Review Reports

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Chen et al raise the question of the role of autophagy in the infection of trophoblasts by the hepatitis e virus (HEV) and identify the role of the capsid ORF2 protein in decreasing the levels of the autophagy LC3B protein, as well as several upstream regulatory proteins. They show the interaction between ORF2 and LC3B

This question is important to understand better the HEV cell cycle and the host process hijacked by the virus. However, the methods should be fully described and data should be consolidated to support the conclusions:

 

Main points

 

• Methods

The HEV strain from which the ORF2 and ORF3 sequences come is not indicated

The protocol for immunofluorescence experiments is not described

The extraction protocol for RT-PCR is incompletely described

The statistical tests used in this this study are not mentioned

 

• Results

The nuclear localization of ORF2 cannot be appreciate without indications about the immunofluorescence protocol and used microscope (permeabilization method?, confocal microscopy?)

 

The antibody used to target p-ULK1 recognized the protein phosphorylated at Ser-757, that is when phosphorylated by mTOR, the negative regulator of autophagy. This has to be indicated because ULK1 can also be phosphorylated on other Ser residues by AMPK, an autophagy activator.

 

Figure 1 and 2. The post-transfection time at which the western blots were performed should be indicated.

 

The decrease of markers of the PI3K/AKT/mTOR pathway should lead to an activation of autophagy at the initiation step and is concordant with the decrease of LC3B-II levels. I agree that there is a discrepancy with the p62 increase observed upon ORF2 overexpression. However, this does not definitely assess a blockage of the autophagic flux and has to be confirmed by other methods (western blots at different time points, RFP-LC3B-GFP reporter…)

 

The interaction of proteins with LC3 is described as depending on the presence of a LIR domain. The aminoacids identified with Alphafold 3 do not seem to belong to such a domain. This is not discussed.

The coimmunoprecipitation experiment confirm the interaction between the two proteins but not their direct interaction.

 

Other points

Several assertions should be supported either by a review (ref 2, 11, 25) or by the corresponding princeps work (ref 22, 29)

Comments for author File: Comments.pdf

Author Response

Comments 1: [Methods]The HEV strain from which the ORF2 and ORF3 sequences come is not indicated.

Response 1: Thank you for this important comment. We apologize for the oversight. As suggested, we have now clearly indicated the source of the viral gene sequences in the revised Materials and methods section under “Plasmids”. Specifically, the inserted ORF2 and ORF3 genes were codon-optimized sequences based on Hepatitis E virus genotype 3 (GenBank: XMQ88656.1 for ORF2; GenBank: AXF47753.1 for ORF3) (lines 115-117). Please also refer to S2 Fig for the schematic diagrams of the plasmid vectors and S1 Table for detailed restriction sites and inserted sequences.

 

Comments 2: [Methods]The protocol for immunofluorescence experiments is not described.

Response 2: Thank you for raising this point. We sincerely apologize for the omission in the initial manuscript. In accordance with your suggestion, we have now provided the complete protocol for the immunofluorescence assay (IFA) in the revised “Materials and methods” section. The detailed procedure includes cell culture, transfection, fixation, permeabilization, DAPI staining, mounting, and image acquisition using a Nikon A1HD25 confocal microscope (lines 189-199). This information is now available for your review.

 

Comments 3: [Methods]The extraction protocol for RT-PCR is incompletely described.

Response 3: Thank you for this constructive feedback. We apologize for the incomplete description. Following your comment, we have revised the Materials and methods section to provide a more detailed protocol for RNA extraction. Specifically, the RNA extraction procedure using Trizol reagent (Invitrogen) is now fully described, including the key steps of lysis, chloroform phase separation, isopropanol precipitation, ethanol washing, and dissolution. RNA quality control using a NanoDrop spectrophotometer is also specified (lines 165-170). We believe the added details address your concern.

 

Comments 4: [Methods]The statistical tests used in this this study are not mentioned.

Response 4: Thank you for this valuable comment. We have now incorporated a dedicated “Statistical Analysis” subsection within the Materials and methods section of the revised manuscript. This subsection specifies the software used (GraphPad Prism 8 and SPSS Statistics 25), states that data are presented as mean ± SD, and details the application of Student’s t-test and one-way ANOVA for group comparisons, with a p-value < 0.05 defined as statistically significant (lines 211-215). We appreciate your guidance in improving the methodological clarity of our work.

 

Comments 5: [Results]The nuclear localization of ORF2 cannot be appreciate without indications about the immunofluorescence protocol and used microscope (permeabilization method?, confocal microscopy?)

Response 5: We sincerely thank you for this important comment, which helped us improve the clarity and context of our findings. We apologize for the initial omission in the methods. A complete and detailed protocol for the immunofluorescence assay (IFA), including cell permeabilization and specification that images were acquired using a Nikon A1HD25 confocal microscope, has now been added to the Materials and methods section (lines 189-199). This provides the full technical context for evaluating the subcellular localization data presented in Fig 1A.

 

Comments 6: [Results]The antibody used to target p-ULK1 recognized the protein phosphorylated at Ser-757, that is when phosphorylated by mTOR, the negative regulator of autophagy. This has to be indicated because ULK1 can also be phosphorylated on other Ser residues by AMPK, an autophagy activator.

Response 6: We sincerely thank you for this expert and critical comment, which is crucial for the precise interpretation of our data. We fully agree with you that clearly specifying the phosphorylation site detected by the antibody is essential, as it distinguishes between mTOR-mediated inhibition and AMPK-mediated activation signals. Accordingly, we have revised the text in the Results section to explicitly state that the p-ULK1 antibody used specifically recognizes phospho-Ser757 (the mTOR-dependent site) (lines 277-280). This clarification has also been added to the legend for Figure 1E in the revised manuscript (lines 253-256).

 

Comments 7: [Results]Figure 1 and 2. The post-transfection time at which the western blots were performed should be indicated.

Response 7: Thank you for your comment regarding methodological clarity. As suggested, we have now explicitly indicated in the legends for Figures 1 and 2 that all Western blot analyses were performed at 48 hours post-transfection (lines 242-244). This key time point is also mentioned in the plasmid transfection protocol within the Materials and methods section for consistency (lines 144-146).

 

Comments 8: [Results]The decrease of markers of the PI3K/AKT/mTOR pathway should lead to an activation of autophagy at the initiation step and is concordant with the decrease of LC3B-II levels. I agree that there is a discrepancy with the p62 increase observed upon ORF2 overexpression. However, this does not definitely assess a blockage of the autophagic flux and has to be confirmed by other methods (western blots at different time points, RFP-LC3B-GFP report...).

Response 8: We sincerely thank you for this insightful and critical comment, which highlights a central paradox in our findings. We fully agree that the observed decrease in PI3K/Akt/mTOR markers alongside p62 accumulation necessitates direct evidence for a blockage of autophagic flux.

In response, we have significantly expanded the discussion of this point in the revised manuscript (please see the updated Discussion section, lines 296-308). First, we explicitly acknowledge this paradox and the limitation of interpreting flux from static markers alone. Crucially, we now cite our prior, direct evidence: in our previous study(Yang et al., EMERGING MICROBES & INFECTIONS, 2023), we employed the mCherry-GFP-LC3 tandem reporter system and conclusively demonstrated that HEV infection blocks autophagic flux at the degradative stage in JEG-3 cells. The core objective of the current study was to identify the specific viral protein responsible for this effect and to elucidate the underlying molecular mechanism.

Therefore, the direct binding of HEV ORF2 to LC3B, as discovered in the present study, provides the long-sought mechanistic explanation for the autophagic flux blockade documented in our prior work. It suggests that ORF2 sequesters LC3B, physically interfering with autophagosome formation/maturation and thereby overriding the upstream pro-autophagic signals. We believe this logically integrates our new mechanistic data with the established functional evidence from our prior work, offering a coherent narrative to resolve the apparent discrepancy.

 

Comments 9: [Results]The interaction of proteins with LC3 is described as depending on the presence of a LIR domain. The aminoacids identified with Alphafold 3 do not seem to belong to such a domain. This is not discussed. The coimmunoprecipitation experiment confirm the interaction between the two proteins but not their direct interaction.

Response 9:We sincerely thank you for these two incisive and highly relevant comments, which together address a crucial nuance in our interaction data. We fully agree with both points: (1) that Co-IP confirms association but not necessarily a direct, binary interaction, and (2) that the AlphaFold 3-predicted interface notably lacks a canonical LIR motif, a key observation we had initially overlooked.

In response, we have revised the Discussion to explicitly acknowledge that while Co-IP confirms an interaction, it does not prove direct binding. We also discuss that the AlphaFold 3-predicted binding interface lacks a canonical LIR domain, suggesting a potentially novel interaction mode (lines 433-453). We therefore present this computational prediction as a working hypothesis for a direct interaction, which must be and will be rigorously tested through future experiments, rather than treating it as conclusive evidence.

We believe these additions provide a more accurate, rigorous, and forward-looking interpretation of our interaction data, directly engaging with the your insightful critique.

 

Comments 10: [Other points]Several assertions should be supported either by a review (ref 2, 11, 25) or by the corresponding princeps work (ref 22, 29)

Response 10: We have revised the manuscript as suggested by you to strengthen the support for key assertions by adding citations to appropriate review articles or original research. The references have been updated. The following specific changes were made in the revised manuscript:

1. The statement about the genome encodes four open reading frames (ORFs) - ORF1, ORF2, ORF3 and ORF4 is now cited a review (ref. 11) (lines 63-64).
2. The description of the JEG-3 cell line as a well-established in vitro model for human trophoblast biology and autophagy research is now explicitly supported by both an original study (ref. 2) and a methodological/review-style reference (ref. 22,23) (lines 89-91).
3. The assertion that HEV infection inhibits all three stages of autophagy in JEG-3 cells is now clearly cited to our corresponding previous work (ref. 26) (lines 271-273).
4. The reference to our prior study using the mCherry-GFP-LC3 reporter system has been corrected to cite the relevant work (ref. 26) (lines 296-308).
5. In the Discussion, the observation that ORF2 inhibits PI3K/Akt/mTOR and activates AMPK yet suppresses ULK1 expression is now accompanied by a citation to a relevant review (ref. 27) that discusses ULK1 stability and degradation pathways, providing a plausible mechanistic context (lines 397-399).

We believe these revisions have adequately addressed the your concern, and thank you for this helpful suggestion.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This manuscript investigates how HEV proteins affect autophagy in human trophoblast JEG-3 cells, with the goal of explaining HEV-associated miscarriage. The authors generated plasmids to overexpress the HEV structural protein ORF2 and the small protein ORF3 in JEG-3 cells. They found that both ORF2 and ORF3 reduced LC3B levels, but only ORF2 caused accumulation of p62, indicating a block in autophagic flux. Surprisingly, ORF2 inhibited the PI3K/Akt/mTOR signaling pathway while upregulating AMPK phosphorylation and TFEB nuclear translocation. To explain this paradox, the authors used co-transfection (mCherry-LC3 and GFP-ORF2), co-localization microscopy, AlphaFold 3 modeling, and co-IP. These experiments support a model in which ORF2 directly binds the LC3B autophagy protein to block autophagosome formation. In summary, the data indicate that HEV ORF2 but not ORF3 is the key viral factor suppressing autophagy in trophoblasts, and that it does so by physically interacting with LC3B. This finding provides a plausible molecular mechanism for HEV-induced miscarriage via autophagy inhibition. Overall, the manuscript is well-written and the results are clearly presented. However, before publication, the authors should address the following points to strengthen the manuscript.

• Introduction section: I suggest to provide a brief summary of why JEG-3 cells were used as a model.
• Importantly, the study is limited to one cell type (JEG-3) and uses overexpression of single proteins rather than HEV infection. While this strategy isolates the role of ORF2, it may not capture all effects of viral infection. The authors should validate key findings in an HEV infection model or another trophoblast model in future work. Ideally, the findings are supported by a small animal model. At the very least, the authors should verify their findings in a second cell line, such as HTR-8/Svneo.
• Whole scans of the Western Blots should be added to the Supporting Information.
• Figures 1&2: Some, but not all p-values are provided in the graphs. Please make sure this is consistent.
• Figure 3: The description of panel B does not match what is shown. Also, the font of the graph on the right (“Gray-Value”) is very small and should be enlarged, and the resolution increased.

 

Comments on the Quality of English Language

• There are minor issues, e.g., line 348 “results is”, that should be addressed by careful proofreading.

Author Response

Comments 1: Introduction section: I suggest to provide a brief summary of why JEG-3 cells were used as a model.

Importantly, the study is limited to one cell type (JEG-3) and uses overexpression of single proteins rather than HEV infection. While this strategy isolates the role of ORF2, it may not capture all effects of viral infection. The authors should validate key findings in an HEV infection model or another trophoblast model in future work. Ideally, the findings are supported by a small animal model. At the very least, the authors should verify their findings in a second cell line, such as HTR-8/Svneo.

Response 1:We thank you for these insightful comments regarding the model system and experimental design, which allow us to clarify the positioning of our study.

(1) Rationale for using JEG-3 cells: As suggested, we have added a brief justification in the Introduction section (lines 89-91). We now state that the JEG-3 cell line is a well-characterized in vitro model for human trophoblast biology and autophagy studies, thereby providing a relevant human-derived cellular context for our investigation.

(2) Context and progression of our research program: We appreciate the your point regarding the inherent limitations of a reductionist approach. We wish to clarify that the present study is a direct mechanistic follow-up to our prior work (Yang et al., EMERGING MICROBES & INFECTIONS, 2023), where we first established—using the same JEG-3 model and a mouse model—that HEV infection impairs autophagic flux and causes pregnancy disorders. The core objective here was to transition from that phenotypic discovery to identifying the specific viral protein (ORF2) and elucidating the precise molecular mechanism (direct LC3B binding). This strategy is essential to isolate causality within the complexity of a full viral infection.

(3) Public health relevance and future directions: We fully agree that validation in additional models is a critical next step. From a public health perspective, understanding HEV pathogenesis in human trophoblasts is paramount, as HEV is a zoonotic virus that poses a significant threat to pregnant women. Our findings in the JEG-3 model provide a crucial human-relevant foundation. We fully agree that validation in additional models is a critical next step. From a public health perspective, understanding HEV pathogenesis in human trophoblasts is paramount. Therefore, our future work will be directed along three key paths to bridge our mechanistic discovery with broader relevance: (i) confirming ORF2's role in the context of full HEV infection in trophoblasts; (ii) extending the findings to primary human trophoblasts or alternative cell lines (e.g., HTR-8/Svneo) to strengthen generalizability; and (iii) ultimately verifying the therapeutic potential of targeting the ORF2-LC3B interaction in relevant animal models. This stepwise approach aims to translate our current molecular insight into translational strategies with potential benefits for both human and animal health.

 

Comments 2: Whole scans of the Western Blots should be added to the Supporting Information.

Response 2: Thank you for this comment. In full compliance with journal policy and to ensure data transparency, all scans of all Western blots have now been compiled and uploaded as Supporting Information. This file provides immediate and direct access to all underlying raw data for review. We trust this will provide the requested full transparency.

 

Comments 3: Figures 1&2: Some, but not all p-values are provided in the graphs. Please make sure this is consistent.

Response 3: Thank you for this meticulous comment. We have carefully reviewed and revised Figures 1 and 2 to ensure that all relevant statistical comparisons are now consistently annotated with their corresponding significance indicators. Throughout the figures, statistical significance is denoted as follows: *p < 0.05 and **p < 0.01.

 

Comments 4: Figure 3: The description of panel B does not match what is shown. Also, the font of the graph on the right (“Gray-Value”) is very small and should be enlarged, and the resolution increased.

Response 4: Thank you for these specific and constructive comments regarding Figure 3. We have carefully revised the figure to address all points raised.

(1)Correction of panel description: The figure legend has been updated. The corrected and complete description for panel B is now: (B) Co-IP analysis of HEV ORF2 and LC3B in JEG-3 cells. Representative confocal microscopy images of JEG-3 cells co-transfected with mCherry-LC3B (red) and eGFP-ORF2 (green) plasmids. Nuclei were stained with DAPI (blue). The merged panel shows the co-localization of ORF2 and LC3B signals (yellow). Scale bar: 20 μm. This now accurately matches the data presented.

(2)Improved readability and resolution: In response to your feedback, we have significantly enlarged the font size in the “Gray-Value” graph and increased the overall resolution of Figure 3. These adjustments ensure optimal clarity for all graphic elements. We believe these revisions have fully resolved the identified issues.

 

Comments 5: Comments on the Quality of English Language: There are minor issues, e.g., line 348 “results is” , that should be addressed by careful proofreading.

Response 5: We thank the you for highlighting the need for language polishing. We have carefully proofread the entire manuscript and corrected grammatical and typographical errors, including the specific instance on lines 397-399 (“results is”). The language has been further refined to ensure clarity and readability throughout the text.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have responded to all my comments and made changes to the manuscript accordingly. From my point of view the manuscript is now eligible for publication.