Neutralizing antibodies (NAbs) are likely to be a critical component of the protective immunity required for an HIV-1 vaccine to be effective.1 However, an inability to induce broadly reactive NAbs by immunization with HIV-1 envelope proteins has proved one of the major obstacles to the development of a successful vaccine. HIV-1 has evolved multiple mechanisms to protect itself against the binding of neutralizing antibodies. Trimerization of the gp120-gp41 structure occludes much of the surface of the monomeric gp120 molecule, limiting the accessibility of potential target epitopes to neutralizing antibodies.2 The exposed surface of gp120 is heavily glycosylated, rendering it less visible to the immune system.3 The virus can also shift the position of sugar moeities in vivo, generating a protective dynamic glycan shield preventing antibody binding by steric hindrance.4-6
Despite the multiple immune evasion mechanisms observed in vivo, primary isolates of HIV-1 of different subtypes can be neutralized in vitro by the broadly NAbs found in the sera of rare HIV-1-infected patients, suggesting that HIV-1 strains share cross-neutralization epitopes that humans can respond to. This broad response is stronger and more frequent in patients with nonprogressive HIV-1 infection-long-term nonprogressors (LTNPs)-than in other HIV-1-infected patients.7-10 Only a few broadly neutralizing monoclonal antibodies (NmAbs) have been generated from such individuals.11 The first NmAb identified was b12, which binds to a conformational epitope overlapping the CD4 receptor on gp120, thereby interfering with virus attachment to target cells.12 A second NmAb, 2G12, recognizes a carbohydrate-dependent epitope and is thought to prevent interactions between the envelope and the host cell by steric hindrance.13-16 The 2F5 and 4E10 NmAbs bind to adjacent, but different, epitopes on the membrane-proximal external region (MPER) of gp41; they probably act by inhibiting fusion.17-22 Passive immunization, with immunoglobulins purified from serum samples obtained from HIV-1-infected individuals or with 1 or several NmAbs, completely prevents infection in animal models.23-27 However, the contribution of broadly NAbs to the prevention of disease progression after infection has been established remains unclear. Poignard et al showed that the treatment of HIV-1-infected SCID mice with a cocktail of anti-HIV-1 mAbs had little sustained effect on viral load because Ab neutralization escape mutants were rapidly selected.28 Trkola et al recently reported a delay in HIV-1 rebound after the cessation of antiretroviral therapy that was due to the passive transfer of human NAbs, suggesting that NAbs could, in principle, contain viremia in patients with established HIV infection.29 However, as in mice, virus escape was rapidly observed under antibody pressure.
So, although viral challenges were performed with a limited number of isolates in the cited studies in animals, it is clear that the presence of a high titer of broadly NAbs before infection can provide protection. However, it remains difficult to induce such antibodies, and additional knowledge is required concerning the nature of the envelope proteins potentially involved in the induction of broadly NAbs. In this study, we investigated the possible association between the presence of broadly NAbs in rare patients and the particular molecular properties of the envelope glycoproteins of the variants found in these patients. We selected LTNPs from the French ANRS cohort CO15 who developed broadly NAbs and compared the molecular characteristics of the envelope glycoproteins of the viruses infecting these patients with those present in patients of the same cohort who did not develop such broadly NAbs.
We analyzed serum samples from 67 HIV-1-infected LTNPs from the French “Asymptomatiques à Long Terme” cohort (ALT ANRS CO15) for the presence of broadly neutralizing antibodies.30 The criteria for inclusion in the ALT cohort were HIV-1 seropositivity for at least 8 years, stable CD4+ T-cell count >600 cells/mm3 over the previous 5 years, no clinical symptoms, and no antiretroviral therapy.31-34 All subjects were infected with subtype B virus. Plasma viral load and DNA viral load at entry were determined as previously described.31,35 Patients were included in the cohort between 1994 and 1996.
In Vitro Neutralization Assays
Neutralization experiments were carried out with 4 heterologous primary isolates (KON, FRO, GIL, and MBA) using the HeLa-CD4+-CXCR4+-CCR5+ cells, as previously described.36,37 The primary isolates were of 4 different clades or circulating recombinant forms (CRF02-AG for KON, clade B for FRO, CRF01-AE for MBA, and clade F for GIL) and of different tropisms (X4 for FRO and KON, R5X4 for MBA, and R5 for GIL). These isolates were selected based on their relative resistance to neutralization. Assays were carried out at least in duplicate, on serial 2-fold dilutions (from 1:10 to 1:320) of heat-inactivated serum samples from LTNPs collected on inclusion in the cohort. The 90% and 50% inhibitory concentrations (IC90 and IC50) of NAbs were expressed as mean values of the reciprocal serum dilution required to decrease by 90% and 50% the number of infected cells 2 days after infection with 100 TCID50.
Nucleic Acid Extraction, PCR, and Cloning
Genomic DNA was extracted from peripheral blood mononuclear cells of selected subjects, collected on inclusion in the cohort, using the QIAamp DNA Blood Midi kit according to the manufacturer's instructions (Qiagen, Courtaboeuf, France). A 1276 bp fragment of the env gene (named V1-V5 because it spanned loops V1 to V5; nt 5941 to 7216 in the HXB2 genome) encompassing most of the gp120 coding sequence was amplified by nested polymerase chain reaction (PCR) using subtype B env-specific primers (Fig. 1). The outer primers pair was S1ext (5′-TGGGTCACAGTCTATTATGGGG-3′) and AS1ext (5′-TTCTCTTTGCCTTGGTGGGTGCTA-3′) and the inner primer pair was S1int (5′-TGCTAAAGCATATGATACAGARGYACA-3′) and AS1int (5′-TTCACTTCTCCAATTGTCCCTCATATY-3′). For each sample, we added 1 μg of template DNA to a PCR mixture containing 0.4 mmol/L of each outer primer, 200 mmol/L dNTP, 1.5 mmol/L MgCl2, and 2.5 U of Expand High FidelityPLUS enzyme blend (Roche Diagnostics, Mannheim, Germany). Amplifications were carried out with the following cycling conditions: 4 minutes at 94°C, followed by 30 cycles of 30 seconds at 95°C, 30 seconds at 50°C, and 1 minute 30 seconds at 72°C, with an additional extension phase of 7 minutes at 72°C. A 5-μL aliquot of the products of the first round of PCR was then transferred to a new reaction mixture containing the inner primer pair, and a second round of amplification was performed under the same cycling conditions. For 1 DNA sample from which the 1276 bp fragment could not be amplified, 2 overlapping fragments encompassing this fragment-a 451 bp fragment (V1-V2; nt 5941 to 6391 in the HXB2 genome) and a 847 bp fragment (V3-V5; nt 6370 to 7216)-were amplified by nested PCR (Fig. 1). For the V1-V2 fragment, the outer primer pair used was S1ext (as above) and AS2ext (5′-ATGTATGGGAATTGGCTCAAAGG-3′), and the inner primer pair was S1int and AS2int (5′-CTTTGGACAGGCCTGTGTAATGR-3′). For the V3-V5 fragment, the outer primer pair was S2ext (5′-ATAAGTTGTAACACCTCAGTCATT-3′) and AS1ext, and the inner primer pair was S2int (5′-CATTACACAGGCCTGTCCAAAGGTA-3′) and AS1int. Amplifications were carried out as described above, with the same cycling conditions. All PCR products were inserted into pCR2.1 (Topo TA cloning kit; Invitrogen, Paisley, UK). At least 11 pCR2.1-env clones for each subject were selected for further analysis.
All PCR2.1-env clones were sequenced, using a set of env-specific internal primers, according to the Dye Terminator cycle sequencing protocol (Applied Biosystems, Foster City, CA). Nucleotide sequences were assembled with the BioEdit package version 5.0.9 (Ibis Biosciences, Carlsbad, CA) and aligned using ClustalW (European Bioinformatics Institute, Cambridge, UK).38,39 Phylogenetic analysis and tree reconstructions were performed by the neighbor-joining method, with MEGA version 3.1 (The Biodesign Institute, Tempe, AZ).40,41 The distance matrix was calculated with the 2-parameter Kimura algorithm (transition-to-transversion ratio of 2.0).42 Approximate confidence limits for individual branches were assigned by bootstrap resampling with 500 replicates. The dS/dN ratios were determined using SNAP (http://www.hiv.lanl.gov/content/sequence/SNAP/SNAP.html).43,44 Amino-acid sequences were deduced from aligned nucleotide sequences, using GeneCutter (http://www.hiv.lanl.gov/content/sequence/GENE_CUTTER/cutter.html). Potential N-linked glycosylation sites (N-gly) were identified by N-GlycoSite (http://www.hiv.lanl.gov/content/sequence/GLYCOSITE/glycosite.html).45
The Mann-Whitney test was used to compare RNA and DNA viral loads in subjects who developed broadly NAbs with those in subjects who did not, and to compare the sequence properties of env clones derived from patients from these 2 groups.
Nucleotide Sequence Accession Numbers
All env sequences have been submitted to GenBank and assigned accession numbers EF179866 through EF179896, EF179924 through EF180024, and EU214586 through EU214600.
Neutralizing Activity of Sera From LTNPs
Sera from 24 of 67 LTNPs (36%) displayed no neutralizing activity, whereas 11 (16%) displayed broadly neutralizing activity, defined as the ability to decrease the infectivity of each of 4 heterologous primary isolates of different clades by at least 90%. Twenty-six (39%) samples neutralized at least 2 strains, and 19 (28%) samples neutralized at least 3 strains. The most susceptible strain was FRO (subtype B), which was neutralized by 33 samples (49%). The most resistant strain was MBA (CRF01-AE), which was neutralized by only 13 samples (19%). Intermediate resistance was observed for KON (CRF02-AG) and GIL (subtype F), which were neutralized by 24 (36%) and 29 (43%) samples, respectively.
We explored the relationships between the neutralizing activity of sera from LTNPs, their plasma RNA viral load, and their peripheral blood mononuclear cell (PBMC)-associated DNA viral load. We found that plasma HIV-1 RNA levels were higher among subjects who developed broadly NAbs than among those who did not, with a median of 34,000 copies/mL (interquartile range [IQR]: 5800 to 120,000 copies/mL) versus 4300 copies/mL (IQR: 124 to 34,500 copies/mL) (P = 0.04, Mann-Whitney test). Similarly, DNA viral loads were higher among subjects who developed broadly NAbs than among those who did not, with a median of 685 copies/106 cells (IQR: 178 to 1362 copies/106 cells) versus 139 copies/106 cells (IQR: 30 to 663 copies/106 cells) (P = 0.02, Mann-Whitney test).
Cloning of env Genes Isolated From Selected Patients
We selected the 5 patients with the highest NAb titers (patients 04063, 05005, 05008, 08003, and 10001) from the 11 LTNPs who developed broadly NAbs (NAb+). We also selected 4 patients who developed no NAbs (patients 04050, 06006, 11005, and 11024) for comparison (NAb-) (Table 1). Genomic DNA was extracted from PBMC of selected subjects. A 1276 bp fragment of the env gene encompassing most of the gp120 coding sequence (V1-V5) was amplified by nested PCR, and then cloned into pCR2.1. A total of 15 PCR2.1-env clones containing the V1-V5 fragment was obtained for patient 04063, 16 clones for patient 05005, 15 clones for patient 05008, 13 clones for patient 08003, 12 clones for patient 10001, 15 clones for patient 04050, 15 clones for patient 06006, and 16 clones for patient 11,005. The V1-V5 fragment could not be amplified from the proviral DNA of the last patient, 11024. For this NAb- patient, 2 smaller overlapping fragments, V1-V2 and V3-V5, encompassing the V1-V5 region, were amplified (Fig. 1). Fourteen PCR2.1-env clones containing the V1-V2 fragment and 16 clones containing the V3-V5 fragment were obtained for patient 11024.
Phylogenetic Analysis of env Sequences
All pCR2.1-env clones were sequenced, giving 117 V1-V5 sequences, 14 V1-V2 sequences, and 16 V3-V5 sequences. We constructed 2 different phylogenetic trees to ensure that env sequences from all 9 patients were included in a global phylogenetic analysis: one tree was based on 131 V1-V2 sequences (5′-part of the 117 V1-V5 sequences and the 14 V1-V2 sequences), and the other was based on 133 V3-V5 sequences (3′-part of the 117 V1-V5 sequences and the 16 V3-V5 sequences) (Fig. 2). This was rendered necessary because the gp120 sequences from patient 11024 were not derived from single continuous fragments. We found that sequences from all patients who developed broadly NAbs formed monophyletic groups, with the exception of those from patient 08003, which formed 2 different groups clustering in different branches of the trees. In contrast, sequences from 2 of the 4 patients who did not develop NAbs formed 2 separate groups, clustering in different branches of the trees (2 groups each for the V1-V2 and V3-V5 regions from patients 04050 and 11024). The env sequences from the last 2 NAb- patients (06006 and 11005) formed monophyletic groups in the 2 trees. The presence of env sequences from a single patient on 2 clearly separated branches is suggestive of coinfection or superinfection, with the presence of 2 different strains. This suggests that 1 of the 5 NAb+ LTNPs and 2 of the 4 NAb- LTNPs had multiple infections. Interestingly, these 3 patients have the lowest viral RNA and DNA loads (Table 1).
Group 1 sequences from patient 11024 clustered next to (V1-V2) or within (V3-V5) the group 2 sequences of patient 08003. We excluded possible contamination of these samples for several reasons. First, the PCR products obtained from the proviral DNA of these 2 patients were not identical. The V1-V5 fragment was amplified for patient 08003, whereas this entire fragment could not be amplified for patient 11024 and the sequences obtained for patient 11024 corresponded to amplicons (V1-V2 and V3-V5 fragments) obtained in different runs. Second, all sequences from patient 08003 (group 2) differed from the sequences from patient 11024 (group 1) by a deletion of 6 nucleotides in the V2 region. If contamination had occurred, this deletion would have been present in at least some of the sequences from patient 11024 and would not have been found in all sequences from patient 08003. Database searches (blastn) revealed that the sequences from these 2 patients were highly similar to those of HXB2-LAI-IIIB-BRU isolates.46 Indeed, the 4 group 2 clones from patient 08003 displayed 98% nucleotide sequence identity to the HXB2-LAI-IIIB-BRU env sequence. The 6 V1-V2 and 9 V3-V5 group 1 clones from patient 11024 displayed 97% to 99% nucleotide sequence identity to the HXB2-LAI-IIIB-BRU env sequence. Interestingly, the V3-V5 sequences from patients 11024 and 08003 were also highly similar (95% to 96% identity) to the env sequences (C2-C5 regions) of 3 HIV-1 isolates from Spanish long-term nonprogressors found in the GeneBank database (accession numbers AY501235, AY501253, and AY501288).
We quantified intrapatient env diversity to document the evolution of quasispecies. For each patient, we used the Kimura method to calculate average genetic distances between all sequences of a monophyletic group, or between the sequences within each group when clustering into different groups was observed.42 We studied the entire V1-V5 sequences from each patient, except for patient 11024, for whom env diversity was calculated separately based on the V1-V2 and V3-V5 sequences. The diversity of the env sequences of viruses from patients who developed broadly NAbs was high. It varied from 3.3% to 5.6%, except for group 2 sequences from patient 08003 (0.3%) (Fig. 3). In contrast, the env sequence diversity of viruses from NAb- patients was more variable. It did not exceed 1.0% in the various groups of sequences of the 2 patients with the lowest viral loads (04050 and 11024) but reached 3.9% and 5.2% in the 2 patients with higher viral loads (06006 and 11005). We also calculated the average genetic distance between intrapatient env sequences from different groups. It varied from 8.2% to 15.1%. These large genetic distances between groups of env sequences for a given patient provide further evidence for the occurrence of multiple infections in these patients.
The ratio of the rate of synonymous mutations per potential site of synonymous mutations (dS) to the rate of nonsynonymous mutations (NS) per potential site of nonsynonymous mutation (dN) was determined for all patients. The average for all pairwise comparisons did not differ significantly between NAb+ and NAb- patients. It varied from 1.03 to 3.35 in NAb- patients and from 1.19 to 1.81 in NAb+ patients, depending on both the patient and the group of sequences considered (Mann-Whitney test, P = 0.31).
Analysis of env Amino Acid Sequences
We analyzed the deduced Env amino acid sequences. Mutations leading to frameshifts and/or premature stop codons were observed in some clones (defective clones), with various prevalences. The proportion of defective V1-V5 clones for each patient is shown in Figure 4. For patient 11024, from whom only V1-V2 and V3-V5 clones were obtained, we estimated the proportion of defective V1-V5 clones by adding the proportions of defective V1-V2 and V3-V5 clones. This estimation was based on the observation that only 1 defective mutation, located in either V1-V2 or V3-V5, occurred in each of the defective V1-V5 clones from the other 8 patients. The proportion of defective env gene clones was higher in patients who did not develop broadly NAbs (25% to 93%) than in patients who developed broadly NAbs (7% to 19%) (P = 0.02, Mann-Whitney test). The proportion of defective clones was highest in patients 04050 and 11024, who had the lowest viral load and the lowest env divergence. In these patients, 88% of defective mutations were deletions of 1 or 2 nucleotides.
We compared the amino acid sequences of the envelope proteins of viruses from patients who developed broadly NAbs with those from patients who did not, trying to identify specific features common to those associated with broadly NAbs. All the nucleotide sequences found in each patient or in each group of patients, NAb+ and NAb-, were aligned. Amino acid sequences were deduced from aligned nucleotide sequences using GeneCutter, and consensus sequences were derived for each patient or each group of patients. The consensus sequences were then compared (Fig. 5A-C). The sequences were highly conserved within each patient, within each group of patients (Fig. 5A-B), and between the 2 groups of patients (Fig. 5C), with the exception of the V1, V2, V4, and V5 variable regions. Twenty-four N-gly sites were conserved in the 2 consensus sequences derived from each group of patients. Two significant differences were observed. The first major difference between the NAb+ and NAb- groups concerned the length of the V1 region, which was longer in the consensus sequence derived from patients who developed broadly NAbs. The second major difference was the presence of additonal N-gly sites in V1 in the consensus sequence derived from patients who developed broadly NAbs. A detailed analysis of the lengths of the variable regions and of the number of N-gly sites in these regions, in all sequences from the 2 populations, confirmed these observations (Fig. 6). The V1 region was highly variable between patients, but both the length of this region and the number of N-gly sites within it were significantly higher in sequences from patients who developed broadly NAbs than in sequences from patients who did not (median length values of 24 amino acids [IQR: 21 to 28] versus 18 amino acids [IQR: 16 to 22; Mann-Whitney test, P < 0.01], and median of 3 N-gly sites [IQR: 3 to 4] versus 2 [IQR: 2 to 3; Mann-Whitney test, P < 0.01]). A statistically significant difference was also observed in the V4 region for the number of amino acids and number of N-gly sites. However, the trend was less clear than for V1, because NAb+ patients harbored viruses with a higher median number of N-gly sites in V4 (5 vs. 4) but a shorter length of V4.
Our aim was to investigate the specific properties of HIV-1 envelope glycoproteins in rare individuals developing broadly NAbs during natural infection, to facilitate identification of the molecular determinants that potentially contribute to the immunogenic properties of these proteins. We focused on a homogeneous population of 67 HIV-1 subtype B-infected individuals with long-term nonprogressive infection, because previous studies have shown that LTNPs have the highest frequency of broadly NAbs.7-10 The breadth of the neutralizing activity of the sera of these patients was evaluated on 4 primary isolates of different clades, selected on the basis of their high resistance to neutralization. Although we cannot exclude the possibility of these 4 isolates being atypical, they appeared to be representative of neutralization-resistant strains, in our experience.30,36,37 We found that 19% to 49% of samples contained NAb, depending on the primary isolate. Only 16% of samples displayed broad neutralizing activity, defined as the ability to decrease the infectivity of each of the 4 strains by at least 90%. We selected the 5 patients with broadly neutralizing sera who presented the higher NAbs titers. For comparison, we also selected 4 patients who did not develop any NAbs. We cloned at least 11 env genes amplified from the proviral DNA of each patient and compared the molecular characteristics of the envelope glycoproteins of the viruses infecting these patients. To obtain more complete information about evolution of the env gene, we focused on the proviral DNA present in PBMC rather than the RNA present in plasma. Indeed, analyses including archived proviral material in addition to the predominant contemporary variants are more representative of the overall evolution of the virus in a given patient.47
DNA quasispecies analysis revealed a significant difference in the evolution of env genes in NAb+ and NAb- patients. Mean intrapatient quasispecies diversity varied from 3.3% to 5.6% in the 5 NAb+ patients (except for 1 cluster of sequences in 1 of them). In contrast, env genes diversity was more variable in the 4 NAb- patients. It was below 1.0% in the 2 patients with the lowest viral loads but reached 3.9% and 5.2% in the 2 patients with higher viral loads. A few studies have reported highly homogeneous quasispecies in some LTNPs, as in our 2 NAb- patients.48-50 The higher viral diversity observed in NAb+ patients was associated with higher viral RNA and DNA loads, suggesting that the virus was replicating and evolving continuously in these patients. In contrast, the 2 NAb- patients with low quasispecies diversity had low viral RNA loads (≤135 copies/mL) and DNA loads (≤28 copies/mL), consistent with viral latency or low replication rates and the arrest or slowing of viral evolution. This association between the development of broadly NAbs and high viral loads was not limited to the patients for whom extensive molecular analysis was carried out. Instead, it was observed throughout the cohort. Plasma HIV-1 RNA levels and DNA viral loads were higher in the 11 subjects who developed broadly NAbs than in the 56 patients who did not (median of 34,000 versus 4300 RNA copies/mL, P = 0.04, Mann-Whitney test; median of 685 versus 139 DNA copies/106 cells, P = 0.02, Mann-Whitney test; data not shown). A similar positive association between viremia and heterologous neutralization was recently reported in subjects with untreated chronic infection.51 Interestingly, the low rate of viral evolution observed in our 2 NAb- patients was associated with a large number of defective env clones, which may have lowered the replicative potential of these strains. The low quasispecies diversity observed in these patients may be a direct consequence of these defects. Defective envelope proteins were also observed in LTNPs in the study by Connor et al, who focused their analysis on the functional properties of env genes from 6 LTNPs.52 They found that only 30% to 50% of env clones from 4 of the 6 LTNPs expressed gp160, suggesting the presence of only a small proportion of proviral env genes with intact open reading frames. In addition, gp160 was processed incorrectly in 75% to 100% of clones expressing gp160. The other 2 LTNPs presented a higher proportion (87%) of clones expressing gp160, correctly processed in most cases. One of these 2 LTNPs progressed to AIDS after more than 10 years of stability, and the second presented an increase of viremia suggestive of a risk of subsequent evolution. These 2 cases may be representative of a subset of LTNPs who progress to AIDS after a prolonged period of clinical stability. If we compare the progression to AIDS of patients included in the ANRS cohort of LTNPs, 82% of NAb+ patients (9 of 11) versus only 58% of NAb- (14 of 24) patients progressed to AIDS 5 years after being enrolled in the cohort. This suggests that a high proportion of NAb- patients are infected with viruses that produce mostly defective virions, whereas NAb+ patients tend to be infected with more replication-competent viruses. Wang et al attributed the lack of viral replication and sequence evolution in a nonprogressor to the presence of stop codons in the structural gag gene (p17 and p24) and in pol RT, due to G→A hypermutations.50 We observed no G→A hypermutations in the env clones from our NAb- patients containing stop codons. Defective mutations mostly involved the deletion of 1 nucleotide. However, we did not analyze full-length genomes, and we cannot exclude the possibility that G→A hypermutations occurred in other viral genes.
Viral replication thus seems to drive the production of broadly NAbs, as shown by the consistent association between viremia and responses to heterologous primary isolates. However, it is difficult to determine whether the diversification of the viral envelope observed in our 5 NAb+ patients is the cause or the consequence of broadly NAb production. Envelope diversification may be a prerequisite for the expansion of antigen-specific B-cell activation leading to a broad response, but it may be also a consequence of selective pressure from the host immune response from which the virus escapes. The dS/dN (average of all pairwise comparison of sequences) ratios of all 5 NAb+ and 4 NAb- patients were >1, suggesting that evolution in LTNPs was not associated with a selective advantage for viruses carrying changes in the envelope protein. Markham et al reported that nonprogressors with low viral loads selected against the nonsynonymous mutations that might have resulted in viruses with higher levels of replication.49 In contrast, they observed that viral evolution was associated with selection that favored nonsynonymous mutations in individuals with rapid or moderate disease progression. However, 2 of our observations are consistent with viral escape in patients developing broadly NAbs. The V1 region was highly variable between patients, but both V1 length and the number of N-gly sites in V1 were higher in sequences from patients who developed broadly NAbs than in sequences from patients who did not. As reported in several recent studies, expansion and higher levels of variable-loop glycosylation are associated with a mechanism used by HIV-1 to evade neutralizing antibody responses to autologous viruses during the early stages of HIV infection.5,6,53 Our observation, after a much longer course of infection (>8 years), suggests that NAb+ patients may have a high initial autologous NAb response that drives viral evolution. The picture was less clear for the V4 region that harbored higher N-gly sites but was shorter in NAb+ patients. Taken together, all these elements may allow us to speculate that immune selective pressure such as broadly NAbs drives the viral genetic diversity in vivo. This hypothesis will be answered by additional analyses of autologous neutralization using pseudotyped virus particles that contain representative envelope clones from each patient.
With the exception of the major differences described above, our analysis revealed no signature sequence specific to viral envelopes derived from NAb+ patients that might be responsible for their induction. This suggests that the induction of broadly neutralizing antibodies is not restricted to a few highly conserved epitopes, but instead involves multiple responses to isolate-specific epitopes that accumulate over time in a given patient.
A comparison of envelope sequences from NAb+ and NAb- patients unexpectedly showed that envelope sequences from 2 of the 4 NAb- patients, but from only 1 of the 5 NAb+ patients, formed 2 or 3 different groups separated by large genetic distances and clustering in different branches of the phylogenetic tree. These large genetic distances between groups of env sequences in a single patient are consistent with the occurrence of multiple infections (coinfections or superinfections) in these patients. Deeks et al recently reported higher levels of neutralizing activity against heterologous viruses during chronic infection than during acute infection.51 They speculated that these NAb responses against heterologous viruses might protect against HIV superinfection, as most well-documented cases of HIV superinfection seem to have occurred in cases of recent infection,54-58 whereas superinfections seem to be very rare in patients with advanced HIV disease.59,60 Smith et al showed that recently infected individuals with HIV superinfection have fewer crossprotective and autologous NAb responses than their nonsuperinfected controls.61 Although we studied envelope sequences from only a small number of patients, these data suggest that the NAb responses against heterologous viruses found in NAb+ LTNPs may have protected them against HIV superinfection. The single NAb+ patient with suspected coinfection or superinfection (patient 08003) presented 2 clusters of sequences: a first cluster displaying 3.3% diversity and a second cluster with only 0.4% diversity. This patient may initially have been infected with a virus that did not diversify and therefore did not induce broadly NAbs, subsequently being superinfected with a virus that diversified over time and induced a broad neutralizing response. The high frequency of putative multiple infections in our limited number of patients may seem surprising. However, all of our LTNPs were enrolled in the ANRS cohort between 1994 and 1996, and were therefore infected at least 8 years previously, early in the epidemic and at a time when little was known about the etiologic agent of AIDS and when the prevention of HIV transmission was still inefficient. Interestingly, a similar observation was recently made by Lamine et al, who reported 2 distinct env sequence clusters for 2 of 4 HIV controllers.62 The presence of viruses highly similar to the ancestral HXB2-LAI-IIIB-BRU viruses isolated in France in 198363 in 2 LTNPs supports this hypothesis. Interestingly, env sequences isolated from these 2 LTNPs were also highly similar (95% to 96% identity) to the env sequences of 3 HIV-1 isolates from Spanish LTNPs found in the GenBank database (accession numbers AY501235, AY501253, and AY501288), suggesting that our observation is not unique. Moreover, by dating the viral populations present in HIV-1 Spanish LTNPs, Bello et al recently reported the presence of ancestral sequences in the viral population of a subset of LTNPs.48 These LTNPs displayed a very limited virus replication and a very slow or arrested viral evolution maintaining a close relationship of the viral population to the transmitted virus. Taken together, these results suggested that “ancestral sequences” may commonly be found in LTNPs.
In conclusion, comparisons of the envelope genes of viruses from NAb+ and NAb- LTNPs did not lead to the identification of a signature sequence specific to sequences derived from NAb+ patients. In constrast, we found that the development of broadly NAbs by some LTNPs was associated with higher gp120 diversity in the viruses infecting these patients and with longer V1 sequences and additional N-gly sites in V1. It remains unclear whether these characteristics are necessary for the induction of NAbs or result from viral escape from a broad response.
We thank all the patients and clinicians who participated in the Asymptomatiques à Long Terme ANRS cohort (ANRS CO15), Catherine Gaudy for helpful discussions concerning phylogenetic analysis, and Alain Moreau for technical assistance.
1. Burton DR, Desrosiers RC, Doms RW, et al. HIV vaccine design and the neutralizing antibody problem. Nat Immunol
2. Wyatt R, Kwong PD, Desjardins E, et al. The antigenic structure of the HIV gp120 envelope glycoprotein. Nature
3. Chen B, Vogan EM, Gong H, et al. Structure of an unliganded simian immunodeficiency virus gp120 core. Nature
4. Dacheux L, Moreau A, Ataman-Onal Y, et al. Evolutionary dynamics of the glycan shield of the human immunodeficiency virus envelope during natural infection and implications for exposure of the 2G12 epitope. J Virol
5. Wei X, Decker JM, Wang S, et al. Antibody neutralization and escape by HIV-1. Nature
6. Derdeyn CA, Decker JM, Bibollet-Ruche F, et al. Envelope-constrained neutralization-sensitive HIV-1 after heterosexual transmission. Science
7. Cao Y, Qin L, Zhang L, et al. Virologic and immunologic characterization of long-term survivors of human immunodeficiency virus type 1 infection. N Engl J Med
8. Carotenuto P, Looij D, Keldermans L, et al. Neutralizing antibodies are positively associated with CD4+ T-cell counts and T-cell function in long-term AIDS-free infection. AIDS
9. Cecilia D, Kleeberger C, Munoz A, et al. A longitudinal study of neutralizing antibodies and disease progression in HIV-1-infected subjects. J Infect Dis
10. Pilgrim AK, Pantaleo G, Cohen OJ, et al. Neutralizing antibody responses to human immunodeficiency virus type 1 in primary infection and long-term-nonprogressive infection. J Infect Dis
11. Binley JM, Wrin T, Korber B, et al. Comprehensive cross-clade neutralization analysis of a panel of anti-human immunodeficiency virus type 1 monoclonal antibodies. J Virol
12. Roben P, Moore JP, Thali M, et al. Recognition properties of a panel of human recombinant Fab fragments to the CD4 binding site of gp120 that show differing abilities to neutralize human immunodeficiency virus type 1. J Virol
13. Kunert R, Ruker F, Katinger H. Molecular characterization of five neutralizing anti-HIV type 1 antibodies: identification of nonconventional D segments in the human monoclonal antibodies 2G12 and 2F5. AIDS Res Hum Retroviruses
14. Sanders RW, Venturi M, Schiffner L, et al. The mannose-dependent epitope for neutralizing antibody 2G12 on human immunodeficiency virus type 1 glycoprotein gp120. J Virol
15. Scanlan CN, Pantophlet R, Wormald MR, et al. The broadly neutralizing anti-human immunodeficiency virus type 1 antibody 2G12 recognizes a cluster of alpha1→2 mannose residues on the outer face of gp120. J Virol
16. Trkola A, Purtscher M, Muster T, et al. Human monoclonal antibody 2G12 defines a distinctive neutralization epitope on the gp120 glycoprotein of human immunodeficiency virus type 1. J Virol
17. Barbato G, Bianchi E, Ingallinella P, et al. Structural analysis of the epitope of the anti-HIV antibody 2F5 sheds light into its mechanism of neutralization and HIV fusion. J Mol Biol
18. Muster T, Steindl F, Purtscher M, et al. A conserved neutralizing epitope on gp41 of human immunodeficiency virus type 1. J Virol
19. Ofek G, Tang M, Sambor A, et al. Structure and mechanistic analysis of the anti-human immunodeficiency virus type 1 antibody 2F5 in complex with its gp41 epitope. J Virol
20. Purtscher M, Trkola A, Grassauer A, et al. Restricted antigenic variability of the epitope recognized by the neutralizing gp41 antibody 2F5. AIDS
21. Stiegler G, Kunert R, Purtscher M, et al. A potent cross-clade neutralizing human monoclonal antibody against a novel epitope on gp41 of human immunodeficiency virus type 1. AIDS Res Hum Retroviruses
22. Zwick MB, Labrijn AF, Wang M, et al. Broadly neutralizing antibodies targeted to the membrane-proximal external region of human immunodeficiency virus type 1 glycoprotein gp41. J Virol
23. Baba TW, Liska V, Hofmann-Lehmann R, et al. Human neutralizing monoclonal antibodies of the IgG1 subtype protect against mucosal simian-human immunodeficiency virus infection. Nat Med
24. Hofmann-Lehmann R, Vlasak J, Rasmussen RA, et al. Postnatal passive immunization of neonatal macaques with a triple combination of human monoclonal antibodies against oral simian-human immunodeficiency virus challenge. J Virol
25. Mascola JR, Stiegler G, VanCott TC, et al. Protection of macaques against vaginal transmission of a pathogenic HIV-1/SIV chimeric virus by passive infusion of neutralizing antibodies. Nat Med
26. Parren PW, Marx PA, Hessell AJ, et al. Antibody protects macaques against vaginal challenge with a pathogenic R5 simian/human immunodeficiency virus at serum levels giving complete neutralization in vitro. J Virol
27. Nishimura Y, Igarashi T, Haigwood N, et al. Determination of a statistically valid neutralization titer in plasma that confers protection against simian-human immunodeficiency virus challenge following passive transfer of high-titered neutralizing antibodies. J Virol
28. Poignard P, Sabbe R, Picchio GR, et al. Neutralizing antibodies have limited effects on the control of established HIV-1 infection in vivo. Immunity
29. Trkola A, Kuster H, Rusert P, et al. Delay of HIV-1 rebound after cessation of antiretroviral therapy through passive transfer of human neutralizing antibodies. Nat Med
30. Braibant M, Brunet S, Costagliola D, et al. Antibodies to conserved epitopes of the HIV-1 envelope in sera from long-term non-progressors: prevalence and association with neutralizing activity. AIDS
31. Candotti D, Costagliola D, Joberty C, et al. Status of long-term asymptomatic HIV-1 infection correlates with viral load but not with virus replication properties and cell tropism. French ALT Study Group. J Med Virol
32. Magierowska M, Theodorou I, Debre P, et al. Combined genotypes of CCR5, CCR2, SDF1, and HLA genes can predict the long-term nonprogressor status in human immunodeficiency virus-1-infected individuals. Blood
33. Martinez V, Costagliola D, Bonduelle O, et al. Combination of HIV-1-specific CD4 Th1 cell responses and IgG2 antibodies is the best predictor for persistence of long-term nonprogression. J Infect Dis
34. Ngo-Giang-Huong N, Candotti D, Goubar A, et al. HIV type 1-specific IgG2 antibodies: markers of helper T cell type 1 response and prognostic marker of long-term nonprogression. AIDS Res Hum Retroviruses
35. Rouzioux C, Hubert JB, Burgard M, et al. Early levels of HIV-1 DNA in peripheral blood mononuclear cells are predictive of disease progression independently of HIV-1 RNA levels and CD4+ T cell counts. J Infect Dis
36. Barin F, Brunet S, Brand D, et al. Interclade neutralization and enhancement of human immunodeficiency virus type 1 identified by an assay using HeLa cells expressing both CD4 receptor and CXCR4/CCR5 coreceptors. J Infect Dis
37. Barin F, Jourdain G, Brunet S, et al. Revisiting the role of neutralizing antibodies in mother-to-child transmission of HIV-1. J Infect Dis
38. Hall TA. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symp Ser
39. Thompson JD, Higgins DG, Gibson TJ. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res
40. Kumar S, Tamura K, Nei M. MEGA3: Integrated software for Molecular Evolutionary Genetics Analysis and sequence alignment. Brief Bioinform
41. Saitou N, Nei M. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol Biol Evol
42. Kimura M. A simple method for estimating evolutionary rates of base substitutions through comparative studies of nucleotide sequences. J Mol Evol
43. Nei M, Gojobori T. Simple methods for estimating the numbers of synonymous and nonsynonymous nucleotide substitutions. Mol Biol Evol
44. Korber B. HIV signature and sequence variation analysis. In: Rodrigo AG, Learn, GH, eds. Computational Analysis of HIV Molecular Sequences
. Dordrecht, Netherlands: Kluwer Academic Publishers; 2001;55-72.
45. Zhang M, Gaschen B, Blay W, et al. Tracking global patterns of N-linked glycosylation site variation in highly variable viral glycoproteins: HIV, SIV, and HCV envelopes and influenza hemagglutinin. Glycobiology
46. Ratner L, Haseltine W, Patarca R, et al. Complete nucleotide sequence of the AIDS virus, HTLV-III. Nature
47. Ghosn J, Pellegrin I, Goujard C, et al. HIV-1 resistant strains acquired at the time of primary infection massively fuel the cellular reservoir and persist for lengthy periods of time. AIDS
48. Bello G, Casado C, Sandonis V, et al. A subset of human immunodeficiency virus type 1 long-term non-progressors is characterized by the unique presence of ancestral sequences in the viral population. J Gen Virol
49. Markham RB, Wang WC, Weisstein AE, et al. Patterns of HIV-1 evolution in individuals with differing rates of CD4 T cell decline. Proc Natl Acad Sci USA
50. Wang B, Mikhail M, Dyer WB, et al. First demonstration of a lack of viral sequence evolution in a nonprogressor, defining replication-incompetent HIV-1 infection. Virology
51. Deeks SG, Schweighardt B, Wrin T, et al. Neutralizing antibody responses against autologous and heterologous viruses in acute versus chronic human immunodeficiency virus (HIV) infection: evidence for a constraint on the ability of HIV to completely evade neutralizing antibody responses. J Virol
52. Connor RI, Sheridan KE, Lai C, et al. Characterization of the functional properties of env genes from long-term survivors of human immunodeficiency virus type 1 infection. J Virol
53. Sagar M, Wu X, Lee S, et al. Human immunodeficiency virus type 1 V1-V2 envelope loop sequences expand and add glycosylation sites over the course of infection, and these modifications affect antibody neutralization sensitivity. J Virol
54. Altfeld M, Allen TM, Yu XG, et al. HIV-1 superinfection despite broad CD8+ T-cell responses containing replication of the primary virus. Nature
55. Jost S, Bernard MC, Kaiser L, et al. A patient with HIV-1 superinfection. N Engl J Med
56. Koelsch KK, Smith DM, Little SJ, et al. Clade B HIV-1 superinfection with wild-type virus after primary infection with drug-resistant clade B virus. AIDS
57. Ramos A, Hu DJ, Nguyen L, et al. Intersubtype human immunodeficiency virus type 1 superinfection following seroconversion to primary infection in two injection drug users. J Virol
58. Smith DM, Wong JK, Hightower GK, et al. Incidence of HIV superinfection following primary infection. JAMA
59. Gonzales MJ, Delwart E, Rhee SY, et al. Lack of detectable human immunodeficiency virus type 1 superinfection during 1072 person-years of observation. J Infect Dis
60. Tsui R, Herring BL, Barbour JD, et al. Human immunodeficiency virus type 1 superinfection was not detected following 215 years of injection drug user exposure. J Virol
61. Smith DM, Strain MC, Frost SD, et al. Lack of neutralizing antibody response to HIV-1 predisposes to superinfection. Virology
62. Lamine A, Caumont-Sarcos A, Chaix ML, et al. Replication-competent HIV strains infect HIV controllers despite undetectable viremia (ANRS EP36 study). AIDS
63. Barre-Sinoussi F, Chermann JC, Rey F, et al. Isolation of a T-lymphotropic retrovirus from a patient at risk for acquired immune deficiency syndrome (AIDS). Science
Keywords:© 2008 Lippincott Williams & Wilkins, Inc.
neutralizing antibody; long-term nonprogressors; gp120 envelope glycoprotein