Figures
Abstract
Due to constitutive expression in cells targeted by human immunodeficiency virus (HIV), and immediate mode of viral restriction upon HIV entry into the host cell, APOBEC3G (A3G) and APOBEC3F (A3F) have been considered primarily as agents of innate immunity. Recent bioinformatic and mouse model studies hint at the possibility that mutation of the HIV genome by these enzymes may also affect adaptive immunity but whether this occurs in HIV-infected individuals has not been examined. We evaluated whether APOBEC-mediated mutations within common HIV CD8+ T cell epitopes can potentially enhance or diminish activation of HIV-specific CD8+ T cells from infected individuals. We compared ex vivo activation of CD8+ T lymphocytes from HIV-infected individuals by wild type HIV peptide epitopes and synthetic variants bearing simulated A3G/F-induced mutations by measuring interferon-γ (IFN-γ) production. We found that A3G/F-induced mutations consistently diminished HIV-specific CD8+ T cell responses against the common epitopes we tested. If this reflects a significant trend in vivo, then adaptation by HIV to enrich sequences that are favored for mutation by A3G/F (A3G/F hotspots) in portions of its genome that encode immunogenic CD8+ T cell epitopes would favor CTL escape. Indeed, we found the most frequently mutated A3G motif (CCC) is enriched up to 6-fold within viral genomic sequences encoding immunodominant CD8+ T cell epitopes in Gag, Pol and Nef. Within each gene, A3G/F hotspots are more abundant in sequences encoding epitopes that are commonly recognized due to their HLA restriction. Thus, in our system, mutations of the HIV genome, mimicking A3G/F activity, appeared to abrogate or severely reduce CTL recognition. We suggest that the physiological significance of this potential effect in facilitating CTL escape is echoed in the adaptation of the HIV genome to enrich A3G/F hotspots in sequences encoding CTL epitopes that are more immunogenic at the population level.
Citation: Monajemi M, Woodworth CF, Zipperlen K, Gallant M, Grant MD, Larijani M (2014) Positioning of APOBEC3G/F Mutational Hotspots in the Human Immunodeficiency Virus Genome Favors Reduced Recognition by CD8+ T Cells. PLoS ONE 9(4): e93428. https://doi.org/10.1371/journal.pone.0093428
Editor: Johan K. Sandberg, Karolinska Institutet, Sweden
Received: November 14, 2013; Accepted: March 5, 2014; Published: April 10, 2014
Copyright: © 2014 Monajemi et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by CIHR (MOP111132 and OCH131580) and Canadian Foundation for AIDS Research (CANFAR) operating grants, the Wallace Ingram Award, the Research and Development Corporation (Newfoundland) funding to ML, and by CIHR operating grants (HOP-111725, HOP-93428) and CIHR catalyst grant (HC1-112568) to MG. ML holds a CIHR New Investigator award. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
APOBEC3G (A3G) and APOBEC3F (A3F) mutate deoxycytidine (dC) to deoxyuridine (dU) in the reverse transcribed nascent DNA of human immunodeficiency virus (HIV) [1]–[5]. A3G/F are constitutively expressed in CD4+ T lymphocytes, macrophages, and dendritic cells, the targets of HIV [6]–[12], but are also induced by interferon (IFN) in these and other cell types [11]. If a host cell expresses A3G/F, these enzymes infiltrate the virions produced by said host. In the next infection cycle, they deaminate dC in signature trinucleotide motifs (CCC and less often TCC for A3G; TTC for A3F) in the minus strand reverse-transcribed viral DNA, registering as deoxyguanosine (dG) to deoxyadenosine (dA) substitutions in the plus strand. These mutations alter reading frames, terminate translation, and/or produce mutated proteins [13]–[17]. In addition to mutating the HIV genome, A3G/F have also been reported to physically interfere with reverse transcription and viral genome integration into the host cell [18]–[20].
In the physiological setting, the effectiveness of A3G/F against HIV is restricted by several factors. First, the HIV auxiliary protein viral infectivity factor (Vif) binds to A3G/F and promotes their degradation [21]–[24]. Second, entrapment of A3G/F in high molecular mass (HMM) ribonuclear complexes renders a major fraction of the cellular pool of enzymes inactive [25]–[28]. Third, even in cells where A3G/F introduce a high mutational load into the invading viral genomes, the final mutational profile in assembled virions is limited by selection for viral fitness over successive stages of the viral life cycle, a phenomenon termed “purifying selection” [29]. As a result of these viral counter measures as well as the filtering process through selection for fitness, the majority of circulating viruses bear a lesser (sub-lethal) mutational load in their genome than would reflect the full enzymatic potential of A3G/F. Regardless of the level of their anti-viral effectiveness in vivo, APOBECs have classically been thought of as host restriction innate immune factors largely owing to their expression pattern and immediate modes of viral restriction. They presumably play a particularly important role in the early phases of HIV infection to limit viral propagation whilst long-lasting and more specific adaptive immune responses takes weeks to months to develop fully.
Following the initial innate immune barriers, adaptive immune responses in the form of anti-viral antibodies and cytotoxic T cells (CTL) develop and act to contain viral propagation throughout the course of infection [30]. Robust anti-viral CTL responses are associated with slower disease progression and lower viremia levels [31]–[35]. Mutations in CTL epitopes of HIV favor immune evasion by reducing CTL recognition and activation [36]–[40]. Although A3G/F are intrinsic anti-viral barriers that function early in infection, once the CTL response escalates to effective levels, the limited action of A3G/F may actually aid HIV by mutating its CTL epitopes towards immune evasion [19], [41]. In support of this possibility, the most highly mutating sites in HIV genomes mediating CTL escape are somewhat enriched in A3G hotspots [41]. However, viral genomes derived from infected individuals are inherently biased towards immune escape mutations because mutations that enhance CTL recognition would be rapidly eliminated. Mutations that might conversely improve CTL recognition, like mutations that would otherwise decrease viral fitness, would not be reflected in the isolated pool of viral genomes. On the other hand, a mouse-model study found that A3G-induced mutations outside HIV peptide epitopes can enhance the HIV-specific CTL response [42]. This study measured the response of MHC-matched CTL clones bearing transgenic peptide-specific T cell receptors (TCR) against peripheral blood mononuclear cells (PBMCs) infected by HIV produced in A3G-expressing or deficient cells. Results indicated that A3G-induced mutations outside CTL epitopes could alter the relative types of peptides available for CTL recognition. This study indicated for the first time that A3G can, in principle, modulate the CTL response by affecting the efficiency of epitope processing and presentation to CTL. However, the results of this study do not inform as to whether A3G/F mutations directly within CTL epitopes enhance or diminish the CTL response or whether the effect is measurable in the physiological setting of human HIV infection.
In this study, we simulated A3G/F mutations of CTL epitopes of HIV and measured their effect on recognition by circulating CD8+ T cells of HLA-matched, HIV-infected individuals ex vivo. We found that HIV-specific CD8+ T cell recognition was generally diminished by A3G/F-induced mutations. In silico analysis revealed the greatest enrichment of A3G/F hotspot motifs occurs within the viral genomic sequences that encode for the most immunogenic CTL epitopes. Our ex vivo data are consistent with the potential of A3G/F mutations to promote CTL escape. Our observation that HIV has evolved strategically-placed A3G/F mutable hotspots in its genome to maximally direct A3G/F activity towards CTL escape is further evidence that A3G/F mutations likely impact adaptive immunity in vivo.
Materials and Methods
In silico Demarcation and Simulation of A3G/F Hotspot Mutations in CTL Epitopes
The 9229 nucleotide-long genome of HIV isolate BRU was retrieved from the National Center for Biotechnology Information (NCBI) website. Isolate BRU was selected due to its representation of clade B, which is most prevalent in western countries [43]. Portions of the HIV genome encoding known CTL epitopes were identified using the HIV Molecular Immunology Database (http://www.hiv.lanl.gov/content/immunology/tables/ctl_summary.html). A3G/F mutations on the 5′ dG located in A3G/F hotspots in the CTL epitope-coding genomic DNA were simulated in silico to generate mutated epitopes. For epitopes with multiple hotspots or those with potential for the generation of novel hotspots as a result of first-round mutation, multiple mutated versions were considered. Wild type (wt) and mutant (mut) forms of all selected epitopes were synthesized to ∼90% purity (Peptide 2.0 Inc., VA-US) and used to measure peptide-specific CTL responses from HLA-matched HIV+ individuals.
Measurement of HIV-specific CD8+ T Cell Recognition
HIV-infected study participants were recruited through the Newfoundland and Labrador Provincial HIV Clinic in St. John’s, Newfoundland, and selected on the basis of their HLA class I types determined for previous immunological studies [44], [45]. Informed written consent was obtained for blood collection and this study received ethical approval from the Health Research Ethics Authority of Newfoundland and Labrador for the consent and experimental protocols (reference #02.157). Cryopreserved PBMC from appropriate HLA-matched subjects were used as responder CD8+ T cells. The ELISPOT assay to measure the frequency of HIV-specific CD8+ T cells has been previously described [46]. The average of duplicate wells was obtained and the number of background spots in the negative control well subtracted. The total number of IFN-γ spot forming cells (SFC) in each well was expressed as IFN-producing SFC/106 PBMC. Even though the background frequency of IFN-producing unstimulated PBMC was negligible (typically <10/106 PBMC), we only considered >50 SFC/106 PBMC as a positive response. (In all, 27/77 subjects tested showed a significant response.). Differences of >25% in SFC between wt and mut were considered significant.
In silico Analysis of A3G/F Hotspot Frequencies in CTL Epitopes
The number of hotspots (for A3G: GGG, GGT or GGA; for A3F: GAA) inside vs. outside CTL epitopes in each HIV gene was determined by scanning the BRU sequence. To normalize for gene size, the frequencies of hotspot occurrence were divided by the total number of nucleotides. These frequencies were calculated both for all A3G/F hotspots, and also solely for the most frequently targeted A3G hotspot (GGG). Subsequently, the ratio (R) of normalized hotspot frequencies inside to outside epitopes was calculated for each gene. In order to evaluate differences in R amongst various genes, an index was generated by determining the average of the R of all genes and dividing the R of each gene by the average. Thus, an index of >1 indicates an above average ratio of A3G/F hotspots inside to outside CTL epitopes. We further conducted the same type of analysis within each gene in order to compare hotspot frequencies between epitopes restricted to different HLA alleles (A1, A2, A3, A11, A24, B7, B8, B35, B40, B44, B53, B57).
Statistical Analysis
The wilcoxon signed-rank test was used to evaluate the statistical likelihood that introduction of APOBEC-related mutations into HIV CTL epitopes reduces recognition of those epitopes. Paired values for all Elispot comparisons of wt vs. mut epitopes were used to calculate statistical significance using Graphpad Prism 5.0.
Results
Demarcation of A3G/F Hotspots within Genomic Sequences Encoding CTL Epitopes and Simulation of A3G/F-induced Mutations
We identified CTL epitopes in all HIV proteins using the experimentally-verified HIV molecular immunology database (http://www.hiv.lanl.gov/content/immunology/tables/ctl_summary.html). This analysis for Gag is shown in Figure S1A (see Figure S2 for all genes). We analyzed a number of epitopes restricted to HLA-A2, B44 and B57. These HLAs were chosen for several reasons: strong CTL responses to B57-restricted epitopes correlate inversely with disease progression, and HLA-A2 and B44 are common in the population and present numerous HIV epitopes [32], [47]–[59]. Moreover, between them, these HLA alleles cover a majority of the population (http://www.allelefrequencies.net). We examined CTL epitopes in Gag, Pol and Nef because these contain the HIV peptides most prominently recognized by CD8+ T cells [60]–[62]. The genomic sequences of 123 epitopes from the aforementioned database were identifiable in the BRU isolate sequence (Gag shown as an example in Figure S1B; see Figure S3 for all genes). Of these, 98 epitopes (80%) contained ≥1 A3G/F hotspot in which A3G/F mutations were simulated and translated in silico. Epitopes for which the only possible mutation led to a stop codon or non-conservative changes in HLA anchor residues (thus impeding presentation by MHC class I) were excluded. Using these criteria, we selected and synthesized 27 wild type (wt) epitopes and 64 derivative mutants (mut) thereof (Table 1). Of the wt epitopes, 7 contained a single A3G or A3F hotspot yielding 1 mut each (example shown in Figure S1c, left panel). The other 20 contained >1 A3G/F hotspot (multiple hotspots separated by >3 nucleotides were considered independently; for example see Figure S1C, middle panel). These yielded multiple muts, some of which could occur because the initial A3G/F mutation created a new A3G/F hotspot that could be mutated again in the same or subsequent round(s) of viral replication (example shown in Figure S1C, right panel). Finally, we considered whether genomes containing these mut epitope-encoding sequences have been reported in a database of submitted HIV sequences (http://www.hiv.lanl.gov/content/sequence/QUICK_ALIGN/QuickAlign.html). We found that nearly all mut epitopes occur in the population. The selected wt and mut epitopes thus considered (Table 1) were synthesized and evaluated by ELISPOT for recognition by HIV-specific CD8+ T cells from HLA-matched subjects.
CD8+ T Cell Recognition of HLA-B57-restricted Epitopes is Diminished by A3G/F-simulated Mutations
We compared the CD8+ T cell response against 3 HLA-B57-restricted epitopes in Gag and the corresponding variants with APOBEC-induced mutations. Of 5 HLA-B57+ subjects (P7, P45, P71, P166, P197) tested against TSTLQEQIGW and 1 mut, P197 alone had a positive response that was 3-fold higher against the wt peptide (280 vs. 85 SFC/106 PBMC) (Figure 1a). Three of the same 5 subjects had positive responses against KAFSPEVIP (P71: 397; P166: 50; P197: 90 SFC/106 PBMC), whilst none had responses against the mut (Figure 1b). Eight subjects (P7, P20, P45, P71, P76, P166, P185, P197) were tested for responses against AISPRTLNAW and 1 mut. Four (P7, P71, P166, P197) had positive responses. P7 and P166 responded only against the wt peptide, while P71 and P197 exhibited responses that were 3- and 1.4-fold higher against wt (P71: 895 vs. 273; P197: 160 vs. 112 SFC/106 PBMC) (Figure 1c). Responses against the Nef epitope GPGVRYPLTFGWCY and 7 muts were compared in 6 subjects (P20, P45, P68, P76, P166, P185). Only P68 had a response, and only against the wt peptide (Figure 1d). Taken together, 9 HLA-B57 subjects responded only against the wt epitopes, whilst 3 responded against both the wt and corresponding mut for which the responses against wt epitopes were 1.4- to 3-fold higher. Thus, for the HLA-B57-restricted epitopes and subjects we studied, introduction of signature A3G/F-induced mutations into the epitopes consistently abrogated or reduced CD8+ T cell recognition.
Responses of PBMC from HLA-matched subjects to wild type and mutant epitopes are shown on the Y-axis, as IFN-γ SFC (Spot Forming Cells)/106 PBMC. A. Comparison of IFN-γ induction between wild type and a mutant form of the B57-restriced Gag epitope TSTLQEQIGW. B. Comparison of IFN-γ induction between wild type and 2 mutant forms of the B57-restriced Gag epitope KAFSPEVIP. C. Comparison of IFN-γ induction between wild type and a mutant form of the B57-restriced Gag epitope AISPRTLNAW. D. Comparison of IFN-γ induction between wild type and 7 mutant forms of the B57 restricted Nef epitope GPGVRYPLTFGWCY.
A3G/F-simulated Mutations Diminish CD8+ T Cell Recognition of HLA-A2-restricted Epitopes
We measured recognition of the HLA-A2-restricted Gag epitope FLGKIWPS and 3 muts by 9 subjects (P3, P18, P30, P35, P64, P71, P78, P98, P234). Five had positive responses exclusively against the wt peptide (Figure 2a). Recognition of the immunodominant Nef epitope FLKEKGGLEGL epitope and 15 muts was assessed in 30 subjects, of whom 4 responded against the wt peptide. Of these, P211 responded against a single mut with 1.6-fold less magnitude than against the wt peptide (1512 vs. 910 SFC/106 PBMC) (Figure 2b-e). Recognition of 4 HLA-A2-restricted Pol epitopes and corresponding variants was compared. The Pol epitope IYQYMDDLYV and 1 mut were tested in 20 subjects, of whom 3 responded with moderately higher responses against the wt than mut peptides (for wt vs. mut: P35: 147 vs. 92; P35′: 190 vs. 160; P214: 225 vs. 90; P233: 570 vs. 495 SFC/106 PBMC) (Figure 2f). Recognition of ILKEPVHGV and 1 mut was tested in 17 subjects, of whom 3 responded, comparably against wt and mut peptides (for wt vs. mut: P35: 2675 vs. 2442; P35′: 1380 vs. 1305; P71: 90 vs. 88; P105: 305 vs. 405 SFC/106 PBMC) (Figure 2g). LVGPTPVNII and 1 mut were tested in 8 subjects, with 2 responding only against wt (Figure 2h). ALVEICTEM and 1 mut were tested in 11 subjects and only the wt peptide was recognized in 1 subject (Figure 2i). Altogether, for 6 HLA-A2-restricted epitopes in Gag, Nef and Pol, 4 were better recognized by CD8+ T cells as wt peptides, whilst 2 were recognized comparably well as wt or mut peptides. A3G/F-simulated mutations in HLA-A2-restricted epitopes thus diminished recognition by CD8+ T cells in the majority of cases.
Responses of PBMC from HLA-matched subjects to wild type and mutant epitopes are shown on the Y-axis, as IFN-γ SFC/106 PBMC. A. Comparison of IFN-γ induction between wild type and 3 mutant forms of the HLA-A2 restricted Gag epitope FLGKIWSPHK. B-E. Comparison of IFN-γ induction between wild type and 15 mutant forms of the HLA-A2 restricted Nef epitope FLKEKGGLEGL. F. Comparison of IFN-γ induction between wild type and a mutant form of the HLA-A2 restricted Pol epitope IYQYMDDLYV. G. Comparison of IFN-γ induction between wild type and a mutant form of the HLA-A2 restricted Pol epitope ILKEPVHGV. H. Comparison of IFN-γ induction between wild type and a mutant form of the HLA-A2 restricted Pol epitope LVGPTPVNII. I. Comparison of IFN-γ induction between wild type and a mutant form of the HLA-A2 restricted Pol epitope ALVEICTEM.
A3G/F-simulated Mutations Reduce the CD8+ T cell Response to the Majority of HLA-B44-restricted Epitopes
Recognition of 2 HLA-B44-restricted wt epitopes and corresponding muts in Gag was examined. Sixteen HLA-B44+ subjects were tested for CD8+ T cell responses against LSEGATPQDL and 3 muts. Five responded against the wt peptide (P35: 4245; P57: 712; P67: 775; P133: 1137; P242: 610 SFC/106 PBMC) and 2 had lesser responses against the G to R mut (P35: 3645; P57: 192 SFC/106 PBMC) (Figure 3a). Five subjects were tested for CD8+ T cell recognition of the Nef epitope KEKGGLEGL and 5 muts, 4 of whom responded to wt only (Figure 3b). Overall, 9 subjects responded to 2 epitopes restricted to HLA-B44. Of these, 7 responded only to wt, whilst 2 responded to both wt and mut. In all cases, the response to wt was significantly higher. Therefore, for the HLA-B44-restricted epitopes and subjects we studied, introduction of signature A3G/F mutations into the epitopes consistently abrogated or reduced CD8+ T cell recognition. In order to evaluate the statistical significance of our results, we conducted a Wilcoxon signed-rank test, considering the Elispot results obtained for all HLA-57, A2- and B44- restricted wt/mut pairs. We obtained a p-value of 0.0001, which supports statistical significance of our finding that the response to mut epitopes was diminished as a result of APOBEC-simulated mutations.
Responses of PBMC from HLA-matched subjects to wild type and mutant epitopes are shown on the Y-axis as IFN-γ SFC/106 PBMC. A. Comparison of IFN-γ induction between wild type and 3 mutant forms of the HLA-B44 restricted Gag epitope LSEGATPQDL. B. Comparison of IFN-γ induction between wild type and 6 mutant forms of the HLA-B44 restricted Nef epitope KEKGGLEGL.
Analysis of A3G/F Hotspot Frequency inside CTL Epitopes
The ex vivo reduction of HIV-specific CD8+ T cell responses by A3G/F mutations introduced into common HLA-A2, B44 and B57-restricted epitopes suggests that it would be advantageous for HIV to adapt towards maximizing A3G/F hotspots in genomic sequences encoding highly immunogenic or broadly presented CTL epitopes. To test this, we calculated the ratio (R) of A3G/F hotspot frequency inside vs. outside CTL epitopes within each HIV gene, normalizing for sequence length. We considered the frequencies of all A3G/F hotspots, but because A3G mutates HIV several fold more potently than A3F, and CCC (GGG on the plus DNA strand) is its preferentially targeted motif [13], [27], [63], [64], we also considered the frequency of GGG independently (table S1). Relative enrichment of hotspots inside vs. outside CTL epitopes for each gene is reflected by an index that constitutes the ratio of the R value of each gene to the average R of all genes. Considering the index values for all hotspots, as well as GGG alone, Gag, Pol and Nef exhibited the highest indices (for GGG: 1.3, 2.1 and 2.0 respectively) (Figure 4a).
A. For each HIV gene, the frequency of A3G/F hotspots (either all hotspots: GGG, GGA, GGT, GAA, or GGG alone) was calculated in genomic sequences that encode CTL epitopes and sequences that do not encode for CTL epitopes. Normalized frequencies were calculated by taking into account the total nucleotide length of the sequence. For each HIV gene, the normalized frequency of A3G/F hotspots inside sequences that encode CTL motifs was divided by the normalized frequency of A3G/F hotspots in sequences that do not encode CTL epitopes to generate a Ratio (R-value). The average of R-values amongst all HIV genes was determined and each R-value was divided by the average to obtain the index value, which reflects enrichment of hotspots within CTL epitopes, relative to other HIV genes. B. The same analysis as panel A was conducted within each HIV gene to compare the relative enrichment of A3G/F hotspots in sequences that encode CTL epitopes restricted to different HLA alleles. C. Comparison of A3G/F enrichment in CTL epitopes vs. the portions of the Env gene that encode antibody recognition epitopes. Average R-value for the CTL epitopes of all genes (average of all R values from panel A) was compared to the R value of A3G/F hotspots inside/outside sequences encoding antibody recognition motifs within Env. Index value of 1 is shown as a scale because an index >1 indicates relative enrichment.
We next addressed the suggestion that enrichment of A3G/F hotspots amongst various epitopes in each gene correlates with immunogenicity, as defined by the ability to elicit a robust (or broad, at the population level) CTL response. To examine this, we categorized epitopes on the basis of their restricting HLA (A1, A2, A3, A11, A24, B7, B8, B35, B44, B57). For each gene, we tallied hotspots inside vs. outside epitopes restricted to each HLA allele (Figure 4b). We observed a strong trend. For genes with the highest number of experimentally verified CTL epitopes (Gag, Pol, Env and Nef), the indices for HLA-B57 were >1, with GGG being over-represented by 1.7- to 4.9-fold. In contrast, the frequency of GGG motifs in HLA-B35 was >2-fold lower than that of other HLA-B alleles. The most striking enrichment was observed in Nef, where all A3G/F hotspots and GGG were respectively 5.7- and 17.6-fold enriched in B44-restricted as compared to B35-restricted epitopes. Nef contains a single B44-restricted epitope that induces robust CTL responses in a broad proportion of the population [32]. We noted that its coding sequence can potentially generate >20 muts. We also noted a higher frequency of GGG motifs in A3-restricted epitopes of Gag and Pol (2.1- to 2.3-fold enriched over epitopes restricted to other HLAs). These results may reflect the evolution of the viral genome in line with a higher number of CTL epitopes in these gene segments as well as the prevalence of the A3 allele in the population.
In order to test if there is broad A3G/F hotspot enrichment within CTL epitopes, we compared the R of A3G/F hotspots inside to outside CTL epitopes of each gene to the R of hotspots inside to outside antibody epitopes in Env, which we considered as a control. The R for antibody epitopes in Env is 0.71 and 0.9 for all A3G/F hotspots or GGG, respectively. These are 1.3- to 3-fold lower than the R values for CTL epitopes of Pol (1.06 and 1.63), Gag (1.16 and 2.60) and Nef (0.89 and 2.45), but interestingly, up to 7-fold higher than the R values found in the CTL motifs of other HIV genes (Vif, Vpr, Tat, Rev and Vpu). The average R value of GGG inside to outside CTL epitopes in all genes is 2-fold higher than its counterpart for Ab epitopes across all genes (Figure 4c). Collectively, this analysis suggests 2 levels of enrichment of A3G/F hotspots of significance for viral escape from the CTL response: first, enrichment in CTL epitopes of Gag, Pol and Nef, and second, enrichment in epitopes within each gene restricted to HLAs that present more immunogenic epitopes (e.g. HLA-B57) and HLAs that occur more broadly at the population level (e.g. HLA-B44, A3).
Discussion
The anti-HIV response by CD8+ CTL plays a key role in controlling viremia as illustrated by the correlation between disease progression and class I HLA genotype [47], [49], [65]–[68]. A mouse model study found that A3G-induced mutations can potentially improve the CTL response against HIV-infected cells in vitro by producing truncated peptides that are more effectively processed and presented [42]. In contrast, another study proposed that A3G hotspots in CTL escape sites reduce CTL recognition of HIV [41]. It is now appreciated that A3G/F may benefit HIV by contributing to drug resistance and other contextual fitness-enhancing mutations [19], [69], [70]. Here we show that A3G/F mutations consistently compromise adaptive immune responses in HIV-infected individuals by favoring CTL escape. Nevertheless, since we did not directly test the effect of random mutations in this study, we cannot conclude that A3G/F mutations specifically diminish the CTL response, nor can we extend our observations beyond the known set of CTL epitopes that have been identified in HIV, to date. However, we note that multiple previous studies of the CD8+ T cell response in HIV+ cohorts to epitope variants adapted at the population level found that >50% of mut epitopes including hypothetical variants (i.e. like our muts: not sequenced from the HIV+ individual whose CTL response was tested) retained their ability to stimulate CD8+ T cells [35], [44], [45].
Several non-mutually exclusive scenarios can account for our observations that in our system the simulated A3G/F-induced reduced CD8+ T cell recognition of HIV. First, it is possible that some in silico-simulated A3G/F mutations may not have arisen in the subjects we tested. This is unlikely since we verified that all of our mut epitopes have been reported at the population level in a database of HIV CTL epitope variants (http://www.hiv.lanl.gov/content/sequence/QUICK_ALIGN/QuickAlign.html). Even if the occurrence of the mutation elicited a response in as low as ∼1 in 10,000 CD8+ T cells, it should be detectable by ELISPOT. It is possible that the decrease in CTL response to mut epitopes was a product of our particular experimental procedures. Our finding that several mut epitopes within Pol were equally well recognized (Figure 2f, g) argues against this. Finally, it is possible that our observations reflect the physiological effect of A3G/F-induced mutations within CTL epitopes. In support of this, our observation that A3G/F hotspots are selectively enriched in CTL epitope-encoding portions of the HIV genome indicates that the reduced CTL recognition of A3G/F-mutated epitopes represents an in vivo phenomenon pronounced enough to leave an evolutionary footprint on the viral genome.
The most direct interpretation of our data is that the majority of A3G/F mutations within CD8+ T cell epitopes reduce peptide recognition by CD8+ T cells originally activated by the wild type HIV peptide epitope. Cross-reactive recognition of the mutated peptide seems unlikely in the case of non-conservative amino acid substitutions associated with A3G/F mutations such as G to R. If the mutated peptide were itself immunogenic and present at high enough levels, then a non-cross-reactive set of CD8+ T cells would be expected to arise against the new epitope. The reduced but apparent reactivity with some mutated epitopes could thus stem either from partial cross-reactivity of CD8+ T cells activated by the wild type epitope or from a smaller, non-overlapping CD8+ T cell response selectively activated by the new, A3G/F-mutated epitope. Studies of the T cell response at the clonal level and of endogenous viral sequences found in each HIV+ individual would be required to further address the issue of cross-reactivity vs. activation of a non-overlapping set of T cells against a de novo A3G/F-derived epitope. Based on previous studies, cross-reactivity is quite likely in the case of conservative or semi-conservative mutations introduced into epitopes by A3G/F [35]. Rare conservative A3G/F-induced mutations, such as the M to I mutation in the HLA A2-restricted epitope IYQYMDDLYV, were more likely to elicit comparable responses. It is probable that A3G/F-mutated epitopes are less efficient at activating CTL “selected” to recognize the wt epitope due to non-conservative mutations and the effect thereof on recognition by the T cell receptor. In the setting within which we tested the impact of A3G/F mutations on activation of CD8+ T cells from HIV-infected individuals, reduction or abrogation of CD8+ T cell activation was the rule.
The significance of our results lies in demonstration of viral access to multiple avenues of mutational escape from immune recognition. Infidelity and lack of proof-reading for reverse transcription results in 1 to 2 mutations/viral genome/round of replication [71]–[75]. While this level of mutation ensures that escape mutations occur at the level of the overall virus population, it would still leave the majority of viruses susceptible until selective expansion of the mutated virus allowed it to become predominant. The need to achieve expeditious escape from a diverse T cell response raises a higher mutational barrier. In this context, A3G is a highly processive enzyme that by conservative estimates, can introduce tens of mutations in a viral genome in a single replication round, and hundreds of mutations in a viral genome in a replication round when its activity is unhindered by Vif [13], [76]. Indeed, turning off (and likely also turning down) the expression and/or A3G-restricting effectiveness of Vif has been shown to be a means by which HIV fine-tunes the increase in A3G activity in order to overcome other selective pressures such as anti-viral drugs [77], [78]. Taken together with selective positioning of A3G motifs within CD8+ T cell epitopes over most other portions of the HIV genome, this provides HIV with a substantially more robust mechanism of generating mutations that affect immune recognition. Furthermore, the specificity of A3G/F, which often led to non-conservative amino acid changes in CTL epitopes that we examined, positively skews the likelihood of mutation contributing to immune escape.
Our findings contrast with those of a recent mouse model study showing that A3G mutations promote the CTL response through introducing stop codons leading to the generation of truncated proteins [42]. This study demonstrated the potential of A3G/F to increase HIV immunogenicity through an indirect mechanism involving mutation of sequences outside of actual CTL epitopes, which modulates the efficacy of epitope production and/or presentation. Here we report that the direct action of A3G/F within sequences encoding CD8+ T cell epitopes most often has the opposite effect in HIV-infected individuals. Whilst the mouse study utilized transgenic T cell receptor-bearing T cells in a well defined animal model and elegantly showed that APOBEC action can impact the CTL response, the work presented here takes into account a diverse set of human subjects with respect to differences in HLA makeup, anti-viral drug treatment and stage of HIV infection. In this context, we note that our data reflect a general physiological effect of A3G/F-induced mutations within CTL epitopes for the cohort of HIV-infected subjects we studied. What exceptions there may be at the individual level remains to be tested. Our observation that A3G/F hotspots are enriched in CTL epitope-encoding sequences of the HIV genome argues that the magnitude of CTL escape as a result of A3G/F-induced mutations is significant enough to have left its evolutionary footprint on the viral genome. Further work is required to shed light on the molecular and cellular pathways through which these DNA-mutating enzymes alter CTL recognition of infected cells, as well as to better understand the role of A3G/F-induced mutations outside CTL epitopes in affecting the processing and presentation of HIV peptides to T cells.
Supporting Information
Figure S1.
Delineation of the amino acid and DNA sequences of CTL epitopes. A. locations of CTL epitopes in Gag, in the HIV-1 Bru isolate sequence of HIV. The brackets indicate epitopes on the peptide sequence. Epitopes are shown in different colors depending on HLA allele restriction, which is indicated above the bracket. Dashed lines display epitopes that are devoid of A3G/F hotspots and continuous lines show epitopes that harbor A3G/F hotspots. B. locations of the viral genomic sequences that encode CTL epitopes in Gag in the HIV-1 Bru isolate plus-sense strand. Colors delineate the sequences encoding CTL epitopes from the surrounding DNA, but do not otherwise correspond to any common features amongst genes or epitopes. C. Simulation of A3G/F-mediated mutations of CTL epitopes. Wild-type epitopes and their encoding DNA sequence are shown. A3G/F targeting hotspots in the plus-sense epitope encoding sequence are underlined and colored in blue. Simulated A3G/F-mediated G to A mutations in the viral genomic sequence and the resulting amino acid mutations in CTL epitopes are shown below in red. Left panel shows a typical epitope with a single possible mutation. Middle panel shows a typical epitope with Multiple independent mutations. Hotspots separated by >3 nucleotides were considered as independent and mutant epitopes bearing either mutation or combinations of multiple mutations were considered. Right panel shows a typical epitope with multiple sequential mutations. For hotspots where an initial A3G/F-mediated mutation can generate a new A3G/F hotspot which may be mutated in the same or a subsequent replication cycle, variants bearing different combinations of mutations were considered.
https://doi.org/10.1371/journal.pone.0093428.s001
(PDF)
Figure S2.
Map of all HIV-1 Bru isolate CTL epitopes across the viral proteome. HIV proteins are shown individually (A: Pol, B: Rev, C: Vif, D: Nef, E: Vpr, F: Env, G: Gag, H: Tat). The brackets indicate epitopes on the peptide sequence. Epitopes restricted to different HLA alleles are shown in different colors and the restricting HLA is indicated above the bracket. Broken lines display epitopes with no A3G/F hotspots and continuous lines show epitopes that harbor A3G/F hotspots.
https://doi.org/10.1371/journal.pone.0093428.s002
(PDF)
Figure S3.
Map of the viral genomic sequences that encode CTL epitopes in the HIV-1 Bru isolate plus-sense coding DNA. Sequence of each gene is shown (A: Pol, B: Rev, C: Vif, D: Nef, E: Vpr, F: Env, G: Gag, H: Tat). Colors differentiate sequences encoding CTL epitopes from the surrounding DNA. Colors do not otherwise correspond to any common features amongst genes or epitopes.
https://doi.org/10.1371/journal.pone.0093428.s003
(PDF)
Table S1.
Relative enrichment of A3G/F hotspot inside/outside CTL epitopes in HIV genes. For each HIV gene, the frequency of A3G/F hotspots (either all hotspots: GGG, GGA, GGT, GAA, or GGG alone) was calculated in genomic sequences that encode for CTL epitopes and those that do not encode for CTL epitopes. Normalized frequencies were calculated by taking into account the total nucleotide length of the sequence. For each HIV gene, the normalized frequency of A3G/F hotspots inside sequences that encode CTL motifs was divided by the normalized frequency of A3G/F hotspots in sequences that do not encode CTL epitopes to generate a Ratio (R-value). The average of R-values was determined and each R-value was divided by the average to obtain the index value and allow for determination of the relative enrichment of hotspots within the CTL epitopes of each gene, as compared to the average ratio of hotspots inside to outside CTL epitopes. An index value >1 indicates relative enrichment.
https://doi.org/10.1371/journal.pone.0093428.s004
(PDF)
Author Contributions
Conceived and designed the experiments: ML MDG MM. Performed the experiments: MM. Analyzed the data: MM ML MDG. Contributed reagents/materials/analysis tools: KZ MG. Wrote the paper: ML CW MDG MM.
References
- 1. Sheehy AM, Gaddis NC, Choi JD, Malim MH (2002) Isolation of a human gene that inhibits HIV-1 infection and is suppressed by the viral Vif protein. Nature 418: 646–650.
- 2. Harris RS, Petersen-Mahrt SK, Neuberger MS (2002) RNA editing enzyme APOBEC1 and some of its homologs can act as DNA mutators. Mol Cell 10: 1247–1253.
- 3. Wedekind JE, Dance GS, Sowden MP, Smith HC (2003) Messenger RNA editing in mammals: new members of the APOBEC family seeking roles in the family business. Trends Genet 19: 207–216.
- 4. Bishop KN, Holmes RK, Sheehy AM, Davidson NO, Cho SJ, et al. (2004) Cytidine deamination of retroviral DNA by diverse APOBEC proteins. Curr Biol 14: 1392–1396.
- 5. Liddament MT, Brown WL, Schumacher AJ, Harris RS (2004) APOBEC3F properties and hypermutation preferences indicate activity against HIV-1 in vivo. Curr Biol 14: 1385–1391.
- 6. Chiu YL, Soros VB, Kreisberg JF, Stopak K, Yonemoto W, et al. (2005) Cellular APOBEC3G restricts HIV-1 infection in resting CD4+ T cells. Nature 435: 108–114.
- 7. Peng G, Lei KJ, Jin W, Greenwell-Wild T, Wahl SM (2006) Induction of APOBEC3 family proteins, a defensive maneuver underlying interferon-induced anti-HIV-1 activity. J Exp Med 203: 41–46.
- 8. Peng G, Greenwell-Wild T, Nares S, Jin W, Lei KJ, et al. (2007) Myeloid differentiation and susceptibility to HIV-1 are linked to APOBEC3 expression. Blood 110: 393–400.
- 9. Pido-Lopez J, Whittall T, Wang Y, Bergmeier LA, Babaahmady K, et al. (2007) Stimulation of cell surface CCR5 and CD40 molecules by their ligands or by HSP70 up-regulates APOBEC3G expression in CD4(+) T cells and dendritic cells. J Immunol 178: 1671–1679.
- 10. Stopak KS, Chiu YL, Kropp J, Grant RM, Greene WC (2007) Distinct patterns of cytokine regulation of APOBEC3G expression and activity in primary lymphocytes, macrophages, and dendritic cells. J Biol Chem 282: 3539–3546.
- 11. Koning FA, Newman EN, Kim EY, Kunstman KJ, Wolinsky SM, et al. (2009) Defining APOBEC3 expression patterns in human tissues and hematopoietic cell subsets. J Virol 83: 9474–9485.
- 12. Pion M, Granelli-Piperno A, Mangeat B, Stalder R, Correa R, et al. (2006) APOBEC3G/3F mediates intrinsic resistance of monocyte-derived dendritic cells to HIV-1 infection. J Exp Med 203: 2887–2893.
- 13. Yu Q, Konig R, Pillai S, Chiles K, Kearney M, et al. (2004) Single-strand specificity of APOBEC3G accounts for minus-strand deamination of the HIV genome. Nat Struct Mol Biol 11: 435–442.
- 14. Lecossier D, Bouchonnet F, Clavel F, Hance AJ (2003) Hypermutation of HIV-1 DNA in the absence of the Vif protein. Science 300: 1112.
- 15. Mangeat B, Turelli P, Caron G, Friedli M, Perrin L, et al. (2003) Broad antiretroviral defence by human APOBEC3G through lethal editing of nascent reverse transcripts. Nature 424: 99–103.
- 16. Zhang H, Yang B, Pomerantz RJ, Zhang C, Arunachalam SC, et al. (2003) The cytidine deaminase CEM15 induces hypermutation in newly synthesized HIV-1 DNA. Nature 424: 94–98.
- 17. Klarmann GJ, Chen X, North TW, Preston BD (2003) Incorporation of uracil into minus strand DNA affects the specificity of plus strand synthesis initiation during lentiviral reverse transcription. J Biol Chem 278: 7902–7909.
- 18. Holmes RK, Koning FA, Bishop KN, Malim MH (2007) APOBEC3F can inhibit the accumulation of HIV-1 reverse transcription products in the absence of hypermutation. Comparisons with APOBEC3G. J Biol Chem 282: 2587–2595.
- 19. Monajemi M, Woodworth CF, Benkaroun J, Grant M, Larijani M (2012) Emerging complexities of APOBEC3G action on immunity and viral fitness during HIV infection and treatment. Retrovirology 9: 35.
- 20. Belanger K, Savoie M, Rosales Gerpe MC, Couture JF, Langlois MA (2013) Binding of RNA by APOBEC3G controls deamination-independent restriction of retroviruses. Nucleic Acids Res 41: 7438–7452.
- 21. Yu XF (2006) Innate cellular defenses of APOBEC3 cytidine deaminases and viral counter-defenses. Curr Opin HIV AIDS 1: 187–193.
- 22. Santa-Marta M, Aires da Silva F, Fonseca AM, Rato S, Goncalves J (2007) HIV-1 Vif protein blocks the cytidine deaminase activity of B-cell specific AID in E. coli by a similar mechanism of action. Mol Immunol 44: 583–590.
- 23. Sheehy AM, Gaddis NC, Malim MH (2003) The antiretroviral enzyme APOBEC3G is degraded by the proteasome in response to HIV-1 Vif. Nat Med 9: 1404–1407.
- 24. Mehle A, Strack B, Ancuta P, Zhang C, McPike M, et al. (2004) Vif overcomes the innate antiviral activity of APOBEC3G by promoting its degradation in the ubiquitin-proteasome pathway. J Biol Chem 279: 7792–7798.
- 25. Chiu YL, Greene WC (2009) APOBEC3G: an intracellular centurion. Philos Trans R Soc Lond B Biol Sci 364: 689–703.
- 26. Wichroski MJ, Robb GB, Rana TM (2006) Human retroviral host restriction factors APOBEC3G and APOBEC3F localize to mRNA processing bodies. PLoS Pathog 2: e41.
- 27. Chiu YL (2011) Biochemical fractionation and purification of high-molecular-mass APOBEC3G complexes. Methods Mol Biol 718: 185–206.
- 28. Gallois-Montbrun S, Kramer B, Swanson CM, Byers H, Lynham S, et al. (2007) Antiviral protein APOBEC3G localizes to ribonucleoprotein complexes found in P bodies and stress granules. J Virol 81: 2165–2178.
- 29. Russell RA, Moore MD, Hu WS, Pathak VK (2009) 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. Retrovirology 6: 16.
- 30. Koup RA, Safrit JT, Cao Y, Andrews CA, McLeod G, et al. (1994) Temporal association of cellular immune responses with the initial control of viremia in primary human immunodeficiency virus type 1 syndrome. J Virol 68: 4650–4655.
- 31. Borrow P, Lewicki H, Hahn BH, Shaw GM, Oldstone MB (1994) Virus-specific CD8+ cytotoxic T-lymphocyte activity associated with control of viremia in primary human immunodeficiency virus type 1 infection. J Virol 68: 6103–6110.
- 32. Perez CL, Larsen MV, Gustafsson R, Norstrom MM, Atlas A, et al. (2008) Broadly immunogenic HLA class I supertype-restricted elite CTL epitopes recognized in a diverse population infected with different HIV-1 subtypes. J Immunol 180: 5092–5100.
- 33. Shankar P, Russo M, Harnisch B, Patterson M, Skolnik P, et al. (2000) Impaired function of circulating HIV-specific CD8(+) T cells in chronic human immunodeficiency virus infection. Blood 96: 3094–3101.
- 34. Harrer T, Harrer E, Kalams SA, Elbeik T, Staprans SI, et al. (1996) Strong cytotoxic T cell and weak neutralizing antibody responses in a subset of persons with stable nonprogressing HIV type 1 infection. AIDS Res Hum Retroviruses 12: 585–592.
- 35. Almeida CA, Bronke C, Roberts SG, McKinnon E, Keane NM, et al. (2011) Translation of HLA-HIV associations to the cellular level: HIV adapts to inflate CD8 T cell responses against Nef and HLA-adapted variant epitopes. J Immunol 187: 2502–2513.
- 36. Leslie AJ, Pfafferott KJ, Chetty P, Draenert R, Addo MM, et al. (2004) HIV evolution: CTL escape mutation and reversion after transmission. Nat Med 10: 282–289.
- 37. Leslie A, Kavanagh D, Honeyborne I, Pfafferott K, Edwards C, et al. (2005) Transmission and accumulation of CTL escape variants drive negative associations between HIV polymorphisms and HLA. J Exp Med 201: 891–902.
- 38. Geels MJ, Cornelissen M, Schuitemaker H, Anderson K, Kwa D, et al. (2003) Identification of sequential viral escape mutants associated with altered T-cell responses in a human immunodeficiency virus type 1-infected individual. J Virol 77: 12430–12440.
- 39. Borrow P, Lewicki H, Wei X, Horwitz MS, Peffer N, et al. (1997) Antiviral pressure exerted by HIV-1-specific cytotoxic T lymphocytes (CTLs) during primary infection demonstrated by rapid selection of CTL escape virus. Nat Med 3: 205–211.
- 40. Barouch DH, Letvin NL (2002) Viral evolution and challenges in the development of HIV vaccines. Vaccine 20 Suppl 4A66–68.
- 41. Wood N, Bhattacharya T, Keele BF, Giorgi E, Liu M, et al. (2009) HIV evolution in early infection: selection pressures, patterns of insertion and deletion, and the impact of APOBEC. PLoS Pathog 5: e1000414.
- 42. Casartelli N, Guivel-Benhassine F, Bouziat R, Brandler S, Schwartz O, et al. (2010) The antiviral factor APOBEC3G improves CTL recognition of cultured HIV-infected T cells. J Exp Med 207: 39–49.
- 43. Alizon M, Wain-Hobson S, Montagnier L, Sonigo P (1986) Genetic variability of the AIDS virus: nucleotide sequence analysis of two isolates from African patients. Cell 46: 63–74.
- 44. Mason RD, Bowmer MI, Howley CM, Gallant M, Myers JC, et al. (2004) Antiretroviral drug resistance mutations sustain or enhance CTL recognition of common HIV-1 Pol epitopes. J Immunol 172: 7212–7219.
- 45. Gladney KH, Pohling J, Hollett NA, Zipperlen K, Gallant ME, et al. (2012) Heteroclitic peptides enhance human immunodeficiency virus-specific CD8(+) T cell responses. Vaccine 30: 6997–7004.
- 46. Pohling J, Zipperlen K, Hollett NA, Gallant ME, Grant MD (2010) Human immunodeficiency virus type I-specific CD8+ T cell subset abnormalities in chronic infection persist through effective antiretroviral therapy. BMC Infect Dis 10: 129.
- 47.
Brennan CA, Ibarrondo FJ, Sugar CA, Hausner MA, Shih R, et al. (2012) Early HLA-B*57-Restricted CD8+ T Lymphocyte Responses Predict HIV-1 Disease Progression. J Virol.
- 48. Cao K, Hollenbach J, Shi X, Shi W, Chopek M, et al. (2001) Analysis of the frequencies of HLA-A, B, and C alleles and haplotypes in the five major ethnic groups of the United States reveals high levels of diversity in these loci and contrasting distribution patterns in these populations. Hum Immunol 62: 1009–1030.
- 49. Altfeld M, Kalife ET, Qi Y, Streeck H, Lichterfeld M, et al. (2006) HLA Alleles Associated with Delayed Progression to AIDS Contribute Strongly to the Initial CD8(+) T Cell Response against HIV-1. PLoS Med 3: e403.
- 50. Flores-Villanueva PO, Yunis EJ, Delgado JC, Vittinghoff E, Buchbinder S, et al. (2001) Control of HIV-1 viremia and protection from AIDS are associated with HLA-Bw4 homozygosity. Proc Natl Acad Sci U S A 98: 5140–5145.
- 51. Rohowsky-Kochan C, Skurnick J, Molinaro D, Louria D (1998) HLA antigens associated with susceptibility/resistance to HIV-1 infection. Hum Immunol 59: 802–815.
- 52. Pelak K, Goldstein DB, Walley NM, Fellay J, Ge D, et al. (2010) Host determinants of HIV-1 control in African Americans. J Infect Dis 201: 1141–1149.
- 53. Migueles SA, Sabbaghian MS, Shupert WL, Bettinotti MP, Marincola FM, et al. (2000) HLA B*5701 is highly associated with restriction of virus replication in a subgroup of HIV-infected long term nonprogressors. Proc Natl Acad Sci U S A 97: 2709–2714.
- 54. Lambotte O, Boufassa F, Madec Y, Nguyen A, Goujard C, et al. (2005) HIV controllers: a homogeneous group of HIV-1-infected patients with spontaneous control of viral replication. Clin Infect Dis 41: 1053–1056.
- 55. Fellay J, Shianna KV, Ge D, Colombo S, Ledergerber B, et al. (2007) A whole-genome association study of major determinants for host control of HIV-1. Science 317: 944–947.
- 56. Fellay J, Ge D, Shianna KV, Colombo S, Ledergerber B, et al. (2009) Common genetic variation and the control of HIV-1 in humans. PLoS Genet 5: e1000791.
- 57. Emu B, Sinclair E, Hatano H, Ferre A, Shacklett B, et al. (2008) HLA class I-restricted T-cell responses may contribute to the control of human immunodeficiency virus infection, but such responses are not always necessary for long-term virus control. J Virol 82: 5398–5407.
- 58. Poropatich K, Sullivan DJ Jr (2011) Human immunodeficiency virus type 1 long-term non-progressors: the viral, genetic and immunological basis for disease non-progression. J Gen Virol 92: 247–268.
- 59. Sette A, Sidney J (1999) Nine major HLA class I supertypes account for the vast preponderance of HLA-A and -B polymorphism. Immunogenetics 50: 201–212.
- 60. Brumme ZL, John M, Carlson JM, Brumme CJ, Chan D, et al. (2009) HLA-associated immune escape pathways in HIV-1 subtype B Gag, Pol and Nef proteins. PLoS One 4: e6687.
- 61. Novitsky V, Cao H, Rybak N, Gilbert P, McLane MF, et al. (2002) Magnitude and frequency of cytotoxic T-lymphocyte responses: identification of immunodominant regions of human immunodeficiency virus type 1 subtype C. J Virol. 76: 10155–10168.
- 62. Karlsson AC, Iversen AK, Chapman JM, de Oliviera T, Spotts G, et al. (2007) Sequential broadening of CTL responses in early HIV-1 infection is associated with viral escape. PLoS One 2: e225.
- 63. Bishop KN, Holmes RK, Malim MH (2006) Antiviral potency of APOBEC proteins does not correlate with cytidine deamination. J Virol 80: 8450–8458.
- 64. Wang X, Dolan PT, Dang Y, Zheng YH (2007) Biochemical differentiation of APOBEC3F and APOBEC3G proteins associated with HIV-1 life cycle. J Biol Chem 282: 1585–1594.
- 65. Betts MR, Krowka JF, Kepler TB, Davidian M, Christopherson C, et al. (1999) Human immunodeficiency virus type 1-specific cytotoxic T lymphocyte activity is inversely correlated with HIV type 1 viral load in HIV type 1-infected long-term survivors. AIDS research and human retroviruses 15: 1219–1228.
- 66. Huang XL, Fan Z, Kalinyak C, Mellors JW, Rinaldo CR Jr (2000) CD8+ T-cell gamma interferon production specific for human immunodeficiency virus type 1 (HIV-1) in HIV-1-infected subjects. Clinical and diagnostic laboratory immunology 7: 279–287.
- 67. Altfeld M, Rosenberg ES, Shankarappa R, Mukherjee JS, Hecht FM, et al. (2001) Cellular immune responses and viral diversity in individuals treated during acute and early HIV-1 infection. J Exp Med 193: 169–180.
- 68. Letvin NL, Walker BD (2001) HIV versus the immune system: another apparent victory for the virus. Journal of Clinical Investigation 107: 273–275.
- 69. Smith HC (2011) APOBEC3G: a double agent in defense. Trends Biochem Sci 36: 239–244.
- 70. Sadler HA, Stenglein MD, Harris RS, Mansky LM (2010) APOBEC3G contributes to HIV-1 variation through sublethal mutagenesis. J Virol 84: 7396–7404.
- 71. Roberts JD, Bebenek K, Kunkel TA (1988) The accuracy of reverse transcriptase from HIV-1. Science 242: 1171–1173.
- 72. Ji J, Loeb LA (1994) Fidelity of HIV-1 reverse transcriptase copying a hypervariable region of the HIV-1 env gene. Virology 199: 323–330.
- 73. Mansky LM, Temin HM (1995) Lower in vivo mutation rate of human immunodeficiency virus type 1 than that predicted from the fidelity of purified reverse transcriptase. J Virol 69: 5087–5094.
- 74. Preston BD, Poiesz BJ, Loeb LA (1988) Fidelity of HIV-1 reverse transcriptase. Science 242: 1168–1171.
- 75. Yamanaka R, Termini J (2007) Nucleotide sequence context influences HIV replication fidelity by modulating reverse transcriptase binding and product release. Biosci Trends 1: 52–61.
- 76. Chelico L, Pham P, Calabrese P, Goodman MF (2006) APOBEC3G DNA deaminase acts processively 3′–>5′ on single-stranded DNA. Nat Struct Mol Biol 13: 392–399.
- 77. Fourati S, Malet I, Binka M, Boukobza S, Wirden M, et al. (2010) Partially active HIV-1 Vif alleles facilitate viral escape from specific antiretrovirals. Aids 24: 2313–2321.
- 78. Yebra G, Holguin A (2011) Mutation Vif-22H, which allows HIV-1 to use the APOBEC3G hypermutation to develop resistance, could appear more quickly in certain non-B variants. J Antimicrob Chemother 66: 941–942.