Review Reports

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Dear authors,

This manuscript provides a useful overview of the structural basis of intermolecular interactions between APOBEC3 proteins and HIV-1 Vif. The topic is of clear interest to the HIV-1 field, particularly considering the potent antiviral activity of some APOBEC3 family members. The review is generally well organised and presents a substantial body of structural and mutagenesis data. However, several sections require clarification, and some sentences need rewriting while others would benefit from careful rephrasing to improve  readability. In particular:

Several points in the manuscript are currently difficult to follow because of poor phrasing. For example, the sentence stating that “it is anticipated that its vif gene iteratively evolved…” (lines 53-54) would read more naturally as “the vif gene is thought to have evolved…”. Similarly, in the Introduction, the phrase “these determined structures” (lines 66-67) would read better simply as “these structures”.

The description of the A30-Vif interfaces also need revision. The sentence beginning “When residues crucial for A3-Vif binding were mapped…” (lines 79-82) is odd and needs clarity. For instance, the authors might write something of the sort of: “When residues crucial for A3-Vif binding were mapped in the reported structures of Vif and the A3 domains, the A3-Vif interfaces could be identified.” This would make  the sentence easier to follow.

The figure legends require attention. In Figure 2, the distinction between the structures derived from mutagenesis and those derived from cryo-EM is not sufficiently clear from the legend alone. I suggest labelling the panels for example as “A” and “B” to make this more apparent. It would also be helpful to include column headings indicating Vif-A3F CTD, Vif-A3G, and Vif-A3H. In addition, the legend should describe all the residues and domains indicated by arrows in the figure. A similar issue applies to Figure 3, where the authors should explicitly state that important residues for the interactions are indicated with arrows. Also in this figure, if the authors clearly list how the important residues are labelled, it is unnecessary to add that they followed the same colour scheme as in Figure 2. Figure 4 also appears before it is cited in the main text, therefore, the figure should either be moved or cited earlier. In addition, it titles reads “…and PPP2R5”, when it should say “… and PPP2R5A”.

In the A3F-Vif section, the phrase “have clarified the detailed machinery” (line 108) should be revised. “Have clarified the molecular details” would be more appropriate. In addition, the sentence stating that reciprocal mutagenesis analysis suggested that Vif R15 interacts with A3F E289 “in a contradictory manner” (lines 126-127) is not understandable. The authors should explain what is contradictory: the mutagenesis data conflict with the cryo-EM structure?, with previous biochemical interpretation?, or with the proposed salt-bridge model?

The A3G-Vif section contains several sentences difficult to comprehend. The opening sentence (line 143) would be better as “Several cryo-EM structures of A3G-Vif complexes have been reported.” Likewise, the sentence “the structure revealed that it additionally contains…” (lines 145-146) is also odd and should be rephrased, since it is not clear whether “it” refers to the structure, the complex, or A3G. The discussion of A3G D128/K128 and Vif adaptation in different primate lineages (lines 158-163) is difficult to follow and should be rewritten. In particular, the sentences beginning “For example, the A3G D128 and K128 residues…” and “In contrast, Vif variants of HIV-1 and SIV…” (line 161) require clarification.

The sentence “human A3G interacts with HIV-1 Vif not only directly but also through the intermediary of ssRNA” (lines 164-165) should be reevaluated. A clearer version would be: “HIV-1 Vif interacts with A3G not only directly but also through ssRNA as an intermediary.” Likewise, the RMSD sentence would be clearer if the authors wrote “for all pairwise comparisons among the three structures” rather than “between any two among the three structures” (line 172).

The A3H-Vif section is the part of the manuscript most difficult to follow. The phrase “We adopted chimpanzee A3H” (line 186) is not idiomatic, instead, “We used” or “We selected chimpanzee A3H” would be more appropriate. The sentence “the dsRNA keeps remains bound…” (193) is grammatically incorrect and should be corrected (I suggest to remove "keeps"). More importantly, the surrounding paragraph is difficult to follow. The authors should clarify whether dsRNA remains bound to the proximal A3H protomer while no RNA-mediated bridging interaction between A3H and Vif is observed, in contrast to the A3G-ssRNA-Vif complex.

The explanation of the distal A3H protomer is also unclear. The manuscript states that the distal protomer was clearly determined, “likely because of occasional dissociation or relatively high flexibility” (lines 196-197), but the reasoning behind this statement is not obvious, and if something contradictory: flexibility/dissociation hinders structural determination. The authors should explain  how this observation supports the conclusion that only the proximal A3H protomer is selectively subjected to Vif-mediated degradation. In the sentence beginning “The intermolecular interactions are constituted primarily…” (lines 198-201), it is also unclear which interactions are meant: A3H-Vif, A3H-dsRNA, or interactions within the dimeric assembly.

The paragraph between lines 204-208 is also incongruent: it starts by stating that it is unclear how Vif H48 mediates degradation of human A3H hapII, to then continue saying “This is because the Vif residue at position 48 is distal from the A3H interface, according to the determined structures”. How something that is unclear has an immediate explanation?

The paragraph describing the “unique dimeric assembly of the A3H-Vif complex” (lines 209-215) should define what the dimer is. At its current state, it is not clear whether the authors refer to an A3H-Vif/A3H-Vif dimer, an A3H dimer within the Vif-bound complex, or another assembly. The final sentence in this paragraph would also benefit from rephrasing.For example, I propose: “This interaction is likely driven by the ability of the positively charged surface around the Vif α1 helix to bind the negatively charged phosphate backbone of RNA.”

Finally, the statement in the similarities/differences section that “positively charged interfaces on both A3G and Vif result in the formation of a negatively charged ssRNA molecule” (lines 242-244) is incongruent. Positively charged protein interfaces can't “result in the formation” of ssRNA. I believe that the authors mean that a negatively charged ssRNA molecule is accommodated between positively charged surfaces on A3G and Vif. This sentence should be rewritten to avoid implying that the protein surfaces generate the RNA molecule.

Comments on the Quality of English Language

I have detected many instances where the language employed is not idiomatic correct. I have tried to point all these instances in my review, but I belive that the manuscript would benefit from a thorough proofreading.

Author Response

Comment 1: This manuscript provides a useful overview of the structural basis of intermolecular interactions between APOBEC3 proteins and HIV-1 Vif. The topic is of clear interest to the HIV-1 field, particularly considering the potent antiviral activity of some APOBEC3 family members. The review is generally well organised and presents a substantial body of structural and mutagenesis data. However, several sections require clarification, and some sentences need rewriting while others would benefit from careful rephrasing to improve readability.

Response 1: We deeply appreciate the reviewer’s careful reading of our manuscript and providing valuable comments. We revised the manuscript according to the reviewer’s suggestions.

Comment 2: In particular: Several points in the manuscript are currently difficult to follow because of poor phrasing. For example, the sentence stating that “it is anticipated that its vif gene iteratively evolved…” (lines 53-54) would read more naturally as “the vif gene is thought to have evolved…”. Similarly, in the Introduction, the phrase “these determined structures” (lines 66-67) would read better simply as “these structures”.

Response 2: As indicated by the reviewer, we corrected the two sentences. (Lines 53-54 and 66-67)

Comment 3: The description of the A30-Vif interfaces also need revision. The sentence beginning “When residues crucial for A3-Vif binding were mapped…” (lines 79-82) is odd and needs clarity. For instance, the authors might write something of the sort of: “When residues crucial for A3-Vif binding were mapped in the reported structures of Vif and the A3 domains, the A3-Vif interfaces could be identified.” This would make the sentence easier to follow.

Response 3: We revised the sentence as follows: “When residues crucial for A3-Vif binding were mapped onto the reported structures of the Vif (reference) (PDB 4N9F) and the A3 domains (references), the A3-Vif interfaces could be identified.” (Lines 81-82)

Comment 4: The figure legends require attention. In Figure 2, the distinction between the structures derived from mutagenesis and those derived from cryo-EM is not sufficiently clear from the legend alone. I suggest labelling the panels for example as “A” and “B” to make this more apparent. It would also be helpful to include column headings indicating Vif-A3F CTD, Vif-A3G, and Vif-A3H. In addition, the legend should describe all the residues and domains indicated by arrows in the figure. A similar issue applies to Figure 3, where the authors should explicitly state that important residues for the interactions are indicated with arrows. Also in this figure, if the authors clearly list how the important residues are labelled, it is unnecessary to add that they followed the same colour scheme as in Figure 2. Figure 4 also appears before it is cited in the main text, therefore, the figure should either be moved or cited earlier. In addition, it titles reads “…and PPP2R5”, when it should say “… and PPP2R5A”.

Response 4: We added the labels “A” and “B” in each of the Figures 2 and 3. Accordingly, we also corrected the labels of the Figures the main text and the figure legends. In addition, we corrected the labels of “+A3F CTD”, “+A3G”, and “+A3H” at the bottom of the Figure 2 and “+Vif/CBF-b” at the bottom of the Figure 3. The color bars for electrostatic potentials were added in the Figures 2 and 3. Moreover, we revised the legends of the Figures 2 and 3 to explicitly list the important residues and helices, and to clearly show the color scheme in the Figure 3. The “PPP2R5” in the Figure 4 legend was also changed to “PPP2R5A”.

Comment 5: In the A3F-Vif section, the phrase “have clarified the detailed machinery” (line 108) should be revised. “Have clarified the molecular details” would be more appropriate. In addition, the sentence stating that reciprocal mutagenesis analysis suggested that Vif R15 interacts with A3F E289 “in a contradictory manner” (lines 126-127) is not understandable. The authors should explain what is contradictory: the mutagenesis data conflict with the cryo-EM structure?, with previous biochemical interpretation?, or with the proposed salt-bridge model?

Response 5: As the reviewer suggested, we rephrased “have clarified the detailed machinery” into “have clarified the molecular details”. (Line 111) In addition, we revised the sentence regarding the contradictory findings between biochemical and cryo-EM studies as follows: “Moreover, reciprocal mutagenesis analysis in a previous biochemical study has suggested that Vif R15 interacts with A3F E289 [36], which is contradictory with the observation from the cryo-EM structure.” (Lines 129-131)

Comment 6: The A3G-Vif section contains several sentences difficult to comprehend. The opening sentence (line 143) would be better as “Several cryo-EM structures of A3G-Vif complexes have been reported.” Likewise, the sentence “the structure revealed that it additionally contains…” (lines 145-146) is also odd and should be rephrased, since it is not clear whether “it” refers to the structure, the complex, or A3G. The discussion of A3G D128/K128 and Vif adaptation in different primate lineages (lines 158-163) is difficult to follow and should be rewritten. In particular, the sentences beginning “For example, the A3G D128 and K128 residues…” and “In contrast, Vif variants of HIV-1 and SIV…” (line 161) require clarification.

Response 6: We revised the first sentence of the section, as suggested by the reviewer. (Line 150) In addition, the “it” was replaced with “the complex”. (Line 152). Moreover, we carefully revised the sentences that the reviewer pointed out in the same paragraph (Lines 193-202). They are now described as follows: “For example, the A3G D128 and K128 residues, which naturally occur in hominoids and Old World monkeys (OWMs), respectively [38], are critical for phenotypes sensitive or resistant to HIV-1 Vif-mediated degradation [39-41]. In contrast, Vif variants of HIV-1 and SIV infected in chimpanzees can adapt to A3G D128 by bearing Q83 or H83, whereas it is likely that the presence of Tyr (Y) at the corresponding position of SIV Vif adapted in OWMs is suitable for A3G K128 [42].”

Comment 7: The sentence “human A3G interacts with HIV-1 Vif not only directly but also through the intermediary of ssRNA” (lines 164-165) should be reevaluated. A clearer version would be: “HIV-1 Vif interacts with A3G not only directly but also through ssRNA as an intermediary.” Likewise, the RMSD sentence would be clearer if the authors wrote “for all pairwise comparisons among the three structures” rather than “between any two among the three structures” (line 172).

Response 7: We corrected the two phrases as suggested by the reviewer. (Lines 171-172, 179)

Comment 8: The A3H-Vif section is the part of the manuscript most difficult to follow. The phrase “We adopted chimpanzee A3H” (line 186) is not idiomatic, instead, “We used” or “We selected chimpanzee A3H” would be more appropriate. The sentence “the dsRNA keeps remains bound…” (193) is grammatically incorrect and should be corrected (I suggest to remove "keeps"). More importantly, the surrounding paragraph is difficult to follow. The authors should clarify whether dsRNA remains bound to the proximal A3H protomer while no RNA-mediated bridging interaction between A3H and Vif is observed, in contrast to the A3G-ssRNA-Vif complex. The explanation of the distal A3H protomer is also unclear. The manuscript states that the distal protomer was clearly determined, “likely because of occasional dissociation or relatively high flexibility” (lines 196-197), but the reasoning behind this statement is not obvious, and if something contradictory: flexibility/dissociation hinders structural determination. The authors should explain  how this observation supports the conclusion that only the proximal A3H protomer is selectively subjected to Vif-mediated degradation. In the sentence beginning “The intermolecular interactions are constituted primarily…” (lines 198-201), it is also unclear which interactions are meant: A3H-Vif, A3H-dsRNA, or interactions within the dimeric assembly.

Response 8: We thoroughly changed this paragraph as follows: Initially, the determination of three X-ray crystal structures of pig-tail macaque, chimpanzee and human hapII A3Hs revealed that these primate A3Hs commonly form homodimers bridged by a short double-stranded RNA (dsRNA) [34,54,55]. Chimpanzee A3H (cpzA3H), which exhibits higher solubility than other primate A3Hs, is more sensitive to HIV-1 Vif than human A3H hapII due to an amino acid substitution at position 97 (Q97 in chimpanzee versus K97 in humans) [21]. Based on the cpzA3H homodimer structure, the putative Vif-interaction interface in the A3H dimer was predicted to reside within a single A3H unit, remaining accessible to Vif without potential steric hindrances, despite the dsRNA-mediated dimerization. Therefore, it was hypothesized that two units of the Vif– CBF- complex could bind simultaneously to one A3H dimer.

Recently, our group and others resolved the complex structures of A3H with HIV-1 Vif, CBF-ß, CUL5, ELOB, and ELOC [27,28] (PDB 8FVI and 9E93), revealing a consistent architecture of A3H-Vif interactions. Unexpectedly, within the stable dsRNA-mediated homodimer, Vif directly binds with only one of the two A3H protomers (the proximal A3H). In these complex structures, the dsRNA remains bound to the proximal A3H protomer, whereas there is no obvious ssRNA-mediated interaction between A3H and Vif unlike the A3G-Vif complex [24-26]. Notably, the structure of the other A3H protomer (the distal A3H) was poorly determined, likely due to occasional dissociation or high flexibility. These observations might explain why only the proximal A3H protomer is selectively disassembled from the stable dimer for subsequent proteasomal degradation [28].

The intermolecular interactions between A3H and Vif are constituted primarily by the a3 and a4 helices in A3H and the b-sheet (b2-b6 strands) in HIV-1 Vif, which is in agreement with previous reports based on mutagenesis experiments [21,56] (Figs. 2 and 3). Intriguingly, the A3H Q97 residue forms intermolecular hydrogen bonds with Vif K63. Therefore, the K97 residue in wild-type human A3H hapII reduces Vif susceptibility

Comment 9: The paragraph between lines 204-208 is also incongruent: it starts by stating that it is unclear how Vif H48 mediates degradation of human A3H hapII, to then continue saying “This is because the Vif residue at position 48 is distal from the A3H interface, according to the determined structures”. How something that is unclear has an immediate explanation?

Response 9: We corrected the sentence as “This is because the Vif residue at position 48 is distal from the A3H interface, according to the determined structures.” (Lines 220-221)

Comment 10: The paragraph describing the “unique dimeric assembly of the A3H-Vif complex” (lines 209-215) should define what the dimer is. At its current state, it is not clear whether the authors refer to an A3H-Vif/A3H-Vif dimer, an A3H dimer within the Vif-bound complex, or another assembly. The final sentence in this paragraph would also benefit from rephrasing. For example, I propose: “This interaction is likely driven by the ability of the positively charged surface around the Vif α1 helix to bind the negatively charged phosphate backbone of RNA.”

Response 10: As the reviewer commented, this paragraph is too complicated for the readers to understand the main point. Therefore, this paragraph was removed. Instead, we explain more details about the issue about the A3H-Vif dimers in the “5. A3H-Vif interaction” section (lines 215-255).

Comment 11: Finally, the statement in the similarities/differences section that “positively charged interfaces on both A3G and Vif result in the formation of a negatively charged ssRNA molecule” (lines 242-244) is incongruent. Positively charged protein interfaces can't “result in the formation” of ssRNA. I believe that the authors mean that a negatively charged ssRNA molecule is accommodated between positively charged surfaces on A3G and Vif. This sentence should be rewritten to avoid implying that the protein surfaces generate the RNA molecule.

Response 11: As the reviewer suggested, we corrected the sentence. (Lines 248-249)

Comment 12: Comments on the Quality of English LanguageI have detected many instances where the language employed is not idiomatic correct. I have tried to point all these instances in my review, but I belive that the manuscript would benefit from a thorough proofreading.

Response 12: We deeply appreciate the reviewer’s helpful comments and suggestions. According to the suggestions, we corrected inappropriate phrases and sentences. In addition, we carefully proofread the revised manuscript and now think it appropriately corrected.

 

 

Reviewer 2 Report

Comments and Suggestions for Authors

In this review, the authors provide an overview of the current understanding of the structure and molecular interactions between the APOBEC3 proteins and HIV-1 Vif protein. In particular, the review focuses on describing how the HIV-1 Vif protein recognizes three different APOBEC3 proteins (A3G, A3F, A3H) through distinct interaction interfaces. It also discusses the differences and similarities in these A3-Vif recognition mechanisms.

Suggestions:

1. Lines 63-66: Please cite the key studies supporting this statement, particularly those reporting the cryo-EM structures of HIV-1 Vif in complex with APOBEC3 proteins and E3 ubiquitin ligase components.
2. Figs. 2 and 3: Please include the color scale for the electrostatic surface potential (e.g., ranging from +5 kT/e in blue to -5 kT/e in red), to facilitate interpretation of the structural models.
3. The manuscript would benefit from a more detailed discussion of the druggability of the Vif–APOBEC3 interaction interface. Expanding this section to discuss the current evidence, challenges, and future therapeutic perspectives would provide readers with a more comprehensive overview of this important aspect.

Author Response

Comment 1: Lines 63-66: Please cite the key studies supporting this statement, particularly those reporting the cryo-EM structures of HIV-1 Vif in complex with APOBEC3 proteins and E3 ubiquitin ligase components.

Response 1: We thank the reviewer for providing us helpful comments. As the reviewer suggested, we added six references regarding the key cryo-EM studies (reference #23-28) at the end of the sentence (Line 66).

Comment 2: Figs. 2 and 3: Please include the color scale for the electrostatic surface potential (e.g., ranging from +5 kT/e in blue to -5 kT/e in red), to facilitate interpretation of the structural models.

Response 2: As suggested by the reviewer, we include the color bars at the bottoms in Figures 2 and 3.

Comment 3: The manuscript would benefit from a more detailed discussion of the druggability of the Vif–APOBEC3 interaction interface. Expanding this section to discuss the current evidence, challenges, and future therapeutic perspectives would provide readers with a more comprehensive overview of this important aspect.

Response 3: In the summary, we added a statement citing a review article regarding drug development against Vif or A3 (Olson et al. 2018, reference#67) as follows: “This information will be valuable for developing a new class of anti-HIV-1 drugs that interfere with interactions at the A3-Vif interfaces. Although the Vif-binding pockets of A3s appear shallow for the design high-affinity small-molecule inhibitors, multiple groups have nevertheless identified candidate inhibitors targeting at the Vif-A3 interaction [67].” (Lines: 275-277)

Reviewer 3 Report

Comments and Suggestions for Authors

HIV-1 Vif functions as a pleiotropic hub proteins that counteracts human APOBEC3 antiviral restriction through three spatially distinct, non-overlapping binding interfaces that recognize three structurally different types of APOBEC3 proteins (A3G, A3C/D/F, and A3H) via distinct molecular mechanism, including direct protein-protein contacts, RNA-mediated interactions, and selective homodimer recognition, ultimately hijacking the cellular Cul5-E3 ubiquitin ligase complex to promote APOBEC3 degradation.

The review is well written, comprehensive, and presents a balanced perspective that will be of broad interest to the virology community. I have only two minor comments:

- Section 4 (A3F–Vif interactions, from line 176): The authors state that "A3G–Vif interaction is most likely due to A3G's strong binding capacity to ssRNA...". While I agree with this interpretation, it would be worth mentioning that Vif is itself an RNA-binding protein, with well-documented RNA-binding activity that plays important functional roles during the HIV-1 replication cycle (DOI: 10.1128/jvi.74.19.8938-8945.2000 (Dettenhofer JVI 2000); DOI: 10.1128/JVI.75.16.7252-7265.2001 (Khan JVI 2001); DOI: 10.1016/j.jmb.2005.09.025 (Henriet JMB 2005); DOI: 10.1074/jbc.M703122200 (Bernacchi JBC 2007)). Briefly acknowledging this point would provide a more complete view of the molecular basis of the A3G–Vif interaction.

- Figure 2: For consistency with Figure 3, I suggest indicating A3F–Vif, A3G–Vif, and A3H–Vif at the bottom (or alternatively in the center) of the figure to facilitate interpretation.

Author Response

Comment 1: - Section 4 (A3F–Vif interactions, from line 176): The authors state that "A3G–Vif interaction is most likely due to A3G's strong binding capacity to ssRNA...". While I agree with this interpretation, it would be worth mentioning that Vif is itself an RNA-binding protein, with well-documented RNA-binding activity that plays important functional roles during the HIV-1 replication cycle (DOI: 10.1128/jvi.74.19.8938-8945.2000 (Dettenhofer JVI 2000); DOI: 10.1128/JVI.75.16.7252-7265.2001 (Khan JVI 2001); DOI: 10.1016/j.jmb.2005.09.025 (Henriet JMB 2005); DOI: 10.1074/jbc.M703122200 (Bernacchi JBC 2007)). Briefly acknowledging this point would provide a more complete view of the molecular basis of the A3G–Vif interaction.

Response 1: We appreciate that the reviewer commented on our manuscript and provided us with important insights. As the reviewer indicated, we have revised the last sentence in the section “4. A3G-Vif interactions” as follows, and have cited all the references that the reviewer recommended.

“Hence, this evidence suggests that Vif recognizes the ssRNA-binding state of A3G rather than recruiting ssRNA to the A3G-Vif complex, although Vif is an RNA-binding protein as well [50-53].” (Lines 189-191)

Comment 2: - Figure 2: For consistency with Figure 3, I suggest indicating A3F–Vif, A3G–Vif, and A3H–Vif at the bottom (or alternatively in the center) of the figure to facilitate interpretation.

Response 2: We amended “+A3F CTD”, “+A3G”, and “+A3H” at the bottom in Figure 2 and “+Vif/CBF-b” at the bottom in Figure 3. Thank you very much.

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have adequately addressed my previous comments in the revised manuscript.