SIV-specific neutralizing antibody induction following selection of a PI3K drive-attenuated nef variant

Reviewer #1 (Public review):

Human and simian immunodeficiency viruses (HIV and SIV, respectively) evolved numerous mechanisms to compromise effective immune responses but the underlying mechanisms remain incompletely understood. Here, Yamamoto and Matano examined the humoral immune response in a large number of rhesus macaques infected with the difficult-to-neutralize SIVmac239 strain and identified a subgroup of animals showing significant neutralizing Ab responses. Sequence analyses revealed that in most of these animals (7/9) but only a minority in the control group (2/19) SIVmac variants containing a CD8+ T-cell escape mutation of G63E/R in the viral Nef gene emerged. Functional analyses revealed that this change attenuates the ability of Nef to stimulate PI3K/Akt/mTORC2 signalling. The authors propose that this improved induction of SIVmac239 nAb is reciprocal to antibody dysregulation caused by a previously identified human PI3K gain-of-function mutation associated with impaired anti-viral B-cell responses. Altogether, the results suggest that PI3K signalling plays a role in B-cell maturation and generation of effective nAb responses. Preliminary data indicate that Nef might be transferred from infected T cells to B cells by direct contact. However, the exact mechanism and the relevance for vaccine development requires further studies

Strengths of the study are that the authors analyzed a large number of SIVmac-infected macaques to unravel the biological significance of the known effect of the interaction of Nef with PI3K/Akt/mTORC2 signaling. This is interesting and may provide a novel means to improve humoral immune responses to HIV. In the revised version the authors made an effort to address previous concerns. Especially, they provide data supporting that Nef might be transferred to B cells by direct cell-cell contact. In addition, the provide some evidence that G63R that also emerged in most animals does not share the disruptive effect of G63G although experimental examination and discussion why G63R might emerge remains poor. Another weakness that remains is that some effects of the G63E mutation are modest and effects were not compared to SIVmac constructs lacking Nef entirely. The evidence for a role of Nef G63E mutation on PI3K and the association with improved nAb responses was largely convincing and it is appreciated that the authors provide additional evidence for a potential impact of "soluble" Nef on neighboring B cells. However, the experimental set-up and the results are difficult to comprehend. It seems that direct cell-cell contact is required and membranes are exchanged. Since Nef is associated with cellular membranes this might lead to some transfer of Nef to B cells. However, the immunological and functional consequences of this remain largely elusive. Alternatively, Nef-mediated manipulation of helper CD4 T cells might also impact B cell function and effective humoral immune responses. As previously noted, the presentation of the results and conclusions was in part very convoluted and difficult to comprehend. While the authors made attempts to improve the writing parts of the manuscript are still challenging to follow. This applies even more to the rebuttal (complex words combined with poor grammar), which made it difficult to assess which concerns have been satisfactory addressed.

Author response:

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

Reviewer #1 (Public Review):

This work describes the induction of SIV-specific NAb responses in rhesus macaques infected with SIVmac239, a neutralization-resistant virus. Typically, host NAb responses are not detected in animals infected with SIVmac239. In this work, seventy SIVmac239-infected macaques were retrospectively screened for NAb responses and a subset of nine animals were identified as NAb-inducers. The viral genomes from 7/9 animals that induced NAb responses were found to encode nonsynonymous mutation in the Nef gene (amino acid G63E). In contrast, Nef G63E mutation was found only in 2/19 NAb non-inducers - implicating that the Nef G63E mutation is selected in NAb inducers. Measurement of Nef G63E frequencies in plasma viruses suggested that Nef G63E selection preceded NAb induction. Nef G63E mutation was found to mediate escape from Nef-specific CD8+ T-cell responses. To examine the functional phenotype of Nef G63E mutant, its effect on downmodulation of Nef-interacting host proteins was examined. Infection of rhesus and cynomolgus macaque CD4+ T cell lines with WT or Nef G63E mutant SIV suggested that Nef mutant reduces S473 phosphorylation of AKT. Using flow cytometry-based proximity ligation assay, it was shown that Nef G63E mutation reduced binding of Nef to PI3K p85/p110 and mTORC2 GβL/mLST8 and MTOR components - kinase complex responsible AKT-S473 phosphorylation. In vitro B-cell Nef invasion and in vivo imaging/flow cytometry-based assays were employed to suggest that Nef from infected cells can target Env-specific B cells. Lastly, it was determined that NAb inducers have significantly higher Env-specific B-cells responses after Nef G63E selection when compared to NAb non-inducers. Finally, a corollary was drawn between the Nef G63E-associated B-cell/NAb induction phenotype and activated PI3K delta syndrome (APDS), which is caused by activating GOF mutations in PI3K, to suggest that Nef G63E-meidated induction of NAb response is reciprocal to APDS.

Strengths:

This study aims to understand the viral-host interaction that governs NAb induction in SIVmac239-infected macaques - this could enable identification of determinants important for induction of NAb responses against hard-to-neutralize tier-2/3 HIV variants. The finding that SIV-specific B-cell responses are induced following Nef G63E CD8+ T-cell escape mutant selection argue for an evolutionary trade-off between CTL escape and NAb induction. Exploitation of such a cellular-humoral immune axis could be important for HIV/AIDS vaccine efforts.

Although more validation and mechanistic basis are needed, the corollary between PI3K hyperactive signaling during autoimmune disorders and Nef-mediated abrogated PI3K signaling could help identify novel targets and modalities for targeting immune disorders and viral infections.

We are grateful for the supportive and insightful comments. The work did seem to unintendedly highlight a conceptual link between extrinsic and intrinsic immune perturbations. We will keep working on both wings, aiming to evoke synergisms.

Weaknesses:

Although the paper does have strengths in principle, the weaknesses of the paper are that the mechanistic basis of Nef-mediated induction of NAb responses are not directly examined. For example, it remains unclear whether SIVmac239 with engineered G63E mutation in Nef would induce faster and potent NAb responses. A macaque challenge study is needed to address this point.

We appreciate the point. We do have certain difficulties in availability of macaques for de novo experiments. As partially discussed in ver1, the identified Nef phenotype selected post-acute infection confers an enhanced CD4+ T cell-killing effect (revised Fig 4F), and it is likely that de novo infection with the mutant would redirect the trajectory of infection to rapid disease/AIDS progression accompanying generalized immune failure by boosting acute-phase CD4 destruction. In other words, mutant de novo infection may not necessarily be directly discussable as an attempt for reconstitution. It appears equally critical to understand the mutant in vitro on an immunosignaling basis, and in the current work we have focused on depicting this as the first step. We will work on reconstitution experiments with emphasis on pharmacology in our future study.

As presented, the central premise of the paper involves infected cell-generated Nef (WT or G63E mutant) being targeted to adjacent Env-specific B cells. However, it remains unclear how this is transfer takes place. A direct evidence demonstrating CD4+ T cell-associated and/or cell-free Nef being transferred to B-cell is needed to address this concern.

We appreciate the point, also pointed out by Reviewers 2 and 3. We have performed three sets of in vitro reconstitution experiments graphically/functionally addressing how Nef transfer from CD4+ T cells to B cells can be modulated (new Fig 6) and edited text accordingly.

The interaction between Nef and PI3K signaling components (p85, p110, GβL/mLST8, and MTOR) has been explored using PLA assay, however, this requires validation using additional biochemical and/or immunoprecipitation-based approaches. For example, is Nef (WT or mutant form) sufficient to affect PI3K-induced phosphorylation of Akt in an in vitro kinase assay? Moreover, the details regarding the binding events of WT vs mutant Nef with PI3K signaling components is lacking in this study. Lastly, it is unclear whether the interaction of Nef with PI3K signaling components is a conserved function of all primate lentiviruses or is this SIV-specific phenotype.

We appreciate the point. Co-immunoprecipitation analysis via pulldown with the mTORC2-intrinsic cofactor Sin1 (revised Fig 4E), showing decreased G63E-Nef binding, should confer robustness to the statement combined with initial manipulation results (Fig 4C). As Sin1 is mTORC2- and not mTORC1-intrinsic, results should be strengthened. Phosflow may be a standard readout nowadays for pAkt itself. Related with sequence variation, conservation will be addressed in studies ahead. We concisely mentioned on this in the revision (Lines 390-391).

It has been previously reported that the region of Nef encoding glycine at position 63 is not conserved in HIV-1 (Schindler et al, Journal of Virology 2004). Thus, does HIV-1 Nef also function in induction of NAb responses in humans? or the observed phenotype specific to SIV?

We appreciate the point, and do not have an answer at the moment. We will explore in our HIV-1-infected patient cohort (Hau et al, AIDS 2022) and other occasions whether corresponding phenotypes may exist. We have mentioned on this point in the revised manuscript (Line 392-393).

Reviewer #2 (Public Review):

It is well known that human and simian immunodeficiency viruses (HIV and SIV, respectively) evolved numerous mechanisms to compromise effective immune responses but the underlying mechanisms remain incompletely understood. Here, Yamamoto and Matano examined the humoral immune response in a large number of rhesus macaques infected with the difficult-to-neutralize SIVmac239 strain. They identified a subgroup of animals that showed significant neutralizing Ab responses. Sequence analyses revealed that in most of these animals (7/9) but only a minority in the control group (2/19) SIVmac variants containing a CD8+ T-cell escape mutation of G63E/R in the viral Nef gene emerged. They further show that this change attenuates the ability of Nef to stimulate PI3K/Akt/mTORC2 signaling. The authors propose that this induction of SIVmac239 nAb induction is reciprocal to antibody dysregulation caused by a previously identified human PI3K gain-of-function (Ref). Altogether, the results suggest that PI3K signaling plays a key role in B-cell maturation and generation of effective nAb responses.

Strengths of the study are that the authors analyzed a large number of SIVmac-infected macaques to unravel the biological significance of the known effect of the interaction of Nef with PI3K/Akt/mTORC2 signaling. This is interesting and may provide a novel means to improve humoral immune responses to HIV. Weaknesses are that only G63E and not G63R that also emerged in most animals was examined in most functional assays. Some effects of the G63E mutation seem modest and comparison to a grossly nef-defective SIVmac construct would be desirable to better assess to impact of the mutation of Nef-mediated stimulation of PI3K. While the impact of this Nef mutations on PI3K and the association with improved nAb responses is largely convincing, the results on the potential impact of soluble Nef on neighboring B cells is much less clear. SIVmac239 infects and manipulates helper CD4 T cells and these are essential for the activation and differentiation of B cells into antibody-producing plasma cells and effective humoral immune responses. Without additional functional evidence that Nef indeed specifically targets and manipulated B cells these results and conclusions should be made with much greater caution. Finally, the presentation of the results and conclusions is partly very convoluted and difficult to comprehend. Editing to improve clarity is highly recommended.

We are very grateful for the supportive and visionary review and suggestions. Experiments have been performed to improve the points raised. This work inevitably involved interdisciplinary factors to even hit on the schematic (NAbs, B cells, CD4+T, CD8+T, viral escape, immunosignaling, IEI as extrapolation & microscopy implementations) and convoluted sections should have existed. We attempted streamlining of certain portions and edited writing throughout, and hope that it became more straightforward.

Reviewer #2 (Recommendations For The Authors):

As outlined in the public review, I found the results potentially very interesting but parts of the manuscript much more complex and confusing than necessary. In addition, the methods on the potential impact of soluble Nef on neighboring B cells in vivo was difficult to assess but altogether this part was not convincing. Have the following specific suggestions:

We are very grateful for the scholarly review, and encouraging and suggestive comments on this orphan work. In the revision we designed experiments to address the properties of Nef transfer to append understanding on the in vivo B-cell data. Recommendations have been addressed as follows.

(1) Title: "AIDS virus-neutralizing antibody induction reciprocal to a PI3K gain-of-function disease". Think this title hardly reflects the data; SIVmac cause simian AIDS and is not the "AIDS virus" the 2nd part is more appropriate for discussion than for the title (and the abstract).

We appreciate the point. The original intent of the title was to conceptually bridge two differing fields of virus-host interaction and inborn errors of immunity/immunosignaling on an original article basis. Certain papers (Mudd et al, Nature 2012 etc) do utilize the term AIDS virus, and we similarly chose the term for simplification to non-virologists at initial submission.

That being said, we understand the scholarly point raised, and feel that the initial aim can be well attained by retaining the key host effector PI3K in the title, as in the revised submission titled “SIV-specific neutralizing antibody induction following selection of a PI3K drive-attenuated nef variant”.

(2) Abstract and throughout: As the authors show, SIVmac is not generally "neutralization resistant"; difficult to neutralize is more appropriate and should be used throughout. Also, the abstract and other parts are more complicated than necessary.

We appreciate the point. HIV/SIV Env immunology work utilizes “neutralization-resistant” for SIVmac239 (e.g., Mason et al, PLoS Pathog 2016), and autologous titer positivity of ~10% at this size of examination does appear low amongst lentiviruses. Nevertheless, as recommended, “difficult-to-neutralize” better describes the nature, and we have switched the term accordingly.

Linked with title modification, we reflected the comment on abstract structure and switched the main introductory sentence (Here we…) to a more data-based one instead of depicting extrapolation, and have modified phrasings in the latter half.

(3) The intro seems a bit biased. Immune evasion due to mutations and proviral integration that play key roles in viral persistence are not mentioned. nAbs are not known to efficiently control HIV or SIV replication in vivo (not even in the present study). Thus, a more "balanced" presentation of the role of nAbs in vivo is desirable.

We agree with the comment. Introduction in ver1 submission was compressed to just display humoral immune perturbation examples across persistence-prone viral infections, and indeed it should be much better to layout the multiscale strategies of lentiviruses in manifesting viral persistence. We have appended two sets of texts, one on the fundamental integrating retroviral life cycle and another on the wide spectrum of accessory protein-driven perturbation. As pointed out, the current endogenous induction is of course not early enough to exert suppressive impact on replication as like in exogenous Ab passive infusions. We have accordingly modulated text to improve the balance.

(4) Lines 73-76: rephrase for clarity.

We acknowledge the comment and have rephrased accordingly.

(5) Line 92: "linked with sustained Env-specific B-cell responses after the mutant Nef selection". After or during in one case; the time frame varies enormously and this should be discussed.

We appreciate the comment. The six Nef-G63E mutant-selecting NAb inducers subjected to B-cell analysis were the ones that showed precedence in Fig 2D (mutant before induction). That being said, we modified text as suggested (Line 104 in revised uploaded text). Text related to temporal deviation has been appended (Lines 378-383 in revised uploaded text).

(6) The authors should discuss G63R and include it in the functional analyses.

We appreciate the comment. Discussion on Nef-G63R in ver1 submission was kept minimal because statistical significance for selection was marginal. We generated a Nef-G63R mutant and results are appended in Fig 4-Figure Supplement 2.

(7) Lines 124/5: conservation only applies to SIVsmm/mac Nefs and this region is also frequently deleted/length-variable in primary HIV-1 Nefs.

We appreciate the comment. We modified description of the region accordingly (Lines 139-141 in revised text).

(8) Lines 153-155: Statement doesn't seem to make sense. The triple mutant Nef SIVmac construct was not attenuated for replication but specifically disrupted in CD3 down-modulation.

We acknowledge the comment. It had meant that the consequent plasma viral load showed a trend of decrease (as in the Graphical Abstract of the work) which should (in a simplistic view) influence antigenicity for humoral immune responses. Yet it is very true that virological replicative capacity was comparable with wild-type as in Fig.1. We have taken down the related text and rephrased it (Ref remains cited in introduction).

(9) Lines 178/9: levels in PI3K gain-of-function mice "with full disease phenotype (Avery et al., 2018)". This needs more information, e.g. what disease exactly are they talking about?

We are grateful for the correction, and have appended text and introduced the mentioned congenital disease in the Introduction section in advance. In-detail description is also appended in the Discussion section.

(10) Lines 186/7: "Env-stimulating high-MOI infection also accelerated phenotype appearance, with enhanced 50% reduction (Figure 4C, right)". Modify text and corresponding figure for clarity.

We acknowledge the comment. We revised as: “A high-MOI SIV infection, comprising higher initial concentration of extracellular Env stimuli, also accelerated phenotype appearance from day 3 to day 1 post-infection with stronger pAkt reduction”.

(11) The validity of the results described in the section "Targeting of lymph node Env-specific B cells by Nef in vivo" was difficult to assess. Altogether, however, I didn't find them convincing, especially since a negative control (e.g. macaques infected with nef-deleted SIVmac) are missing.

We acknowledge the comment. As a pure experimental control, whole-Nef deletion may assist for subtracted baselines. Within this work, the staining per se at least should be highly specific (mAb multiply verified in other applications and cytometry panel also designed for minimal spillover into AF488 channel). On in vivo basis, direct comparison may be somewhat frustrated by the fact that reduction in other pleiotropic effects of Nef seem to more dominate upon Nef deletion, as a set of reduced viremia, robust CD8 responses, killer CD4 responses and increased binding Ab titers (Johnson et al, J Virol 1997, Gauduin et al, J Exp Med 2006, Fukazawa et al, Nat Med 2012, Adnan et al, PLoS Pathog 2016 etc) leading to altered trajectory. We promise that we will work on refinement of the methodology in studies ahead.

(12) Lines 309-319: This paragraph made little sense to me (as did lines 328-331).

We acknowledge the comment and have edited both sections.

Reviewer #3 (Additional Reviewer):

In this manuscript, Hiroyuki Yamamoto et al examined virus-specific antibody responses and identified a subgroup of nine individuals, out of seventy SIVmac239 rhesus macaques of Burmese origin infected with SIVmac239, that develop neutralizing antibodies (NAb). The authors propose the emergence of a nef mutant (Nef-G63E) that impacts on B cell maturation resulting in PI3K gain-of-function.

My major concerns are:

The authors by different aspect addressed the role of the emergence of Nef-G63E mutant in individuals developing NAb. The manuscript is confused and the rational not always clearly stated. This reflects the two aspects of the manuscript (i) NAb identification in a subgroup of macaque and (2) the identification this nef mutation.

We are grateful for the comprehensive and scholarly comments. As pointed out, the work did need to confront potential bifurcation of the influence of the obtained viral immunosignaling phenotype for CD4-intrinsic (which might be your specialty) and B-cell-intrinsic impact. Based on your suggestions we have acquired additional data and revised the manuscript as attached.

The authors used both males (n=57) and females (n=13). However, there is no indication related to the sex regarding NAb inducers versus non-NAb Inducers. The notion of "highly pathogenic" is certainly not correct (see the introduction). Pathogenicity is also depending on monkey origin. Thus, cynomolgus are less sensitive to SIVmac239 or SIVmac251 compared to rhesus macaques (Ling B Aids 2002; Reimann KA, J Virol 2005; Cumont MC, J Virol 2008), or to pigtails used in US. Indeed, the authors used Burmese macaques, and therefore the dynamics of pathogenicity is different to rhesus macaque (Indian origin) housed in US. How many animals have been sacrificed out of the 61 animals? Herein, the animals are surviving longer (more than one year), and therefore the notion of "highly pathogenic" merits to be modulated.

We appreciate the comment. We have accordingly appended sex information (M/F: 8/1 versus 49/12 in NAb inducers vs non-inducers, p > 0.99 by Fisher’s exact test) in the methods section. As pointed out there are differences in the frequency and rate of AIDS progression among macaques of differing origin, whereas we have also previously reported reproducible AIDS progression dependent on MHC-I genotypes in the Burmese rhesus macaques utilized (Nomura, Yamamoto et al., J Virol 2012). Adhering to advice, we have attenuated the term to “pathogenic” in the revised manuscript and appended one reference showing pathogenesis gradation from a cell-death perspective (Cumont 2008).

Furthermore, no indication is provided regarding CD4 T cell dynamics, or CD8 T cells. In particular, the extent of T cell immunodeficiency may compromise humoral response. Therefore, this data needs to be shown. Indeed, previous reports have indicated that early CD4 T cell depletion is associated with defective humoral response. Furthermore, Tfh cell depletion was reported in several immune tissues, which are essential for B cell immune response like the spleen. Thus, this should be discussed as an alternative mechanism to the absence of NAb. Indeed, the authors found higher and persistent env-specific plasmablast cells in NAb inducers than that observed in non-NAb inducers figure 6. Why to have selected twelve individuals out of 61 individuals for assessing anti-env response (Supplemental S3 for figure 1, panel 1), and only eleven for western blots. The explanation in the text is absent. This requires to be clearly stated. See lines 108-110.

We appreciate the comment. As in other sections, this study utilized available cryopreserved samples from a retrospective cohort, also having heterogeneity in data acquisition along the way. We acknowledge that some supplemental data are particularly limited in information, which is also a reason they are presented in SI. We felt that one important core was to secure samples for Nef-G63E-selecting NAb inducers versus viremic non-inducers, for which we acquired six versus twelve in the B-cell analysis.

We (Nakane et al, PLoS ONE 2013) and others (Hirsch et al, J Virol 2004) have already reported on western blotting-basis that SIV-infected rapid progressors tend to manifest serological failure (impaired binding Ab-WB bands). Therefore, to compare quantitative traits at this basal stage (Fig 1), we judged that NAb inducer comparison with more non-rapid-progressing (>60 wk survival) non-inducers would be a criterion. We have mentioned on this in the revised manuscript (results/methods). Additionally, we have replaced the immunoblotting result with one more non-inducer (n = 12) to enhance results. Please note that there are lot deviations in strip-coated antigen (e.g., gp160) but the result is comparable (now covers 12/13 of animals with >60-wk survival).

The authors indicated the frequencies of Nef-G63E mutant in figure 2 panel C. However. no information is indicated in the legend about the number of NAb non-inducers used to calculate this frequency. The authors indicated line 127, "only in two of the nineteen NAb non-inducers, including one rapid progressor". Thus, different numbers of individuals are used through the manuscript. For the readers, this is clearly a statement that needs to be clarify and to refer to what. This is not homogeneous along the text and the analyses performed.

We appreciate the comment, and have appended the number in the revised Fig 2C. As aforementioned, heterogeneity of sample number in different sections is indeed a limitation of the work, and have mentioned this in the Discussion.

The rational related to the sentence lines 140-142. Please clarify.. "NAb induction is not associated with these MHC-I genotypes (P = 0.25 by Fisher's exact test, data not shown) but with the Nef-G63E mutation itself".

We appreciate the comment. We have rephrased it as:

“Ten of nineteen NAb non-inducers also had either of these alleles (Figure 1-figure supplement 1). This did not significantly differ with the NAb inducer group (P = 0.25 by Fisher’s exact test, data not shown), indicating that NAb induction was not simply linked with possession of these MHC-I genotypes but instead required furthermore specific selection of the Nef-G63E mutation.” (Lines 159-162).

In supplemental figure 3, only 7 individuals have been tested, while the authors indicated "Ten of nineteen NAb non-inducers also had either of these alleles". Why only seven? In NAb Burmese monkeys, the authors indicate specific T cells capable to recognize WT nef peptide, but not G63E peptide mutant. Thus, nef is immunogenic in vivo generating T cells despite to be mutated.

In contrary, non-NAb-inducers demonstrate the absence of nef specific T cells (supplemental figure 3, excepted R01-011 panel A). Although, the authors propose an escape mutant for CD8 T cells, this is not associated with the absence of immunogenicity and not with a difference in viral load in comparison to NAb inducers (panel C). Therefore, the conclusions merit to be revised. Thus, this part of the manuscript is confusing. Please clarify the rational to link NAb and Nef specific CD8 T cells.

We appreciate the comment. 7 out of 8 non-inducers positive for the allele and not selecting for the Nef-G63E mutant was available for analysis. The relative contribution of this single Nef62-70 epitope-specific CTL response is speculated not to be largely impacting viral control, among the many induced. This is basally discussed in a previous paper (Nomura, Yamamoto et al., J Virol 2012), more suggestive of an MHC-I haplotype-level correlation with plasma viral load. We assume that the CTL pressure-driven selection of Nef-G63E mutant was a rather pure immunosignaling trigger under persistent viremia. We appended this in the revised text (Line 172).

In the next part of the manuscript, the authors assessed the function of this Nef-G63E mutant. The rational to introduce Ferritin in this part of the document is not clear for the reader. Furthermore, a subgroup for each (NAb+ versus NAb-) is shown: 4 for NAbneg versus 6 for NAbpos.

We appreciate the point. As introduced, Swingler et al Cell Host Microbe 2008 reported HIV-infected macrophage-derived ferritin as a potentially B cell-disrupting factor. In that paper, viral load, ferritin and binding antibody titers positively correlated. Current data shows that SIVmac239-specific NAb induction is distinct from such kinetics already versus viral load (Fig 3-Supplement 1C), and ferritin levels were measured for some available samples more simply for confirmation. We appended three more available samples in the NAb- group. (The six NAb+/G63E animals correspond to the ones with B-cell data in Figure 7.) Statistical results appear unaffected and robust, as shown in this version. The revised manuscript incorporates appended explanation for the former.

Similarly, whereas the authors observed a role of nef mutant on pAkt Ser473 (less induced) in comparison to WT, the authors suggest that this may have an impact on T cell survival.

We appreciate the point. In the first submission we obtained peripheral memory Tfh decrease, whereas it is true that this is indirect. In the current revision we have addressed apoptotic cell death, shown to increase with Nef-G63E mutation (Figure 4F).

The rational to analyze CXCR3-CXCR5+PD-1+ memory follicular Th (Tfh) is not clear. Moreover, the references used are not the adequately cited. Indeed, these papers show an expansion. See the literature for a depletion (Xu H, J Immunol. 2015; Moukambi F, PLoS Pathog. 2015; Yamamoto T, Sci Transl Med. 2015; Xu H, J Immunol. 2018 Moukambi F, Mucosal Immunol. 2019).

We appreciate these points on in vivo CD4+ T cells.

Peripheral memory Tfh was reported to correlate with Ab cross-reactivity in one human cohort (Locci et al, Immunity 2013) and we concisely examined the subset in the current NAb induction. We mentioned this in the revised manuscript.

Moukambi F et al, PLoS Pathog 2015 & Mucosal Immunol 2019 are demonstrative work on acute-phase destruction. We have cited non-neonatal/vaccine-related ones suggested, including these two, in the revised manuscript. The biphasic dysregulation of Th (acute-phase destruction and chronic-phase adverse hyper-expansion) may indeed have a unique role with the current phenotype, which is beyond aim of the current analysis. We have concisely mentioned on this in the Discussion.

Then, the authors assess the potential B-cell-intrinsic influence of the G63E-Nef phenotype. The rational here is clearly indicated, making sense with figure 1. Furthermore, this part is clearer. The dot-plots merit to be revised and the markers used better stated. The authors indicate that Nef invasion upregulates pAkt Ser473 assuming aberrant PI3K/mTORC2 signaling. What is the impact of Nef-G63E mutant on pAkt Ser473 using in vitro model of transfer. This is not addressed for comparison.

We appreciate the remarks/suggestions, also pointed out by Reviewers 1 and 2. We have performed three sets of in vitro reconstitution experiments visually and functionally addressing how Nef transfer to B cells can be modulated (new Fig 6), and edited text accordingly.

Minor points are:

- the presence of references in the legend.

-some Ab clones are in the table, however they are not used such CD38 and CD138, which are well known to be non-valid B cell markers for monkeys."

We appreciate the suggestions.

Mentioning on reference have been removed from the legend (Fig.1, Fig. 3) and moved to the corresponding Methods section (Fig. 1).

We also understood this well in advance (CD38/CD138), and incorporated them in the memory B-cell panel just to check whether they ever behave in a specific pattern. As expected, no notable behavior was observed in these NAb inducers.