Geospatial HIV-1 subtype C gp120 sequence diversity and its predicted impact on broadly neutralizing antibody sensitivity

Abstract

Evolving diversity in globally circulating HIV-1 subtypes presents formidable challenge in defining and developing neutralizing antibodies for prevention and treatment. HIV-1 subtype C is responsible for majority of global HIV-1 infections. Broadly neutralizing antibodies (bnAbs) capable of neutralizing distinct HIV-1 subtypes by targeting conserved vulnerable epitopes on viral envelope protein (Env) are being considered as promising antiviral agents for prevention and treatment. In the present study, we examined the diversity in genetic signatures and attributes that differentiate region-specific global HIV-1 subtype C gp120 sequences associated with virus neutralization outcomes to key bnAbs having distinct epitope specificities. A total of 1814 full length HIV-1 subtype C gp120 sequence from 37 countries were retrieved from Los Alamos National Laboratory HIV database (www.hiv.lanl.gov). The amino acid sequences were assessed for their phylogenetic association, variable loop lengths and prevalence of potential N-linked glycosylation sites (pNLGS). Responses of these sequences to bnAbs were predicted with a machine learning algorithm ‘bNAb-ReP’ and compared with those reported in the CATNAP database. Phylogenetically, sequences from Asian countries including India clustered together however differed significantly when compared with pan African subtype C sequences. Variable loop lengths within Indian and African clusters were distinct from each other, specifically V1, V2 and V4 loops. Furthermore, V1V2 and V2 alone sequences were also found to vary significantly in their charges. Pairwise analyses at each of the 25 pNLG sites indicated distinct country specific profiles. Highly significant differences (p<0.001***) were observed in prevalence of four pNLGS (N130, N295, N392 and N448) between South Africa and India, having most disease burden associated with subtype C. Our findings highlight that the distinctly evolving clusters within global intra-subtype C gp120 sequences are likely to influence the disparate regionspecific sensitivity of circulating HIV-1 subtype C to bnAbs.Author SummaryBroadly neutralizing antibodies (bnAbs) that act on viral envelope protein and prevent establishment of infection have emerged as an attractive choice for HIV-1 treatment and prophylaxis over and above existing antiretroviral drug therapy. While HIV-1 subtype C accounts for nearly half of the global infection, the degree of intra-clade C variability across different geographical boundaries with particular reference to env gene (which is the target of neutralizing antibodies) is poorly understood. Through a systematic approach and by applying standardized computational algorithms, we carried out comprehensive comparison of the genetic properties of the gp120 sequences that represent globally circulating HIV-1 subtype C towards accurately predicating their degree of susceptibility to most broad and potent neutralizing monoclonal antibodies having different target specificities across gp120. Our data not only revealed significant intra-clade C diversity but also highlighted neutralization diversities. These observations provide insights that warrant greater and continued surveillance of evolving HIV-1 globally circulating subtype C variants that will inform selection of bnAbs that will best dissect the region-specific genetic and neutralization diversities towards selecting appropriate bnAb combination for clinical use.

Neutralization diversity of HIV-1 Indian subtype C envelopes obtained from cross sectional and followed up individuals against broadly neutralizing monoclonal antibodies having distinct gp120 specificities

“…1 a; top left). This observation also corroborates with our earlier finding [ 39 ]. The neutralization scores (IC 50 values) against VRC01, CAP256-VRC26.25, PGDM1400 and PGT121 of most of the same retrieved gp160 sequences of all subtypes were next obtained from the CATNAP database for the purpose of comparing with that obtained in the current study.…”

“…Subtype C being the most predominant subtype circulating majorly in Africa and India, a greater understanding of the antigenic and neutralization diversity of HIV-1 subtype C env would facilitate understanding of optimal bnAb combination that could potentially overcome region-specific the intra-clade C diversity. Our recent work has demonstrated genotypic distinctness of HIV-1 subtype C envelope (gp120) sequences reported from Indian and African epidemics [ 39 ]. In that study, we observed these sequences to be distinct in terms of variable loop lengths (V1, V2 and V4), an abundance of potential N-linked glycosylation sites and entropy at bnAb contact sites (including PGDM1400 and CAP256-VRC26.25).…”

“…Potential N-linked glycosylation sites were predicted using N-Glycosite tool available at the LANL HIV database [ 45 ]. Statistics regarding the antibody resistance-associated signature residues were generated as described recently [ 39 ]. A phylogenetic tree was generated for all the HIV-1 viral clone sequences available in the CATNAP database along with those generated in the present study (Total N = 1020) with iqtree under ‘HIVb’ model with estimated Ƴ parameters and number of invariable sites [ 46 ].…”

Neutralization Diversity of HIV-1 Indian Subtype C Envelopes Obtained From Cross Sectional and Followed up Individuals Against Broadly Neutralizing Monoclonal Antibodies Having Distinct Gp120 Specificities

“…1A; top left). This observation also corroborates with our earlier nding [39]. The neutralization scores (IC 50 values) against VRC01, CAP256-VRC26.25, PGDM1400 and PGT121 of most of the same retrieved gp160 sequences of all subtypes were next obtained from the CATNAP database for the purpose of comparing with that obtained in the current study.…”

“…Potential N linked glycosylation sites were predicted using N-Glycosite tool available at the LANL HIV database [45]. Statistics regarding the antibody resistance associated signature residues were generated as described recently [39]. A phylogenetic tree was generated for all the HIV-1 viral clone sequences available in the CATNAP database along with those generated in the present study (Total N= 1020) with iqtree under 'HIVb' model with estimated parameters and number of invariable sites [46].…”