Conformational Epitope-Specific Broadly Neutralizing Plasma Antibodies Obtained from an HIV-1 Clade C-Infected Elite Neutralizer Mediate Autologous Virus Escape through Mutations in the V1 Loop

ABSTRACT

Broadly neutralizing antibodies isolated from infected patients who are elite neutralizers have identified targets on HIV-1 envelope (Env) glycoprotein that are vulnerable to antibody neutralization; however, it is not known whether infection established by the majority of the circulating clade C strains in Indian patients elicit neutralizing antibody responses against any of the known targets. In the present study, we examined the specificity of a broad and potent cross-neutralizing plasma obtained from an Indian elite neutralizer infected with HIV-1 clade C. This plasma neutralized 53/57 (93%) HIV pseudoviruses prepared with Env from distinct HIV clades of different geographical origins. Mapping studies using gp120 core protein, single-residue knockout mutants, and chimeric viruses revealed that G37080 broadly cross-neutralizing (BCN) plasma lacks specificities to the CD4 binding site, gp41 membrane-proximal external region, N160 and N332 glycans, and R166 and K169 in the V1-V3 region and are known predominant targets for BCN antibodies. Depletion of G37080 plasma with soluble trimeric BG505-SOSIP.664 Env (but with neither monomeric gp120 nor clade C membrane-proximal external region peptides) resulted in significant reduction of virus neutralization, suggesting that G37080 BCN antibodies mainly target epitopes on cleaved trimeric Env. Further examination of autologous circulating Envs revealed the association of mutation of residues in the V1 loop that contributed to neutralization resistance. In summary, we report the identification of plasma antibodies from a clade C-infected elite neutralizer that mediate neutralization breadth via epitopes on trimeric gp120 not yet reported and confer autologous neutralization escape via mutation of residues in the V1 loop.

IMPORTANCE A preventive vaccine to protect against HIV-1 is urgently needed. HIV-1 envelope glycoproteins are targets of neutralizing antibodies and represent a key component for immunogen design. The mapping of epitopes on viral envelopes vulnerable to immune evasion will aid in defining targets of vaccine immunogens. We identified novel conformational epitopes on the viral envelope targeted by broadly cross-neutralizing antibodies elicited in natural infection in an elite neutralizer infected with HIV-1 clade C. Our data extend our knowledge on neutralizing epitopes associated with virus escape and potentially contribute to immunogen design and antibody-based prophylactic therapy.

INTRODUCTION

Broadly neutralizing antibodies (BNAbs) target trimeric envelope glycoprotein (Env) spikes of human immunodeficiency virus type 1 (HIV-1). Characterization of the BNAbs has provided key clues toward the design and development of both prophylactic and therapeutic vaccines (1–6). A small proportion of individuals chronically infected with HIV-1 develop BNAbs (7–14), and the isolation of several broad and potent neutralizing monoclonal antibodies (MAbs) from such individuals with distinct molecular specificities to viral envelope protein has been reported (15–23). The cross-neutralizing serum antibodies obtained from such individuals (also referred to as elite neutralizers), which have considerable breadth, target epitopes on structurally conserved regions of Env such as the CD4 binding site (CD4bs) (22, 24–26), V1V2, including glycan moieties (19, 20, 27, 28), the gp120-gp41 interface (18, 29), and the membrane-proximal external regions (MPER) (16, 30–32). Several studies have indicated that the variable regions within HIV-1 gp120 contain epitopes targeted by autologous antibodies as well as BNAbs (33–40). Recently the V1V2 region has been linked to the development of broadly cross-neutralizing (BCN) antibodies (35, 41), and the residues between 160 and 172 (notably R166S/K or K169A) in V1V2 have been demonstrated to be associated with virus escape from autologous antibody response (35). Recent studies have further indicated that BCNAb development in vivo is associated with antibody affinity maturation and coevolution of virus, resulting in a considerable degree of somatic hypermutations (19, 20, 23, 26, 35, 42–50). Such information is crucial for the design and development of suitable Env-based immunogens capable of eliciting broad and potent cross-neutralizing antibodies through vaccination.

While a number of studies on the molecular specificities of broadly neutralizing antibodies obtained from African clade C-infected individuals have been reported (9, 37, 51–62), knowledge on immune evasion in Indian clade C-infected elite neutralizers is very limited (63).

In the present study, we examined plasma samples obtained from two hundred asymptomatic and antiretroviral therapy (ART) naive Indian HIV-infected donors and identified plasma with cross-neutralizing antibodies. The molecular specificities of plasma antibodies obtained from an HIV-1 clade C-infected elite neutralizer was characterized in detail that displayed exceptional neutralization breadth across clades of different geographical origins. Interestingly, we found that neutralization breadth was associated with the presence of unique epitopes on the trimeric gp120.

MATERIALS AND METHODS

Ethics statement.

The blood samples were collected under the IAVI Protocol G study from slow-progressing ART naive HIV-1-positive donors from Nellore District of the state of Andhra Pradesh, southern India, by trained clinicians at the YRG Care Hospital following approval and clearance from the Institutional Review Board (IRB) and the Ethics Committee. The serum and plasma samples collected were shipped to the HIV Vaccine Translational Research Laboratory, Translational Health Science and Technology Institute, for further assessment and research on the neutralizing antibody response.

Plasmids, viruses, antibodies, proteins, and cells.

Plasmids encoding HIV-1 envelopes representing distinct clades are shown in Table 1. Monoclonal antibodies used in the study and TZM-bl cells were procured from the NIH AIDS Research and Reagents Reference program and from the IAVI Neutralizing Antibody Consortium (NAC). 293T cells were purchased from the American Type Culture Collection (ATCC). Plasmid DNA encoding BG505-SOSIP.664-D7324, its purified cleaved trimeric protein (64), and pcDNA5-FRT BG505 furin A (65) were kindly provided by John Moore, Weill Cornell Medical College, New York. Purified gp120 TripleMut core protein (66) was obtained from Richard Wyatt, The Scripps Research Institute, through the NAC. HIV-2 7312A and its chimeric constructs were provided by Lynn Morris, NICD, Johannesburg, South Africa.

TABLE 1.

Neutralization breadth of Protocol G G37080 plasma samples collected at two different time points tested against panel of 57 Env-pseudotyped viruses

Purification of monomeric and trimeric Env proteins.

Codon-optimized gp120 plasmid encoding clade C 4-2.J41 (67, 68) gp120 was cloned in pcDNA 3.1/V5-His-TOPO vector and transfected into 293T cells using polyethyleneimine (PEI). Supernatants containing soluble gp120 were filtered through 0.45-μm-pore-size filters and subsequently purified using nickel-nitrilotriacetic acid (Ni-NTA) agarose matrix (Qiagen Inc.) by elution with phosphate-buffered saline (PBS) containing 300 mM imidazole (pH 8.0). The purified monomeric gp120 protein was extensively dialyzed with PBS (pH 7.4), concentrated using Amicon ultracentrifugal filters (Millipore Inc.) with a 30-kDa cutoff, and stored at −80°C until further use.

The trimeric BG505-SOSIP.664 protein was purified using 293F cells essentially as described by Sanders et al. (69). Briefly, the 293F cells were transfected with plasmid DNA encoding both BG505-SOSIP.664 gp140 envelope and furin (65). Supernatant containing soluble BG505-SOSIP.664 gp140 was harvested 72 to 96 h posttransfection, filtered, and passed through a lectin agarose column obtained from Galanthus nivalis (Sigma Inc.). The nonspecifically bound proteins then were washed in PBS (pH 7.4) supplemented with 0.5 M NaCl. The bound proteins then were eluted using 0.5 M methyl α-d-mannopyranoside, extensively dialyzed with 1× PBS, and concentrated. BG505-SOSIP.664 was further purified by Sephadex G-200 size exclusion chromatography (AKTA; GE). Trimeric protein fractions were collected and pooled, their quality was assessed by running in blue native polyacrylamide gel electrophoresis (BN-PAGE), and they were favorably assessed for their ability to bind to only neutralizing and not to nonneutralizing and MPER-directed monoclonal antibodies as described elsewhere (64) by enzyme-linked immunosorbent assay (ELISA).

Depletion of plasma antibodies by monomeric gp120 and trimeric gp140 Env proteins.

Purified soluble monomeric 4-2.J41 gp120 and trimeric BG505 SOSIP.664 proteins, in addition to the MPER peptide (C1C; encoding clade C sequence) (70), were used for the depletion of plasma antibodies where purified proteins were covalently coupled to tosylactivated MyOne Dynabeads (Life Technologies Inc.) according to the manufacturer's protocol. Briefly, 30 mg of beads was coupled with 1 mg of both monomeric and trimeric Env proteins in coupling buffer [0.1 M NaBO₄, 1 M (NH₄)₂SO₄; pH 9.4] overnight at 37°C for 16 to 24 h. Proteins bound to magnetic beads were separated from unbound proteins using a DynaMag 15 magnet (Life Technologies, Inc.). Beads bound to Env proteins next were incubated with blocking buffer (PBS [pH 7.4], 0.1% bovine serum albumin [BSA; Sigma], and 0.05% Tween 20) at 37°C to block the unbound sites. The antigenic integrity of both 4-2.J41 monomeric gp120 and BG505-SOSIP.664 bound to the beads was assessed for their ability to bind VRC01 and 4E10 MAbs (for monomeric gp120) and PGT121, F105, and 4E10 MAbs (for BG505-SOSIP.664) by flow cytometry (FACSCanto; Becton and Dickinson, Inc.).

For depletion studies, G37080 plasma was diluted to 1:50 in Dulbecco's modified Eagle's medium (DMEM) containing 10% fetal bovine serum (FBS), and 500 μl of diluted plasma was incubated with 20 μl of beads at room temperature for 45 min. Unbound plasma antibodies were separated from those that were bound to protein-coated beads using a DynaMag 15 magnet as described above. This step was repeated 4 to 5 times for the depletion of plasma antibodies by monomeric gp120 and 10 to 12 times in the case of BG505-SOSIP.664-coated beads. As a negative control, G37080 plasma antibodies were depleted with uncoated beads in parallel. In addition to ELISA, the percent depletion of G37080 plasma antibodies was assessed by examining the sequential decrease in binding of protein-coated beads with depleted plasma antibodies by fluorescence-activated cell sorting (FACS). PGT121 MAb was taken as a positive control for checking depletion by BG505-SOSIP.664 trimeric Env.

gp120 and gp140 ELISA.

For gp120 ELISA, a high-binding polystyrene microtiter plate (Nunc, Inc.) was coated with 100 μl of monomeric 4-2.J41 gp120 (1 μg/ml) in binding buffer containing 0.1 M NaHCO₃ (pH 8.6) and incubated overnight at 4°C. The gp120-plate was washed once with 1× PBS (pH 7.4) and blocked with 5% nonfat milk for 90 min at 37°C. The plate then was washed three times with 1× PBS followed by the addition of 100 μl of MAbs, as well as the depleted and undepleted plasma antibodies at different dilutions, and incubated for 1 h at room temperature. The wells of the ELISA plate were washed four times with PBS containing 0.1% Tween 20 (PBST), followed by the addition of 100 μl of 1:3,000-diluted horseradish peroxidase (HRP)-conjugated anti-human IgG (Jackson ImmunoResearch, Inc.), and further incubated for 45 min at room temperature. Unbound conjugates were removed by washing with PBST, and color was developed by the addition of 100 μl of 3,3′,5,5′-tetramethylbenzidine (TMB) (Life Technologies, Inc.) substrate. Absorbance was measured at 450 nm in a spectrophotometer.

Binding of antibodies to BG505-SOSIP.664-D7324 trimeric protein was assessed essentially as described by Sanders et al. (69) in a sandwich ELISA. Briefly, a high-binding microtiter plate (Nunc, Inc.) first was coated with D7324 antibody at 10 μg/ml (Aalto Bio Reagents, Dublin, Ireland) followed by blocking extra unbound sites with 5% nonfat milk for 90 min at 37°C. One hundred microliters of BG505.664-D7324 trimeric protein (300 ng/ml) then was added and incubated for 45 min at room temperature. The extent of binding of G37080 plasma antibodies compared to that of known neutralizing monoclonal antibodies was assessed by the addition of primary and HRP-conjugated secondary anti-human antibody as described above.

Neutralization assay.

Neutralization assays were carried out using TZM-bl cells as described before (68). Briefly, Env-pseudotyped viruses were incubated with various dilutions of depleted plasma antibodies and incubated for an hour at 37°C in a CO₂ incubator under humidified conditions, and subsequently 1 × 10⁴ TZM-bl cells were added to the mixture in the presence of 25 μg/ml DEAE-dextran (Sigma, Inc.). The plates were further incubated for 48 h, and the degree of virus neutralization was assessed by measuring relative luminescence units (RLU) in a luminometer (Victor X2; PerkinElmer Inc.).

Amplification, cloning, and mutagenesis of autologous HIV-1 env genes.

Autologous complete env genes were obtained from G37080 plasma as described previously, with slight modifications (68). Briefly, viral RNA was extracted using a high-pure viral RNA kit (Roche Inc.) by following manufacturer's protocol, and cDNA was prepared by reverse transcription-PCR (RT-PCR) using a Superscript III first-strand synthesis kit (Invitrogen Inc.). rev-gp160 env genes were amplified using a Phusion high-fidelity DNA polymerase (New England BioLabs Inc.). The gp160 amplicons were purified and ligated into pcDNA 3.1/V5-His-TOPO (Invitrogen Inc.) vector. Chimeric Env proteins were prepared by overlapping PCR, and point substitutions were made with a QuikChange II kit (Agilent Technologies Inc.) by following the manufacturer's protocol and as described previously (71).

Preparation of envelope-pseudotyped viruses.

Pseudotyped viruses were prepared by the cotransfection of envelope-expressing plasmid with an env-deleted HIV-1 backbone plasmid (pSG3ΔEnv) into 293T cells in 6-well tissue culture plates using a FuGENE6 transfection kit (Promega Inc.). Cell supernatants containing pseudotyped viruses were harvested 48 h posttransfection and then stored at −80°C until further use. The infectivity assays were done in TZM-bl cells (1 × 10⁵cells/ml) containing DEAE-dextran (25 μg/ml) in 96-well microtiter plates, and the infectivity titers were determined by measuring the luciferase activity using Britelite luciferase substrate (PerkinElmer Inc.) with a Victor X2 luminometer (PerkinElmer Inc.).

RESULTS

Identification of an elite neutralizer with HIV-1 clade C infection whose plasma showed exceptional neutralization breadth.

The present study, under IAVI Protocol G, was designed (i) to screen and identify plasma antibodies obtained from Indian donors chronically infected with HIV-1 clade C, with substantial breadth toward neutralizing cross-clade HIV-1 primary variants, and (ii) to elucidate the molecular specificities associated with neutralization breadth. Our hypothesis was that given the genetic distinctness of clade C viruses of Indian and non-Indian origin, as well the likely differences in host genetics between populations and their ancestral origins associated with the modulation of humoral immune responses, the specificities of antibodies developed in vivo associated with neutralization breadth and potency would be different.

Through screening of two hundred plasma samples obtained from chronically infected ART naive Indian patients against a panel of 57 pseudoviruses containing Envs of distinct clades and geographical origins (Fig. 1A), we identified one donor (G37080) whose plasma showed exceptional neutralization breadth. Donor G37080 serum neutralized >90% of the 57 different pseudoviruses tested, with a median 50% inhibitory dose (ID₅₀) value of 533.03 (Table 1 and Fig. 1B).

FIG 1.

(A) Genetic divergence of amino acid sequences of 57 HIV-1 Env (gp160) pseudoviruses used to assess neutralization breadth and potency of G37080 BCN plasma. The maximum likelihood bootstrapped consensus phylogenetic tree was constructed using the Jones-Taylor-Thornton (JTT) substitution model with 50 bootstrapped replicates in Mega 5.2. Bootstrapped values are shown at the nodes of each branch. Hollow circles represent envelopes (16055-2.3 and 92TH021) resistant to neutralization by G37080 BCN plasma. (B) Neutralization breadth of the G37080 BCN plasma obtained at visit 1 and visit 2 were assessed against pseudotyped viruses expressing HIV-1 Env representing different clades and origins. Neutralization titers (median ID₅₀ values) were obtained by titrating Env-pseudotyped viruses against G37080 plasma samples. Values at the top of each bar graph indicate the number of viruses belonging to each clade/origin tested.

Follow-up plasma sample from this donor (G37080) subsequently was obtained after 8 months to assess whether the neutralization breadth and potencies, along with their molecular specificities, were retained and/or improved, as we expected that during the course of disease, the breadth and potency of neutralizing antibodies broadens through somatic hypermutations (72) and/or clonal selection processes. As shown in Fig. 1B and Table 1, follow-up plasma antibodies of donor G37080 (referred to as the visit 2 samples) were found to exhibit neutralization breadth comparable to that of visit 1 plasma. Overall, G37080 BCN plasma was found to potently neutralize pseudoviruses containing Indian clade C Env with a neutralization score of 2.5 (13). Furthermore, the neutralization sensitivity of Env-pseudotyped viruses was found to be correlated with the serum IgG (data not shown), suggesting that the broad neutralization was associated with IgG-specific responses. Taken together, our data indicate that a strong humoral immune response to HIV-1 was mounted in donor G37080 and was maintained over time.

Evidence that G37080 BCN plasma antibodies do not target epitopes in CD4bs, MPER, and known glycan and nonglycan residues in variable domains of Env.

We first examined whether the G37080 BCN plasma contains antibodies directed to the CD4bs on Env. Plasma samples obtained from both visits were pretreated with 25 μg/ml of TripleMut core protein (66), which was a concentration that we found to inhibit the neutralization of 25711-2.4 pseudovirus by VRC01 MAb by >95%. Pretreated plasma subsequently was used to neutralize pseudovirus 25711-2.4 Env, and as shown in Fig. 2, no perturbation of G37080 neutralizing activity was observed against pseudovirus 25711-2.4. A similar observation was made when these plasma antibodies were pretreated with RSC3 core protein (22). In addition, the G37080 BCN plasma antibodies were found to efficiently neutralize IgG1b12- and VRC01-resistant viruses (data not shown). Our data indicated that the G37080 BCN plasma antibodies do not contain CD4bs-directed neutralizing antibodies.

FIG 2.

Assessing dependence of G37080 BCN antibodies to CD4 binding site (CD4bs) region of HIV-1 Env. G37080 BCN plasma samples and VRC01 MAb (concentrations that neutralized 25711-2.4 by >80%) preincubated with different concentrations, as indicated, with TripleMut core (A) and RSC3 (B) proteins were examined for their ability to neutralize 25711-2.4 Env pseudotyped virus in a TZM-bl cell neutralization assay. Note that while VRC01 preabsorbed with both TripleMut and RSC3 proteins showed inhibited neutralization of 25711-2.4 in a dose-dependent manner, no such effect was observed with G37080 BCN plasma, indicating the absence of CD4bs-directed neutralizing antibodies.

To elucidate whether the BCN plasma antibodies are directed to MPER in gp41, we used HIV-2/HIV-1 chimeric viruses (73) that expressed minimal residues of HIV-1 MPER containing epitopes required for MPER-directed MAbs, such as 2F5, 4E10, Z13e, and 10E8. As shown in Table 2, the G37080 BCN plasma from both visits was found to show modest neutralization of HIV-2 expressing HIV-1 clade C MPER (7312-C1C), with ID₅₀ values of 306.42 and 371.02, respectively. We also found that the depletion of G37080 plasma with a clade C MPER peptide (C1C) completely abolished the sensitivity of 7312A-C1C virus to G37080 plasma (Table 3). Our data suggest that although the G37080 BCN plasma neutralized 7312-C1C, the presence of MPER-directed antibodies was not associated with neutralization breadth.

TABLE 2.

Examination of specificity of G37080 plasma antibodies obtained at both visits to HIV Env

HIV type	Region	Fold decrease in ID50a for plasma at visit:
		1	2
HIV-1 Env mutants
HIV-1 25711-2.4 N160A	V2	1.02	<1
HIV-1 25711-2.4 R166A		<1	<1
HIV-1 25711-2.4 K169E		<1	<1
HIV-1 93IN905 R166A		<1	<1
HIV-1 93IN905 K169A		<1	<1
HIV-1 25711-2.4 N332A	V3	1.52	<1
HIV-1 CAP239.G3 N332A		1.35	1.32
HIV-2/HIV-1 chimera	Region of HIV-1	ID50	ID50
HIV-2 7312A	HIV-2 wild type	<20	<20
HIV-2 7312A-C1C	Clade C MPER	306.42	371.02
HIV-2 7312A-C3	2F5 epitope	<20	<20
HIV-2 7312A-C4	4E10, Z13e1, and 10E8 epitopes	334.34	371.27
HIV-2 7312A-C6	4E10 minimal epitope	<20	223.90
HIV-2 7312A-C7	2F5 minimal epitope	<20	<20

^aID₅₀ values refer to the reciprocal dilution that conferred 50% neutralization of viruses in a TZM-bl assay. Assays were done in duplicate and were repeated more than three times.

TABLE 3.

Degree of shift in sensitivity of Env-pseudotyped viruses to G37080 BCN plasma depleted with monomeric and trimeric Envs and C1C peptide

Env-pseudotyped virus	Fold reduction in neutralization (ID50)a
	gp140 trimer (BG505-SOSIP.664)	gp120 monomer (4-2.J41)	MPER (C1C peptide)
25710-2.3	>10.30	1.3	0.83
25711-2.4	>8.52	1.4	1.44
3-5.J25	>7.85	0.9	0.84
4-2.J41	12.11	1.1	1.04
IAVI_C22	>15.92	1.2	1.18
92BR020	>35.08	1.1	1.34
93IN905	3.41	1.2	0.94
JRCSF	>8.75	0.5	0.93
Q23.17	>23.28	1.0	0.98
Du156.12	>15.73	0.8	1.61
HVTR-PG80v1.eJ7	>10.03	0.9	1.12
HVTR-PG80v1.eJ19	>15.60	0.5	1.18
HIV-2 7312A-C1C			>10

^aFold reduction in virus neutralization was obtained by comparing the neutralization titer (ID₅₀ values) of panel viruses against undepleted and depleted G37080 plasma. ID₅₀ values are reciprocal dilutions at which the undepleted and depleted plasma conferred 50% neutralization of the Env-pseudotyped viruses in TZM-bl cells.

We next investigated whether the plasma antibodies of the donor G37080 target residues in variable loops, particularly in V1V2 and V3 regions, which have been shown in several studies to be epitopes targeted by BCN antibodies on HIV-1 Env. We first tested the extent of neutralization by G37080 BCN plasma antibodies of Env-pseudotyped viruses lacking glycans at positions 160 (N160) and 332 (N332) in the V2 region and V3 base, respectively, and also R166 and K169 in the V2 region, which are major targets of recently identified broad and potent neutralizing monoclonal antibodies. In order to test this, two clade C Envs (25711-2.4 and CAP239.G3), containing N160A and N332A substitutions, were tested (Table 2). Our data indicate that the pseudoviruses containing Env expressing the N160 or N332 substitution have sensitivities identical to those of G37080 plasma antibodies. Similar observations were found with R166A and K169A in the 93IN905 Env backbone. Taken together, our observations indicate that G37080 BCN plasma antibodies did not utilize these residues in V2 and V3 regions for neutralization breadth; these have been identified as important epitopes recognized by broadly neutralizing antibodies elicited in clade C infection as described before (35, 41, 56).

Association of neutralization breadth of G37080 plasma with recognition of conformational epitopes on cleaved trimeric Env but not with that in monomeric gp120 or MPER.

In order to examine whether broad neutralization conferred by the G37080 plasma antibodies was through the recognition of epitopes on monomeric gp120 or cleaved near-native Env trimers, we tested the binding of G37080 serum IgG to monomeric 4-2.J41 gp120 and soluble gp140 (BG505-SOSIP.664) by ELISA. We found that in addition to the monomeric 4-2.J41 gp120 (Fig. 3A), G37080 serum polyclonal IgG was found to efficiently bind to the BG505 SOSIP.664-D7324 soluble trimeric Env (Fig. 3B), indicating that the G37080 plasma primarily contains neutralizing antibodies that target epitopes on cleaved Env trimers.

FIG 3.

Binding of G37080 BCN plasma IgG to 4-2.J41 monomeric gp120 (A) and BG505-SOSIP.664-D7324 cleaved trimeric gp140 (B) soluble proteins was assessed by ELISA. IgG purified from HIV-negative healthy donor and known MAbs were used as controls. The extent of binding of the depleted and undepleted G37080 BCN plasma with magnetic beads coated with 4-2.J41 monomeric gp120 (C) and BG505-SOSIP.664 cleaved trimeric gp140 to their respective proteins by ELISA. Note that binding to trimeric protein by ELISA was assessed by using BG505-SOSIP.664 tagged with the D7324 epitope to maintain the native conformation of trimeric Env as described before (69).

We next examined whether binding of the G37080 plasma antibodies to epitopes on cleaved BG505-SOSIP.664 trimeric envelope was associated with neutralization breadth. For this, we tested the ability of G37080 plasma antibodies depleted of both monomeric and trimeric Envs, as well as of MPER peptides, to neutralize a set of Env-pseudotyped viruses, which were found to be sensitive to this particular plasma sample. Purified 4-2.J41 monomeric gp120, BG505-SOSIP.664 trimeric gp140, and C1C MPER peptide bound to the magnetic beads were used to deplete G37080 plasma antibodies as described in Materials and Methods. The depleted BCN G37080 antibodies first were assessed for their binding to 4-2.J41 gp120 monomers, BG505-SOSIP.664-D7324, and C1C peptide and compared to undepleted plasma antibodies by ELISA. As shown in Fig. 3C and D, G37080 plasma depleted with monomeric gp120 and trimeric gp140, respectively, had significantly reduced binding activity against the respective soluble proteins. Similar observation was made with MPER peptide (data not shown). The depleted plasma antibodies subsequently were assessed for neutralization activity using a panel of 12 Env-pseudotyped viruses that were susceptible to untreated G37080 plasma antibodies as mentioned above. As shown in Table 3, depletion with 4-2.J41gp120 monomer and C1C peptide did not show any change in neutralization breadth of G37080 plasma antibodies, while depletion with BG505-SOSIP.664 showed a significant reduction in virus neutralization. Similar observations were made with the BG505-SOSIP.664-depleted PGT121 and C1C peptide-depleted 4E10 MAbs, which lost the ability to efficiently neutralize Env-pseudotyped viruses (16055 and ZM233.6) and HIV-2/HIV-1 (7312A-C1C) chimeric virus compared to their undepleted counterparts (data not shown), validating our data. Interestingly, C1C peptide-depleted G37080 plasma failed to neutralize HIV-2/HIV-1 (7312A-C1C) chimeric virus, indicating that the presence of residual traces of MPER-directed antibodies (as shown in Table 2) is not responsible for neutralization breadth. Furthermore, the examination of sensitive (25711-2.4) and resistant (16055-2.3 and CAP45.G3) chimeric Envs indicated that the BCN G37080 plasma antibodies predominantly target epitopes in the V1V2 region (Table 4) in gp120. Our data clearly indicate a correlation between neutralization breadth and binding of the G37080 BCN plasma antibodies to the conformational epitopes on cleaved trimeric gp120, likely in the V1V2 region; however, we do not rule out the possibility that this BCN plasma targets other discontinuous epitopes in gp120 but not in MPER.

TABLE 4.

Dissection of specificity for autologous neutralization resistance

Chimera and point mutant	Neutralization potency (ID50)
	Fold changea	Effectb
PG80v1.eJ7 Env backbone
V1V2 loop
PG80v2.eJ38 (V1V2) in v1.eJ7	45.75	Decrease
PG80v2.eJ38 (V1) in v1.eJ7	27.97	Decrease
PG80v2.eJ38 (V2) in v1.eJ7	1.35	No effect
Point mutations
PG80v1.eJ7 (D133N)	3.25	Decrease
PG80v1.eJ7 (S143G)	0.87	No effect
PG80v1.eJ7 (D133N + S143G)	2.88	Decrease
PG80v1.eJ7 (T147P)	8.21	Decrease
PG80v1.eJ19 Env backbone
V1V2 loop
PG80v2.eJ38(V1V2) in v1.eJ19	23.05	Decrease
PG80v2.eJ38 (V1) in v1.eJ19	28.61	Decrease
PG80v2.eJ38 (V2) in v1.eJ19	1.87	Increase
Point mutations
PG80v1.eJ19 (D133N)	2.51	Decrease
PG80v1.eJ19 (T139A)	0.99	No effect
PG80v1.eJ19 (T139A + T140D)	2.64	Increase
PG80v1.eJ19 (N143G)	1.38	No effect
PG80v1.eJ19 (T139A + T140D + N143G)	2.16	Increase
PG80v1.eJ19 (T145P)	3.24	Decrease
PG80v2.eJ38 Env backbone
V1V2 loop
PG80v1.eJ7 (V1V2) in v2.eJ38	26.56	Increase
PG80v1.eJ7 (V1) in v2.eJ38	49.62	Increase
PG80v1.eJ7 (V2) in v2.eJ38	1.07	No effect
PG80v1.eJ19(V1V2) in v2.eJ38	12.81	Increase
PG80v1.eJ19 (V1) in v2.eJ38	37.60	Increase
PG80v1.eJ19 (V2) in v2.eJ38	0.94	No effect
Other regions in gp120
PG80v1.eJ7 (V3C3) in v2.eJ38	0.84	No effect
PG80v1.eJ7 (V3C3V4C4) in v2.eJ38	0.90	No effect
PG80v1.eJ7 (C4V5C5) in v2.eJ38	1.08	No effect
PG80v1.eJ19 (V3C3) in PG80v2.eJ38	1.09	No effect
PG80v1.eJ19 (V3C3V4C4) in PG80v2.eJ38	1.15	No effect
PG80v1.eJ19 (C4V5C5) in PG80v2.eJ38	0.96	No effect
Point mutations
PG80v1.eJ19 V1 (T139A+T140D) in v2.eJ38	53.91	Increase
PG80v2.eJ38 (N133D)	4.72	Increase
PG80v2.eJ38 (G143S)	0.94	No effect
PG80v2.eJ38 (N133D+G143S)	3.63	Increase
PG80v2.eJ38 (P147T)	4.11	Increase
Heterologous Env chimera
V1V2 loop
16055-2.3 (25711-2.4 V1V2)	18.38	Increase
25711-2.4 (16055-2.3 V1V2)	2.03	Decrease
CAP45 (25711-2.4 V1V2)	16.84	Increase
25711-2.4 (CAP45-V1V2)	10.54	Decrease

^aFold changes in reciprocal dilution of plasma mediating 50% virus neutralization (ID₅₀).

^bFold increase or decrease in neutralization titer (ID₅₀ values).

Mutations in V1 region confer resistance to autologous viruses to the G37080 plasma antibodies.

In order to decipher the specificity of the G37080 plasma antibodies, we examined the degree of susceptibility of pseudoviruses prepared using env genes amplified from contemporaneous autologous G37080 plasma obtained at the baseline and follow-up visits. As shown in Fig. 4A, both of the Env proteins obtained from visit 2 plasma (HVTR-PG80v2.eJ38 and HVTR-PG80v2.eJ41) were found to be resistant to its contemporaneous plasma antibodies, while Env proteins obtained from visit 1 plasma (HVTR-PG80v1.eJ7 and HVTR-PG80v1.eJ19) were found to be modestly sensitive to visit 2 autologous G37080 plasma antibodies. To facilitate mapping G37080 BCN antibody specificity, we prepared chimeric Envs between a sensitive (HVTR-PG80v1.eJ7 and HVTR-PG80v1.eJ19) and a resistant (HVTR-PG80v2.eJ38) autologous Env by first swapping the V1V2 regions, as their amino acid sequences differed maximally in this region (Fig. 4B). As shown in Table 4, the insertion of the V1V2 sequences of HVTR-PG80v1.eJ7 and HVTR-PG80v1.eJ19 into HVTR-PG80v2.eJ38 conferred Env-pseudotyped viruses expressing HVTR-PG80v2.eJ38 Env with sensitivity to G37080 visit 2 plasma antibodies enhanced by >25- and >12-fold, respectively. Conversely, the neutralization susceptibilities of the Env-pseudotyped viruses expressing HVTR-PG80v1.eJ7 and HVTR-PG80v1.eJ19, which contained HVTR-PG80v2.eJ38 V1V2 sequence corresponding to visit 2 G37080 plasma, were found to be reduced by >45- and >23-fold, respectively. We noted that alterations of regions other than the V1V2 loop in the autologous Env did not confer any change in neutralization sensitivity (Table 4). To further specify residues in the V1V2 loop associated with neutralization sensitivity and resistance of autologous Envs, chimeric Envs and point mutants were prepared and tested for their degree of modulation in susceptibility to autologous G37080 plasma obtained from the second visit. As shown in Table 4, we found that the V1 sequence, but not the V2 sequence, of the sensitive Envs (HVTR-PG80v1.eJ7 and HVTR-PG80v1.eJ19) increased sensitivity to G37080 BCN plasma antibodies by >50 and >37-fold, respectively, when transferred to the resistant HVTR-PG80v2.eJ38 Env. In agreement with this result, V1 of HVTR-PG80v2.eJ38, when transferred into the sensitive Envs described above, increased neutralization resistance by >27- and >28-fold, respectively, to the G37080 visit 2 BCN plasma antibodies. We observed that the removal of a glycan at the 138 position in V1 (T140D) in HVTR-PG80v1.eJ19 mediated the enhanced sensitivity of this Env to G37080 plasma by 2.64-fold (Table 4). Concurrent with this observation, we found that the alteration of the V1 region of PG80v1.eJ19 with the T140D substitution in PG80v2.eJ38 Env exhibited enhanced susceptibility compared to that of the PG80v2.eJ38 Env chimera containing the PG80v1.eJ19 V1 loop, as shown in Table 4. Our data indicate that N138 glycan masks the PG80v1.eJ19 Env from being efficiently neutralized by the autologous plasma compared to that of its contemporaneous counterpart, PG80v1.eJ7 Env. Fine scanning of V1 regions of the autologous Envs further revealed that the N133 glycan motif and P147 residues in the PG80v2.eJ38 Env played a significant role in neutralization resistance to G37080 BCN autologous plasma antibodies (Fig. 4B). Interestingly, all of the V1 chimeras as well as the point mutants showed sensitivities to PG9 MAb comparable to those of their wild types (Table 5), indicating that the shifts in neutralization susceptibilities were not due to changes in Env conformation. Moreover, we noted that both the sensitive and the resistant autologous Envs contain T332 in the V3 base, clearly indicating that the absence of N332 was not associated with resistance to autologous neutralization. Similar observations were made with respect to N160, R166, and K169 amino acid residues, further consolidating that the neutralization conferred by G37080 BCN plasma antibodies was not associated with antibody targeting these epitopes in autologous Envs, and is likely the case for all of the Envs tested against G37080 plasma antibodies.

FIG 4.

(A) Neutralization susceptibility of autologous Envs to contemporaneous G37080 BCN plasma and its follow-up sample from the same donor. Neutralization titers (median ID₅₀) were obtained by titrating pseudotyped viruses expressing autologous Envs obtained from visit 1 and follow-up G37080 plasma to contemporaneous plasma antibodies. Note that both of the Envs obtained from follow-up G37080 plasma (visit 2) were found to be resistant to contemporaneous autologous plasma, while Envs obtained from visit 1 G37080 plasma were found to be sensitive to follow-up plasma antibodies. (B) Alignment of V1V2 amino acid sequences of sensitive and resistant autologous Envs obtained at both visits was done by using seqpublish, available at the HIV Los Alamos database (www.hiv.lanl.gov). Key residues that mediate autologous neutralization resistance are highlighted.

TABLE 5.

Sensitivity of wild type, chimera, and point mutants of autologous Envs to PG9 MAb

Env chimera and mutant	ID50
PG80v1.eJ7 (wild type)	0.12
PG80v1.eJ19 (wild type)	0.97
PG80v1.eJ19 (T139A + T140D)	0.77
PG80v2.eJ38 (wild type)	0.02
PG80v1.eJ7 (V1) in v2.eJ38	0.04
PG80v1.eJ19 (V1) in v2.eJ38	0.05
PG80v1.eJ19 (V1) (T139A + T140D) in v2.eJ38	0.06
PG80v2.eJ38 (N133D)	0.01
PG80v2.eJ38 (P147T)	0.04

DISCUSSION

The identification of the molecular specificities of antibodies elicited in natural infection and that mediate neutralization breadth and potency is key in the design and development of suitable Env-based immunogens capable of eliciting similar antibody responses upon vaccination. In the present study, we characterized the molecular specificity of plasma antibodies obtained from an Indian elite neutralizer (G37080) infected with HIV-1 clade C that displayed exceptional cross-neutralization of different clades of distinct geographical origins. The G37080 plasma was found to contain the most broad and potent cross-neutralizing antibodies among the two hundred plasma samples obtained from Indian patients chronically infected with HIV-1. Plasma samples collected from the G37080 donor at two time points at 8 months apart showed similar neutralization breadth with modest increase in potency in the follow-up visit, indicating an association with the sustained maturation of antibody-producing B cells in this individual.

Since polyclonal plasma antibodies are not suitable for epitope mapping, we examined the specificity of the G37080 BCN plasma by making use of mutant viruses with specific point substitutions of known neutralizing epitopes with nonspecific amino acids and via depletion with monomeric and trimeric Envs in addition to MPER peptide. The G37080 plasma antibodies did not show dependence on the N160/K169 and N332 epitopes in the V2 apex and V3 base, respectively. Our data also are consistent with the target epitopes of the G37080 BCN antibodies being distinct from those which are recognized by 2G12 (74), PGT121-128 (17), and PGT130-131 and PGT135 (19) (e.g., residues at the following positions: 295, 297, 301, 332, 334, 386, 388, 392, 394, 448, and 450); thus, BCN G37080 antibodies appear to target a new epitope. Our data highlighting the N332-independent development of neutralizing antibodies in clade C-infected donor G37080 also differ from recent findings (57, 75–77) associating N332 with the development of broad and potent neutralizing antibody, especially in clade C infection in African donors. Moreover, recent studies indicating the role of K169 as a target of BCN antibodies obtained from a clade C-infected South African donor (41, 56) and the observation that vaccine-induced protection in the RV144 vaccine trial was associated with antibodies targeting epitopes, including K169 in the V2 apex (27, 78), prompted us to examine whether broad neutralization of the G37080 plasma antibodies also was dependent on the K169 epitope. In the present study, the neutralization potency of G37080 was unaffected by N160A/K169A knockout mutations, and we also observed that both sensitive and resistant autologous Envs obtained from both visits contain N160 and K169 in the V2 region. Hence, owing to the lack of association of neutralization breadth of the G37080 BCN antibodies with N160, K169, and N332 dependencies, our study further highlighted that there is a likelihood of differences in the development pathway of elicitation of broadly neutralizing antibodies in individuals infected with HIV-1 clade C, particularly those with ethnically distinct variants.

Wibmer et al. (41) recently demonstrated an association between the evolution of a broadly neutralizing antibody response in a clade C-infected donor with shifts in antibody specificities from the recognition of epitopes in V2 to the CD4bs. In the present study, the G37080 neutralizing plasma antibodies obtained from both visits were found not to be absorbed by the TripleMut (66, 79) and RSC3 (22) core proteins, which effectively absorb antibodies directed to the CD4bs. This result indicates a lack of development of CD4bs-directed neutralizing antibodies during the disease course in the G37080 donor. Additionally, the absence of MPER-directed antibodies from G37080 plasma was found, although a negligible antibody titer (1:300 reciprocal dilutions) to the HIV2/HIV1 (C1C) chimera was observed with plasma samples from both visits. However, the neutralization breadth of the G37080 plasma was not found to be associated with the presence of MPER-directed antibody. Nonetheless, we do not rule out the possibility that in the further course of infection, this donor may be able to develop MPER-directed antibodies.

Recent studies have shown that neutralizing antibodies that target conformational epitopes bind exclusively to the cleaved near-native trimeric Envs (15, 64, 80, 81). In the present study, we found that the absorption of G37080 plasma antibodies to soluble trimeric BG505-SOSIP.664 Env was associated with the depletion of neutralizing activity in G37080 BCN plasma. However, we do not rule out the possibility of the presence of 39F-, 19b-, and 14e-like nonneutralizing antibodies that were reported to bind to BG505-SOSIP.664 trimeric Env (69). Our findings indicate that the G37080 BCN antibodies target conformational epitopes in gp120. Our observation also highlights that native-like trimeric Envs, such as BG505-SOSIP.664, can be utilized in selecting antigen-specific memory B cells, as reported earlier (82), from donor G37080 toward isolation of MAb correlating with broad neutralization displayed by the plasma antibodies.

We made use of env clones obtained from autologous G37080 plasma from both time points to refine the fine specificity of the G37080 BCN plasma antibodies. By examining chimeric Envs and mutant viruses, we identified key residues in the V1 loop associated with neutralization resistance. Interestingly, the Env chimera and mutant viruses showed susceptibility to PG9 MAb comparable to that of their respective wild-type Envs, indicating that they did not alter Env conformation. We identified a glycan at the 133 position and a proline residue at the 147 position within the V1 loop of the resistant Env (PG80v2.eJ38) that were found to be associated with neutralization escape, which indicated that these are contact sites for the G37080 BCN plasma antibodies. Thus, from our study we conclude that changes in V1 loop sequence are associated with the escape of autologous viruses to the BCN G37080 plasma. Additionally, an examination of the degree of susceptibilities of pseudoviruses expressing chimeric heterologous Envs to the G37080 plasma revealed that the BCN plasma antibodies predominantly target epitopes in the V1V2 region in gp120. However, we do not rule out the possibility of the contribution of other discontinuous epitopes in gp120 in mediating neutralization breadth. The isolation and identification of monoclonal antibodies from this elite neutralizer donor (G37080) will help precisely map specific epitopes associated with neutralization breadth and potency.

In summary, we identified an HIV-1-infected elite neutralizer whose plasma showed exceptional neutralization breadth, and we provided evidence that it targets novel conformational epitopes on trimeric Env, predominantly in the V1V2 region, not reported previously. Moreover, the neutralization resistance of the autologous Envs to G37080 plasma is associated with substitutions of novel residues within the V1 loop that form the key contact points of the BCN plasma antibody. The identification of novel epitopes associated with broad neutralization of HIV-1, in particular the major circulating clade C strains, will significantly contribute to efforts toward effective immunogen design.

Funding Statement

This study was made possible by the generous support of the American people through the United States Agency for International Development (USAID) through the IAVI, support from a THSTI-IAVI HIV Vaccine Design Program grant through the Department of Biotechnology, Government of India, in part by a grant from the Department of Science and Technology, Government of India (DST/INT/SAFR/Mega-P3/2011 to J.B.), and in part by a DBT National Bioscience Research Award [BT/HRD/NBA34/01/2012-13(iv) to J.B.]. The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.