HIV-2 infection represents a unique model of attenuated retroviral infection, characterized by a slow disease progression , a low viral production , and a low rate of transmission [3,4].
In HIV-1, APOBEC3F/3G cellular proteins act as a potent host restriction factor of retroviral replication through cytidine deaminase activity . The introduction of G-to-A substitutions might produce stop codons and also hypermutated viruses when occurring at excessive levels [6–8]. The Vif viral protein is able to counteract APOBEC3F/3G antiretroviral activity by inducing its proteasomal degradation [9–11]. HIV-2 Vif determinants involved in the interaction with APOBEC3F/3G are distinct from those of HIV-1 Vif . APOBEC3F/3G editing level in HIV-2 sequences remains unknown.
The objective of this study was to assess the level of in-vivo APOBEC3F/3G editing in sequences from HIV-2-infected antiretroviral-naive patients.
Patients and methods
We assessed antiretroviral-naive patients, included in the French Agence Nationale de Recherches sur le SIDA et les hépatites virales (ANRS) CO5 HIV-2 cohort and for whom peripheral blood mononuclear cells (PBMC) were collected. Written informed consent was obtained from all patients. In some patients, paired plasma samples were available. We also assessed HIV-1 subtype B PBMC sequences from antiretroviral-naive patients available in GenBank .
HIV-2 RNA and HIV-2 total DNA quantification
HIV-2 plasma viral load was assessed by a previously described in-house method with a limit of detection of 100 copies/ml . HIV-2 total DNA was assessed by a previously in-house real-time PCR assay with a limit of detection of 5 copies/105 PBMC .
Amplification and sequencing of HIV-2 vif and pol regions
DNA was isolated from PBMC with QIAsymphony DSP-DNA-MiniKit (Qiagen, Hilden, Germany). Firstly, vif gene was amplified by PCR (AmpliTaqGoldDNA Polymerase; Applied Biosystems, Foster City, California, USA) following the manufacturer's instructions using the primers: MBHIVVIF5’ (5’- position 4645 ROD-AAAAGAAGGGGAGGAATAGGG) and VPX3’New (5’-5826-CAAGCRGGRTCAAAATGC). Then, a nested-PCR was performed using the primers: PFD (5’-4761-GGTCTATTWCAGAGAAGGCAG) and MBHIVVIF3’ (5’-5690-GAGGAGGRGGAGGTCC). Sequencing reactions were run on an ABI Prism BigDye Terminator kit with the nested-primers using an automated sequencer (ABI Prism 3130 XL; Applied Biosystems). GenSearch software (PhenoSystems, Lillois, Belgium) was used to edit and align the nucleotidic sequences.
HIV-2 protease and reverse transcriptase sequencing were performed using in-house methods previously described .
Hypermutation was assessed using the Hypermut2.0 program [17,18]. ROD and EHO sequences were used as reference for HIV-2 groups A and B, respectively. This program performed a Fisher's exact test comparing the number of G-to-A changes in the trinucleotidic APOBEC3F/3G context (GGD or GAD [D = A, G, or T]) vs. the control context (GYN or GRC, Y = C or T, R = G or A, N = all), and a sequence with a P value < 0.05 was considered as hypermutated.
In our study, we considered as defective sequences all hypermutated sequences and those containing at least one stop codon.
Mann–Whitney and Fisher exact tests were used for comparisons with a P value of 0.05 significance level .
Hypermutation detection in Vif and protease–reverse transcriptase HIV-2 proviral sequences
Vif sequencing was successful in 82 samples among 144 tested, including 52 group A and 30 group B. Characteristics were similar between patients infected with HIV-2 group A and those infected with group B (Table 1).
Fifteen (28.8%) of the 52 group A Vif proviral sequences were defective including five hypermutated and 10 containing stop codons sequences. No difference was observed in the level of APOBEC3F/3G editing between patients with viral load below 100 copies/ml (n = 37) and those with viral load above 100 copies/ml (n = 15). Protease–reverse transcriptase proviral sequences were successfully amplified in 30 samples showing four (13.3%) defective sequences including one hypermutated (Table 2).
Among the 30 HIV-2 group B Vif proviral sequences, five were defective (16.7%) including two hypermutated. Protease–reverse transcriptase sequences were obtained in 13 samples and none was defective. Only six of the 30 patients had detectable viral load, limiting comparison of the level of APOBEC3F/3G editing with patients displaying viral load below 100 copies/ml.
Hypermutation detection in HIV-2 RNA Vif plasma sequences
Paired plasma and PBMC samples were available in 20 patients. None Vif sequence was defective. The median number of APOBEC-type G-to-A mutations was 7 (IQR = 6–7) and 9 (IQR = 8–12), in HIV-2 group A (n = 11) and B (n = 9) sequences, respectively.
A significantly higher number of APOBEC-type G-to-A mutations (9 vs. 6, P = 0.00008) was observed in HIV-2 group B Vif sequences than in HIV-1 GenBank sequences (n = 107), but not with group A Vif sequences (7 vs. 6, P = 0.21).
Comparison of APOBEC3F/3G footprint between HIV-2 groups A and B sequences
A significantly higher median number of APOBEC-type G-to-A mutations was observed in HIV-2 group B than in group A sequences, both in Vif and protease–reverse transcriptase regions (P = 0.02 and P = 0.006, respectively) (Table 2) and also in plasma (7 vs. 9, P = 0.001).
In group B sequences APOBEC3F contributed in a significantly higher proportion of G-to-A mutations in Vif than in protease–reverse transcriptase region (68 vs. 51%, P = 0.00005). In contrast, there was no difference in group A sequences (55 vs. 45%). In addition, APOBEC-type G-to-A mutations in Vif region were more significantly generated by APOBEC3F in group A than in group B sequences (68 vs. 55%, P = 0.0001).
Comparison of APOBEC3F/3G footprint between HIV-2 and HIV-1 sequences
We assessed HIV-1 proviral sequences from GenBank database in Vif (n = 295) and in protease–reverse transcriptase (n = 237) region.
A significantly higher proportion of hypermutated viruses was observed in HIV-2 group A than in HIV-1 sequences (P = 0.0049) and a trend was observed in HIV-2 group B sequences (P = 0.09) (Table 2). A higher proportion of proviral sequences with stop codons was observed among HIV-2 than HIV-1 sequences in Vif region in both groups A (P = 0.00001) and B sequences (P = 0.013), and also in group A protease–reverse transcriptase sequences (P = 0.02). The number of APOBEC-type G-to-A mutations was found significantly higher in HIV-2 than in HIV-1 sequences, whatever the group or the genomic region assessed (P = 0.00001 in all comparisons) (Table 2).
Comparison of immunovirological parameters between patients harboring or not Vif defective viruses
No difference was observed between patients harboring Vif defective viruses (n = 20) and those harboring Vif not defective viruses (n = 62) regarding the following: proportion of patients with viral load below the limit of detection (65 vs. 77%, P = 0.15); median viral load in patients with detectable viremia (1476 vs. 1452 copies/ml, P = 0.85); proportion of patients with undetectable HIV-2 total DNA load (0 vs. 16%, P = 0.18); median HIV-2 total DNA level in patients with detectable HIV-2 total DNA (2.79 vs. 2.71 log10 copies/106 PBMC, P = 0.98); and median CD4+ cell count (451 vs. 538/mm3, P = 0.11).
In the present study, we showed a high level of in-vivo APOBEC3F/3G editing in HIV-2 proviral sequences from 82 antiretroviral-naive patients.
We evidenced 9.6% hypermutated sequences in PBMC issued from antiretroviral-naive HIV-2-infected group A patients, showing a higher proportion than in HIV-1 proviral sequences from the GenBank database. The median number of APOBEC-type G-to-A mutations was significantly higher in HIV-2 than in HIV-1 sequences in both genomic regions assessed. These findings are not the consequence of the different nucleotidic context between HIV-1 and HIV-2 genomes, as the proportion of potential sites of APOBEC3F/3G editing and the proportion of tryptophan residues were similar between HIV-1 and HIV-2 sequences in both regions assessed. We found a higher proportion of defective viruses in the Vif region than in the protease–reverse transcriptase region. Such gradient of hypermutation along viral genome has also been described in HIV-1 sequences, depending on the time during which the region remains single-stranded during reverse transcriptase reaction .
In HIV-1, Vif polymorphisms could be important in vivo, as some Vif allelic variants display less anti-APOBEC activity than others [20–23]. We can hypothesize that the higher in-vivo level of APOBEC3F/3G editing we observed in HIV-2 sequences might result from Vif polymorphisms that are less able to counteract APOBEC antiretroviral activity. Mainly, the HIV-1 Vif mutation K22H, recently described as decreasing the efficacy of Vif function , was naturally present in all HIV-2 patients and reference sequences.
HIV-2 groups A and B viruses represent independent host transfers of simian immunodeficiency virus infecting sooty mangabey in West Africa and display an heterogeneity at nucleotidic level . We showed a HIV-2 group effect with a significantly higher level of APOBEC3F/3G editing in group B than in group A sequences. We also showed a differential contribution of APOBEC3F and APOBEC3G in the generation of G-to-A mutations between group B and group A sequences. To our knowledge, no difference was reported in the degree of hypermutation depending of the HIV-1 subtype.
In HIV-1 infection, the proportion of patients displaying hypermutated sequences reported in the literature largely varies resulting of the different contexts in which the degree of hypermutation was assessed [20,21,25–28], with major differences regarding patients’ therapeutic status, sequencing technique, viral genome region studied, and identification method of hypermutated viruses, rendering unreliable any comparison. Strength of our study is to have assessed the level of APOBEC3F/3G editing in HIV-1 and HIV-2 sequences using similar criteria of sequences selection and the same tools.
A gradient of hypermutation has been described in HIV-1 showing the highest levels of mutations in the viral DNA and the lowest ones in plasma viral RNA . No defective virus was detected in Vif region of plasma viruses, suggesting a similar gradient in HIV-2 sequences, even though we only assessed 20 plasma sequences.
A higher level of APOBEC3F/3G editing was observed in HIV-1 successfully antiretroviral-treated patients [28,30–32]. Thus, we can hypothesize that the low viral production of HIV-2 infection, with 74% of antiretroviral-naive patients showing undetectable viral load in our study, might have an impact on APOBEC3F/3G editing level. However, we did not evidence difference between viremic and aviremic patients, suggesting no major impact of the viral replication status on APOBEC3F/3G editing in HIV-2 infection.
No association between immunovirological parameters and the level of hypermutation could be evidenced in this cross-sectional analysis, and may be linked to a lack of power in statistical analyses. A longitudinal study might help to assess the impact of defective viruses on disease progression. In HIV-1 infection, data concerning the correlation of in-vivo hypermutation phenomenon with clinical parameters are contradictory. Some studies showed association between the level of hypermutation and a lower viral load , a lower size of reservoir , or higher CD4+ cell count . Others studies did not find association between the level of hypermutation and viral load or CD4+ cell count [21,33].
One limitation of our study is the use of Sanger sequencing, indeed using ultra-deep sequencing would provide more information and might allow quantifying the proportion of defective virus. Another limitation is the use of HIV-1 GenBank sequences that are not matched to HIV-2 sequences and for whom immunovirological characteristics were not available. Finally, the selection of antiretroviral-naive HIV-2-infected patients from the ANRS CO5 HIV-2 cohort, mainly based on the availability of PBMC samples and the small sample size, represent a limitation.
In conclusion, we showed a high level of APOBEC3F/3G editing in vif and pol regions of HIV-2 proviral sequences from antiretroviral-naive patients with a higher level of APOBEC3F/3G footprint in HIV-2 group B than group A. Further studies are needed to assess the impact of defective viruses on HIV-2 disease progression.
The authors would like to thank Pr Vincent Calvez and Dr Slim Fourati for their helpful discussion. We thank all clinical and virological investigators of the HIV-2 Cohort Study (ANRS CO5 HIV-2).
Authors’ contributions: M.B., C.C., S.M., F.B.V., and D.D. contributed to this study's concept. M.B., B.V., and A.S. performed population sequencing. M.B., G.C., and L.L. performed HIV-2 DNA quantification. M.B., C.C,. F.D., S.M,. F.B.V., and D.D. contributed to the analysis and interpretation of the data. M.B., C.C., S.M., F.B.V., and D.D. contributed to writing the manuscript. All authors contributed to the critical review of the manuscript.
Source of funding: French National Agency for Research on AIDS and viral hepatitis [Agence Nationale de Recherches sur le SIDA et les hépatites virales (ANRS)] and from the European Community's Seventh Framework Programme (FP7/2007–2013) under the project ‘Collaborative HIV and Anti-HIV Drug Resistance Network (CHAIN)’ (grant no. 223131).
Presentation: This work was presented as an Oral Communication at the International Workshop on Antiviral Drug Resistance (abstract # 26), 3–6 June 2014, Berlin, Germany.
Conflicts of interest
There are no conflicts of interest.
1. 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
2. Popper SJ, Sarr AD, Travers KU, Guèye-Ndiaye A, Mboup S, Essex ME, et al. Lower human immunodeficiency virus (HIV) type 2 viral load reflects the difference in pathogenicity of HIV-1 and HIV-2
. J Infect Dis
3. Kanki PJ, Travers KU, MBoup S, Hsieh CC, Marlink RG, Gueye-NDiaye A, et al. Slower heterosexual spread of HIV-2 than HIV-1
4. Burgard M, Jasseron C, Matheron S, Damond F, Hamrene K, Blanche S, et al. Mother-to-child transmission of HIV-2 infection from 1986 to 2007 in the ANRS French Perinatal Cohort EPF-CO1
. Clin Infect Dis
5. Sheehy AM, Gaddis NC, Choi JD, Malim MH. Isolation of a human gene that inhibits HIV-1 infection and is suppressed by the viral Vif protein
6. Harris RS, Bishop KN, Sheehy AM, Craig HM, Petersen-Mahrt SK, Watt IN, et al. DNA deamination mediates innate immunity to retroviral infection
7. Yu Q, König R, Pillai S, Chiles K, Kearney M, Palmer S, et al. Single-strand specificity of APOBEC3G accounts for minus-strand deamination of the HIV genome
. Nat Struct Mol Biol
8. Bishop KN, Holmes RK, Sheehy AM, Davidson NO, Cho S-J, Malim MH. Cytidine deamination of retroviral DNA by diverse APOBEC proteins
. Curr Biol
9. Sheehy AM, Gaddis NC, Malim MH. The antiretroviral enzyme APOBEC3G is degraded by the proteasome in response to HIV-1 Vif
. Nat Med
10. Marin M, Rose KM, Kozak SL, Kabat D. HIV-1 Vif protein binds the editing enzyme APOBEC3G and induces its degradation
. Nat Med
11. Conticello SG, Harris RS, Neuberger MS. The Vif protein of HIV triggers degradation of the human antiretroviral DNA deaminase APOBEC3G
. Curr Biol
12. Smith JL, Izumi T, Borbet TC, Hagedorn AN, Pathak VK. HIV-1 and HIV-2 Vif interact with human APOBEC3 proteins using completely different determinants
. J Virol
14. Damond F, Collin G, Descamps D, Matheron S, Pueyo S, Taieb A, et al. Improved sensitivity of human immunodeficiency virus type 2 subtype B plasma viral load assay
. J Clin Microbiol
15. Damond F, Descamps D, Farfara I, Telles JN, Puyeo S, Campa P, et al. Quantification of proviral load of human immunodeficiency virus type 2 subtypes A and B using real-time PCR
. J Clin Microbiol
16. Charpentier C, Eholié S, Anglaret X, Bertine M, Rouzioux C, Avettand-Fenoël V, et al. Genotypic resistance profiles of HIV-2-treated patients in West Africa
17. Hypermut hypermutation analysis page. http://www.hiv.lanl.gov/content/sequence/HYPERMUT/hypermut.html
. [Accessed 15 January 2015]
18. Rose PP, Korber BT. Detecting hypermutations in viral sequences with an emphasis on G -> A hypermutation
19. AdScience. Statistical software for common statistical testing, medical statistical testing and data analysis: StatEL. http://www.adscience.eu/software/statel
. [Accessed 15 January 2015]
20. Pace C, Keller J, Nolan D, James I, Gaudieri S, Moore C, et al. Population level analysis of human immunodeficiency virus type 1 hypermutation and its relationship with APOBEC3G and vif genetic variation
. J Virol
21. Piantadosi A, Humes D, Chohan B, McClelland RS, Overbaugh J. Analysis of the percentage of human immunodeficiency virus type 1 sequences that are hypermutated and markers of disease progression in a longitudinal cohort, including one individual with a partially defective Vif
. J Virol
22. Fourati S, Malet I, Binka M, Boukobza S, Wirden M, Sayon S, et al. Partially active HIV-1 Vif alleles facilitate viral escape from specific antiretrovirals
23. Simon V, Zennou V, Murray D, Huang Y, Ho DD, Bieniasz PD. Natural variation in Vif: differential impact on APOBEC3G/3F and a potential role in HIV-1 diversification
. PLoS Pathog
24. Peeters M, Jung M, Ayouba A. The origin and molecular epidemiology of HIV
. Expert Rev Anti Infect Ther
25. Janini M, Rogers M, Birx DR, McCutchan FE. Human immunodeficiency virus type 1 DNA sequences genetically damaged by hypermutation are often abundant in patient peripheral blood mononuclear cells and may be generated during near-simultaneous infection and activation of CD4(+) T cells
. J Virol
26. Land AM, Ball TB, Luo M, Pilon R, Sandstrom P, Embree JE, et al. Human immunodeficiency virus (HIV) type 1 proviral hypermutation correlates with CD4 count in HIV-infected women from Kenya
. J Virol
27. Fourati S, Lambert-Niclot S, Soulie C, Wirden M, Malet I, Valantin MA, et al. Differential impact of APOBEC3-driven mutagenesis on HIV evolution in diverse anatomical compartments
28. Fourati S, Lambert-Niclot S, Soulie C, Malet I, Valantin MA, Descours B, et al. HIV-1 genome is often defective in PBMCs and rectal tissues after long-term HAART as a result of APOBEC3 editing and correlates with the size of reservoirs
. J Antimicrob Chemother
29. Russell RA, Moore MD, Hu W-S, Pathak VK. APOBEC3G induces a hypermutation gradient: purifying selection at multiple steps during HIV-1 replication results in levels of G-to-A mutations that are high in DNA, intermediate in cellular viral RNA, and low in virion RNA
30. Kieffer TL, Kwon P, Nettles RE, Han Y, Ray SC, Siliciano RFG. A hypermutation in protease and reverse transcriptase regions of human immunodeficiency virus type 1 residing in resting CD4+ T cells in vivo
. J Virol
31. Eyzaguirre LM, Charurat M, Redfield RR, Blattner WA, Carr JK, Sajadi MM. Elevated hypermutation levels in HIV-1 natural viral suppressors
32. Josefsson L, von Stockenstrom S, Faria NR, Sinclair E, Bacchetti P, Killian M, et al. The HIV-1 reservoir in eight patients on long-term suppressive antiretroviral therapy is stable with few genetic changes over time
. Proc Natl Acad Sci
33. Ulenga NK, Sarr AD, Hamel D, Sankale J-L, Mboup S, Kanki PJ. The level of APOBEC3G (hA3G)-related G-to-A mutations does not correlate with viral load in HIV type 1-infected individuals
. AIDS Res Hum Retroviruses