Viral Characteristics Associated with Maintenance of Elite Neutralizing Activity in Chronically HIV-1 Clade C-Infected Monozygotic Pediatric Twins

Herein, we report the longitudinal development of bnAbs in a pair of chronically HIV-1 clade C-infected monozygotic pediatric twins, AIIMS_329 and AIIMS_330, who acquired the infection by vertical transmission. The plasma from both donors, sharing a similar genetic makeup and infecting virus, showed the evolvement of bnAbs targeting common epitopes in the V2 and V3 regions of the envelope, suggesting that bnAb development in these twins may perhaps be determined by specific sequences in the shared virus that can guide the development of immunogens aimed at eliciting V2 and V3 bNAbs. Characterization of the neutralization-sensitive and -resistant viruses coevolving with bNAbs in the contemporaneous AIIMS_330 plasma provides information toward understanding the viral alterations that may have contributed to the development of resistance to bnAbs. Further longitudinal studies in more monozygotic and dizygotic twin pairs will help in delineating the role of host and viral factors that may contribute to the development of bnAbs.

ABSTRACT

Broad and potent neutralizing antibodies (bnAbs) with multiple epitope specificities evolve in HIV-1-infected children. Herein, we studied two antiretroviral-naive chronically HIV-1 clade C-infected monozygotic pediatric twins, AIIMS_329 and AIIMS_330, with potent plasma bnAbs. Elite plasma neutralizing activity was observed since the initial sampling at 78 months of age in AIIMS_330 and persisted throughout, while in AIIMS_329 it was seen at 90 months of age, after which the potency decreased over time. We evaluated potential viral characteristics associated with the varied immune profiles by generating single genome-amplified pseudoviruses. The AIIMS_329 viruses generated from the 90-month time point were neutralization sensitive to bnAbs and contemporaneous plasma antibodies, while viruses from the 112-month and 117-month time points were resistant to most bnAbs and contemporaneous plasma. AIIMS_329 viruses developed resistance to plasma neutralizing antibodies (nAbs) plausibly by N160 glycan loss and V1 and V4 loop lengthening. The viruses generated from AIIMS_330 (at 90 and 117 months) showed varied susceptibility to bnAbs and autologous contemporaneous plasma antibodies, while the viruses of the 112-month time point, at which the plasma nAb specificities mapped to the V2 glycan, V3 glycan, and CD4 binding site (CD4bs), were resistant to contemporaneous plasma antibodies as well as to most bnAbs. Chimeric viruses were constructed from 90-month-time-point PG9-sensitive AIIMS_329 and AIIMS_330 viruses with swapped V1V2 regions of their respective evolved viruses (at 112 and 117 months), which led to higher resistance to neutralization by PG9 and autologous plasma antibodies. We observed the evolution of a viral pool in the AIIMS_330 donor comprising plasma antibody neutralization-sensitive or -resistant diverse autologous viruses that may have contributed to the development and maintenance of elite neutralizing activity.

IMPORTANCE Herein, we report the longitudinal development of bnAbs in a pair of chronically HIV-1 clade C-infected monozygotic pediatric twins, AIIMS_329 and AIIMS_330, who acquired the infection by vertical transmission. The plasma from both donors, sharing a similar genetic makeup and infecting virus, showed the evolvement of bnAbs targeting common epitopes in the V2 and V3 regions of the envelope, suggesting that bnAb development in these twins may perhaps be determined by specific sequences in the shared virus that can guide the development of immunogens aimed at eliciting V2 and V3 bNAbs. Characterization of the neutralization-sensitive and -resistant viruses coevolving with bNAbs in the contemporaneous AIIMS_330 plasma provides information toward understanding the viral alterations that may have contributed to the development of resistance to bnAbs. Further longitudinal studies in more monozygotic and dizygotic twin pairs will help in delineating the role of host and viral factors that may contribute to the development of bnAbs.

INTRODUCTION

Disease progression in HIV-1-infected children is observed to develop in a biphasic manner. More than 50% of the infected children, if they are not initiated on antiretroviral therapy (ART), progress to AIDS within a couple of years of infection due to an immature immune system and higher viremia (1, 2). Infected children who survive beyond the initial years mostly develop chronic disease (3) with slow progression. HIV-1-infected children have been shown to mount potent plasma broadly neutralizing antibodies (bnAbs) earlier than adults and more frequently with multiple bnAb specificities than adults (4–6). The bnAb BF520.1 was recently isolated from an infant at 1 year postinfection (p.i.) (4, 7). The presence of epitopes targeted by the second-generation bnAbs on the functional native-like HIV-1 envelope trimers identifies such envelopes as candidates for HIV-1 immunogen design. The native-like HIV-1 gp140 trimeric envelope glycoprotein BG505.T332.SOSIP.664.C2 T332N (8–10), generated from the circulating virus of a clade A-infected infant, is the best immunogen documented so far. A recent study of 303 HIV-1 transmission pairs showed that among the majority of infecting viral strains, only a few had the ability to induce bnAb responses (11). Tracing the virological characteristics of such rare viral antigens capable of initiating and shaping potent antibody responses to HIV-1 will provide key insights for future vaccine strategies capable of eliciting similar bnAb responses (11, 12). Thus, pediatric HIV-1 infection, with its unique characteristics of the rapid and frequent induction of potent bnAbs with multiple specificities, provides a unique model to understand the virological characteristics that can be utilized for future HIV-1 vaccine candidates capable of inducing bnAbs.

Chronically HIV-1-infected children have been shown to have potent plasma bnAbs with diverse epitope specificities (5, 6). In our pediatric cohort of antiretroviral-naive HIC-1 clade C (HIV-1C)-infected children previously characterized for plasma antibody responses (6, 13–18), we identified a pair of chronically infected antiretroviral-naive HIV-1C-infected children, AIIMS_329 and AIIMS_330, defined herein as genetically identical twins by their identical HLA phenotypes, who had acquired the infection at birth by vertical transmission. Both twins AIIMS_329 and AIIMS_330 developed potent plasma neutralizing antibodies (nAbs), with the latter showing elite neutralizing activity since the first sampling. Elite neutralizers are a rare subset of the top 1% of HIV-1-infected individuals with the ability to neutralize pseudoviruses belonging to multiple clades at 50% inhibitory dose (ID₅₀) titers of 300 or more (19). A systematic analysis of the circulating viral variants in elite neutralizers over time, their antibody imprinting ability (11), and their correlation with the evolving autologous plasma neutralization response (19) can provide useful information for immunogen design.

We evaluated AIIMS_329 and AIIMS_330 for a period of 60 months with time-matched sampling since their baseline sampling in 2013 to decipher the viral characteristics responsible for the induction and maintenance of potent plasma bnAbs. One of the twins, AIIMS_330, showed elite neutralizing activity at the first sampling that improved with time. The AIIMS_330 plasma polyclonal antibodies showed the presence of bnAbs with multiple epitope specificities that increased in potency at the follow-up time points, and AIIMS_330 had multiple circulating viral variants with varied susceptibility to contemporaneous autologous plasma antibodies and bnAbs. The plasma of the second twin, AIIMS_329, developed elite neutralizing activity at the 90-month sampling, which, however, decreased with time. In order to determine the virological traits associated with the diverse plasma bnAb response, we generated envelope pseudoviral clones of the circulating viruses from the twins’ plasma samples at different time points and tested their susceptibility to neutralization by the contemporaneous plasma antibodies as well as to neutralization by the known bnAbs to identify the neutralizing determinants on these viruses of pediatric origin. After the 90-month time point, AIIMS_329 showed the presence of circulating viral variants that were resistant to neutralization by the autologous contemporaneous plasma and V2 glycan-targeting bnAbs, whereas a mixture of neutralization-sensitive and -resistant viruses was observed in AIIMS_330. Taken together, our data show the presence of multiple and distinct circulating viral variants with varied susceptibility to autologous plasma antibodies and bnAbs that coevolved with the polyclonal bnAbs in the plasma of the HIV-1C-infected pediatric twins, despite their similar genetic background and source of infecting virus.

(This article was submitted to an online preprint archive [20].)

RESULTS

Evolution of broadly neutralizing antibodies (bnAbs) in plasma of genetically identical twins against diverse clades of HIV-1.

In this study, we evaluated a total of nine independent and time-matched plasma samples from AIIMS_329 and AIIMS_330, pediatric HIV-1 clade C-infected donor twins earlier identified to be long-term nonprogressors (6). Both the twins were asymptomatic and antiretroviral naive throughout the duration of the study. High-resolution HLA genotyping revealed that AIIMS_329 and AIIMS_330 were identical at the HLA-A, -B, and -DRB1 loci and contained wild-type CCR5 alleles (data not shown). The detailed two field HLAs were HLA-A*02:11, *24:02; HLA-B*35:03*40:06; HLA-DRB1*14:04*15:01. The baseline (first) sampling of the identical twins was done at the age of 78 months, and they were then longitudinally followed up for a period of 60 months up to the age of 138 months. The CD4⁺ T cell counts and viral loads at the respective time points are given in Table 1.

TABLE 1.

Clinical and immune profiles of AIIMS_329 and AIIMS_330 through the duration of 90 to 138 months of age

Visit code (mo p.i.)	Age (mo)		No. of CD4+ T cells/μl		HIV-1 viral load (no. of RNA copies/ml)
	AIIMS_329	AIIMS_330	AIIMS_329	AIIMS_330	AIIMS_329	AIIMS_330
78	78	78	1,280	1,174	39,000	27,500
82	82	82	1,120	1,045	45,210	66,350
90	90	90	984	1,203	47,810	74,330
102	102	102	1,069	1,152	42,360	85,400
112	112	112	1,432	1,137	52,130	1102,00
117	117	117	783	789	56,470	151,210
124	124	124	827	788	38,450	141,250
131	131	131	682	578	46,520	124,560
138	138	138	743	734	41,130	185,850

Longitudinal assessment of the virus-neutralizing activity of the plasma antibodies in AIIMS_329 and AIIMS_330 was done over a period of 60 months utilizing a multiclade panel of 50 HIV-1 envelope pseudoviruses representing strains circulating across the globe (21). The HIV-1-neutralizing activity was assessed by performing neutralization assays using a TZM-bl cell-based luciferase reporter assay (22). The potency of the plasma antibodies was determined by the use of neutralization scores, defined as the weighted average of the log-transformed ID₅₀ titers across the 50-virus panel (19). Breadth was determined by the percentage of viruses neutralized in the 50-virus panel at ID₅₀ titers of >50. For the 78-month baseline sample, the plasma antibodies of both the twins showed neutralizing activity (ID₅₀ titers, >100) against the majority of the pseudoviruses in the panel (36 out of 50 for AIIMS_329 and 38 out of 50 for AIIMS_330) (Fig. 1), though the inclusion of several tier 1 viruses may have overstated the breadth in the plasma samples. The AIIMS_329 plasma antibodies demonstrated an increase in potency (geometric mean titers [GMT], 268 to 379; neutralization score, 2.42 to 2.58) up to 90 months, followed by a reduction in neutralizing antibody potency (GMT, 197; neutralization score, 2.29) until the final sampling time point (Fig. 2). The AIIMS_330 plasma antibodies demonstrated an increase in neutralization breadth (from 76% to 100% of viruses were neutralized) and potency (GMT, 306 to 786; neutralization score, 2.49 to 2.89) with time (Fig. 2). On log transformation of the ID₅₀ titers, a neutralization score of 2.47 corresponds to an ID₅₀ of 300, and therefore, using a cutoff 2.5, the plasma samples from AIIMS_329 and AIIMS_330 were assessed for elite neutralizing activity. AIIMS_329 plasma showed elite neutralizing activity at 90 months, after which the plasma potency decreased over time (Fig. 2A). At the baseline sampling at 78 months, AIIMS_330 plasma showed elite neutralizing activity, and unlike AIIMS_329 plasma, the elite neutralizing activity consistently increased over time until the final sampling time point. As AIIMS_329 plasma showed a loss of potency and elite neutralizing activity, we next assessed whether the loss of neutralizing activity was across all clades or specific to certain clades. We observed the loss of potency in AIIMS_329 plasma against all clades tested (Fig. 2B); interestingly, the increase in potency in AIIMS_330 plasma was seen against all clades (Fig. 2C).

FIG 1.

The evolution of the neutralization breadth of AIIMS_329 and AIIMS_330 plasma antibodies against a multiclade panel of 50 pseudoviruses over a period of 60 months (78 months to 138 months) was assessed, with the plasma neutralization ID₅₀ titers being color coded according to the key given in the figure. Murine leukemia virus (MuLV) was used as a negative control.

FIG 2.

Elite neutralizing activity in AIIMS_329 and AIIMS_330 was assessed by their neutralization score against the multiclade 50-virus panel, with a cutoff of 2.5 depicting elite neutralizing activity. (A) AIIMS_330 showed elite neutralizing activity since 78 months of age, which increased in potency over time, while AIIMS_329 showed elite neutralizing activity at the 90-month time point, which then decreased over time. (B) The reduction in the neutralization scores of AIIMS_329 observed was similar across all clades. (C) The increase in the neutralization scores of AIIMS_330 observed was distributed similarly across all clades.

AIIMS_330 developed bnAbs targeting V2 glycan, V3 glycan, and CD4bs epitopes, showing an increase in potency of V2 glycan-directed antibodies with time.

In order to identify the neutralizing specificities of AIIMS_329 and AIIMS_330 plasma antibodies that may have contributed to their potent neutralizing activity, we tested the plasma neutralization of mutant pseudoviruses harboring mutations at key residues of V2 glycan and V3 glycan to assess glycan dependence (5, 6); binding reactivity with the RSC3 wild-type probe and its mutant, RSC3Δ371I/P363N (23), for the presence of CD4 binding site (CD4bs) dependence; and MPER peptides (MPER-B and -C) (24, 25) for MPER dependence.

To test for V2 glycan dependence, we used Indian clade C pseudovirus 25710 and its N156K, N160K, K169E, and K171E single-base-mutant and K169E-K171E double-mutant pseudoviruses. Dependence was assigned when the ID₅₀ was at least 3-fold less than the ID₅₀ for the wild-type virus. PG9, a V2 glycan-directed antibody, was used as a positive control, and HIV-1-seronegative healthy donor plasma samples were used as a negative control. Based on these criteria, V2 glycan-directed neutralizing antibodies were found in AIIMS_330 plasma at all time points, including the baseline, while AIIMS_329 showed the emergence of V2 glycan-directed plasma neutralizing antibodies at 102 months. The V2 glycan antibodies initially showed dependence solely on N160. Dependence on K169 and K171 was observed from 124 months onwards (Fig. 3). In both twins, plasma V2 glycan-dependent neutralizing activity increased with time. Testing for the N332 dependence using the N332K mutant of the 25710 virus showed that V3 supersite glycan-directed nAbs evolved from 117 months onwards in the plasma of both twins (Fig. 3).

FIG 3.

The specificities of the plasma antibodies in AIIMS_329 and AIIMS_330 were mapped using single-base-mutant pseudoviruses and probes. The dependence on key residues within the V2 glycan bnAb epitope (N156, N160, K169, K171, and K169-K171) (a) and the V3 glycan bnAb epitope (N301, N332, and N301-N332) (b) was mapped using the 25710 wild-type pseudovirus and its single-base mutants. Dependence was assigned when the ID₅₀ for the mutant viruses was 3-fold or lower than the ID₅₀ for the wild-type virus. (c) The presence of CD4bs antibodies was mapped by a binding ELISA using the proteins of RSC3 and its mutant, RSC3Δ371I/P363N. Shown are the fold change in the optical density at 450 nm at a plasma dilution of 1:100 for RSC3 compared to its mutant, RSC3Δ371I/P363N, with a cutoff of 3 suggesting the presence of CD4bs antibodies. PG9, PGT121, and VRC01 were used as positive controls, and plasma from seronegative healthy donors was used as a negative control. UD, undetermined; NR for plasma samples from healthy controls, nonreactivity.

The presence of CD4bs-directed antibodies was determined by performing enzyme-linked immunosorbent assays (ELISAs) using wild-type RSC3 protein and its mutant, RSC3Δ371I/P363N. The CD4bs-directed bnAb VRC01 was used as a positive control. Plasma samples from two healthy HIV-1-seronegative donors were used as negative controls. A reduction of 3-fold or more in binding to RSC3Δ371I/P363N compared to the binding to RSC3 was scored as positive for the presence of CD4bs-directed antibodies. AIIMS_330 plasma showed the presence of CD4bs antibodies at 112 and 117 months, which at later time points waned, and no CD4bs dependence was seen at 131 and 138 months (Fig. 3). AIIMS_329 did not show a significant difference in binding to RSC3 and RSC3Δ371I/P363N at any of the time points tested, indicating the absence of CD4bs-directed antibodies in the plasma. For the presence of antibodies against MPER, a binding ELISA using MPER peptides was used. The MPER-directed bnAb 2F5 was used as a positive control, and two HIV-1-seronegative healthy donor plasma samples were used as a negative control. We did not observe MPER-directed plasma antibodies at any of the time points tested for either of the twins’ plasma samples (data not shown).

Development of resistance to V2 glycan-directed bnAbs as well as autologous plasma antibodies in the evolving viruses of AIIMS_329 and AIIMS_330.

As seen in plasma neutralization and mapping assays, the most striking differences observed in both the twins were in the time frame of 90 months to 117 months, with a distinct reduction in AIIMS_329 and an increase in AIIMS_330 plasma neutralization potency and a dependence on multiple epitopes. In order to determine the potential viral characteristics associated with the varied immune profiles in the two twins, we generated pseudoviruses by single-genome amplification (SGA) of the envelope gene from HIV-1 RNA, isolated from the plasma. Further, all the amplicons generated by SGA were directly sequenced to assess the viral diversity. At least 20 amplicons from each time point were sequenced to give a 90% confidence interval, to cover most of the circulating strains with a population frequency of 10%. The sequences of the viral amplicons were checked for their clade (by use of the REGA HIV subtyping tool), pairwise distance to a reference sequence (by use of the PhyloPlace program), and coreceptor usage (by use of the Web PSSM tool). All amplicons belonged to clade C, had the highest phylogenetic relatedness to the reference sequence C.IN.95.95IN21068.AF067155 (GenBank accession number AF067155) with a median pairwise sequence distance of 0.093, and utilized CCR5 as the coreceptor. Attempts were made to generate functional pseudoviruses from all the SGA-amplified viral envelope clones; however, we were unable to generate functional pseudoviruses from all SGA amplicons. The AIIMS_329 pseudoviruses generated from the sample collected at the 90-month time point were susceptible to neutralization by the majority of the second-generation bnAbs as well as contemporaneous plasma antibodies (Fig. 4); viruses from later time points (112 months and 117 months) were, however, resistant to most of the V2 glycan-dependent bnAbs and less susceptible to neutralization by the contemporaneous plasma (Fig. 4). The AIIMS_330 viruses obtained at 90, 112, and 117 months showed varied susceptibility to V2 glycan-dependent bnAbs (except for one pseudovirus from the 117-month time point, the 330.16.E6 virus, which demonstrated a high level of susceptibility to neutralization by most of the bnAbs tested), were resistant to both VRC01 and PGT151, and were susceptible to neutralization by autologous contemporaneous plasma antibodies. The AIIMS_330 pseudoviruses from the 112-month time point, at which plasma neutralizing antibody specificities mapped to the V2 glycan, V3 glycan, and CD4bs, were less susceptible to neutralization by V2 glycan-directed antibodies and resistant to neutralization by VRC01 and PGT151 as well as contemporaneous plasma antibodies (Fig. 4). All the viruses generated from AIIMS_329 and AIIMS_330 within the time frame of 90 to 117 months remained susceptible to V3 glycan-dependent antibodies 10-1074, BG18, AIIMS-P01 PGT121, and PGT128.

FIG 4.

Neutralization susceptibilities of envelope pseudoviruses prepared from plasma samples from both the twins to broadly neutralizing antibodies recognizing major epitopes of viral envelope and contemporaneous autologous plasma. The 50% inhibitory concentrations (IC₅₀) of the bnAbs and the ID₅₀ of the plasma samples were determined in a TZM-bl cell-based neutralization assay by titrating the pseudoviruses with serially diluted bnAbs and plasma. The values given are the means from three independent replicates.

Development of viral resistance in both the twins showed distinct mechanisms of escape from contemporaneous plasma antibodies and bnAbs.

In order to understand the alterations in the evolving viruses that led to the development of neutralization resistance to V2 glycan-directed bnAbs, we evaluated the viral sequence changes within the epitopes targeted by these bnAbs. The sequences of SGA-amplified envelope genes representing the circulating strains of both the twins at 90 months, 112 months, and 117 months were used to generate the population frequency of amino acid residues in the V2 glycan epitope. For AIIMS_329, the sequence logo generated showed the loss of the N160 glycan in evolving viruses. The lysine residue at position 171 was absent, while the lysine at position K169 was retained in all the SGA amplicons (Fig. 5A). The core epitope for V3 glycan-directed bnAbs remained unchanged in the evolving viruses (Fig. 5B), while the core epitope for CD4bs-directed bnAbs showed mutations in the evolving viruses within hypervariable loop 5 (Fig. 5C). No known escape mutations were observed for the MPER-directed bnAbs (Fig. 5D). For AIIMS_330, no changes were observed at the N160 glycan or K171 in the evolving viruses, while the K169 residue was absent (Fig. 6A). The core epitope for V3 glycan-directed bnAbs remained unchanged (Fig. 6B), while the core epitope for CD4bs-directed bnAbs showed mutations in loop D (particularly the N276 glycan), β20/21, and hypervariable loop 5 (Fig. 6C). No known escape mutations were observed for MPER-directed bnAbs in the evolving viruses of AIIMS_330 either (Fig. 6D).

FIG 5.

Sequence logos depict the diversity of amino acids in the key bnAb epitopes of V2 glycan bnAbs (A), V3 glycan bnAbs (B), CD4bs bnAbs (C), and MPER bnAbs (D) in the viral clones sequenced in AIIMS_329 at 90 months, 112 months, and 117 months. Positions are numbered based on the HXB2 reference sequence.

FIG 6.

Sequence logos depict the diversity of amino acids in the key bnAb epitopes of V2 glycan bnAbs (A), V3 glycan bnAbs (B), CD4bs bnAbs (C), and MPER bnAbs (D) in the viral clones sequenced in AIIMS_330 at 90 months, 112 months, and 117 months. Positions are numbered based on the HXB2 reference sequence.

V1V2 loop lengthening has been implicated as one of the potential mechanisms for the development of viral resistance to V2 glycan-directed bnAbs; therefore, we next aligned the sequences of the V1V2 loops of AIIMS_329 and AIIMS_330 viruses to assess whether loop lengthening was responsible for the development of viral resistance to V2 glycan-directed bnAbs as well as the contemporaneous plasma. In the functional AIIMS_329 pseudoviruses, the key glycan residue N160 was mutated to Y, thereby leading to the loss of the N160 glycan. As the plasma V2 glycan-directed antibodies in AIIMS_329 were predominantly dependent on N160, the N160Y mutation may perhaps be one of the potential escape mechanisms employed by AIIMS_329 viruses to develop neutralization resistance to contemporaneous plasma antibodies. Although the key residues at N156, N160, and K171 were preserved in the AIIMS_330 viruses that were resistant to contemporaneous plasma antibodies, these viruses had longer V1 and V2 loops with an increased number of potential N-linked glycosylation sites (PNGS) (Fig. 7). Distinct mechanisms in AIIMS_329 viruses, i.e., by the loss of critical epitopes like N160 (Fig. 7), and AIIMS_330 viruses, i.e., by V1V2 loop lengthening (26, 27) and an increased number of PNGS, may have led to the development of resistance to neutralization by contemporaneous plasma antibodies (Table 2).

FIG 7.

Alignment showing the sequence of the V1V2 loops of AIIMS_329 (A) and AIIMS_330 (B) functional pseudoviruses and the distinct mechanism employed by them to escape autologous plasma neutralization. The V1 and V2 loops are shown by the horizontal bars, and key residues (N160, K169, K171) involved in neutralization by V2 glycan bnAbs are labeled with vertical bars.

TABLE 2.

Variable region characteristics of AIIMS_329 and AIIMS_330 in the time frame of 90 to 117 months of age^a

Twin and age (mo)	No. of sequences (SGA amplicons)	V1		V2		V3		V4		V5
		Length	No. of PNGS	Length	No. of PNGS	Length	No. of PNGS	Length	No. of PNGS	Length	No. of PNGS
AIIMS_329
90	26	26	2	42	2	36	1	30	3	16	2
112	24	28	5	56	2	36	1	37	6	12	2
117	29	40	5	50	2	36	1	38	5	12	1
P value		****	****	****				****	****
AIIMS_330
90	24	24	4	44	3	37	1	37	5	17	2
112	20	23	4	55	5	37	1	36	3	17	2
117	21	34	4	59	5	37	1	36	5	16	2
P value		****		****	****

^aThe variable loop length (in nucleotides) and number of PNGS of longitudinal SGA-amplified amplicons of AIIMS_329 and AIIMS_330 were calculated using the Variable Region Characteristics tool available at the HIV Database (https://www.hiv.lanl.gov/content/sequence/VAR_REG_CHAR/index.html). The Kruskal-Wallis test was used to compare the difference in variable loop length and number of PNGS of the SGA amplicons in the time frame of 90 months to 117 months. Given are the average loop length and PNGS of all SGA amplicons for the respective time point. ****, P < 0.0001.

V1V2 loop lengthening conferred resistance to autologous plasma antibodies in chimeric viruses of AIIMS_329 and AIIMS_330.

In order to evaluate whether V1V2 loop lengthening in the evolving viruses of AIIMS_329 and AIIMS_330 contributed to the development of resistance to neutralization by V2 glycan-directed bnAbs and autologous plasma antibodies, we swapped the V1V2 loop from the evolved viruses from later time points (112 months and 117 months) into the preceding pseudoviral envelope backbone of the respective viruses from 90 months, followed by testing their susceptibility to neutralization by PG9 and autologous plasma antibodies (Table 3). 329.14.B1 and 330.14.A20 were the 90-month viral clones used as the envelope backbone (as these clones were sensitive to PG9), in which the V1V2 loops from their evolved viruses (resistant to bnAbs) from the respective later time points were swapped. As shown in Table 3, on swapping the V1V2 loop from the evolving viruses into AIIMS_329 and AIIMS_330 pseudoviruses from 90 months, the chimeric viruses showed higher resistance than their respective wild-type viruses to neutralization by the autologous plasma antibodies from 117 months.

TABLE 3.

Neutralization susceptibility of chimeric envelope pseudoviruses prepared from AIIMS_329 and AIIMS_330 to autologous plasma from 117 months and V2 glycan bnAb PG9^a

Twin and V1V2 loop	Envelope backbone	117-mo autologous plasma			PG9
		ID50		Fold change in ID50	IC50		Fold change in IC50
		Envelope backbone	Chimeric envelope		Envelope backbone	Chimeric envelope
AIIMS_329
329.15.A1	329.14.B1	1,124	151	7.42	<0.02	>10	>500
329.15.B1	329.14.B1	1,124	345	3.25	<0.02	6.91	>345
329.16.C1	329.14.B1	1,124	137	8.21	<0.02	7.42	>371
329.16.D5	329.14.B1	1,124	156	7.21	<0.02	>10	>500
329.14.B1	329.16.D5	<100	363	>3.63	>10	2.65	>3.77
AIIMS_330
330.15.C36	330.14.A20	1,427	108	13.21	1.11	8.92	8.03
330.15.D36	330.14.A20	1,427	112	12.81	1.11	7.91	7.12
330.15.E36	330.14.A20	1,427	<100	>14.27	1.11	9.96	8.97
330.16.A2	330.14.A20	1,427	<100	>14.27	1.11	3.64	3.28
330.16.D4	330.14.A20	1,427	<100	>14.27	1.11	7.69	6.92
330.16.E6	330.14.A20	1,427	699	2.04	1.11	0.86	0.77
330.16.H2	330.14.A20	1,427	<100	>14.27	1.11	3.99	3.59
330.14.A20	330.16.D4	151	635	4.21	>10	3.34	>2.99

^aGiven are the respective ID₅₀ (50% inhibitory concentrations [IC₅₀] for PG9) and fold change in ID₅₀ (50% inhibitory concentrations for PG9) for the V1V2 loop-swapped chimeric envelopes of AIIMS_329 and AIIMS_330.

DISCUSSION

Pediatric HIV-1 infection and its biphasic mode of progression are distinct from the infection and pattern of disease progression in adults. Significantly higher viral loads in the context of normal CD4⁺ T cell counts, rapid disease progression, and the generation of a potent and polyclonal antibody response earlier in infection in the pediatric population than in their adult counterparts are some of the differences between pediatric and adult HIV-1 infection. Recently, we and others observed the longitudinal evolution of the plasma cross-neutralization antibody response with multiple epitope specificities in select antiretroviral-naive chronically HIV-1 clade C (HIV-1C)-infected children (5, 6). The clade C primary isolates that we established from chronically infected children were relatively resistant to some of the tested second-generation adult bnAbs (18). We identified herein and characterized the coevolution of virus and the plasma antibody response in a pair of monozygotic pediatric twins from this cohort of chronically HIV-1-infected antiretroviral-naive children that were longitudinally studied. The presence of plasma bnAbs in both twins prompted us to map the neutralizing determinants and viral diversity. The isolation and characterization of single-genome-amplified functional pseudoviruses provided information toward understanding the viral evolutionary mechanisms that may have plausibly contributed to the distinct differences in the development and persistence of plasma bnAbs in the twins over time.

Since the first sampling at 78 months postinfection (p.i.), AIIMS_330 plasma neutralizing activity, tested using a multiclade panel of 50 HIV-1 envelope pseudoviruses (21), showed a consistent increase in potency over a period of 60 months and a median ID₅₀ neutralization titer and GMT which peaked at 112 months p.i. and then reached a plateau throughout the duration of the study. AIIMS_329, on the other hand, showed the highest GMT, median ID₅₀ neutralization titer, and potency at 90 months p.i., which then declined at 112 months p.i., beyond which the titers stabilized. A persistent antigenic stimulus due to relatively high viremia may have contributed to the sustained elite neutralizing activity seen in AIIMS_330, as a long exposure to a high viral load with declining CD4⁺ T cell counts has earlier been reported to be associated with the development of plasma breadth and potency over time (28).

The plasma bnAbs evolved in a time-dependent manner, with the initial development of V2 glycan-directed antibodies, which is in accordance with several studies showing V2 glycan bnAbs to be the most common and potent (29, 30). Initially, these antibodies showed dependence solely on N160 glycan in both twins. Neutralization dependence on K169 and K171 was observed at 124 months p.i., and this was confirmed by the reduced neutralization susceptibility of the K169E-K171E double mutant to the AIIMS_329 and AIIMS_330 plasma. The binding interactions of the V2 glycan-targeting antibodies have been suggested to be influenced by the alignment of K169 and K171 residues at the 3-fold axis of the trimer, since the lysine residues on adjacent protomers can contribute to the binding by V2 apex-targeting bnAbs (31). Likewise, it is plausible that the V2 glycan-dependent plasma neutralizing antibodies in both the twins may have shown more dependence on the interaction with these lysine residues over time. At the 112-month time point, AIIMS_330 plasma antibodies demonstrated CD4bs binding reactivity; however, neutralization dependence could be confirmed only by depletion analysis.

Many studies have shown that children mount a broader and more potent response against HIV-1 earlier than adults, with one report showing the presence of broad neutralizing activity in an infant at 1 year p.i. (4, 7). Polyclonal bnAb responses have been reported in both adult and pediatric infections (32–34), with chronic infection and higher antigenic stimulation often being identified as factors responsible for neutralization dependence on multiple specificities. Our observation of dual epitope specificity in the AIIMS_329 polyclonal plasma and antibodies against 3 distinct epitopes in AIIMS_330 plasma that developed by 112 months of age is in consonance with the findings of a recent study wherein children who had been infected with HIV-1 for a duration of 10 years, on average, were found to have multiple HIV-1 nAb specificities (5). Based on the multiple epitope specificities observed in AIIMS_330 plasma at 112 months p.i. and the highest GMT and median ID₅₀ values obtained from the plasma mapping analysis using the 50-virus panel, we infer that the spike in AIIMS_330 plasma broad cross-neutralization activity observed at the 117-month time point was probably due to the emergence of V3 glycan- and CD4bs-directed plasma bnAbs. Overall, the HIV-1-infected monozygotic twins, sharing a similar genetic makeup, showed the evolvement of bnAbs at different time points that consistently targeted common epitopes, such as the V2 and V3 regions of the virus envelope, suggesting that bnAb development in natural infection in these twins may have been determined by specific sequences in the shared virus that could guide the development of immunogens aimed at eliciting V2 and V3 bNAbs.

Both the virus and the antibody responses in the host are known to coevolve, and a continuous interplay between the HIV-1 envelope-specific antibody response and viral escape influences the development of bnAbs (12, 35–40). We assessed the circulating viral strains in both the twins around the time period (90 months to 117 months) when they developed bnAbs with multiple specificities, in order to delineate the viral properties that may be associated with the altered plasma neutralization activity. To negate the cloning bias often associated with the conventional bulk PCR amplification approach, functional pseudoviruses representing circulating HIV-1 strains were generated via single-genome amplification of envelope genes from HIV-1 RNA isolated from plasma samples from the twins, and their susceptibility to neutralization by contemporaneous plasma as well as bnAbs was evaluated. To further ensure that the functional pseudoviruses generated represented the currently circulating viral strains, we sequenced more than 20 SGA PCR amplicons from each time point studied (41). In AIIMS_330 plasma, we observed diverse circulating viral strains, including viruses that showed varied neutralization susceptibility to autologous plasma antibodies and bnAbs as well as resistant viruses, throughout the time frame of 90 months to 117 months p.i. In AIIMS_329, the viruses generated at 90 months p.i. were susceptible to the majority of the bnAbs as well as contemporaneous autologous plasma antibodies, whereas the viruses generated from 112 months and 117 months p.i. were resistant to the majority of the V2 glycan-targeting bnAbs and contemporaneous autologous plasma. A similar neutralization susceptibility to bnAbs was observed for AIIMS_330, with the clones generated at 90 months p.i. being sensitive, while the evolving viruses developed resistance to the majority of V2 glycan-targeting bnAbs. All the viruses in AIIMS_329 and AIIMS_330 remained sensitive to V3 glycan-targeting bnAbs, suggesting that the selection pressure on the V3 loop by the autologous plasma antibodies might have been less than that mediated by V2 glycan-targeting plasma antibodies. One major limitation of the study was the inability to generate functional pseudoviruses from all the SGA amplicons that represented the overall viral diversity in AIIMS_329 and AIIMS_330, but based on sequence similarity, we speculate that the bnAb susceptibility profile of the circulating variants may have been more or less similar to that of the functional pseudoviruses that we were able to generate. We observed two distinct evolutionary pathways of viral escape in response to increasing plasma potency and breadth in the twins. By generating chimeric pseudoviruses using the viruses sensitive and resistant to autologous plasma antibodies, we confirmed that the viruses of AIIMS_329 developed resistance to autologous plasma bnAbs by previously established mechanisms (26, 27, 42–44) of V1V2 loop lengthening, an increased number of PNGS, and a loss of critical epitopes like N160. The AIIMS_330 viruses that were resistant to contemporaneous autologous plasma also had longer V2 loops but maintained all epitopes. In AIIMS_329, all the viruses sampled at 112 months and 117 months p.i. were resistant, whereas AIIMS_330 viruses formed a pool, with some viruses being sensitive and others being resistant to the contemporaneous plasma antibodies. Taken together, these observations suggest that in both AIIMS_329 and AIIMS_330, the plasma bnAb activity was predominately targeted at residues in the V1V2 region of gp120, as a result of which the viruses in AIIMS_329 and AIIMS_330 plausibly developed resistance to most V2 glycan-targeting bnAbs and not V3 glycan-targeting bnAbs. Taken together, we observed that the plasma antibodies of AIIMS_330, the donor with coevolving plasma neutralizing activity and a mixture of sensitive and resistant viruses, showed high potency, a high GMT, and multiple antibody specificities in the plasma, suggesting that the coevolution of viruses and antibodies favored the development of plasma breadth and potency.

To summarize, in this study, we describe for the first time the diverse evolution of plasma bnAbs in a pair of genetically identical twins over a period of 60 months, even though they were infected at the same time point, had the same source of infection, and had similar CD4⁺ T cell counts, with an added advantage being assessments at matched time points between two infected individuals belonging to one transmission pair. The evolution of bnAbs that consistently targeted common epitopes in the V2 and V3 region of the viral envelope in this pair of monozygotic twins indicates that the specific neutralizing determinants in the shared virus can contribute information toward designing immunogens directed to elicit V2 and V3 bnAbs. In addition, we observed two distinct potential viral evolution mechanisms in genetically identical twins AIIMS_329 and AIIMS_330, of which the AIIMS_330 donor harbored a viral pool constituting autologous plasma-resistant and -sensitive viruses that may have contributed to the development and maintenance of elite neutralizing activity.

MATERIALS AND METHODS

Study population.

AIIMS_329 and AIIMS_330 were recruited from the Outdoor Patient Department of the Department of Pediatrics, All India Institute of Medical Sciences (AIIMS), for this study at the age of 9 years and were followed for a total period of 60 months. Blood was drawn in 5-ml EDTA vials, and plasma was aliquoted for plasma neutralization assays, viral RNA isolation, and viral load determinations. The study was approved by the Institute Ethics Committee of the All India Institute of Medical Sciences (IEC/NP-295/2011 and RP-15/2011).

HLA typing.

High-resolution HLA genotyping was performed by a reverse sequence-specific oligonucleotide probe (rSSO) method using LABType SSO HD kits (One Lambda, Inc., USA). Briefly, 40 ng DNA was utilized, specific amplification of each HLA class I (HLA-A and -B) and class II (HLA-DRB1) locus was performed, and individual amplicons for each locus were hybridized with a bead mixture coated with hundreds of SSO probes for specific HLA loci (covering exons 2 and 3 of class I and exon 2 of class II). Samples were acquired on a Luminex-based LABscan 100 platform. Allele assignment was performed based on the hybridization reaction pattern observed and analyzed through the use of HLA Fusion software (version 2.0). Additionally, confirmatory HLA typing was performed for all HLA loci using sequencing-based typing (SBT). Initially, 100 to 120 ng DNA was taken for long-range PCR using locus-specific amplification and primers for each locus, HLA-A, HLA-B, and HLA-DRB1 (SBTexcellerator reagents; GenDx). After an amplification check on agarose gels, ExoSAP cleanup of the amplicons was performed using exonuclease 1 and shrimp alkaline phosphatase, and unincorporated primers and nucleotides were removed. Further, exon-specific primers were utilized for the second amplification (forward and reverse reactions for exons 2, 3, and 4 for HLA-A and -B and exons 2 and 3 for the HLA-DRB1 locus). The sequencing product was further cleaned using Sephadex G50, and the product was finally sequenced using an ABI 3130XL genetic analyzer. Raw sequencing data were analyzed using SBT Engine software.

Plasmids, viruses, monoclonal antibodies, peptides, and cells.

Plasmids encoding HIV-1 env genes representing different clades, monoclonal antibodies, and TZM-bl cells were procured from the NIH AIDS Reagent Program. Plasmids carrying the RSC3 wild-type probe and its mutant, RSC3Δ371I/P363N, were kindly provided by John Mascola, National Institute of Allergy and Infectious Diseases, USA. The MPER-B and -C peptides were commercially synthesized from Sigma Genosys, USA (24, 25). HEK293T cells were purchased from the American Type Culture Collection (ATCC).

Cloning of autologous HIV-1 envelope genes and production of replication-incompetent pseudoviruses.

Autologous replication-incompetent envelope pseudoviruses were generated from AIIMS_329 and AIIMS_330 as described previously (22). Briefly, viral RNA was isolated from 140 μl of plasma using a QIAamp viral RNA minikit and reverse transcribed, using gene-specific primer OFM19 (5′-GCACTCAAGGCAAGCTTTATTGAGGCTTA-3′) and SuperScript III reverse transcriptase, into cDNA, which was used in a two-round nested PCR for amplification of the envelope gene using high-fidelity Phusion DNA polymerase (New England Biolabs). The envelope amplicons were purified and ligated into the pcDNA3.1D/V5-His-TOPO vector (Invitrogen). Pseudoviruses were prepared by cotransfecting 1.25 μg of an HIV-1 envelope-containing plasmid with 2.5 μg of an envelope-deficient HIV-1 backbone (PSG3Δenv) vector at a molar ratio of 1:2, using PEI-MAX as the transfection reagent, into HEK293T cells seeded in a 6-well culture plates. Culture supernatants containing pseudoviruses were harvested at 48 h posttransfection, filtered through 0.4-μm-pore-size filter, aliquoted, and stored at −80°C until further use. The 50% tissue culture infective dose was determined by infecting TZM-bl cells with serially diluted pseudoviruses in the presence of DEAE-dextran and lysing the cells at 48 h postinfection. Infectivity titers were determined by measuring luminescence activity in the presence of the Bright Glow reagent (Promega).

HIV-1 envelope sequences and phylogenetic analysis.

HIV-1 envelope genes were PCR amplified from plasma viral RNA by single-genome amplification (41) and directly sequenced commercially. Individual sequence fragments of SGA-amplified amplicons were assembled using Sequencher (version 5.4) software (Gene Code Corporation). Nucleotide sequences were aligned by use of the MEGA (version X) program (45).

Generation of chimeric envelope pseudoviruses.

Chimeric envelopes were generated by swapping the V1V2 regions between sensitive and resistant viruses by individually amplifying the V1V2 loop via conventional PCR and the remaining backbone without V1V2 by inverse PCR. The V1V2 loops were then ligated into the backbone without V1V2 with an infusion HD Cloning Plus kit per the manufacturer’s instructions. The primers utilized for amplifying the V1V2 loops from AIIMS_329 were 329.V1.In (5ʹ-TGGGATCAAAGCCTAAAGCCATGTG-3ʹ) and 329.V2.In (5ʹ-CATAACCAGCTGGAGCACAATAGTG-3ʹ), and the primers utilized for amplifying the V1V2 loops from AIIMS_330 were 330.V1.In (5ʹ-GTGTAAAGTTGACTCCCACTCTGTGTCAC-3ʹ) and 330.V2.In (5ʹ-CAGCTGGAGCACAATAGTGTATAGGAATTGG-3ʹ). The backbone (the envelope clone without V1V2 loops) for AIIMS_329 was amplified with primers 329.V1.Out (5ʹ-CACATGGCTTTAGGCTTTGATCCCA-3′) and 329.V2.Out (5′-CACTATTGTGCTCCAGCTGGTTATG-3ʹ), and for AIIMS_330, the primers were 330.V1.Out (5ʹ-GTGACACAGAGTGGGAGTCAACTTTACAC-3ʹ) and 330.V2.Out (5ʹ-CCAATTCCTATACACTATTGTGCTCCAGCTG-3ʹ).

Neutralization assay.

Neutralization assays were carried out using TZM-bl cells, a genetically engineered HeLa cell line that constitutively expresses CD4, CCR5, and CXCR4 and that contains the luciferase and β-galactosidase genes under the control of the HIV-1 tat promoter, as described before (46). Briefly, envelope pseudoviruses were incubated in the presence of serially diluted bnAbs or heat-inactivated plasma samples for 1 h. After incubation, freshly trypsinized TZM-bl cells were added with 25 μg/ml DEAE-dextran. The plates were incubated for 48 h, the cells were lysed in the presence of the Bright Glow reagent, and luminescence was measured. Using the luminescence of serially diluted bnAbs or plasma, a nonlinear regression curve was generated and titers were calculated as the bnAb concentration or reciprocal dilution of serum that showed a 50% reduction in luminescence compared to that for the untreated virus control. For V2 and V3 glycan dependence, pseudoviruses with key mutations in the V2 and V3 glycans were used with their wild-type counterparts and were incubated with plasma for 1 h at 37°C, followed by addition of TZM-bl cells and readout of the results at 48 h postinfection.

Binding ELISAs.

MPER, RSC3, and RSC3Δ371I/P363N ELISAs were performed as described previously (16). Briefly, 96-well ELISA plates (Corning, USA) were coated with 2 μg/ml of the RSC3 and RSC3Δ371I/P363N proteins and the MPER-B and MPER-C peptides overnight at 4°C. The coated plates were washed with phosphate-buffered saline containing 0.05% Tween 20. The plates were blocked with 5% skimmed milk in blocking buffer. A 50-fold dilution of plasma sample was added, titrated, and incubated at 37°C for 1 h. Unbound plasma antibodies were washed with wash buffer, and the plates were incubated with peroxidase-conjugated goat anti-human IgG at a dilution of 1:1,000. Following secondary antibody incubation, the wells were washed and tetramethylbenzidine substrate was added. After color development, the reaction was stopped with 0.2 M H₂SO₄ and the absorbance was measured at 450 nm.

Statistical analysis.

The Kruskal-Wallis test was used for the comparison of three parameters. All statistical analyses were performed on GraphPad Prism (version 6) software. A P value of <0.05 was considered significant.

Data availability.

The SGA-amplified HIV-1 envelope sequences are available at GenBank with accession numbers MK076582 to MK076724.