Potent and broad HIV-neutralizing antibodies in memory B cells and plasma

Abstract

Induction of broadly neutralizing antibodies (bnAbs) is a goal of HIV-1 vaccine development. Antibody 10E8, reactive with the distal portion of the membrane-proximal external region (MPER) of HIV-1 gp41, is broadly neutralizing. However, the ontogeny of distal MPER antibodies and the relationship of memory B cell to plasma bnAbs are poorly understood. HIV-1-specific memory B cell flow sorting and proteomic identification of anti-MPER plasma antibodies from an HIV-1-infected individual were used to isolate broadly neutralizing distal MPER bnAbs of the same B cell clonal lineage. Structural analysis demonstrated that antibodies from memory B cells and plasma recognized the envelope gp41 bnAb epitope in a distinct orientation compared with other distal MPER bnAbs. The unmutated common ancestor of this distal MPER bnAb was autoreactive, suggesting lineage immune tolerance control. Construction of chimeric antibodies of memory B cell and plasma antibodies yielded a bnAb that potently neutralized most HIV-1 strains.

Fig. 1. Isolation of MPER-directed bnAbs

(A) Fluorescently labeled MPR.03 peptide tetramers were used to decorate PBMCs from donor CH0210. A flow cytometric plot is shown. The square represents frequency of MPR.03 double-positive MPER-specific memory B cells that were sorted for Ig gene amplification and expression. Colored dots within the square show individual cells that yielded MPER-specific mAbs DH511.1 to DH511.6 and DH517, as revealed by index sorting. Memory B cells were gated as live CD16⁻CD14⁻CD3⁻CD235⁻CD19⁺IgD⁻/CD38^all. (B) Phylogenetic tree of V_HD_HJ_H sequences of the DH511 clonal lineage. Ancestral reconstruction of the evolutionary pathway from the inferred UCA to the mature mAbs including six maturational intermediates (I1 to I6, circles). DH511.13 could not be expressed and was not studied further. (C) Neutralization activity of probe-identified MPER antibodies against a panel of 208 cross-clade HIV-1 isolates. Median and geometric mean neutralization potency against viruses neutralized with an IC₅₀/IC₈₀ of <50 µg/ml. Right: The percentage of 208 viruses neutralized by mAbs DH511.2, 10E8, and VRC01 at IC₅₀ or IC₈₀ of <50 µg/ml, <1 µg/ml, and <0.1 µg/ml. (D) Neutralization potency and breadth of DH511.2 compared with 10E8 and VRC01 against a 208-isolate HIV-1 Env pseudovirus panel displayed as potency-breadth curves. The percentage of isolates neutralized at IC₅₀ (top) and IC₈₀ (bottom) values is plotted against mAb concentration. (E) Percent maximum neutralization of each isolate by DH511.2. (F) Identification of MPER-directed broadly neutralizing plasma antibodies by proteomics. Phylogenetic tree of heavy chain sequences identified in the plasma and added to DH511.1 to DH511.6 isolated from the memory B cell compartment [see (B)]. The bar on the right shows the relative abundance of the three identified clonotypes in serum (IV, 95%; II, 4%; III, 1%).

Fig. 2. Structural analysis of the DH511 lineage

(A) Crystal structures of DH511.1 and DH511.2 Fabs in complex with gp41 MPER peptides 656 to 683 and 662 to 683, respectively, oriented on the basis of Cα atom superposition of distal MPER residues. MPER residues 668 to 683 are colored red. (B) Crystal structures of antibodies 10E8 (PDB: 4G6F) and 4E10 (PDB: 2FX7) in complex with MPER, oriented as in (A). Close-up view of DH511.1 and DH511.2 (C) and 10E8 (D) contacts with MPER residues 668 to 683. MPER residues that interact with antibody V_H3–15 region residues, HCDR3 residues, or both are shown in cyan, red, or brown, respectively. (E) Directions of approach to distal gp41 MPER by antibodies DH511.1, DH511.2, 10E8, and 4E10. Shown are superimposed structures of antibody-bound gp41 MPER, with lines representing antibody variable region directions of approach colored as in (A) and (B). (F) Crystal structures of Fabs of plasma-derived variants DH511.11P and DH511.12P in complex with gp41 MPER peptide 662 to 683, with residues 668 to 683 colored red. Antibody residues shown in surface representation differ in sequence from DH511.1 or DH511.2, with those at the interface with gp41 shown in red. (G) Close-up view of DH511.11P and DH511.12P contacts with MPER residues 668 to 683, colored as in (C). (H) Crystal structure of chimeric antibody DH511.2_K3 in complex with MPER peptide 670 to 683, with a close-up view shown in (I) and colored as in (C). (J) Close-up view of the LCDR3 loop of DH511.2.K3 (magenta), rotated along the antibody longitudinal axis relative to (H) and superimposed with the complex structure of DH511.12P.

Fig. 3. Comparison of paratopes of DH511 lineage antibodies with 10E8 MPER bnAb

(A) Surface representations of DH511 lineage antibodies and 10E8, rotated by 60° toward the reader, relative to the orientations in Fig.2. Antibody residues at positions within the heavy chain V_H3–15 region that mediate contacts with gp41 in both the DH511 and 10E8 lineages are shown in cyan. Remaining gp41-contacting residues within the V_H3–15 region of the respective antibodies are shown in dark violet. HCDR3 residues that contact gp41 are shown in red. (B) Sequence alignment of DH511 lineage antibodies from memory B cells, plasma, and inferred intermediates, against antibody 10E8 and their shared V_H3–15 germline gene precursor. Residues that contact gp41 in the crystal structures are labeled with closed circles; somatically mutated residues are shaded. Conserved V_H3–15 germline residues that interact with gp41 are shown in red.

Fig. 4. Interactions of HIV-1 Env gp41 Asn and Thr at position 671 with distal MPER bnAbs 10E8 and DH511.2

(A) 10E8 heavy chain (lavender) is shown in complex with gp41 MPER peptide (gray), with Asn (green, original amino acid in the structure) and Thr (red, modeled) at 671 shown in ball and stick representations. The 10E8 heavy chain volume within 4 Å of Asn⁶⁷¹ is shown in yellow surface representation. Inset: Alignment of the modeled side chain for Thr with the Asn from the crystal structure, with various heavy atoms labeled. The crystal structure from PDB 4G6F (1) was used. (B) View of Asn (green opaque) and Thr⁶⁷¹ (red opaque) surfaces from within the 10E8 heavy chain (HC) surface (lavender transparent). A part of the Thr⁶⁷¹ surface penetrates the 10E8 heavy chain clash volume. A cutoff of 1.4 Å around each heavy atom was used to render the amino acid surfaces. (C) Same as in (A) with DH511.2 heavy chain shown in blue and with DH511.2 heavy chain volume within 4 Å of Asn⁶⁷¹ shown in orange. gp41 peptide is aligned such that Env positions 673 to 685 have the same arrangement as in (A). The relative position of Thr compared with Asn was very similar, as shown in the inset of (A). (D) Same as in (B), with DH511.2 heavy chain volume shown as transparent blue. No part of the Thr⁶⁷¹ amino acid surface penetrated the DH511.2 volume.

Fig. 5. Comparison of DH511.2 binding to MPER peptide liposome conjugates to that of distal MPER bnAbs 4E10 and 10E8

(A) Both DH511.2 and 10E8 mAb binding to MPER liposomes followed a two-step “encounter-docking” model [A + B→ABx (encounter step); ABx→AB (docking step)] and was similar to the mode of binding of the MPER bnAb 4E10. The experimental data for each concentration of the indicated mAbs are shown in black, and the fitted curves derived from global curve fitting to the two-step conformational model are shown in red. Decreasing concentration dose of each mAb is as follows: DH511.2 = 1000, 666.7, 333.3, 166.7, and 66.7 nM; 10E8 = 1333.3, 1000, 666.7, 333.3, 166.7, and 66.7 nM; 4E10 = 533.3, 266.7, 66.7, and 33.3 nM; 13H11 = 2000, 1000, 666.7, 533.3, 333.3, 266.7, 133.3, and 66.7 nM. The inset for each of the indicated MPER bnAbs shows the component curves for the encounter (Abx; green) and docked (AB; blue) complexes. The calculated rate constants for each binding step are shown in the table at the bottom. (B) Raw (control unsubtracted) binding of each mAb to bare (peptide-free) liposomes of the same lipid composition as the MPER liposomes. The control and non-neutralizing MPER 13H11 mAb show no binding to either MPER liposomes or bare (peptide-free) liposomes. MPER656 or MPER656.1 peptide was conjugated to synthetic liposomes for binding to 10E8 or DH511.2, respectively, and binding analysis was performed by biolayer interferometry analysis as described in the Materials and Methods. Data are representative of two independent experiments.

Fig. 6. DH511.2 recognizes a transiently exposed intermediate state of gp41, and the lifetime of DH511.2 epitope exposure is the same as that of 10E8 and 4E10

Time course of neutralization of the tier 2 HIV-1 isolate B.BG1168 was measured by addition of mAbs to TZM-bl cells preincubated with virus. The mean t_1/2 half-life of neutralization ± SD values were derived from four-parametric sigmoid curve fitting analysis using data from two or three independent experiments. Half-life values were similar among the three antibodies.