Distinct region-specific neutralization profiles of contemporary HIV-1 clade C against best-in-class broadly neutralizing antibodies

Abstract

While broadly neutralizing antibodies (bnAbs) have been clinically shown to prevent HIV-1 acquisition, their relative effectiveness against regionally relevant HIV-1 forms is not clear. In the present study, we examined the extent of neutralization susceptibility of contemporary HIV-1 Indian clade C at a population level along with a head-to-head comparison with that from South Africa against a panel of clinically relevant best-in-class bnAbs. Env-pseudotyped viruses encoding HIV-1 India clade C env were found to be best neutralized by the V3 glycan-directed bnAbs (10-1074 and BG18) and select CD4 binding site (CD4bs)-directed bnAbs (VRC07, N6, and 1-18); however, they demonstrated significant resistance to V1/V2 apex-directed bnAbs. Interestingly, the magnitude of the neutralization sensitivity differed between contemporary India and South Africa clade C. Neutralization resistance to key bnAbs was observed to be associated with differences in residues on Env that form bnAb contact sites, gp120 loop lengths, and potential N-linked glycans. Notably, the second generation CD4bs bnAbs (VRC07, N6, 1-18) showed neutralization of VRC01- and 3BNC117-resistant viruses but with two- to sevenfold reduced potency compared to the VRC01-sensitive counterparts, likely due to the enrichment of resistance-associated residues observed in loop D. Predictive analysis indicated that the combination of BG18, N6, and PGDM1400 can provide over 95% neutralization coverage of contemporary India clade C at 1 µg/mL (IC80), an observation distinct from that observed with Africa clade C. Our study clearly highlights that both the complementarity of bnAb classes and the regionally relevant HIV-1 forms are important in achieving clinical effectiveness.IMPORTANCEWhile the development of vaccines to prevent HIV infection remains a global priority, their potential effectiveness is limited by the extraordinarily diversified circulating forms of HIV-1. The prospect of best-in-class broadly neutralizing antibodies (bnAbs) as a potential prevention option has been demonstrated in several studies, including the phase 2b Antibody-Mediated Prevention trials; however, to be broadly applicable, bnAbs will need to overcome the substantial variability of HIV env circulating globally, beyond the regions where efficacy trials are conducted. The present study highlights that the region-specific contemporary HIV-1 clade C viruses not only vary in their degree of susceptibility to the best-in-class clinically relevant bnAbs, but also are evolving at a population level to become increasingly resistant to the best-in-class bnAbs. Overall, the outcome of this study highlights the need for periodic assessment of sequence and neutralization profiles of the circulating regionally relevant HIV-1 forms toward prioritizing the bnAb combination suitable for effective intervention.

Keywords: HIV-1; India; South Africa; bnAb; clade C; contemporary virus; envelope; genetic diversity; neutralizing antibodies; prevention.

Fig 1

Phylogenetic relatedness of the contemporary HIV-1 India clade C at the population level. (A) Surveillance sites built and samples collected across different geographical sites in India between 2020 and 2023. The map was sourced from the map reported by the Department of Science and Technology, Government of India, at https://surveyofindia.gov.in/pages/outline-maps-of-india. (B) Phylogenetic relatedness of HIV-1 clade C Env proteins representing circulating forms in different geographic regions. Phylogenetic trees were generated for 249 HIV-1 envelope amino acid sequences, which included 232 contemporary (obtained between 2020 and 2023) from India and 17 HIV-1 group M reference sequences. These sequences were aligned using MAFFT, and the alignment was manually curated in BioEdit v.7.2.5. The phylogenetic tree was constructed with IQ-TREE under the HIVb model with estimated Ƴ parameters and number of invariable sites. The robustness of the tree topology was further assessed by SH-aLRT as well as 1,000 ultrafast bootstrap replicates implemented in IQ-TREE.

Fig 2

Neutralization profiles of contemporary HIV-1 India clade C to best-in-class existing bnAbs. Pseudoviruses expressing 115 contemporary envs obtained from individuals representing nine geographically distant regions in India and comprising distinct risk groups were assessed for their degree of susceptibility to 14 bnAbs as indicated having distinct epitope specificities on viral Env. IC50 and IC80 refer to the IgG concentrations (μg/mL) at which pseudoviruses demonstrated 50% and 80% neutralizations, respectively. Pseudoviruses that were not neutralized up to 25 µg/mL of IgG were considered as resistant viruses. Neutralization assay was carried out at least three times in duplicate, and the average was used to plot the graph. Neutralization breadth of each bnAb expressed as percent neutralization by IgG up to 25 µg/mL is shown on top of each graph (upper and lower panel).

Fig 3

Comparison of neutralization sensitivity to key bnAbs between historic and contemporary India clade C. (A) Degree of neutralization susceptibility of historical (N = 124; obtained before 2014) and contemporary viruses (N = 115; obtained between 2020-2023) assessed by pseudovirus neutralization assay. IC50 value of 25 µg/mL was considered as the neutralization sensitivity threshold. Statistical analysis to assess significance (P-values) of differences in neutralization sensitivity to a given bnAb by pseudoviruses expressing both historical and contemporary envs was performed by Mann-Whitney U-test. Neutralization assay was repeated at least three times in duplicate, and the average was used to plot the graph. (B) gp120 variable loop characteristics of historical and contemporary env sequences were assessed using the “variable characteristics tool” hosted at the Los Alamos National Laboratory HIV database (LANL-HIVDB, https://www.hiv.lanl.gov/content/sequence/VAR_REG_CHAR/index.html). Potential N-linked glycosylation sites prediction was performed with the tool N-Glycosite at LANL-HIVDB (https://www.hiv.lanl.gov/content/sequence/GLYCOSITE/glycosite.html). Statistical significance was assessed by the Mann-Whitney U-test. P-values between 0.05–0.01, 0.01–0.001, and < 0.001and <0.0001 are depicted as “*,” “**,” “***,” and “****,” respectively.

Fig 4

Comparison of phylogenetic and head-to-head neutralization profiles between contemporary India and South African clade C. (A) Phylogenetic relatedness of env genes obtained from contemporary HIV-1 clade C of India (N = 232) and Africa (N = 73) origins as well as historical India (N = 132) and Africa (N = 138) origins and 17 HIV-1 group M reference sequences. South Africa clade C envs comprised those obtained from the FRESH cohort (N = 41) and AMP placebo arm (N = 32). (B) Comparison of the degree of neutralization susceptibility of pseudoviruses expressing contemporary HIV-1 clade C envs of Indian (N = 115) and South African (N = 40; AMP placebo arm) origins to 12 best-in-class bnAbs with distinct epitope specificities on viral Env. Env expressed as a pseudovirus that showed IC50 value >25 µg/mL against a particular bnAb was considered as resistant. Statistical analysis to assess significance (P-values) of differences in neutralization sensitivity to a given bnAb by pseudoviruses expressing envs of India and South African origins was assessed by Mann-Whitney U-test. Fisher’s exact test (color coded in blue) was used to identify differences in the overall proportion of sensitive and resistant pseudoviruses. (C) Comparison of the magnitude of neutralization sensitivity of India and South Africa clade C viruses to select clinically relevant bnAbs. The neutralization breadth of each bnAb tested against India and South Africa clade C envelopes is expressed in the y-axis as percent neutralization at a given concentration of corresponding antibody (IgG) concentration given in x-axis. The values in the x-axis are the geometric mean of the IC80 values (μg/mL) calculated for each bnAb. Neutralization assay was carried out in duplicate replicates at least three times, and average values were used to plot the graph.

Fig 5

Diversity in gp120 sequence features and contact sites polymorphism between contemporary India clade C Envs sensitive and resistant to V/1/V2 apex clinically relevant bnAbs. Frequency of contact sites associated with CAP256-VRC26.25 and PGDM1400 sensitivity was compared between CAP256-VRC26.25-sensitive and -resistant pseudoviruses (A) and PGDM1400-sensitive and -resistant viruses (B). The gp160 position (based on HXB2 numbering) of the key amino acids in the sequence logo is shown in the x-axis, and their relative abundance expressed as probability is in the y-axis. O has been used to differentiate potential N-linked glycosylated asparagine from potentially unglycosylated asparagine (N). Residues underscored in purple line are direct Ab contact sites. Residues showing statistically significant changes in abundance following a Fisher’s exact test are highlighted with yellow arrows. (C) Variable loop length, PNGs, and net charges of sensitive and resistant envelopes.

Fig 6

Comparison of env sequence features of contemporary India and South Africa clade C viruses. (A) The amino acid sequences of complete envs (gp120 and gp41) of India and South Africa contemporary HIV-1 clade C were analyzed to compare their average variable loop lengths, PNLGs, and net charges in gp120 as well as the length of gp41. These are analyzed using the “variable region characteristics” tool available at the Los Alamos HIV database (https://www.hiv.lanl.gov/content/sequence/VAR_REG_CHAR/index.html) and N-Glycosite (https://www.hiv.lanl.gov/content/sequence/GLYCOSITE/glycosite.html). (B) Comparison of key amino acid residues on India and South Africa clade C envs that are linked with CAP256-VRC26.25 and PGDM1400 resistance is shown in sequence logos. The statistically significant enrichment of key residues for viruses sensitive and resistant to CAP256-VRC26.25 and PGDM1400 is shown on the y-axis. O has been used to differentiate potential N-linked glycosylated asparagine from potentially unglycosylated asparagine (N). Amino acid residues underscored in purple line are direct Ab contact sites for respective bnAbs. Residues showing statistically significant changes in abundance following a Fisher’s exact test are highlighted with yellow arrows.

Fig 7

Neutralization of V1/V2 apex bnAb-resistant contemporary Indian clade C viruses by CD4bs- and V3 glycan supersite-directed bnAbs. (A) Sensitivity of pseudoviruses expressing contemporary Indian clade C envs which were fully resistant to all V1/V2-directed bnAbs to CD4bs (VRC01, VRC07, 3BNC117, N6, and 1-18) and V3 glycan supersite (PGT121, 10-1074, and BG18). Left panel shows percent neutralization of pseudoviruses that were resistant to all V1/V2-directed bnAbs tested (CAP256-VRC26.25, PGDM1400, PG9) (N = 26) by CD4bs- and V3 glycan-directed bnAbs. The right panel shows the same but only to pseudoviruses resistant to CAP256-VRC26.25-resistant envelopes (N = 62). Percent neutralization breadth conferred by CD4bs- and V3 glycan-directed bnAbs was calculated by the number of resistant viruses that showed IC80 values <25 µg/mL. (B) Magnitude of neutralization of V1/V2-directed bnAb-resistant pseudoviruses conferred by each of the CD4bs- and V3 glycan-directed bnAbs. The magnitude of virus neutralization equivalent to potency was measured as the lowest geometric mean titer conferred by each bnAb IgG (μg/mL) that demonstrated 80% neutralization of pseudovirus. Neutralization assay was carried out in duplicate replicates at least three times, and average values were used to plot the graph.

Fig 8

Neutralization efficiency of VRC01- and 3BNC17_resistant contemporary viruses by second-generation CD4bs-directed bnAbs. (A) Proportion of contemporary viruses (N = 115) that were found to be resistant to first- (VRC01 and 3BNC117) and second-generation (VRC07, N6, 1-18) CD4bs-directed bnAbs. Pseudoviruses with a neutralization score of IC80 >25 µg/mL to respective bnAbs were considered resistant. (B) Proportion of VRC01- and 3bnc117-resistant contemporary pseudoviruses that demonstrated sensitivity to second-generation CD4bs bnAbs (VRC07, N6, 1-18). Note that both VRC01- and 3BNC117-resistant viruses were least neutralized by 3BNC117 (11.11%) and VRC01 (17.85%) compared to VRC07, N6, and 1-18, indicating that the viruses resistant to both of them lack common key residues that are essential for both VRC01 and 3BNC117 for comprehensive neutralization. All the second-generation CD4bs bnAbs showed better neutralization (over 50%), with 1-18 demonstrating most (>74%) of VRC01- and 3BNC117-resistant viruses. (C) Comparison of the magnitude of neutralization of VRC01- and 3BNC117-sensitive and -resistant viruses by second-generation CD4bs bnAbs. Left panel shows the differences in the magnitudes of neutralization of VRC01-sensitive and -resistant viruses by all three CD4bs bnAbs (VRC07, N6, 1-18) and the right panel shows the same with 3BNC117-sensitive and -resistant viruses. The fold difference in magnitude of neutralization was obtained by calculating the average (GMT) of IC80 (μg/mL) for each paired set. GraphPad Prism was used to plot all the graphs.

Fig 9

Predictive neutralization coverage of contemporary India clade C viruses by clinically relevant bnAbs. Cumulative neutralization coverage of pseudoviruses carrying contemporary HIV-1 clade C envs by bnAb combination was assessed using the CombiNAber tool using the Bliss-Hill statistical model. (https://www.hiv.lanl.gov/content/sequence/COMBINABER/combinaber.html). CombiNAber analysis of 115 contemporary viruses from India against BG18 + N6 + PGDM1400 and 45 contemporary viruses from Africa against BG18 + 1-18 + PGDM1400, as well as the same combinations with at least two active bnAbs, have been plotted for target bnAb concentrations of 1 µg/mL (A) and 10 µg/mL (B), respectively. Predicted IC80 (μg/mL) combinations have been plotted on the x-axis, while the cumulative breadth of the viruses has been depicted on the y-axis. (C) Pseudoviruses expressing 24 contemporary difficult-to-neutralize envs were assessed for their degree of susceptibility to PGDM1400, N6, BG18, a combination of N6 and BG18, and a combination of PGDM1400, N6, and BG18. We used single bnAb at starting concentrations of 25 µg/mL with subsequent fivefold dilutions up to 0.00032 µg/mL, along with two combinations of 12.5 µg/mL each of BG18 and N6, and three combinations of 8.33 µg/mL each of BG18, N6, and PGDM1400 Abs. IC80 refers to the IgG concentrations (μg/mL) at which pseudoviruses demonstrated 80% neutralization, respectively. Pseudoviruses that were not neutralized up to 25 µg/mL of IgG were considered as resistant viruses.