Potential Associations of Mutations within the HIV-1 Env and Gag Genes Conferring Protease Inhibitor (PI) Drug Resistance
Round 1
Reviewer 1 Report
Ntombikhona Maphumulo and Michelle Gordon propose in this study that potential associations of several mutations within env and gag genes of HIV-1 subtype C clinical strains may be involved in protease inhibitor (LPV/r) resistance. Despite the absence of in vitro and virological experiments, the authors performed useful bioinformatics and in silico analyses of 25 clinical sequences and stored sequences from HIV-1 subtype C patients and observed potential associations of mutations (in Env and Protease proteins) involved in PI resistance.
I have some suggestions to address in order to increase and improve the quality and the usefulness of this paper.
Title : The term “during” in the title appears not to be the most appropriate. It could be worth to change the title (may I propose “Potential associations of mutations within the HIV-1 env and gag genes conferring Protease Inhibitors (PI) drug resistance” or something similar ?).
Abstract: Minor concerns:
- add abbreviation of protease inhibitors (PI) (line 9)
- Change to “suggesting a potential role” (line 10)
- add abbreviation of gag-protease region : gag-PR (line 12)
Throughout the article: there is some typo concerns: the authors should perform an in-depth review of the manuscript because of several double spaces, unnecessary space, space between word and dash, etc.
There are also some confusion with proteins and genes nomenclature (Gag instead of gag, env domain instead of Env domain, and conversely). Clarification of typography of the Proteins and genes should be made throughout the article.
Introduction:
Addition of a paragraph explaining the role of HIV protease proteins in maturation and its cellular localization during viral replication cycle could be very informative.
I think it could be very interesting to add some clinical and virological data (age, gender, VLs, time from infection, CD4 counts, nadir CD4,… ) as well as treatment history (interruption, change of inhibitor, viral escape and blips, …) regarding the 25 patients included in this study. It could be also useful to provide information about potential transmitted PI resistance mutation if it is possible.
Materials and methods:
Minor concerns:
- line 68 : should be “Viral RNAs were extracted…”
- Primers: no brackets, should be 5’-XXXXXXXXXXXX-3’- (line 71 – 91)
- Nested PCR: primers of second PCR appears very long.
- Can the authors provide PCR conditions ? And/or an example of gel electrophoresis to check the bands ?
- NL43∆ ? What is it ? I assume that it is the HIV-1 (subtype B) reference strain NL4-3 but what is the delta for ?
- Concerning the analyses of sequences generated by Sanger sequencing, how the authors manage the double peak (i.e. minority variants) to choose nucleotide identity ? Indeed, it is sometimes a little bit tricky to choose which nucleotide is when double peak is present. It’s just for my information.
Major concerns:
- Sequences analysis: Why did the authors choose the HXB2 reference strain (which is a HIV-1 subtype B strain) ? Would it be possible for the authors to redo a sequence comparison using a HIV-1 subtype C reference strain to check if the results are similar ?
Results:
- Line 138: change “subptye” by subtype
- Figure 1: Is a Fisher’s exact test performed for all positions shown in the graph ? All p<0.05, and never <0.01 or 0.001 ? The authors should clarify this issue and indicate it in the figure legend + asterisks.
- Figure 2: What is the statistical significance for the positive selected sites highlighted in this figure ? The authors should clarify this issue and indicate it in the figure legend + asterisks.
- Line 168: “ha” ?
- An important (“major”) suggestion/issue : Could it be possible for the authors to create – by side-directed mutagenesis – viral particles harboring Env mutation (i.e. the Env T138S alone or in association with PR mutations) by using a HIV-1 subtype C infectious molecular clone (such as the HIV-1 93IN904 or 93IN905, or the HIV-1 Z3576F Transmitted/Founder (T/F) Infectious Molecular Clone from the NIH AIDS reagent program) and perform infections of cultured cells treated or not with LPV/r and check the replicative fitness of the progeny virions in TZM-bl cells ?
Despite the fact that the authors stated it in the abstract, these experiments could be very informative. It could also strengthen and recapitulate the observations presented in this study in order to have an unambiguous and clear conclusion on which mutations (or associations of) in the env and gag genes can be involved in PI resistance.
Conclusion:
Comments on the conclusion will be given based on what the authors will change in the M&M and results sections.
Author Response
Point 1: Title : The term “during” in the title appears not to be the most appropriate. It could be worth to change the title (may I propose “Potential associations of mutations within the HIV-1 env and gag genes conferring Protease Inhibitors (PI) drug resistance” or something similar ?).
Response 1: Thank you, the title has been changed as suggested.
“Potential associations of mutations within the HIV-1 env and gag genes conferring Protease Inhibitors (PI) drug resistance”
Point 2: add abbreviation of protease inhibitors (PI) (line 9)
Response 2: This has been added.
Point 3: Change to “suggesting a potential role” (line 10)
Response 3: This has been changed.
Point 4: add abbreviation of gag-protease region : gag-PR (line 12)
Response 4:This has been added.
Point 5:Throughout the article: there is some typo concerns: the authors should perform an in-depth review of the manuscript because of several double spaces, unnecessary space, space between word and dash, etc.
Response 4: This has been corrected throughout the document.
Point 6: There are also some confusion with proteins and genes nomenclature (Gag instead of gag, env domain instead of Env domain, and conversely). Clarification of typography of the Proteins and genes should be made throughout the article.
Response 5: This has been corrected throughout the document.
Point 7: Addition of a paragraph explaining the role of HIV protease proteins in maturation and its cellular localization during viral replication cycle could be very informative.
Response 7: This paragraph has been added in the introduction (line 30 to 38).
“PIs are known to inhibit the activity of the HIV-1 protease (PR) enzyme responsible for the proteolytic processing of HIV structural Gag and enzymatic Pol polyproteins components(Pettit et al., 1994). The viral protease (PR) enzyme cleaves the Gag precursor protein (Pr55Gag) and the Gag-Pol precursor protein (Pr160Gag-Pol), resulting in virion maturation. (LORI et al., 1988, Rabi et al., 2013, Pettit et al., 1994). This proteolytic process prevents the formation and maturation of infectious HIV particles. The efficacy of the PI is limited by the emergence of resistance mutations that are potentially caused by poor compliance, subtherapeutic systemic levels of the drug or, prolonged treatment with one PI-based regimen during virologic failure (Zdanowicz, 2006).”
Point 8: I think it could be very interesting to add some clinical and virological data (age, gender, VLs, time from infection, CD4 counts, nadir CD4,… ) as well as treatment history (interruption, change of inhibitor, viral escape and blips, …) regarding the 25 patients included in this study. It could be also useful to provide information about potential transmitted PI resistance mutation if it is possible.
Response 8: This information has been added in the results section 3.1( line 166 - 171).
“3.1. Participant characteristics
Twenty-four Gag-Prot-Env sequences from 24 patients failing a PI inclusive treatment regimen were available for analysis, of whom 14 were females and 10 were males with the median age of 35 years (interquartile range (IQR) 17 – 38 years). The median viral load was 4.84 log10 copies/ml (IQR 4.12 – 5.51). All patients were infected with HIV subtype C as established by the Virulign tool.”
Point 9: line 68 : should be “Viral RNAs were extracted…”
Response 9: This has been changed (line 85).
Point 10: Primers: no brackets, should be 5’-XXXXXXXXXXXX-3’- (line 71 – 91)
Response 10: Brackets has been removed as suggest(line 88– 115).
Point 11: Nested PCR: primers of second PCR appears very long.
Response 11: The 2nd round primers are long so that they cover enough of the region preceding the gag-PR gene to facilitate cloning.
Point 12: Can the authors provide PCR conditions ? And/or an example of gel electrophoresis to check the bands ?
Response 12: PCR conditions have been added in the methodology and the gel pictures are in the supplementary.
Gag-PR conditions
first round: at 55ºC for 30 minutes (cDNA synthesis) and 94ºC for 2 minutes (initial denaturation), followed by 35 cycles of 94ºC for 15 seconds (denaturation), 55ºC for 30 seconds (annealing) and 68ºC for 2 minutes (extension), and ended with a 5 minute incubation at 68ºC (final extension),( line 91 – 95).
Second round: 94ºC for 2 minutes (initial denaturation), 40 cycles of 94ºC for 30 seconds (denaturation), 60ºC for 30 seconds (annealing) and 72ºC for 2 minutes (extension) followed by a 7 minute hold at 72ºC (final extension),(line 100 – 103).
Envelope conditiond: 94ºC for 2 minutes (initial denaturation), 35 cycles of 94ºC for 15 seconds (denaturation), 55ºC for 30 seconds (annealing) and 68ºC for 4 minutes (extension) followed by a 10 minute hold at 68ºC (final extension), (line 115 - 121.
Supplementary figure 1. : Gel picture, showing the amplification of Gag-PR. Gag-PR size 1.7 kb (line 326).
Supplementary figure 2.: Gel picture showing the amplification of envelope. Envelope size 3.5 kb (line 328).
Point 13: NL43∆ ? What is it ? I assume that it is the HIV-1 (subtype B) reference strain NL4-3 but what is the delta for ?
Response 13: Thank you for the comment. You are correct that this is an HIV-1 subtype B reference strain. This particular plasmid has a deleted gag-pro region, hence the delta sign. The following has been added to the text for clarity (line 104).
Point 14: Concerning the analyses of sequences generated by Sanger sequencing, how the authors manage the double peak (i.e. minority variants) to choose nucleotide identity ? Indeed, it is sometimes a little bit tricky to choose which nucleotide is when double peak is present. It’s just for my information.
Response 14:Where true double peaks were seen (i.e. minority variants), the major peak was called, unless it occurred at a known drug resistance associated site.
Point 15: Sequences analysis: Why did the authors choose the HXB2 reference strain (which is a HIV-1 subtype B strain) ? Would it be possible for the authors to redo a sequence comparison using a HIV-1 subtype C reference strain to check if the results are similar ?
Response 15: Sequences were analysed using the consensus subtype C sequence downloaded from the Los Alamos database. The reference to HXB2 was in error. All wildtype amino acids are written according to the subtype C consensus. For example, at codon 19, the wildtype in subtype C is an I and in HXB2 the wildtype is a T.
Point 16: Line 138: change “subptye” by subtype
Response 16: This has been changed (line 176).
Point 17: Figure 1: Is a Fisher’s exact test performed for all positions shown in the graph ? All p<0.05, and never <0.01 or 0.001 ? The authors should clarify this issue and indicate it in the figure legend + asterisks.
Response 17: Yes, the fisher’s exact test was performed in all position shown in figure 1.
The p values were clarified in the figure and figure legends as follows: * = p<0.05, ** = p<0.01 and *** = p<0.0001 (line 194 – 195).
Point 18: Figure 2: What is the statistical significance for the positive selected sites highlighted in this figure ? The authors should clarify this issue and indicate it in the figure legend + asterisks.
Response 18: PAML uses Bayesian probability (* p>95%; **p>99%). listed in the figure are all significant with the **p>99%. Also included in the figure legend.
Point 19: Line 168: “ha” ?
Response 19: This typo has been corrected (line 208).
Point 20: An important (“major”) suggestion/issue : Could it be possible for the authors to create – by side-directed mutagenesis – viral particles harboring Env mutation (i.e. the Env T138S alone or in association with PR mutations) by using a HIV-1 subtype C infectious molecular clone (such as the HIV-1 93IN904 or 93IN905, or the HIV-1 Z3576F Transmitted/Founder (T/F) Infectious Molecular Clone from the NIH AIDS reagent program) and perform infections of cultured cells treated or not with LPV/r and check the replicative fitness of the progeny virions in TZM-bl cells ?
Response 20: Thank you for the suggestion. As this article reports data from our exploratory computational study, the intention is to use the information gained from this analysis to guide futher laboratory-based experiments to confirm these findings. As you have mentioned, using a subtype C backbone may provide more subtype C specific information, although several studies have shown that there is no statistical difference between data obtained using a subtype B or subtype C backbone (de Almeida et al 2016).
Point 21: Despite the fact that the authors stated it in the abstract, these experiments could be very informative. It could also strengthen and recapitulate the observations presented in this study in order to have an unambiguous and clear conclusion on which mutations (or associations of) in the env and gag genes can be involved in PI resistance.
Response 21: Thank you for the suggestion. The information gained from this analysis will be used for laboratory-based experiments to confirm these findings.
Author Response File: 
Author Response.docx
Reviewer 2 Report
The manuscript by Maphumulo and Gordon analyzed the Gag, Env and protease sequences from 24 clade C individuals experiencing virological failure on Protease Inhibitor inclusive ART regimen. Using computational and co-evolution analysis, the authors identified clusters of Env-gag and protease residues that may be responsible for exerting the resistance towards the treatment regimen. The study is straightforward, and the discussion of the results is informative. The manuscript is very well written and is well presented. Adding some experimental evidence to prove that the identified residue pairs in different regions are indeed associated with resistance towards Lopinavir/Ritonavir would have strengthened the manuscript. Nonetheless, the study adds to the information relevant to retroviral treatments and their failure in select individuals.
Author Response
Point 1: The manuscript by Maphumulo and Gordon analyzed the Gag, Env and protease sequences from 24 clade C individuals experiencing virological failure on Protease Inhibitor inclusive ART regimen. Using computational and co-evolution analysis, the authors identified clusters of Env-gag and protease residues that may be responsible for exerting the resistance towards the treatment regimen. The study is straightforward, and the discussion of the results is informative. The manuscript is very well written and is well presented. Adding some experimental evidence to prove that the identified residue pairs in different regions are indeed associated with resistance towards Lopinavir/Ritonavir would have strengthened the manuscript. Nonetheless, the study adds to the information relevant to retroviral treatments and their failure in select individuals.
Response 1: Thank you for the suggestion. As this article reports data from our exploratory computational study, the intention is to use the information gained from this analysis to guide further laboratory-based experiments to confirm these findings.
Author Response File: Author Response.docx
Reviewer 3 Report
In this study, the authors examine the potential co-evolution of mutations in Env, Gag, and protease during virological failure to protease inhibitor (PI) therapy. They report an association between mutations in Gag, Env and protease during the development of resistance.
While the topic of mutations outside of protease that are responsible for resistance to PIs is of interest, the current study adds little in the way of useful information.
- No experiments are performed to actually show that the mutations in Env or Gag contribute to PI resistance. The paper is based on modeling and speculation.
- The authors have missed important work in the literature showing that mutations in Env can confer broad antiretroviral resistance by enhancing virus replication capacity (Van Duyne et al., PNAS 2019; Hikichi mBio 2021). This phenomenon may explain the results.
- The authors speculate in the Discussion about cathepsin cleavage of gp120 and CD4-independent infection. There are no data to support the notion that this is relevant to the observations reported here. Mutations in variable loops arise readily in patients in response to the host immune response (this is why they are denoted variable loops).
- The paper contains a number of typos that need to be corrected.
Author Response
Point 1: No experiments are performed to actually show that the mutations in Env or Gag contribute to PI resistance. The paper is based on modeling and speculation.
Response 1: Thank you for your response. You are correct that this article reports on data derived from computational modelling. It has been shown that these data provide valuable information that assist with designing future laboratory-based experiments.
Point 2: The authors have missed important work in the literature showing that mutations in Env can confer broad antiretroviral resistance by enhancing virus replication capacity (Van Duyne et al., PNAS 2019; Hikichi mBio 2021). This phenomenon may explain the results.
Response 2: Information from the two mentioned papers has been added to the introduction (line 59 - 64).
“Furthuremore, Env mutations have been suggested to promote cell to cell transmission leading to high-level drug resistance mutations in ARV target genes, by doing so, they increase the level of resistance to a broad panel of ARV’s in vitro (Van Duyne et al., 2019, Hikichi et al., 2021). HIV-1 Envelope is said to concurrently evolve to escape from both neutralizing antibodies (NABs) and ARVs(Hikichi et al., 2021).”
Point 3: The authors speculate in the Discussion about cathepsin cleavage of gp120 and CD4-independent infection. There are no data to support the notion that this is relevant to the observations reported here. Mutations in variable loops arise readily in patients in response to the host immune response (this is why they are denoted variable loops).
Response 3: This has been removed from the discussion.
Point 4:The paper contains a number of typos that need to be corrected.
Response 4: All typos have been corrected throughout the text.
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
The authors responded to most of the suggestions and comments from the first round of reviewing.
Although the authors do not provide experimental data concerning the presence of PI resistance mutations (or associations of mutations) in the env and gag-PR genes using an infectious molecular clone, the data presented in the article could serve as a basis for further studies (performed by these authors or other authors).
Therefore, this article may pave the way for further studies regarding PI resistance mutations, and this article may therefore be cited a few times in the months or years to come.
This article therefore deserves publication.
Reviewer 3 Report
the authors have addressed my comments
