|Year : 2019 | Volume
| Issue : 4 | Page : 574-583
Interaction of human immunodeficiency virus-1 and human immunodeficiency virus-2 capsid amino acid variants with human tripartite motif 5α protein SPRY domain and its association with pathogenesis
Veena Vadhini Ramalingam1, Suganya Subramanian2, G John Fletcher1, Priscilla Rupali3, George Varghese3, Susanne Pulimood4, Lakshmanan Jeyaseelan5, Balaji Nandagopal2, Gopalan Sridharan2, Rajesh Kannangai1
1 Department of Clinical Virology, Christian Medical College, Vellore, Tamil Nadu, India
2 Sri Narayani Hospital and Research Centre, Sri Sakthi Amma Institute of Biomedical Research, Vellore, Tamil Nadu, India
3 Department of Infectious Diseases, Christian Medical College, Vellore, Tamil Nadu, India
4 Department of Dermatology, Christian Medical College, Vellore, Tamil Nadu, India
5 Biostatistics, Christian Medical College, Vellore, Tamil Nadu, India
|Date of Submission||17-Mar-2020|
|Date of Acceptance||30-Mar-2020|
|Date of Web Publication||18-May-2020|
Dr. Rajesh Kannangai
Department of Clinical Virology, Christian Medical College, Vellore - 632 004, Tamil Nadu
Source of Support: None, Conflict of Interest: None
Purpose: The sequence variation of human immunodeficiency virus (HIV) capsid region may influence and alter the susceptibility to human tripartite motif 5α protein (huTRIM5α). Materials and Methods: Molecular docking was carried out with huTRIM5α SPRY domain by the use of ClusPro and Hex docking program for HIV-1 and HIV-2 capsid sequences. Results: The sequence analysis on HIV-1 and HIV-2 capsid gag gene identified 35 (19.7%) single-nucleotide polymorphisms (SNPs) in HIV-1 and 8 (4.5%) SNPs in HIV-2. The variations observed in the HIV-2 capsid region were significantly lower than HIV-1 (P < 0.001). The molecular docking analysis showed that HIV-1 wild type used V1 loop, while HIV-2 used V3 loop of huTRIM5α for interaction. HIV-1 with A116T SNP and HIV-2 with V81A SNP use V3 and V1 loop of huTRIM5α for interaction respectively. The reduced huTRIM5α inhibition may lead to a faster progression of disease among HIV-1-infected individuals. However, in case of HIV-2, increased inhibition by huTRIM5α slows down the disease progression. Conclusion: Polymorphisms in the capsid protein with both HIV-1- and HIV-2-monoinfected individuals showed the difference in the docking energy from the wild type. This is the first study which documents the difference in the usage of loop between the two HIV types for interaction with huTRIM5α. Variations in the capsid protein result in alteration in the binding to the restriction factor huTRIM5α.
Keywords: Docking, human immunodeficiency virus-1, human immunodeficiency virus-2, human tripartite motif 5α protein, molecular modelling
|How to cite this article:|
Ramalingam VV, Subramanian S, Fletcher G J, Rupali P, Varghese G, Pulimood S, Jeyaseelan L, Nandagopal B, Sridharan G, Kannangai R. Interaction of human immunodeficiency virus-1 and human immunodeficiency virus-2 capsid amino acid variants with human tripartite motif 5α protein SPRY domain and its association with pathogenesis. Indian J Med Microbiol 2019;37:574-83
|How to cite this URL:|
Ramalingam VV, Subramanian S, Fletcher G J, Rupali P, Varghese G, Pulimood S, Jeyaseelan L, Nandagopal B, Sridharan G, Kannangai R. Interaction of human immunodeficiency virus-1 and human immunodeficiency virus-2 capsid amino acid variants with human tripartite motif 5α protein SPRY domain and its association with pathogenesis. Indian J Med Microbiol [serial online] 2019 [cited 2020 Jun 2];37:574-83. Available from: http://www.ijmm.org/text.asp?2019/37/4/574/284521
| ~ Introduction|| |
Human immunodeficiency virus (HIV) infection continues to be a significant problem in both developed and developing countries. There are approximately 36.9 million people who are infected with HIV. Compared to HIV-1, HIV-2 was shown to be less infective based on the clinical and immunological parameters of patients infected with HIV-2. However, the underlying differences in the mechanism of pathogenesis between HIV-1 and HIV-2 are not well outlined. It may be either the viral or the host factors of the people harbouring the virus. There are many cellular antiviral restriction factors that hinder HIV replication including apolipoprotein B editing catalytic polypeptide, human tripartite motif 5α protein (huTRIM5α), tetherin, SAM and HD domain-containing deoxynucleoside triphosphate triphosphohydrolase 1, Moloney leukaemia virus 10 and cellular miRNAs. One of the host factors concerned in early post-entry steps of HIV replication is huTRIM5α. It functions as an antiretroviral restriction factor that blocks HIV replication within the cytoplasm of the host cell by targeting the viral capsid and interferes with the uncoating process. The SPRY domain of huTRIM5α mediates post-entry restriction by recognising and binding to the retroviral capsid. The specificity with which each retrovirus interacts with huTRIM5α is determined by the sequence variation within the PRY-SPRY domain of huTRIM5α.
Compared to HIV-2, HIV-1 is moderately restricted by the antiviral action of huTRIM5α. However, there is a scarcity of data on the susceptibility of clinical isolates of HIV-2 subtype A strains to huTRIM5α or another capsid (CA)-targeting restriction factors as reflected by an amino acid variation within the HIV-2 capsid region. Fewer studies on the susceptibility of HIV-2 to huTRIM5α are limited solely on HIV-2 ROD, a standard laboratory strain of HIV-2 isolate, whereas using only the bioinformatic tool with none analysis of clinical isolates. Recently, one study has proposed the difference in susceptibility to huTRIM5α between strains, or the clinical progression of HIV-2 disease varies with specific polymorphisms within the capsid sequence of HIV-2. The aim of our study was to look at the variations within the capsid protein of HIV-1 subtype C and HIV-2 subtype A and its interactions with the SPRY domain of huTRIM5α. We looked at the variations that may alter the interaction between these proteins by molecular modelling and any association of capsid protein variations with the clinical presentation of infected individuals during this cross-sectional study.
| ~ Materials and Methods|| |
This cross-sectional study was done during the period of January 2014 to October 2018. The study participants were recruited from HIV-1/HIV-2-infected individuals who attended the Infectious Diseases Department and referred to the Clinical Virology Department of a tertiary care hospital in South India. Individuals who match the inclusion criteria were included in the present study after informed consent. HIV-1- or HIV-2-monoinfected individuals who were >18 years of age and treatment naive were included in the study. Individuals below 18 years of age and who were found to be positive for anti-hepatitis C virus and hepatitis B surface antigen and individuals with dual (HIV-1 and HIV-2) infection were excluded from the study. This study was approved by the Institutional Review Board (IRB Min. No. 8798, 19 March 2014).
Amplification and sequencing of human immunodeficiency virus 1/2 capsid genome
RNA was extracted using the QIAamp Viral RNA Mini Kit (Qiagen, Hilden, Germany) from the plasma samples. The extracts were amplified and sequenced for the HIV-1 and HIV-2 capsid gene by a nested polymerase chain reaction (PCR) using Qiagen one-step reverse transcription-PCR kit (Qiagen, Hilden, Germany). This combines both conversions of RNA to cDNA and amplification of the target region using specific primers in the first round of PCR, then followed by the second round of nested PCR. The primers are listed in [Table 1]. The cycling conditions were used for the amplification of HIV-1 gag and HIV-2 capsid gene was the one regularly used for the amplification of the pol gene of HIV-1 genotypic assay as described earlier. The amplified PCR products were sequenced using the ABI 310 HITACHI Genetic Analyser (Applied Biosystems, Foster City, CA) by Sanger's sequencing method at the Department of Clinical Virology.
|Table 1: Different set of primers used in this study to amplify the different viral and host gene targets|
Click here to view
PCR assay was performed with noted positive and negative controls. The instruments used for sequencing the study samples were calibrated, and routine clinical samples were done in the instrument for assays such as HIV-1 genotypic resistance testing with appropriate split and audit sample testing. The HIV-1 and HIV-2 capsid sequences were examined to rule out cross-contamination throughout sample processing or PCR. The phylogenetic tree was constructed using Mega 6 software for the analysis of subtypes.
Human immunodeficiency virus-1 and human immunodeficiency virus-2 capsid variants in the study groups
The sequence variation was assessed in the capsid protein (p24/p26) gene (CA) obtained from HIV-1 (n = 59) and HIV-2 (n = 14) monoinfected individuals. The primers used for amplification of the HIV-1 and HIV-2 capsid gene are shown in [Table 1]. Primers for amplification of the HIV-1 capsid gene were designed from the HIV-1 subtype C consensus which was retrieved from the Los Alamos database. However, for HIV-2 capsid gene amplification, primers were used from an earlier published article. To identify the variations in the capsid region, the translated amino acid sequences of HIV-1 (p24) and HIV-2 (p26) identified from these individuals (HIV-1 [n = 59] and HIV-2 [n = 14]) were aligned. This was compared with the reference strain BAA85225.1 (GenBank accession number) for HIV-1 and ADI44723.1 (GenBank accession number) for HIV-2 using BioEdit software (https://softfamous.com/bioedit/). The phylogenetic tree for HIV-1 and HIV-2 to determine the subtype was constructed using Mega 6 software.
Amplification and sequencing of human tripartite motif 5α protein (SPRY domain)
The extracted DNA using QIAamp DNA Blood Mini Kit (Qiagen, Hilden, Germany) from peripheral blood mononuclear cells (PBMCs) was amplified for huTRIM5α gene using 5 μL of DNA input to a PCR mix containing 2.5 units of Hot start Taq master mix (Qiagen, Hilden, Germany) and 20 picomoles of forward and reverse primers [Table 1] used from previous literature,, in a total reaction volume of 50 μL. The PCR was standardised and also the following cycling conditions were used, 95°C for 15 min, followed by 94°C for 1 min, 57°C for 1 min and 72°C for 1 min for 40 cycles with final extension cycle at 72°C for 7 min. The amplified PCR products were sequenced by Sanger's sequencing technique.
Sequence retrieval and 3D protein modelling
Crystallised reference protein structures of HIV-1 capsid, HIV-2 capsid and huTRIM5α SPRY domain were constructed using computational modelling based on homology prediction. Amino acid sequences of HIV-1 capsid subtype C (GenBank accession number: BAA85225.1), HIV-2 capsid subtype A (GenBank accession number: ADI44723.1) and huTRIM5α SPRY domain (Uniprot: Q9C035) were obtained. Reference sequences for both HIV-1 and HIV-2 were selected based on blast results which showed more than 90% similarity with our strains. These sequences as input were used to model 3D protein structures using I-TASSER online web server (http://zhanglab.ccmb.med.umich.edu/I-TASSER). Among the five models generated by the I-TASSER program, the model that had the very best confidence score and root mean square deviation (RMSD) score was selected for further analysis. Each variant model was generated using the 'mutation tool' in the Swiss-PdbViewer.
Superimposition of reference variant proteins and root mean square deviation calculation
With the use of the 'mutation tool' in the Swiss-PdbViewer, the reference strain amino acid was replaced by the 'best' rotamer of the variant amino acid. Energy minimisation for the predicted models (reference and variant strains) was performed with the GROMOS 43B1 field implementation of the DeepView v4.1 tool (https://spdbv.vital-it.ch/energy_tut.html). This force field was built to evaluate the energy of a protein structure as well as repair distorted geometries through energy minimisation.
The extent of structural deviation between the reference strain and the variant protein structures associated with a functional effect on the protein was predicted by calculating the RMSD by superimposing the reference and variant protein structures. The RMSD values of the atoms upon superimposing the reference strain and the variant protein structure were calculated using the 'Calculate RMS' function in the Swiss-PdbViewer program. If the RMSD values are higher, the structural deviations are more likely to be associated with the altered function of the protein.
Structure validation using Procheck server
In order to ascertain the structural and functional properties of the protein structure, Ramachandran plot was used. The program calculates the dihedral angles of the amino acid residues and predicts the energetically allowed residues. The 3D protein structure models of reference and the variant protein were energy minimised and validated using Procheck online server program (https://servicesn.mbi.ucla.edu/PROCHECK/).,
Molecular docking of the reference and variant strain of HIV-1 and HIV-2 with huTRIM5α structures was performed individually using the ClusPro web server (https://cluspro.bu.edu/publications.php). The binding affinity of both reference and variant strain complexes was assessed based on the most populated pose clusters and lowest binding energy scores. Hex 6.12 server (http://hex.loria.fr/dist612/) is the first Fourier transform (FFT)-based protein docking server to be powered by graphic processors. The reference and variant strains of HIV-1 and HIV-2 with huTRIM5α structures were docked and obtained energy total (e-total) values for all docked complex. PyMol (https://pymol.org/2/) software was used for visual representation and assessing the complex interaction and loop conformation between HIV-1 capsid and HIV-2 capsid with huTRIM5α SPRY domain.
CD 4+ T-cell enumeration
The CD4 + T-cell enumeration was performed using flow cytometry (BD FACSCount CD4 Reagents, San Jose, USA). CD4+ T-cell enumeration was done on samples from HIV-1 and HIV-2 treatment-naive monoinfected individuals during the regular follow-up routine testing.
Quantification of human immunodeficiency virus-1 and human immunodeficiency virus-2 plasma RNA load
HIV-1 RNA viral load quantitation was performed on HIV-1-monoinfected individuals from plasma samples by TaqMan-based real-time PCR using Abbott real-time PCR (Abbott Molecular Inc., Des Plaines, IL, USA). Nucleic acid extraction was done using 0.6 ml of plasma by the Abbott m2000sp sample preparation system using magnetic particle technology and amplification/detection by m2000rt instrument, with a lower limit of detection (LOD) as <40 copies/mL. HIV-2 plasma viral loads in plasma samples were determined using the ExaVir™ Load version 3 assay (HIV RT version 3). The ExaVir assay (Cavidi, Sweden) measures the viral RT enzymatic activity in an ELISA format, and the conversion to RNA copies/ml is by software (ExaViral load analyser). The lower LOD of the assay was <200 copies/ml.
Quantification of human immunodeficiency virus-1 and human immunodeficiency virus-2 proviral DNA load
DNA from PBMCs was extracted using the commercially available QIAamp DNA blood Mini Assay (Qiagen, Hilden, Germany). The eluted DNA was used for the quantification of HIV-1 and HIV-2 proviral DNA. The primers used for amplification are listed in [Table 1] used from previous literature. The cycling conditions were as follows: 95°C for 10 min, followed by 95°C for 30 s and 57°C for 30 s for 50 cycles. The lower limit of detection (LOD) of the HIV-1 and HIV-2 proviral DNA quantitation assay was 30 and 70 copies/reaction, respectively.
Divergence rates and selection pressure of human immunodeficiency virus-1 and human immunodeficiency virus-2
We further looked at the genomic sites under positive and negative selection in HIV-1 and HIV-2 capsid domain by comparison of synonymous (dS, no amino acid change) and non-synonymous (dN, amino acid change) substitution rates using Mega 6 software, and the Kimura test was used for calculating the selection pressure.
Human immunodeficiency virus disease stratification of subjects
The WHO clinical guideline was used to stage HIV disease into Stage 1, 2, 3 and 4 based on the clinical details of the study participants.
Fisher's exact test was done to look at the significance of each variation with the viral load and on susceptibility to huTRIM5α and clinical presentation. The P value for the synonymous and non-synonymous polymorphisms in the HIV-1 and HIV-2 immunodominant genes was calculated using an independent t-test. The P < 0.05 is taken as significant. All the statistical analysis was carried out using MedCalc (version 14.8.1, Belgium).
| ~ Results|| |
List of amino acid variations identified in the human immunodeficiency virus-1 capsid gene (p24)
In total, there were 35 (19.7%) variations in the capsid region of HIV-1-infected study participants, when compared with the reference strain (BAA85225.1, HIV-1 subtype C). Among these, H87Q (30.51%) was frequently identified in the individuals infected with HIV-1 and the least frequently detected variations are M96 L and I91P (1.69%).
List of amino acid variations identified in the human immunodeficiency virus-2 capsid gene (p26)
Among 14 HIV-2 samples sequenced, V81A (50%) followed by P119A (42.86%), P159S (28.57%) and V190I (42.86%) is more frequently identified. The other variations identified in the capsid region of HIV-2 individuals are as follows: I152V (35.71%), V81T (21.43%), P119Q (14.29%) and S162N (14.29%). In total, there were 8 (4.5%) variations identified among HIV-2-infected individuals when compared with the reference strain (ADI44723.1, HIV-2 subtype A).
All the HIV-1 and HIV-2 samples subjected for subtyping were identified as subtype C and subtype A, respectively, based on the phylogenetic analysis [Figure 1] and [Figure 2].
|Figure 1: Phylogenetic tree generated from the human immunodeficiency virus gag region sequences from human immunodeficiency virus-1 (n = 59) monoinfected individuals. Phylogenetic tree (circular view) to determine the human immunodeficiency virus-1 subtypes of the study of clinical samples. Reference sequences from the Gen Bank database are identified with accessions numbers for all human immunodeficiency virus -1 subtypes (A-C, F-H and J). Study sample sequences are shown with spots (•)|
Click here to view
|Figure 2: Phylogenetic tree generated from the human immunodeficiency virus gag region sequences from human immunodeficiency virus-2-monoinfected individuals (n = 14). Phylogenetic tree (circular view) to determine the human immunodeficiency virus-2 subtypes of the study of clinical samples. Reference sequences from the GenBank database are identified with accessions numbers for Human immunodeficiency virus-2 subtypes (A, B and CRF). Study sample sequences are shown with spots (•)|
Click here to view
Identification of human tripartite motif 5α protein SPRY domain variant
To define potential huTRIM5α variations in the SPRY domain, we amplified DNA from a total of 73 (59 HIV-1 and 14 HIV-2) monoinfected individuals. All the samples were successfully sequenced. The obtained sequences were first analysed using the finch TV software to check for the quality of the electropherogram. None of the sequenced samples showed variations in the SPRY domain compared to the reference strain (Uniprot no: Q9C035).,
Upon validation using PROCHECK server, the capsid protein model of HIV-1 and HIV-2 had 84% and 88% of residues, respectively, in the allowed region indicating a good prediction of protein structures. Out of 35 variants identified in the capsid protein of HIV-1, ten variants showed changes in the number of residues in the favoured and allowed region compared to the reference strain. Among 8 variants identified in HIV-2, two variant models (P119Q and P119A) showed a minor change in the number of residues in the favoured and allowed region compared to the reference strain.
The HIV-1 and HIV-2 sequences had 66.5% identical amino acids (https://embnet.vital-it.ch/software/LALIGN_form.html). The superimposed structure is shown in [Figure 3]. On energy minimisation and superimposition of the reference strain protein with each of the variant proteins, the total energy of HIV-1 and HIV-2 reference strain capsid protein showed −11217.2 KJ/mol and −13074.343 KJ/mol, respectively. The difference in the RMSD value for each of the variant proteins compared to reference protein was 0.00Š, indicating that there was no deviation in HIV-1 and HIV-2 capsid protein structures. Among 35 HIV-1 amino acid variants, A92P, Q120A, I91 L and E79D showed an increase in protein stability. However, all eight variants in HIV-2 were shown to have decreased protein stability.
|Figure 3: Superimposition of human immunodeficiency virus-1 and human immunodeficiency virus-2 gag capsid (CA) 3D structures to see the structural identity. Superimposed structure of human immunodeficiency virus-1 (green) and human immunodeficiency virus-2 (yellow). Superimposition structure was generated through PyMol (http://www.pymol.org/). Out of 178 amino acids aligned, 107 amino acids are identical between human immunodeficiency virus-1 and human immunodeficiency virus-2 showing similarity of 60%|
Click here to view
Loop conformation analysis of human tripartite motif 5α protein by ClusPro
The different loops of huTRIM5α SPRY domain involved in the interaction with HIV capsid protein are shown in [Figure 4]. HIV-1 reference strain used nine different amino acids interacting in 18 different positions with huTRIM5α. HIV-2 reference strain used 12 different amino acids interacting in 16 different positions with huTRIM5α. However, no difference in the length of hydrogen bonds was observed for both HIV-1 and HIV-2 capsid protein while interacting with huTRIM5α.
|Figure 4: An image showing the different loops of human tripartite motif 5α protein involved in the interaction of human immunodeficiency virus capsid protein. Structure of human tripartite motif 5α protein showing different loop formations. V1 loop is shown in yellow colour, V2 loop in magenta colour and V3 loop in cyan colour|
Click here to view
In ClusPro analysis, both HIV-1 and HIV-2 reference and variant strains were docked with huTRIM5α SPRY protein. HIV-1 reference strain used V1 loop and HIV-2 reference strain used V3 loop of huTRIM5α SPRY domain for interaction. The amino acid residues involved in the interaction with huTRIM5α also varied between HIV-1 and HIV-2 reference strains. When the variants of HIV-1 were docked with huTRIM5α, we found HIV-1 strains with A116T variation used V3 loop of huTRIM5α for interaction, while the other HIV-1 capsid variants used V1 loop similar to the reference strain. The difference in the interaction between HIV-1 reference and variant strain (A116T) is shown in [Figure 5]a. The image shows the difference in the usage of the huTRIM5α loop by HIV-1 capsid reference and variant strain. Similarly, the HIV-2 strain with the V81A variant used the V1 loop instead of the V3 loop for interaction with huTRIM5α. The difference in the interaction between HIV-2 reference and variant strain (V81A) is shown in [Figure 5]b.
|Figure 5: A representative image showing the molecular interaction of human immunodeficiency virus-1 (a) and human immunodeficiency virus-2 (b) capsid reference and variant with human tripartite motif 5α protein SPRY domain reference type. The receptor (human immunodeficiency virus-1 and human immunodeficiency virus-2 reference and variant type) is shown in green colour and the ligand (human tripartite motif 5α protein SPRY) is represented in cyan colour. The interacting amino acid residues of human immunodeficiency virus 1 and 2 reference and variant type are shown in red colour and amino acid residues of human tripartite motif 5α protein SPRY domain are shown in yellow colour|
Click here to view
Comparison of docking energy using Hex server program
The reference and variant strain of HIV-1 and HIV-2 protein models were docked with huTRIM5α, and the docking energy was compared using the Hex server program. The variations which showed a difference in the Hex-docking energy from the reference strain among both HIV-1 and HIV-2 are shown in [Figure 6]a and 6b. The binding energy of the HIV-1 reference protein structure was − 743 KJ/mol. Among 35 variants, the variant D187E was found to have the lowest negative binding energy of −719.1 KJ/mol (23.9% energy difference). Twenty-four variants of HIV-1 showed higher docking energy ranging from −743.2 to −779.5 compare to reference strain. Variant N183G had the highest binding energy of −779.5 (36.5% energy difference). Eleven variants had negative lower energy values ranging from −742.8 to −725.4 KJ/mol.
|Figure 6: The list of variations in the capsid protein among (a) human immunodeficiency virus-1 (n = 59) and (b) human immunodeficiency virus-2 (n = 14) monoinfected individuals showing the difference in the docking energy from the reference type|
Click here to view
The binding energy of the reference protein structure of HIV-2 was determined to be −728 KJ/mol. Among the 8 variants identified, P159, P119A and I152V found to have the highest binding energy of −735.1KJ/mol compared with the reference structure (7% energy difference). The other five variants such as V190I, P119Q, S162N, V81T and V81A were found to have lower energy values ranging from −701.6 to −727.4 KJ/mol, and energy difference ranged from 0.6% to 26.4%.
Impact of amino acid variations in the human immunodeficiency virus-1 capsid gene to human tripartite motif 5α protein susceptibility and pathogenesis
A few variations identified in the capsid gene of HIV-1 observed to have a major difference in the binding energy but did not correlate with the higher plasma viral load. N183G single-nucleotide polymorphism (SNP) with the highest docking energy was found in participants with both high and low viral loads. The four SNPs (H87P, G120A, A92P and I91 L) showed an increase in protein stability with the highest docking energy but no difference in susceptibility with huTRIM5α. The usage of the huTRIM5α loop for interaction with HIV-1 capsid is slightly different between HIV-1 and HIV-2 and also between reference and variant strain. In our docking analysis, the HIV-1 reference strain used the V1 loop of huTRIM5α for interaction except for strains with A116T which used the V3 loop. The median viral load of individuals with and without the A116T variant was 5.08 and 4.53 log copies/ml, respectively, with no significant difference (P = 0.45) in the susceptibility to huTRIM5α. In addition, every variation in the HIV-1 capsid was compared with the HIV-1 plasma viral load, proviral DNA, CD4+ T-cell count and WHO staging but did not show any significant effect on the susceptibility to huTRIM5α and also with the pathogenesis of HIV-1 infection in terms of HIV disease status.
Impact of amino acid variations in the human immunodeficiency virus-2 capsid gene to human tripartite motif 5α protein susceptibility and pathogenesis
Among HIV-2-infected individuals, the variation identified in the HIV-2 capsid region was observed to have a significant role in determining the level of HIV-2 restriction by huTRIM5α. Out of eight variations identified, variations at position 81, 119, 152 and 159 were of special interest since the difference in the binding energy at those positions had an impact on the HIV-2 plasma viral load. Among the variations identified, P159S, P119A and I152V have the highest docking energy, whereas V81A/T and P119Q have the lowest docking energy. The lowest binding energy showed less interaction between HIV-2 capsid and huTRIM5α. HIV-2 reference strain uses the V3 loop of huTRIM5α for interaction, while variant strain with V81A uses the V1 loop for interaction and had the lowest docking energy; this showed a difference in susceptibility to huTRIM5α. There was no significant association observed with CD4+ T-cell count, proviral DNA status and WHO staging [Table 2]. Among the HIV-2-infected individuals who had V81A/T variation, 66.7% had presented with WHO clinical stage 2 or higher. The difference was statistically insignificant (P = 0.13).
|Table 2: Impact of HIV-2 capsid polymorphism with immunological, biological markers and WHO staging|
Click here to view
None of the variants showed a significant association with the viral and immunological markers except V81A/T in which the median viral load of patients with and without V81A/T variation was 4.22 log copies/ml and 2.91 log copies/ml, respectively (P = 0.08). This is shown in [Table 3].
|Table 3: Comparison of HIV-2 mutations with the CD4+ T-cell count and plasma viral load as measured by reverse transcription activity and WHO clinical stage|
Click here to view
Divergence rates and selection pressure of human immunodeficiency virus-1 and human immunodeficiency virus-2 capsid region
Among the HIV-1- and HIV-2-monoinfected individuals, synonymous (dS) and non-synonymous substitutions (dN) observed in HIV capsid region were as follows: dS (6.15 and 0.09, P < 0.001) and dN (0.84 and 0.12, P < 0.001), respectively. There was a significant difference in the genetic diversity in the capsid region of HIV-1 and HIV-2. The number of variations observed in the HIV-1 capsid sequence was significantly (P < 0.001) higher compared to HIV-2.
| ~ Discussion|| |
HIV has a limited host range which shows the evidence of genetic factors involved in blocking infectivity and virus replication during initial steps before reverse transcription occurs. Our study showed an in-detail analysis of the HIV-1 capsid (p24) and HIV-2 capsid (p26) sequence variation within a set of 59 HIV-1 and 14 HIV-2 sequences. An early study by Stremlau et al. (2004) showed the target for huTRIM5α lies in the retroviral capsid region. Studies on retroviruses from the different groups have established that the huTRIM5α-mediated restriction is dependent on its avidity for interaction with the retroviral capsid region. The sequence analysis data on the HIV-1 and HIV-2 capsid region from our centre identified 35 (17.8%) and 8 (4.5%) amino acid variations. The variation determined within the HIV-2 capsid region was significantly lower than HIV-1 (P < 0.001). To the best of our knowledge, this is the first study in India to identify and demonstrate variations in the capsid region of both HIV-1 and HIV-2 circulating in the country. In India, HIV-1 subtype C and HIV-2 subtype A are prevalent., We also tried to identify the mutational hot spots in the capsid region that may alter the susceptibility to human TRIM5α.
A study done by Yang et al. has demonstrated the crystal structure of Rhesus (Rh) TRIM5α SPRY domain and therefore the essential feature for binding of capsid with huTRIM5α. In our study, the docking analysis showed that HIV-1 capsid protein uses the V1 loop of huTRIM5α SPRY domain in both reference and variant strain except strains with A116T variant, during which V3 loop interacts with the huTRIM5α SPRY domain. However, this particular variation A116T failed to show any significant association with the viral load or clinical staging (P = 0.45).
The docking analysis of HIV-2 with huTRIM5α SPRY domain showed the usage of the V3 loop by HIV-2 capsid protein for the interaction with human huTRIM5α in both reference and variant strain, except strains with V81A variant, where V1 loop of huTRIM5α was involved in the interaction, similar to HIV-1 in which they use the V1 loop of huTRIM5α for interaction and has the lowest docking energy difference of 26.4% that could be one of the explanations for the higher viral load in individuals with this particular variation. We assume that strains with the V81A variant have an impact on the viral susceptibility to huTRIM5α and also on viral replication. A couple of mutagenesisin vitro studies have shown that the single amino acid change at positions 120 either with alanine or glutamine rather than proline affected the sensitivity to cynomolgus monkey TRIM5α and huTRIM5α., The mutations affected the conformation of the neighbouring loop (L4 and L5) therefore alter the sensitivity to huTRIM5α. In our study, amino acid variation (V81A) in the HIV-2 capsid protein alters the huTRIM5α loop interaction which can have an impact on the viral pathogenesis. This also increased the likely preponderance to progress to WHO stage 2 or more.
Our sequence analysis data identified 35 amino acid variations within the capsid region of HIV-1 which was higher compared to variations identified in the capsid region of HIV-2. An earlier study done by Veillette et al. has analysed V86M mutation in the capsid region of HIV-1 and therefore the role of cyclophilin A (cypA) in conferring resistance to huTRIM5α. A couple of studies have mentioned mutations in the capsid region of HIV-1 which may alter the uncoating of the virus in the replication cycle., They described and analysed three different mutations such as A92E, E45A and N74D in the capsid region of HIV-1 subtype B and their effect on the HIV uncoating. However, we illustrated the variations in the capsid gene of HIV-1 subtype C, which is the first study documented from India.
Molecular docking studies revealed that even if there is a difference in the docking energy between reference and variant strain in the case of HIV-1, this did not show a significant association with HIV-1 plasma viral load. This shows that huTRIM5α activity on HIV-1 is less efficient. Our data, in agreement with the previous studies, showed a transparent association of huTRIM5α on HIV-2 pathogenesis. Although there have been several mutational studies on P119A, P159S and P178S that showed a good association with HIV-2 viral load,, our study identified a novel variation V81A/T that showed a marginal significance (P = 0.08) with higher HIV-2 plasma viral load as measured by RT activity. This shows that the variations observed in the HIV-2 capsid gene have an impact on the antiviral activity of huTRIM5α and might have an effect on the pathogenesis of HIV-2 infection. This requires detailed functional study to identify the intricate causal interplay between viruses and hosts which influence pathogenesis. Conversely, none of the variations identified in the HIV-1 capsid region showed an association with the HIV-1 viral load and other virological markers.
Onyango et al. have analysed the variations in the HIV-2 (p26) capsid region and demonstrated that non-proline residues at positions 119, 159 and 178 were more frequent in patients with high viral load. In addition, Onyango et al. constructed variant viruses carrying these variations and analysed thein vitro replication levels suggesting an alteration in susceptibility to huTRIM5α. Although in our study participants none of the strains showed variation at 178 positions, the other two variations (P119A/Q and P159S) identified affected the susceptibility to huTRIM5α though the correlation of p119A/Q with the plasma viral load (RT activity) was insignificant probably due to small sample size.
Even though huTRIM5α in Old World monkeys blocks HIV-1 infection effectively during viral replication in the early entry step,,,, huTRIM5α interferes with HIV-1 infection less efficiently., This can be overcome by partial adaptation of HIV-1 which turn out to be sensitive to huTRIM5α. Our data showed that the difference in the synonymous and non-synonymous polymorphism between HIV-1 and HIV-2 additionally strengthens this hypothesis. Among the HIV-1- and HIV-2-monoinfected individuals, synonymous (dS) and non-synonymous (dN) substitutions were significantly higher among HIV-1-infected individuals compared to HIV-2-infected individuals (dS mean 6.15 and 0.09, P < 0.001 and dN mean 0.84 and 0.12, P < 0.001, respectively). Among the HIV-2-infected individuals, even though the number of non-synonymous polymorphisms was significantly (P < 0.001) lower compared to the individuals infected with HIV-1, the impact of the variations identified in the HIV-2 capsid region affected the susceptibility to huTRIM5α. Compared to HIV-1, they have identified fewer sites under positive selection among HIV-2 patients. We also described the comparable results on the selective pressure on the HIV-1 (p24) and HIV-2 (p26) capsid region showing the lower variations in the HIV-2 sequences compared to HIV-1, presumably explaining the lower rate of replication and attenuated pathogenesis. These findings suggest that compared to HIV-2, HIV-1 evolves faster with greater adaptive potential and better survival rate.
To the best of our knowledge, this is the first study to describe a novel variation (V81A) identified in the capsid region of HIV-2-infected individuals. V81A, found in strains of our study individuals, showed a significant role with huTRIM5α interaction as shown by molecular modelling data and the HIV-2 viral load findings in the infected individuals.
The main limitation of our study is the small number of HIV-2 individuals involved in this study. Since the prevalence of HIV-2 is low in our region, recruitment of individuals infected with HIV-2 was difficult. We tend to additionally face a major challenge to come up with the sequence data for a significant proportion of HIV-2 individuals because of low copy numbers in the blood. However, we consider that our results though not significant are valuable and encourage future research. HLA selection of variations in the capsid protein of HIV-1 and HIV-2 was not done in our study to clearly indicate whether these polymorphisms are specific to a particular HLA haplotype of humans. Compared to HIV-2, variations in the HIV-1 capsid was significantly higher and evolution is faster which makes HIV-1 resistant to huTRIM5α. Our main aim was to look at the difference in the interaction between HIV-1 and HIV-2 with huTRIM5α. We have hexamer pdb structure for HIV-1 subtype B; unfortunately, no pdb structure for HIV-1 subtype C was available, and we were not able to generate the hexamer structure for HIV-1 and HIV-2 capsid gene. Hence, the interaction was done between monomer HIV-1 and HIV-2 capsid with TRIM5α SPRY domain. Trim5α SPRY domain exhibits extraordinarily weak affinities for isolated capsid monomers; however, even with the hexavalent capsid assembly, the SPRY binding affinity remains an equivalent.
| ~ Conclusion|| |
Our data showed that the HIV-1 virus has a higher replication capability compared to HIV-2. The usage of the huTRIM5α interacting loop is different between the HIV types, where the HIV-1 reference strain uses the V1 loop and HIV-2 reference strain uses the V3 loop for interaction. The analysis on the HIV-1 capsid gene shows that though there is a shift in the interacting loop by the strains with A116T variation, there is no difference in the susceptibility to huTRIM5α and pathogenesis of HIV-1 infection. However, the HIV-2 strains with V81A variation reduced huTRIM5α activity because of shifting in the interaction of the V3 loop to the V1 loop as shown by the lowest docking energy and higher HIV-2 plasma viral load. This may be one of the reasons for the difference in the replication of viruses observed in the HIV-1- and HIV-2-infected individuals.
We acknowledge Dr. Sathish Shankar and Dr. Mahesh Babu from Sri Sakthi Amma Institute of Biomedical Research, Sri Narayani Hospital and Research Centre, Sripuram, Vellore - 632055, Tamil Nadu, for their critical review on the paper.
Financial support and sponsorship
We gratefully acknowledge the financial support from the Fluid Research Fund (IRB Ref Min No. 8798) and Special Fund from the Department of Clinical Virology, Christian Medical College, and Vellore. The published work is a part of a Ph.D. thesis of R Veena Vadhini, under of the Tamil Nadu Dr. M.G.R Medical University.
Conflicts of interest
There are no conflicts of interest.
| ~ References|| |
Gayle HD, Hill GL. Global impact of human immunodeficiency virus and AIDS. Clin Microbiol Rev 2001;14:327-35.
Marlink R, Kanki P, Thior I, Travers K, Eisen G, Siby T, et al
. Reduced rate of disease development after HIV-2 infection as compared to HIV-1. Science 1994;265:1587-90.
Chollet-Martin S, Simon F, Matheron S, Joseph CA, Elbim C, Gougerot-Pocidalo MA. Comparison of plasma cytokine levels in African patients with HIV-1 and HIV-2 infection. AIDS 1994;8:879-84.
Zheng YH, Jeang KT, Tokunaga K. Host restriction factors in retroviral infection: promises in virus-host interaction. Retrovirology 2012;9:112.
de Silva TI, Leligdowicz A, Carlson J, Garcia-Knight M, Onyango C, Miller N, et al
. HLA-associated polymorphisms in the HIV-2 capsid highlight key differences between HIV-1 and HIV-2 immune adaptation. AIDS 2018;32:709.
Kovalskyy DB, Ivanov DN. Recognition of the HIV capsid by the TRIM5α restriction factor is mediated by a subset of pre-existing conformations of the TRIM5α SPRY domain. Biochemistry 2014;53:1466-76.
Grütter MG, Luban J. TRIM5 structure, HIV-1 capsid recognition, and innate immune signaling. Curr Opin Virol 2012;2:142-50.
Takeuchi JS, Perche B, Migraine J, Mercier-Delarue S, Ponscarme D, Simon F, et al
. High level of susceptibility to human TRIM5α conferred by HIV-2 capsid sequences. Retrovirology 2013;10:50.
Kandathil AJ, Kannangai R, Verghese VP, Pulimood SA, Rupali P, Sridharan G, et al
. Drug resistant mutations detected by genotypic drug resistance testing in patients failing therapy in clade C HIV-1 infected individuals from India. Indian J Med Microbiol 2009;27:231-6.
] [Full text]
Yang J, Zhang Y. I-TASSER server: New development for protein structure and function predictions. Nucleic Acids Res 2015;43:W174-81.
Weinberg JL, Kovarik CL. The WHO clinical staging system for HIV/AIDS. AMA J Ethics 2010;12:202-6.
Tamura K, Stecher G, Peterson D, Filipski A, Kumar S. MEGA6: Molecular evolutionary genetics analysis version 6.0. Mol Biol Evol 2013;30:2725-9.
Speelmon EC, Livingston-Rosanoff D, Li SS, Vu Q, Bui J, Geraghty DE, et al
. Genetic association of the antiviral restriction factor TRIM5alpha with human immunodeficiency virus type 1 infection. J Virol 2006;80:2463-71.
Guex N, Peitsch MC, Schwede T. Automated comparative protein structure modeling with SWISS-MODEL and Swiss-PdbViewer: A historical perspective. Electrophoresis 2009;30 Suppl 1:S162-73.
Laskowski RA, Rullmannn JA, MacArthur MW, Kaptein R, Thornton JM. AQUA and PROCHECK-NMR: Programs for checking the quality of protein structures solved by NMR. J Biomol NMR 1996;8:477-86.
Ghoorah AW, Devignes MD, Smaïl-Tabbone M, Ritchie DW. Protein docking using case-based reasoning. Proteins Structure Funct Bioinformat 2013;81:2150-8.
Garcia-Diaz A, Labbett W, Clewley GS, Guerrero-Ramos A, Geretti AM. Comparative evaluation of the Artus HIV-1 QS-RGQ assay and the Abbott RealTime HIV-1 assay for the quantification of HIV-1 RNA in plasma. J Clin Virol 2013;57:66-9.
Borrego P, Gonçalves MF, Gomes P, Araújo L, Moranguinho I, Figueiredo IB, et al
. Assessment of the cavidi exavir load assay for monitoring plasma viral load in HIV-2-infected patients. J Clin Microbiol 2017;55:2367-79.
MacNeil A, Sankalé JL, Meloni ST, Sarr AD, Mboup S, Kanki P. Genomic sites of human immunodeficiency virus type 2 (HIV-2) integration: Similarities to HIV-1in vitro
and possible differences in vivo
. J Virol 2006;80:7316-21.
Yamauchi K, Wada K, Tanji K, Tanaka M, Kamitani T. Ubiquitination of E3 ubiquitin ligase TRIM5 alpha and its potential role. FEBS J 2008;275:1540-55.
Battivelli E, Migraine J, Lecossier D, Matsuoka S, Perez-Bercoff D, Saragosti S, et al
. Modulation of TRIM5α activity in human cells by alternatively spliced TRIM5 isoforms. J Virol 2011;85:7828-35.
Kannangai R, Shaji RV, Ramalingam S, Jesudason MV, Abraham OC, George R, et al
. HIV-2 subtype circulating in India (south). J Acquir Immune Defic Syndr 2003;33:219-22.
Kandathil AJ, Ramalingam S, Kannangai R, David S, Sridharan G. Molecular epidemiology of HIV. Indian J Med Res 2005;121:333-44.
Yang H, Ji X, Zhao G, Ning J, Zhao Q, Aiken C, et al
. Structural insight into HIV-1 capsid recognition by rhesus TRIM5α. Proc Natl Acad Sci U S A 2012;109:18372-7.
Miyamoto T, Yokoyama M, Kono K, Shioda T, Sato H, Nakayama EE. A single amino acid of human immunodeficiency virus type 2 capsid protein affects conformation of two external loops and viral sensitivity to TRIM5α. PLoS One 2011;6:e22779.
Song H, Nakayama EE, Yokoyama M, Sato H, Levy JA, Shioda T. A single amino acid of the human immunodeficiency virus type 2 capsid affects its replication in the presence of cynomolgus monkey and human TRIM5αs. J Virol 2007;81:7280-5.
Veillette M, Bichel K, Pawlica P, Freund SM, Plourde MB, Pham QT, et al
. The V86M mutation in HIV-1 capsid confers resistance to TRIM5α by abrogation of cyclophilin A-dependent restriction and enhancement of viral nuclear import. Retrovirology 2013;10:25.
Pacheco B, Menéndez-Arias L, Sodroski J. Characterization of two distinct early post-entry blocks to HIV-1 in common marmoset lymphocytes. Sci Rep 2016;6:37489.
Onyango CO, Leligdowicz A, Yokoyama M, Sato H, Song H, Nakayama EE, et al
. HIV-2 capsids distinguish high and low virus load patients in a West African community cohort. Vaccine 2010;28 Suppl 2:B60-7.
Stremlau M, Perron M, Lee M, Li Y, Song B, Javanbakht H, et al.
Specific recognition and accelerated uncoating of retroviral capsids by the TRIM5α restriction factor. Proc Natl Acad Sci U S A 2006;103:5514-9.
Besnier C, Takeuchi Y, Towers G. Restriction of lentivirus in monkeys. Proc Natl Acad Sci USA 2002;99:11920-5.
Cowan S, Hatziioannou T, Cunningham T, Muesing MA, Gottlinger HG, Bieniasz PD. Cellular inhibitors with Fv1-like activity restrict human and simian immunodeficiency virus tropism. Proc Natl Acad Sci USA 2002;99:11914-9.
Hatziioannou T, Cowan S, Goff SP, Bieniasz PD, Towers GJ. Restriction of multiple divergent retroviruses by Lv1 and Ref1. EMBO J 2003;22385-94.
Wagner JM, Christensen DE, Bhattacharya A, Dawidziak DM, Roganowicz MD, Wan Y, et al
. General model for retroviral capsid pattern recognition by TRIM5 proteins. J Virol 2018;92: pii: e01563-17. doi: 10.1128/JVI.01563-17.
Untergasser A, Cutcutache I, Koressaar T, Ye J, Faircloth BC, Remm M, et al
. Primer3-”new capabilities and interfaces. Nucleic Acids Res 2012;40:e115.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6]
[Table 1], [Table 2], [Table 3]