Decoding Self vs. Non-Self: Alphavirus Cap0 Recognition and Immune Evasion

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

I have read the paper and can confirm that it fits in the scope of Viruses. It is a review paper that reports on how Cap0 structures in alphavirus RNA are recognized by host innate immune sensors, particularly RIG-I and IFIT proteins, and describes viral mechanisms that evade this recognition. The topic is timely and relevant because alphaviruses such as chikungunya virus and other medically important members continue to cause outbreaks. The paper synthesizes current knowledge on RNA cap methylation, innate immune sensing, and viral evasion strategies. The review is generally well organized and readable. However, several issues should be addressed regarding clarity of biological mechanisms, balance of literature coverage, and the strength of some conclusions. Some of my minor concerns are as follows:

1. This review paper provides a good introductory overview of alphavirus biology, and also the explanation of cap methylation differences between host and viral RNA is clear. However, there are some points that need clarification. The mechanistic details of IFIT1 recognition and sequestration could be explained more explicitly. Also, the description of how RNA secondary structure shields the cap could benefit from clearer mechanistic explanation and more specific examples. And also some sections move rapidly between viruses and mechanisms without sufficient transitions. Overall, the biology is generally understandable but would benefit from clearer mechanistic descriptions in some sections.

 

2. As a review paper, it summarizes previously published studies, integrates findings from different alphavirus systems, and it highlights mechanisms of immune recognition and evasion. It presents a conceptual framework around Cap0 recognition and immune evasion. However, the authors should ensure that the review includes the most recent literature, and also clearly (better) distinguishes established findings vs. emerging hypotheses

 

3. After reading the text, I think that the conclusions are mostly consistent with the cited literature, but in several places they appear slightly overstated. For example, some statements imply universal mechanisms across alphaviruses, while many cited studies are performed on specific viruses (e.g., CHIKV, SFV). Also the role of RNA structural shielding of the cap may be presented as more definitive than currently demonstrated.

 

4. On the figures, I have the following advise to the authors. Make more detailed figure legends; do a clearer labeling of molecular steps; and better to indicate more clear which mechanisms are experimentally demonstrated vs. hypothetical.

 

 

Author Response

Dear Reviewers,

We sincerely thank you for reviewing our manuscript, and for your positive and constructive feedback. Below, we provide a point-by-point response to your comments. Corresponding changes in the manuscript are highlighted in yellow.

We believe that the manuscript is now much improved.

Best regards,

Claudia Filomatori

Comment 1: This review paper provides a good introductory overview of alphavirus biology, and also the explanation of cap methylation differences between host and viral RNA is clear. However, there are some points that need clarification. The mechanistic details of IFIT1 recognition and sequestration could be explained more explicitly. Also, the description of how RNA secondary structure shields the cap could benefit from clearer mechanistic explanation and more specific examples. And also some sections move rapidly between viruses and mechanisms without sufficient transitions. Overall, the biology is generally understandable but would benefit from clearer mechanistic descriptions in some sections.

Response 1: We appreciate the reviewer’s suggestion regarding the need for clearer mechanistic explanations. To address this point, we have expanded the description of IFIT1 recognition of capped RNA, incorporating molecular details of its interaction with Cap0 structures. We now describe how IFIT1 accommodates the 5′ cap within its binding pocket and discuss how 2′-O-methylation (Cap1) alters the local geometry and steric environment of the cap, thereby reducing IFIT1 binding affinity (page 6, lines 193-205).

In addition, we have clarified the mechanism by which RNA secondary structure may shield the cap (page 8, lines 237-242). Specifically, it has been proposed that IFIT1 binding requires a 5′ single-stranded RNA overhang to properly position the cap within its binding channel. In contrast, base-pairing at the 5′ end constrains the RNA into a structured conformation that is incompatible with this interaction, thereby impairing cap engagement. In this context, the 5′ stem-loop (5′SL) may act as a structural shield that functionally mimics Cap1-mediated evasion.

Regarding transitions between viruses and mechanisms, to improve clarity, we integrated evasion strategies into a more unified conceptual framework. We guide the reader to more clearly compare the advantages and limitations of different antiviral strategies that evade cap-dependent recognition (page 16, lines 505–523).

Comment 2: As a review paper, it summarizes previously published studies, integrates findings from different alphavirus systems, and it highlights mechanisms of immune recognition and evasion. It presents a conceptual framework around Cap0 recognition and immune evasion. However, the authors should ensure that the review includes the most recent literature, and also clearly (better) distinguishes established findings vs. emerging hypotheses.

Response 2: We thank the reviewer for this insightful comment. To address this point, we have incorporated several recent references into the manuscript covering the following aspects:

1. RNA sensing and signaling (Refs. 8–10: Cryer et al., 2026; Luan et al., 2024; Bartok et al., 2020);
2. A recent review on translational control of alphavirus–host interactions (Ref. 47: Ventoso et al., 2024);
3. Length dependence of viral dsRNA in RIG-I recognition (Ref. 29: Im et al., 2023);
4. The effect of cap-adjacent N⁶,2′-O-dimethyladenosine (m⁶Am) modifications on IFIT binding (Refs. 77–79: Geng et al., 2024; Akichika et al., 2019; Boulias et al., 2019);
5. mRNA vaccine technology (Ref. 82: Overmars et al., 2022);
6. Additional examples of viral immune evasion mechanisms, including Influenza A virus NS1-mediated modulation of innate immunity (Ref. 94: Zhang et al., 2022).

We also agree with the reviewer that, in some sections, the distinction between established findings and emerging hypotheses was not sufficiently clear. To address this, we have revised the text and adjusted the language to more explicitly indicate the level of evidence supporting each statement.

Comment 3: After reading the text, I think that the conclusions are mostly consistent with the cited literature, but in several places, they appear slightly overstated. For example, some statements imply universal mechanisms across alphaviruses, while many cited studies are performed on specific viruses (e.g., CHIKV, SFV). Also, the role of RNA structural shielding of the cap may be presented as more definitive than currently demonstrated.

Response 3: We agree that, in some instances, our original wording could give the impression of overgeneralization across alphaviruses or present certain mechanisms with a level of certainty that is not fully supported by the available evidence. To address this, we have moderated several statements.

In addition, we have incorporated explicit statements identifying aspects that remain unclear. For example:

Page 8, lines 271–272: “Taken together, the model of cap shielding by the 5’SL has been proposed for selected alphaviruses, however, its conservation across the entire genus remains to be established.”

Page 13, lines 391–392: “However, further experimental studies will be required to determine whether these mechanisms are broadly conserved across the genus.”

Page 17, lines 561–564 (Open Questions): “While the 5′SL appears to be conserved across alphaviruses, many functional studies have been performed in a limited number of viral models. It therefore remains unclear whether IFIT evasion mediated by the 5′SL is restricted to specific viruses or represents a more general mechanism shared among alphaviruses that replicate in vertebrate hosts.”

We have also revised the text to present RNA structural shielding of the cap as a proposed model rather than a definitive mechanism. For example:

Page 8, lines 237–242: “As noted above, cap accommodation within the IFIT1 binding pocket requires a ssRNA overhang. In this context, base-pairing at the 5′ end constrains the RNA into a conformation that may be incompatible with insertion into the IFIT1 binding channel, thereby impairing cap recognition. Consequently, the 5′SL has been suggested to act as a structural shield that functionally mimics cap1-mediated evasion.”

Comment 4: On the figures, I have the following advice to the authors. Make more detailed figure legends; do a clearer labeling of molecular steps; and better indicate more clear which mechanisms are experimentally demonstrated vs. hypothetical.

Response 4: We have reworded the legends of Figures 1, 2, and 3 to provide more detailed, self-explanatory descriptions. In addition, we have incorporated references into the legends to better support well-established mechanisms. Specifically, in Figure 3 we have included Ref. 27 (Wang et al., 2022) and Ref. 28 (Kowalinski et al., 2011), and in Figure 4, Ref. 45 (Sonenberg et al., 2009) and Ref. 46 (Hinnebusch et al., 2016).

Finally, we have labeled individual molecular steps within Figures 1 and 4, with corresponding descriptions provided in the legends.

Reviewer 2 Report

Comments and Suggestions for Authors

The review provides a structured overview of how alphavirus Cap0 structures are sensed by host immunity and how 5′ RNA structures contribute to immune evasion; however, several points would benefit from further clarification and critical depth. First, the manuscript emphasizes the role of the 5′SL in shielding Cap0 from IFIT1, but to what extent is this mechanism quantitatively supported across different alphaviruses, and are there comparative experimental data validating this model beyond selected case studies? Second, the discussion of RNA folding shifts from thermodynamic stability to kinetic accessibility, which is conceptually important, but lacks concrete experimental or computational evidence ( some evidences are provided by predicting the delta G values) , but can the authors provide direct support or specific studies demonstrating kinetic folding effects in this context? Third, while the translational implications for mRNA and saRNA vaccine design are interesting, the connection to alphavirus biology appears somewhat speculative; can the authors more clearly delineate which insights are directly supported versus extrapolated? Additionally, the review would benefit from a more critical discussion of limitations, particularly regarding the reliance on predicted RNA structures and the lack of high-resolution structural validation of the 5′SL. Finally, given the broad coverage of viral evasion strategies, the manuscript could be strengthened by better integrating these mechanisms into a unifying framework rather than presenting them as a descriptive list, and by clarifying how these diverse strategies compare in their relative efficiency and evolutionary significance. Further, protons in structure of guanine ( Figure 2) should be at the N1 position and not the N3 position. please correct this. Similarly, it is good that authors have used alphafold3 to predict the structures of RNA structures, but which is the reliability of the structures? No discussion on quality is discussed. How the structures of the mentioned structures change when they interact with surrounding ligands or the protein structure?     

Author Response

Dear Reviewers,

We sincerely thank you for reviewing our manuscript, and for your positive and constructive feedback. Below, we provide a point-by-point response to your comments. Corresponding changes in the manuscript are highlighted in yellow.

We believe that the manuscript is now much improved.

Best regards,

Claudia Filomatori

 

Comment 1: The manuscript emphasizes the role of the 5′SL in shielding Cap0 from IFIT1, but to what extent is this mechanism quantitatively supported across different alphaviruses, and are there comparative experimental data validating this model beyond selected case studies?

Response 1: As the reviewer noted, the role of the 5′SL in shielding the Cap0 structure from IFIT1 has been experimentally validated in selected alphaviruses, including VEEV, CHIKV, SINV, and SFV. However, we agree that our original wording could give the impression of overgeneralization across the genus. To address this, we have incorporated explicit statements to more clearly define the scope and limitations of the available evidence. For example:

Page 8, lines 271–272: “Taken together, the model of cap shielding by the 5’SL has been proposed for selected alphaviruses, however, its conservation across the entire genus remains to be established.”

Page 13, lines 391–392: “However, further experimental studies will be required to determine whether these mechanisms are broadly conserved across the genus.”

Page 17, lines 561–564 (Open Questions): “While the 5′SL appears to be conserved across alphaviruses, many functional studies have been performed in a limited number of viral models. It therefore remains unclear whether IFIT evasion mediated by the 5′SL is restricted to specific viruses or represents a more general mechanism shared among alphaviruses that replicate in vertebrate hosts.”

Regarding quantitative and comparative experimental support across alphaviruses, we acknowledge that such data remain limited. Some studies suggest that differences in 5′SL stability may correlate with IFIT1 sensitivity; for instance, viruses with less stable predicted 5′SL structures often exhibit higher susceptibility to interferon responses (Refs 52 and 55). However, this relationship is not yet fully established and may be influenced by additional factors. Notably, viruses displaying only partial resistance to IFIT1, like EEEV, may compensate through reduced induction of type I IFN in infected cells and animal models (Ref. 56, Aguilar et al., 2008). These observations have been incorporated in page 8, lines 266–272.

Comment 2: Second, the discussion of RNA folding shifts from thermodynamic stability to kinetic accessibility, which is conceptually important, but lacks concrete experimental or computational evidence (some evidence is provided by predicting the delta G values), but can the authors provide direct support or specific studies demonstrating kinetic folding effects in this context?

Response 2: This is an important point, as the discussion of whether RNA folding shifts from thermodynamic stability to kinetic accessibility is novel and conceptually relevant. This idea has been experimentally supported in a recent study of another cis-acting RNA element, the CHIKV frameshifting element (Ref. 66, Lee et al., 2026). In that study, thermodynamic parameters alone failed to predict frameshifting efficiency: mutants with similar predicted free energies exhibited markedly different functional outcomes. Instead, activity correlated with the formation of long-lived folding intermediates and kinetic traps that restricted access to alternative conformations.

Regarding the 5′SL, IFIT1 sensitivity has traditionally been interpreted based solely on predicted ΔG values. However, in light of these findings, the biologically relevant 5′SL structure may not necessarily correspond to the minimum-free-energy conformation, but rather to a structure that is kinetically accessible during folding. Unfortunately, kinetic analyses of the 5′SL are still lacking, and further studies will be required to determine how RNA folding dynamics influence PAMP exposure and the efficiency of immune recognition.

These considerations and the limitations of available studies were incorporated into the discussion (page 15, lines 443-453).

Comment 3: Third, while the translational implications for mRNA and saRNA vaccine design are interesting, the connection to alphavirus biology appears somewhat speculative; can the authors more clearly delineate which insights are directly supported versus extrapolated?

Response 3: To clarify that the potential translational application remains preliminary, we added the following statement: “Understanding how RNA features modulate innate immune recognition may have implications for vaccine design, as minimizing unwanted immune sensing while preserving efficient antigen expression remains a central challenge.” (page 15, lines 454–456).

We also explicitly qualify the connection to alphavirus biology: “In this context, although still speculative, engineering the alphavirus 5′SL may also contribute to shaping innate immune recognition.” (page 16, line 488).

Moreover, throughout the manuscript we have carefully adjusted the language to reflect the level of evidence.

Comment 4: Additionally, the review would benefit from a more critical discussion of limitations, particularly regarding the reliance on predicted RNA structures and the lack of high-resolution structural validation of the 5′SL.

Response 4: To address this criticism, we now discuss the limitations of predicted structures and the lack of experimental high-resolution structure for the 5’SL.

Page 16, lines 496-504. “A key limitation in the understanding of the 5′SL structure lies in the lack of high-resolution experimental validation. Moreover, while computational models provide valuable insights into potential folding and stability, they do not capture the full complexity of RNA dynamics in the cellular context. In addition, it remains unclear how the RNA structure is affected by interactions with ligands or associated proteins. Future high-resolution structural studies of the 5′SL, including approaches such as cryo-electron microscopy or X-ray crystallography, particularly in complex with interacting proteins, may help clarify cap positioning relative to IFIT1 and identify RNA features involved in polymerase recognition.”

Comment 5: Finally, given the broad coverage of viral evasion strategies, the manuscript could be strengthened by better integrating these mechanisms into a unifying framework rather than presenting them as a descriptive list, and by clarifying how these diverse strategies compare in their relative efficiency and evolutionary significance.

Response 5: In response to the reviewer’s comment, we have expanded the discussion to more clearly compare the advantages and limitations of these strategies. In particular, we highlight that direct cap mimicry or acquisition provides highly effective immune evasion but requires either dedicated viral enzymatic machinery or dependence on host-derived substrates. In contrast, cap-independent translation strategies, as well as RNA structure-based mechanisms, may represent more economical solutions that minimize the need for additional viral functions; however, they often involve trade-offs in translational efficiency and dependence on the cellular context.

In addition, we now frame these mechanisms as alternative evolutionary solutions to a common selective pressure: maintaining efficient translation while evading host RNA sensing.

These considerations have been incorporated into the manuscript (page 16, lines 505-523).

Comment 6: Further, protons in structure of guanine (Figure 2) should be at the N1 position and not the N3 position. please correct this.

Response 6: Thank you for pointing this out. The structure of guanine in Figure 2 has been corrected.

Comment 7: Similarly, it is good that authors have used alphafold3 to predict the structures of RNA structures, but which is the reliability of the structures? No discussion on quality is discussed. How the structures of the mentioned structures change when they interact with surrounding ligands or the protein structure?     

Response 7: We thank the reviewer for raising this important point. We have now included a detailed assessment of the confidence of the AlphaFold predictions in the revised manuscript.

Overall, the analyzed alphavirus 5′UTR RNA constructs exhibit low global confidence scores, indicating limited reliability in long-range tertiary organization. This behavior is expected, as current AlphaFold models are not specifically calibrated for RNA structure prediction, particularly for short RNAs whose folding is dominated by secondary structure. Accordingly, these values should not be interpreted as evidence of incorrect local folding. To provide a more appropriate evaluation of model quality, we analyzed Predicted Aligned Error (PAE) matrices and per-residue pLDDT (predicted Local Distance Difference Test) values directly from the full atomic models (see new Figure S1).

PAE matrices provide an estimate of the positional error between pairs of nucleotides. Our analyses show low error values along the diagonal, consistent with reliable local structural features, but increased uncertainty for long-range interactions. In turn, the per-residue pLDDT values are consistently high, indicating strong internal confidence in local geometry.

Importantly, the predicted stem-loop architectures are fully consistent with SHAPE reactivity data, RNAfold predictions, and the expected structural organization of the alphavirus 5′UTR. We have clarified in the manuscript that these models are therefore used as structural representations compatible with experimentally supported secondary structure, rather than as de novo predictors of global RNA fold (page 10, lines 318-334).

Regarding the potential effects of ligands or protein interactions, we agree that such interactions may influence RNA conformation. However, these effects are not captured by the present AlphaFold predictions, which were performed on isolated RNA sequences. We have added statements in the manuscript acknowledging that potential conformational changes upon protein or ligand binding remain an open question (page 16, lines 497-500; page 17, lines 554-556).

 

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have addressed my comments satisfactorily, and the manuscript has improved significantly