Viral escape from neutralizing antibodies in early subtype A HIV-1 infection drives an increase in autologous neutralization breadth

Abstract

Antibodies that neutralize (nAbs) genetically diverse HIV-1 strains have been recovered from a subset of HIV-1 infected subjects during chronic infection. Exact mechanisms that expand the otherwise narrow neutralization capacity observed during early infection are, however, currently undefined. Here we characterized the earliest nAb responses in a subtype A HIV-1 infected Rwandan seroconverter who later developed moderate cross-clade nAb breadth, using (i) envelope (Env) glycoproteins from the transmitted/founder virus and twenty longitudinal nAb escape variants, (ii) longitudinal autologous plasma, and (iii) autologous monoclonal antibodies (mAbs). Initially, nAbs targeted a single region of gp120, which flanked the V3 domain and involved the alpha2 helix. A single amino acid change at one of three positions in this region conferred early escape. One immunoglobulin heavy chain and two light chains recovered from autologous B cells comprised two mAbs, 19.3H-L1 and 19.3H-L3, which neutralized the founder Env along with one or three of the early escape variants carrying these mutations, respectively. Neither mAb neutralized later nAb escape or heterologous Envs. Crystal structures of the antigen-binding fragments (Fabs) revealed flat epitope contact surfaces, where minimal light chain mutation in 19.3H-L3 allowed for additional antigenic interactions. Resistance to mAb neutralization arose in later Envs through alteration of two glycans spatially adjacent to the initial escape signatures. The cross-neutralizing nAbs that ultimately developed failed to target any of the defined V3-proximal changes generated during the first year of infection in this subject. Our data demonstrate that this subject's first recognized nAb epitope elicited strain-specific mAbs, which incrementally acquired autologous breadth, and directed later B cell responses to target distinct portions of Env. This immune re-focusing could have triggered the evolution of cross-clade antibodies and suggests that exposure to a specific sequence of immune escape variants might promote broad humoral responses during HIV-1 infection.

Figure 1. Identification of R880F nAb escape variants.

Twenty-two single-genome amplified subtype A HIV-1 Envs were cloned out of R880F plasma collected at 0-months (A), 2-months (B), 5-months (C), 7-months (D), and 10-months (E) post-seroconversion, pseudotyped, and assayed against autologous plasma contemporaneous to their respective dates of isolation. Two 0-month Envs (0-A6/B24) representative of the transmitted/founder sequence are included in each panel. To demonstrate that humoral escape variants were neutralized during the course of infection, all 22 longitudinal Envs were assayed for neutralization with 16-month plasma (F); it was from PBMC collected at this time point that the two autologous R880F mAbs, 19.3H-L1 and 19.3H-L3, were derived. Percent viral infectivity, as adjusted against wells containing no test plasma, is depicted on the vertical axis; reciprocal plasma dilutions are plotted along the horizontal axis in a logarithmic fashion. Each curve represents a single Env-plasma combination, and error bars demonstrate the standard error of the mean of two independent experiments using duplicate wells (0-month Envs = circles, 2-month Envs = triangles, 5-month Envs = inverted triangles, 7-month Envs = squares, 10-month Envs = diamonds). Colored lines (2-A9/2-A13 in magenta, 2-B31 in red, 2-B12 in cyan, and 5-B52 in green) indicate Envs that succumbed to neutralization, in varying combinations, by the isolated R880F mAbs.

Figure 2. Amino acid alignment of longitudinal R880F Envs.

Longitudinal Env amino acid sequences from 20 R880F contemporaneous plasma escape variants were aligned and examined using Sequencher v5.0 and Geneious v5.0.3 software, with particular emphasis on the three mutational hot spots–C2, the alpha2 helix in C3, and V5–that developed during early infection, as compared to the transmitted/founder Env (0-A6/B24). Dashes represent conserved positions. Subject-specific amino acid numbering labels the demarcated regions at their points of origination. Five residues including 295, 335, 338, 341, and 456 (HXB2 residues 293, 334, 337, 340, and 460) were specifically interrogated to define their contributions to nAb evasion and have been highlighted in black in 0-A6/B24 for easy identification. Colored sequences (2-A9/2-A13 in magenta, 2-B31 in red, 2-B12 in cyan, and 5-B52 in green) indicate Envs that succumbed to neutralization, in varying combinations, by the isolated R880F mAbs. Note that additional differences from the transmitted/founder Env, which occurred over time in these sequences but are not diagrammed here, may be viewed in Figure S1.

Figure 3. Determination of the earliest R880F nAb escape signatures in Env.

Suspected 2-month nAb escape mutations from four different amino acid positions were introduced by site-directed mutagenesis into the transmitted/founder Env 0-B24 or escape Env 2-A3, which differed from the founder only at the site of mutation introduction. The transmitted/founder and wild-type 2-month Envs (solid lines) along with site-directed mutant Envs (dashed lines) were pseudotyped and assayed with 2-month plasma to determine if substitutions at C2 295 (A), alpha2 helix 338 (B), alpha2 helix 341 (C), or V5 456 (D) (HXB2 residues 293, 337, 340, or 460) could individually confer resistance. Percent viral infectivity, as adjusted against wells containing no test plasma, is depicted on the vertical axis; reciprocal plasma dilutions are plotted along the horizontal axis in a logarithmic fashion. Each curve represents a single Env-plasma combination, and error bars demonstrate the standard error of the mean of two independent experiments using duplicate wells (0-month Envs = circles, 2-month Envs and representative point mutants = triangles). Colored lines (2-A9/2-A13/0-B24 I295R in magenta, 2-B31/0-B24 I295T in red, and 2-B12/0-B24 D341N in cyan) indicate Envs that succumbed to neutralization, in varying combinations, by the isolated R880F mAbs.

Figure 4. Homology model of R880F gp120.

A 3-dimensional gp120 monomer (blue) based on the R880F transmitted/founder Env 0-A6/B24 sequence was homology modeled from existing gp120 structures (see Materials and Methods) and spatially oriented using MacPyMOL software to illuminate the region targeted by the earliest nAbs in this subject. Functional domains such as the alpha2 helix (green) and V3 (magenta) are delineated, and subject-specific amino acid numbering indicates positions that mutated at 2-months post-seroconversion to confer nAb escape. These residues (295, 338, and 341 in red) nest together in a putative epitope.

Figure 5. Amino acid alignment of R880F immunoglobulin heavy and light chain variable domains and neutralization by R880F mAbs 19.3H-L1 and 19.3H-L3.

Germline heavy and light chain gene segment utilization was determined by SoDA, a somatic diversification analysis program , and amino acid sequences were aligned and examined using Sequencher v5.0 and Geneious v5.0.3 software. Dashes represent conserved positions. Complementarity-determining regions (CDRs) are highlighted in gray. The two R880F mAbs share a common heavy chain, 19.3H-HC (A), which utilizes V3-30*02, D1-7*01, and J4*02 gene families, while the somatically related light chains 19.3H-L1 and 19.3H-L3 (B) employ V2-14*01 and J2*01 gene families and differ from each other at five positions in and just downstream of CDR1. Heavy chain 19.3H-HC, when paired with either 19.3H-L1 (C) or 19.3H-L3 (D), was evaluated for neutralization against pseudotyped R880F wild-type (solid lines) and site-directed mutant Envs (dashed lines). Percent viral infectivity, as adjusted against wells containing no mAb, is depicted on the vertical axis; mAb concentrations (in µg/ml) are plotted along the horizontal axis in a logarithmic fashion. Each curve represents a single Env-mAb combination, and error bars demonstrate the standard error of the mean of two independent experiments using duplicate wells (0-month Envs = circles, 2-month Envs and representative point mutants = triangles, 5-month Envs and a representative point mutant = inverted triangles, 7-month Envs = squares, 10-month Envs = diamonds). Colored lines (2-A9/2-A13/0-B24 I295R in magenta, 2-B31/0-B24 I295T in red, 2-B12/0-B24 D341N in cyan, and 5-B52/2-A3 K338G D341N in green) indicate Envs that succumbed to neutralization, in varying combinations, by the isolated R880F mAbs.

Figure 6. gp120 binding by and competition of R880F mAbs 19.3H-L1 and 19.3H-L3.

The baseline binding of four biotinylated mAbs, 19.3H-L1, 19.3H-L3, 6.4C, or PGT128, was evaluated by ELISA with three R880F gp120 proteins: (A) wild-type 0-A6/B24, (B) point mutant 0-A6/B24 I295R, and (C) point mutant 0-A6/B24 E338K. R880F mAbs 19.3H-L1 and 19.3H-L3 were then competed with themselves, each other, and the negative control antibody, 6.4C. For the competition ELISAs, plates were coated with wild-type R880F 0-A6/B24 gp120 protein, pre-incubated with serially-diluted 19.3H-L1, 19.3H-L3, or 6.4C, washed, and then incubated with 1 µg/ml biotinylated 19.3H-L1 (D) or 19.3H-L3 (E). From data in (A), 1 µg/ml was selected as a point of non-saturated binding. The horizontal dashed lines in (D) and (E) represent 100% binding of biotinylated 19.3H-L1 or 19.3H-L3, at 1 µg/ml in the absence of competitor, respectively. Optical density values at 450 nm are depicted on the vertical axis; mAb concentrations (in µg/ml) are plotted along the horizontal axis in a logarithmic fashion. Error bars demonstrate the standard error of the mean of two independent experiments.

Figure 7. Crystal structures of R880F mAbs 19.3H-L1 and 19.3H-L3.

(A) CDR loops. A top view looking down at the antigen-binding site of 19.3H-L1 represented by ribbons. The framework regions of the light chain and heavy chain are colored cyan and green, respectively, while each CDR loop is colored separately. The side chains of the three VL CDR1 residues different between 19.3H-L1 and 19.3H-L3 are displayed. (B) ODA analysis of the Fab 19.3H-L1. The size/redness of each sphere is proportional to the binding strength of the region indicated. Note that the antigen-binding site is centered at VL CDR1 and VH CDR3. (C) The electrostatic surface potentials of the antigen-binding site of 19.3H-L1. Red and blue coloration represents the negatively and positively charged regions, respectively, while a dashed line encircles the flat surface of the antigen-binding site. (D) The three VL CDR1 amino acid differences, S27T, G29T, and Y32F, between 19.3H-L1 (cyan) and 19.3H-L3 (yellow).

Figure 8. Escape from mAbs 19.3H-L1 and 19.3H-L3 by glycan addition and/or shifting.

To investigate how longitudinal viruses, namely 7- and 10-month Envs, could harbor the humoral vulnerability associated with mutation D341N whilst maintaining neutralization resistant phenotypes, two potential compensatory mutations were investigated. I295N, which inserts a PNGS near the N-terminus of V3, was introduced by site-directed mutagenesis into two mAb-sensitive Envs, 0-A6 and 2-B12 (light orange). S335N, which shifts a PNGS closer to the N-terminus of the alpha2 helix, was similarly created in 0-A6, 2-B12, and 5-B52 (purple). Wild-type (solid lines) and site-directed mutant Envs (dashed lines) were pseudotyped and assayed with mAbs 19.3H-L1 (A) and 19.3H-L3 (B). To determine if mAb resistance was glycan-dependent, an S335Q substitution that destroyed the N333 PNGS was also created in Env 2-B12 (dark orange). Percent viral infectivity, as adjusted against wells containing no mAb, is depicted on the vertical axis; mAb concentrations (in µg/ml) are plotted along the horizontal axis in a logarithmic fashion. Each curve represents a single Env-mAb combination, and error bars demonstrate the standard error of the mean of two independent experiments using duplicate wells. In (C), the V3/alpha2 helix portion of the labeled Envs has been conceptualized with glycan forks, each of which represents a PNGS in the corresponding amino acid sequences. Glycans of particular interest (N295, N333, and N335) are designated using R880F-specific numbering. In (D), the proposed escape glycans N295 and N335 (red) have been modeled onto the R880F 0-B24 Env gp120 monomer (green) to illustrate how such masking could obscure underlying epitopes and prevent recognition by mAbs 19.3H-L1 and 19.3H-L3.

Figure 9. Heterologous neutralization breadth in R880F.

The two R880F mAbs, 19.3H-L1 (A) and 19.3H-L3 (B), in conjunction with 16-month (C) and 3-year (D) autologous plasma were evaluated for cross-neutralizing capacity against virions pseudotyped with fourteen heterologous HIV-1 Envs from three clades (A/C recombinant and subtype A Envs = lavender, subtype B Envs = coral, subtype C Envs = teal). Two 0-month Envs (0-A6/B24) representative of the transmitted/founder sequence are included in each of these panels. Percent viral infectivity, as adjusted against wells containing no mAb or test plasma, is depicted on the vertical axis; mAb concentrations (in µg/ml) or reciprocal plasma dilutions are plotted along the horizontal axis in a logarithmic fashion. Each curve represents a single Env-mAb or Env-plasma combination, and error bars demonstrate the standard error of the mean of two independent experiments using duplicate wells. V3/alpha2 helix amino acid sequences were aligned and examined using Sequencher v5.0 and Geneious v5.0.3 software (E). Dashes represent conserved positions; dots represent gaps. Significant PNGS sequons (N295, N333, N335) are highlighted in black at their points of origin.

Figure 10. Contribution of specific glycans to heterologous neutralization breadth in R880F.

To gauge if nAb responses continued to exert pressure, specifically against escape glycans, on the N- and C-terminal V3 flanks of envelope during long-term HIV-1 infection, wild-type (solid lines) and site-directed mutant (dashed lines) heterologous Envs were pseudotyped and assayed for neutralization with 3-year R880F autologous plasma (A). Percent viral infectivity, as adjusted against wells containing no test plasma, is depicted on the vertical axis; reciprocal plasma dilutions are plotted along the horizontal axis in a logarithmic fashion. Each curve represents a single Env-plasma combination, and error bars demonstrate the standard error of the mean of two independent experiments using duplicate wells. In (B), the V3/alpha2 helix portion of the labeled Envs has been conceptualized with glycan forks, each of which represents a PNGS in the corresponding amino acid sequences. Glycans of particular interest (N295, N333, and N335) are designated using R880F-specific numbering.