Click on the links below to access all the ArticlePlus for this article.
Please note that ArticlePlus files may launch a viewer application outside of your web browser.
In the absence of antiretroviral therapy, the majority of HIV-infected individuals are unable to adequately control viral replication. This results in persistent viremia, a gradual loss in CD4+ T-cell number, and ultimately immune suppression. Although the correlates of protective immunity remain undefined, HIV-specific cellular immune responses seem to play a critical role in controlling the virus.1 It is generally accepted that cytotoxic T lymphocytes (CTLs) are responsible for viral containment during acute infection, as the decline in HIV plasma viral load is temporally associated with the development of virus-specific CTL.2,3 CD4+ T helper cell (TH) responses are also presumed critical to viral control. During acute infection, robust TH cell responses directed toward the HIV proteins Gag and Nef can be detected in the majority of infected individuals,4 and during untreated chronic infection there is a negative association between Gag-specific CD4+ T-cell responses and plasma HIV RNA viral load.5 Furthermore, the absolute number and percent of Gag-specific CD4+ interferon-gamma+ (IFN-γ+) interleukin-2+ (IL-2+) cells is negatively associated with viral load and significantly higher in HIV-infected individuals able to spontaneously control viral replication.6
HIV infection is characterized not only by a decline in CD4+ T-cell number, but also a loss of TH cell function, the cause of which is still undefined. Some possible mechanisms for the virus-specific CD4+ T-cell impairment include direct or indirect cell death as a result of infection, virus-induced anergy, and ineffective antigen presentation. In addition, sequence variation within virus-encoded CD4 epitopes may contribute to the dysfunction. It has been established that single amino acid substitutions within a CD4 epitope can cause partial T-cell activation, resulting in the loss of some, but not all, effector functions.7 Due to the error-prone nature of HIV replication, changes in the viral sequence are common, especially during periods of high viral replication. We hypothesize that during acute HIV infection, when viral replication is at its peak, sequence variation within CD4 epitopes could result in the development of suboptimal CD4+ T-cell responses and may be partially responsible for the loss of TH function that is characteristic of progressive HIV infection. Loss of CD4+ T-cell function as a result of natural variation within CD4 epitopes has been observed in the context of several infectious diseases, including hepatitis C virus.8-18 Further, it has been speculated that sequence changes within CD4 epitopes could represent TH escape mutations. This has been described in murine lymphocytic choriomeningitis virus (LCMV) infection, where CD8+ T-cell-deficient mice with transgenic CD4+ T cells specific for an immunodominant major histocompatibility complex (MHC) class II, restricted LCMV epitope developed viral mutants that were not recognized by TH cells. Thus, under conditions in which immune pressure is mediated through a single CD4 specificity, in vivo viral sequence variation can result in escape from TH recognition. Sequence variation within HIV-specific CD4 epitopes and an associated loss of CD4+ T-cell function has been described in chronic HIV-infected individuals.13 However, these findings have not been corroborated by others.19
This study was designed to examine the extent to which sequence variation occurs in HIV-specific CD4 epitopes during early infection and to assess the possibility that mutational escape from HIV-specific TH responses can occur. To do this, we examined amino acid changes that occurred in Gag, Nef, and Integrase (Int) in 7 individuals identified during acute HIV infection. We then determined whether there were differences in the peptide-specific CD4+ T-cell response for areas where sequence variation was observed.
Seven HIV-1 infected subjects were chosen from a cohort of individuals with acute HIV infection from Massachusetts General Hospital (MGH) and local clinics in Boston. These subjects were chosen based on the availability of samples within 2 months of presentation in which there was sufficient viral RNA to perform sequencing. Acute HIV infection was defined by either a negative HIV-1/2 enzyme-linked immunosorbent assay (ELISA) or a positive ELISA with 2 or fewer bands in an HIV-1 Western blot plus the presence of detectable HIV-1 RNA. The MGH clinical laboratories performed all HIV RNA viral loads and CD4+ T-cell counts. This study was approved by the MGH Institutional Review Board, and all subjects gave informed consent before participation.
Human Leukocyte Antigen Typing
Genomic DNA was isolated from peripheral blood mononuclear cells (PBMCs) using the Puregene DNA isolation kit (Gentra Systems, Minneapolis, MN). HLA class I and II typing was performed at Royal Perth Hospital by the Department of Clinical Immunology and Biochemical Genetics (Perth, Australia) using sequence-specific primer PCR as previously described.20
Isolation of PBMCs and CD8 Depletion
PBMCs were isolated from whole blood within 24 hours of phlebotomy using a Ficoll-Hypaque (Sigma, St. Louis, MO) density gradient. The plasma layer was removed and stored at -80°C. CD8 depletion was performed with either RosetteSep Human CD8 Depletion (StemCell Technologies, Vancouver, Canada) or Dynabeads CD8 depletion (Dynal Biotech, Oslo, Norway) as per manufacturer's instructions. Depletion using either method results in ≥98% purity.
Synthetic HIV-1 Peptides
Peptides spanning HIV-1 Gag (peptides 1 to 66), Nef (peptides 67 to 93), and Int (peptides 244 to 277) were synthesized at the MGH Peptide Core facility on an automated peptide synthesizer (Advanced Chemtech, Louisville, KY). Each peptide contained 15 to 20 amino acids and overlapped the adjacent peptide by 10 amino acids. Peptide sequences were based on the 2001 HIV clade B consensus sequence obtained from the Los Alamos National Laboratory HIV Sequence Database.21
CD4+ T-cell responses were measured using an IFN-γ enzyme-linked immunosorbent spot (ELISPOT) assay as previously described.4 CD8-depleted PBMCs were stimulated using single peptides at a concentration of 4 μg/mL. Positive (phytohemagglutinin at 5 μg/mL) and negative (media alone) control wells were included on each plate. The number of spots per well was assessed using an automated plate reader (Series 3B CTL Analyzer; CTL Analyzers, Cleveland, OH). Results are expressed as the number of spot-forming cells (SFC) per million CD8-depleted PBMCs after subtracting out background, which averaged <7 SFC per input cells. A peptide-specific response was considered positive when it was >2.5 times background. A change in the response from time point 1 to time point 2 was considered significant if there was a >50% change and a difference of at least 50 SFC/million input cells.
Sequencing of Gag and Nef
Viral RNA was extracted from plasma using QIAmp Viral RNA Mini Kit (QIAgen, Valencia, CA). Complementary DNA (cDNA) was obtained using MuLV Reverse Trascriptase (Applied Biosystems, Foster City, CA) and oligo (dT) or gene-specific primers. Nested PCR was performed in at least duplicate independent amplifications of cDNA. A 2361-bp amplicon encompassing all of Gag and corresponding to bases 652 to 3012 of the HXB2 genome was obtained using Titanium Taq (Clonetech, Mountain View, CA) and the primers Gag 1F (AACAGGGACCTGAAAGCGAAG) and Gag 1R (CCCCACCTCAACAGATGTTGTC) for the first round and Gag 2F (ACCTGAAAGCGAAAGGGAAACC) and Gag 2R (CTTTCCATCCCTGTGGAAGCAC) for the second round.
Nef and int were obtained using AmpliTaq Gold (Applied Biosystems, Foster City, CA) and the following primers: Nef 7e-F (CGACAGGCCCGAAGGAATCGAAG), 7h-Rext 9664 (AGGGATCTCTAGTTACCAGAGTC), Nef 7f-F (CGAGGATTGTGGAACTTCTGGGACG), 7h-Rint 9606 (GACTTAAGGCAAGCTTTATTGAGGCTTA), Int 1F (GGTGGACAGAGTATTGGCAAGC), Int 1R (GAGGGAGCCACACAATGAATGG), Int 2F (AAGCCACCTGGATTCCTGAGTG), and Int 2R (CATTTGGGTCAGGGAGTCTCCA). This results in a 1073-bp Nef amplicon corresponding to bases 8560 to 9632 of the HXB2 genome and a 1531-bp Int amplicon corresponding to bases 3769 to 5299 of the HXB2 genome. Despite multiple attempts, we were unable to amplify Int from subject AC196 using the primers listed above. Therefore the following primer sets were used: F4a/b (YCTGGCATGGGTACCAGC), F5aR (CTCCCTGRCCYARATGCC), F3gF (GGTACCAGCACACAAAGG), and F4dR (ATGCCAKTCTCTTTCTCC), resulting in an 1127-bp amplicon corresponding to bases 4154 to 5280 of the HXB2 genome.
Purified PCR products (QIAquick gel extraction kit; QIAgen, Valencia, CA) were sequenced at the MGH Sequencing Core using an ABI 3730XL DNA Analyzer. In addition to the forward and reverse primers described above, the following internal primers were also used: Gag 1eF (TAATCCACCTATCCCAGTAGG), Gag 1dF (TCTGGGTTCGMATTTTGGACC), Nef 7gF (GCAATAAGATGGGTGGCAAGTGG), and Int 2b (CAGCACACAAAGGAATTGGAGG).
Nucleotide sequences were assembled and edited using Aligner (CodonCode, Dedham, MA). Deduced amino acid sequences for all subjects were aligned to the 2001 Consensus B sequence obtained from the Los Alamos National Laboratory HIV Sequence Database21 using DSGene (Accelrys, San Diego, CA) and prepared for publication using Seqpublish.21 All sequences are available through GenBank (accession numbers DQ996244-DQ996267 and EF119598-EF119615).
The baseline characteristics of the study subjects are given in Tables 1 and 2. All 7 subjects were men with an average age of 36 years (range, 27 to 48 years old). Seroconversion was confirmed for 5 of the 7 subjects during the time period analyzed. The average HIV RNA viral load at the first available time point was 9.9 × 106 RNA copies/mL (range, 190,000 to 66,000,000 copies/mL; median, 6.62 × 105) and 149,370 copies/mL at the second time point tested (range, 2170 to 456,000 copies/mL; median, 24,900). CD4 counts at the first time point averaged 423 cells/mm3 (275 to 657 cells/mm3) and at the second time point averaged 539 cells/mm3 (351 to 765 cells/mm3). Subject AC206 was on antiretroviral therapy (ART) at both time points, having started at week 0. Two of the subjects were placed on ART between the first and second time point. Subject AC184 started therapy at week 7 and subject AC213 started therapy at week 5. Because the relative contribution of therapy is difficult to assess, data from these subjects should be interpreted in the context of therapy.
Identifying Mutations in HIV-Specific CD4 Epitopes
Gag, Nef, and Int were the initial sequences chosen for study because epitopes within these proteins are frequently targeted during HIV infection.4 In the 7 individuals studied, we observed a total of 28 sequence changes, which were distributed equally between the 3 proteins. (Supplemental materials are available via the Article Plus feature at http://www.jaids.com. Locate this article, then click on the Article Plus link on the right.) An average of 4 sequence changes were identified per subject (range, 1 to 5 changes), the majority of which changed the amino acid sequence and were therefore nonsynonymous (Table 3). In subject AC184, we also identified a 13-amino-acid deletion within Int. Eleven of the 19 amino acid changes (4 in Gag, 4 in Nef, and 3 in Int) and the deletion were within regions of the proteins that have previously been described as CD4 epitopes.21 There were 3 instances of mutations occurring within the same described CD4 epitope in different subjects. Subject AC206 and AC213 had mutations within a previously identified epitope in Gag. Similarly, subjects AC184 and AC213 had a mutation within the same Nef epitope, and subjects AC206 and AC184 had a mutation within the same Int epitope. It is notable that these were the only subjects in this study receiving antiretroviral therapy. They also share a common HLA A allele (A*0201), suggesting that the mutations may have been induced by pressure from CTL.
To determine whether the mutations occurred within recognized HIV-specific CD4 epitopes, CD4+ T-cell responses were measured by IFN-γ ELISPOT at matched time points in each subject (Fig. 1). Owing to sample limitations, we were unable to assess Int responses in subjects AC99 and AC192. Four mutations in Gag, 4 mutations in Nef, and 2 mutations in Int were within recognized CD4 epitopes. The only mutation identified in subject AC99 was in an overlapping peptide set in Nef that did not elicit a CD4 response. The other 6 subjects had a CD4 response specific for at least 1 of the regions of the virus containing a mutation. The mutations identified in subjects AC184 and AC196 were all within recognized CD4 epitopes. Subjects AC192, AC206, AC210, and AC213 each had 1 or more mutations within recognized CD4 epitopes. These data indicate that sequence variation does occur within HIV-specific CD4 epitopes during early HIV infection.
Mutations are Coincident With Changes in the CD4 Response
Once we verified that sequence variations occur within HIV-specific CD4 epitopes during early infection, we sought to determine their functional impact. By comparing the consensus peptide-specific CD4 response at both time points in each subject, we were able to determine whether there was a change in IFN-γ production coincident with the observed mutations (Fig. 1). In subjects AC192 and AC210, IFN-γ responses decreased significantly for both of the overlapping peptides in which the sequence change was observed. There was a significant increase in the peptide-specific IFN-γ response in subject AC206 for peptide 274, subject AC213 for peptide 78, and subject AC184 for peptide 77. Interestingly, these were the individuals on antiretroviral therapy. To eliminate the potential of assay variability, we applied conservative criteria to determine significance. The remaining responses did not change in these individuals, nor did the responses in subject AC196. These results suggest that sequence variation within HIV-specific CD4 epitopes may affect TH cell function, in some cases decreasing the TH response and in others increasing it.
In this study, we have shown that mutations occur within HIV-specific CD4 epitopes during early HIV infection. Previous studies looking at sequence variation within HIV-specific CD4 epitopes have produced conflicting results. In 1 study, sequence variation was identified within recognized CD4 epitopes in 2 asymptomatic HIV+ subjects by sequencing proviral DNA at successive time points over the course of infection.13 In another, investigators performed clonal sequencing from plasma of 4 viremic subjects at a single time point and looked for variation within the clones.19 Although they found intrapatient amino acid variation within Gag, there was no sequence variation within the CD4 epitopes they had identified. These disparate results may be a consequence of looking for sequence variation at a single time point versus looking for sequence changes that occur over time. Alternatively, these differences may be due to the difficulty in detecting CD4+ T-cell responses, especially in chronically infected individuals.
We hypothesize that sequence variation within CD4 epitopes may be partially responsible for the TH cell dysfunction that is common in HIV-infected individuals. It has been established that stimulating CD4+ T cells with peptide analogs containing single amino acid substitutions can cause partial T-cell activation. This results in inhibition of some, but not all, effector functions.7 It is less clear, however, whether sequence variation within HIV or other pathogens can result in ineffective CD4+ T-cell responses and functional TH cell escape. Examples of possible TH escape have been observed in infections with parasites,16,17 bacteria,9 and viruses.10,12,14,18 For example, sequence variation within the immunodominant NS3 protein of hepatitis C virus (HCV) is confined to regions recognized by CD4+ T cells.22 Furthermore, variation within the NS3 and Env 2 proteins have been associated with decreased CD4+ T-cell proliferation and a shift from a Th1 to a Th2 cytokine profile.12,23 Similarly, CD4+ T-cell lines and fresh uncultured cells from chronic HIV-infected individuals proliferated when stimulated with some, but not all peptides representing autologous Gag sequences, indicating that natural sequence variation within a CD4 epitope can result in a loss of TH cell function.13 We found 2 mutations that occurred coincident with a decrease in the CD4 response during early HIV infection. This supports the conclusion that sequence variation within HIV-specific CD4 epitopes can contribute to TH cell dysfunction.
We also found mutations that were coincident with an increase in CD4 responses. This is not unexpected, inasmuch as modification of T-cell epitopes to increase peptide-specific responses has been proposed as a potential immunotherapy.24 Although these mutations could have occurred at random, we speculated that other factors might generate a mutation that results in a more robust TH response. For example, in subject AC213, the sequence change observed in peptides 77 and 78 represented reversion to the Consensus B sequence. This region of Nef is commonly recognized by CTL, so it may be that this virus had developed a CTL escape mutation and upon transmission to a new host, when the mutation was no longer advantageous, it was reverted. The question then is whether mutations that are associated with better CD4 responses would be maintained in the virus population, and if so, why. It is notable that we saw increased responses only in those individuals on therapy; however, this may be an artifact. It is well established that ART given during early HIV infection can restore CD4 responses. When responses to other peptides were examined in these individuals, the number of peptide-specific responses that increased was higher than the number of responses that decreased, but still represented less than half of the total breadth of the response (data not shown). Additional longitudinal studies examining the evolution of sequence changes associated with increases in the TH response are necessary and may provide insight into the type of responses that should be generated by a vaccine or immune-based therapy.
We identified several mutations within regions of the virus that failed to elicit a CD4 response. Although we would not expect that every viral peptide would stimulate a response, it is possible that mutations may have occurred in these peptides before the first time point tested, causing them to not be recognized. Alternatively, these peptides may not have elicited CD4 responses because of differences between the autologous virus sequence and the consensus sequence used in the ELISPOT assays, especially if the variation occurred at key residues within the epitope. When we examined this possibility, we found that these peptides differed from consensus by as many as 3 amino acids. Therefore it is possible that we would have detected a response if we had used autologous rather than consensus peptides. We find that measuring CD4+ T-cell responses from fresh rather than frozen cells improves the likelihood of detecting peptide-specific responses during early infection and therefore, whenever possible, fresh cells were used.
Determining the minimal sequence that defines a CD4 epitope is difficult due to the nature of the interaction between the MHC class II molecule and its peptide.25 In this study we did not attempt to determine the precise location of an epitope relative to the observed mutations. Rather, we examined responses to consensus peptides spanning the area of interest. As a consequence, there were instances where a mutation was associated with a change in the IFN-γ response for 1 peptide but not the other within the overlapping peptide set. In these instances it is possible that the epitope was primarily in 1 of the 2 peptides. Alternatively, changes in the IFN-γ response to 1 or both peptides may have been unrelated to the mutation. Additional experiments are required to provide direct evidence that the mutations are responsible for changes in CD4 function.
The mechanisms by which CTL and antibody responses are able to select for viral escape mutations are well described; however, the reasons why mutations would occur and be maintained in CD4 epitopes are less clear. Sequence variation within CD4 epitopes may be the result of random mutations generated during error-prone transcription; however, most mutations we observed were nonsynonymous, suggesting a selective pressure was involved. Because Gag and Nef contain multiple CD8 and CD4 epitopes that overlap, we considered the possibility that the mutations we observed were a result of CTL pressure. We found that the sequence variation within peptides 77 and 78 observed in subjects AC184 and AC213 are within a known MHC class I, restricted A*02 epitope, and both subjects express the A*02 allele. An additional 11 mutations identified in this cohort were within previously described CD8 epitopes,21 and many of the mutations we observed were within CD8 epitopes restricted by the MHC class I allele expressed in that subject. It is therefore possible that CTL can provide the selective pressure necessary to generate mutations within CD4 epitopes. However, we did identify sequence changes that were not within previously identified CD8 epitopes. Although this does not rule out the possibility that CTL pressure is responsible, it leaves open the possibility that it is not the only mechanism at play.
By examining sequence changes that occurred during the earliest phase of HIV infection, when TH cells are functional and viral replication is high, we were able to identify multiple epitopes in which sequence variation was coincident with changes in the CD4 T-cell response. We hypothesize that TH escape may contribute to, but is not completely responsible for, the decline in TH cell function that is common in the majority of HIV-infected individuals. Understanding the effect that sequence variation has on TH cell function during early infection may provide insight into HIV pathogenesis.
1. McMichael AJ, Rowland-Jones SL. Cellular immune responses to HIV. Nature
2. Koup RA, Safrit JT, Cao Y, et al. Temporal association of cellular immune responses with the initial control of viremia in primary human immunodeficiency virus type 1 syndrome. J Virol
3. Schmitz JE, Kuroda MJ, Santra S, et al. Control of viremia in simian immunodeficiency virus infection by CD8+ lymphocytes. Science
4. Kaufmann DE, Bailey PM, Sidney J, et al. Comprehensive analysis of human immunodeficiency virus type 1-specific CD4 responses reveals marked immunodominance of gag and nef and the presence of broadly recognized peptides. J Virol
5. Rosenberg ES, Billingsley JM, Caliendo AM, et al. Vigorous HIV-1-specific CD4+ T cell responses associated with control of viremia. Science
6. Boaz MJ, Waters A, Murad S, et al. Presence of HIV-1 Gag-specific IFN-gamma+IL-2+ and CD28+IL-2+ CD4 T cell responses is associated with nonprogression in HIV-1 infection. J Immunol
7. Sloan-Lancaster J, Allen PM. Altered peptide ligand-induced partial T cell activation: molecular mechanisms and role in T cell biology. Annu Rev Immunol
8. Bouhdoud L, Villain P, Merzouki A, et al. T-cell receptor-mediated anergy of a human immunodeficiency virus (HIV) gp120-specific CD4(+) cytotoxic T-cell clone, induced by a natural HIV type 1 variant peptide. J Virol
9. Brown WC, Brayton KA, Styer CM, et al. The hypervariable region of Anaplasma marginale major surface protein 2 (MSP2) contains multiple immunodominant CD4+ T lymphocyte epitopes that elicit variant-specific proliferative and IFN-gamma responses in MSP2 vaccinates. J Immunol
10. Burt DS, Mills KH, Skehel JJ, et al. Diversity of the class II (I-Ak/I-Ek)-restricted T cell repertoire for influenza hemagglutinin and antigenic drift. Six nonoverlapping epitopes on the HA1 subunit are defined by synthetic peptides. J Exp Med
11. Fenoglio D, Li Pira G, Lozzi L, et al. Natural analogue peptides of an HIV-1 GP120 T-helper epitope antagonize response of GP120-specific human CD4 T-cell clones. J Acquir Immune Defic Syndr
12. Frasca L, Del Porto P, Tuosto L, et al. Hypervariable region 1 variants act as TCR antagonists for hepatitis C virus-specific CD4+ T cells. J Immunol
13. Harcourt GC, Garrard S, Davenport MP, et al. HIV-1 variation diminishes CD4 T lymphocyte recognition. J Exp Med
14. Hunziker L, Recher M, Ciurea A, et al. Antagonistic variant virus prevents wild-type virus-induced lethal immunopathology. J Exp Med
15. Lekutis C, Letvin NL. Substitutions in a major histocompatibility complex class II-restricted human immunodeficiency virus type 1 gp120 epitope can affect CD4+ T-helper-cell function. J Virol
16. Millar AE, Wleklinski-Lee M, Kahn SJ. The surface protein superfamily of Trypanosoma cruzi stimulates a polarized Th1 response that becomes anergic. J Immunol
17. Plebanski M, Lee EA, Hill AV. Immune evasion in malaria: altered peptide ligands of the circumsporozoite protein. Parasitology
18. Wang H, Eckels DD. Mutations in immunodominant T cell epitopes derived from the nonstructural 3 protein of hepatitis C virus have the potential for generating escape variants that may have important consequences for T cell recognition. J Immunol
19. Koeppe JR, Campbell TB, Rapaport EL, et al. HIV-1-specific CD4+ T-cell responses are not associated with significant viral epitope variation in persons with persistent plasma viremia. J Acquir Immune Defic Syndr
20. Bunce M, O'Neill CM, Barnardo MC, et al. Phototyping: comprehensive DNA typing for HLA-A, B, C, DRB1, DRB3, DRB4, DRB5 & DQB1 by PCR with 144 primer mixes utilizing sequence-specific primers (PCR-SSP). Tissue Antigens
21. Korber BT, Brander C, Haynes BF, et al, eds. HIV Molecular Immunology 2005
. Los Alamos, NM: Los Alamos National Laboratory Theoretical Biology and Biophysics; 2005. LA-UR 06-0036.
22. Wang H, Bian T, Merrill SJ, et al. Sequence variation in the gene encoding the nonstructural 3 protein of hepatitis C virus: evidence for immune selection. J Mol Evol
23. Wang JH, Layden TJ, Eckels DD. Modulation of the peripheral T-Cell response by CD4 mutants of hepatitis C virus: transition from a Th1 to a Th2 response. Hum Immunol
24. Abrams SI, Schlom J. Rational antigen modification as a strategy to upregulate or downregulate antigen recognition. Curr Opin Immunol
25. Cosimi LA, Rosenberg ES. The characterization of HIV-1 specific CD4+ T helper epitopes IV-45. In: Korber BC, Haynes BF, Koup R, et al, eds. HIV Molecular Immunology 2000
. Los Alamos, NM: Los Alamos National Laboratory Theoretical Biology and Biophysics; 2000.