analysis would strengthen the authors' claim.
   7. In the experiments assessing viral entry in CD81-KO and control cells, both cells were co-transfected with Tim-1. Have the authors confirmed that Tim-1 expression levels were comparable between KO and control cells?
   8. The VP40-CD81 interaction was assessed only by PLA, but the results were not shown due to high background signals. Other methods, such as co-IP or the BiFC assay used in the manuscript, could yield clear data and deepen the discussion.
   9. In the Fig.4b, both the number of GFP-positive cells and the GFP intensity are noticeably lower than in other similar experiments (e.g., Fig. 1c and Fig. S4a). The "GP ++" population is much smaller and difficult to define or gated. Please clarify this discrepancy.
   10. The statement that "VP40-mediated downregulation of surface CD81 was strongly blocked by MG132 and partially by BafA1 (Fig. 4b)" is not supported by the data shown.
   11. In Fig.4c, assessing the role of GP glycan shield, is there a statistically significant difference between GP and GPΔmucin? It appears that deletion of the mucin domain does not affect the structural shielding of CD81, whereas PNGase treatment does.
   12. In Fig.5c, please enlarge the PLA image for better visibility.
   13. CD81 localization in the presence of GP differs between Fig. 5c and Fig. 5d under similar conditions. In the Fig. 5d, GP redistributes CD81 to both the cytoplasm and the cell surface. Please clarify this discrepancy.
   14. In Fig. 8a, the percentage of GFP-positive cells (infected cells) are very low, up to 5%. What MOI was used, and can the effect of the CD81 antibody on infection be reliably evaluated under this condition? Statistical significance should compare CD81 antibody with the isotype control, not with no antibody.
   15. The authors use the term "multiple" (e.g., "multiple cell lines" and "CD81 inhibits multiple steps throughout the viral life cycle"); however, this wording feels overstated as two cell lines were used and the two steps (entry and replication steps) are inhibited.
   16. Please cite the following article where relevant: Nanoscale organization of tetraspanins during HIV-1 budding by correlative dSTORM/AFM (Nanoscale, 2019).
   17. What primary antibodies are used in the PLA. Please describe them in the method section.
   18. Lines 468-471. Please provide the relevant references for the GP mutations tested.

   Significance

   To identify cell surface receptors modulated by GP, the authors performed a flow cytometry-based screen using the LEGENDScreen Human Cell PE Kit, which targets 332 host cell surface markers. Among the validated hits in two human cell lines, the authors focused on the tetraspanin CD81 as its expression was selectively reduced by EBOV GP and other filovirus GPs, but not by GPs from unrelated viruses, suggesting a filovirus GP-specific mechanism. This is analogous to the downregulation of CD81 expression by viral proteins such as Vpu and Gag from HIV, and NS5A from HCV. The authors also found that VP40 overexpression reduced CD81 levels, potentially enhancing VP40-mediated NF-κB activation. The authors suggest that CD81 reduction may result from degradation led by GP-CD81 interaction. CD81 downregulation was also observed in infected monocyte-derived macrophages (MDMs). Detailed analysis using CD81-KO cells and the transcription- and replication-competent VLP (trVLP) system demonstrated that CD81 is involved in EBOV entry and replication steps. While these data provide key insights, concerns remain regarding their statistical significance and interpretation.

   Read the original source
5. 

   Version published to 10.64898/2026.02.04.703765 on bioRxiv

   Feb 4, 2026