Analysis of memory B cell responses and isolation of novel monoclonal antibodies with neutralizing breadth from HIV-1-infected individuals

Abstract

Background: The isolation of human monoclonal antibodies (mAbs) that neutralize a broad spectrum of primary HIV-1 isolates and the characterization of the human neutralizing antibody B cell response to HIV-1 infection are important goals that are central to the design of an effective antibody-based vaccine.

Methods and findings: We immortalized IgG(+) memory B cells from individuals infected with diverse clades of HIV-1 and selected on the basis of plasma neutralization profiles that were cross-clade and relatively potent. Culture supernatants were screened using various recombinant forms of the envelope glycoproteins (Env) in multiple parallel assays. We isolated 58 mAbs that were mapped to different Env surfaces, most of which showed neutralizing activity. One mAb in particular (HJ16) specific for a novel epitope proximal to the CD4 binding site on gp120 selectively neutralized a multi-clade panel of Tier-2 HIV-1 pseudoviruses, and demonstrated reactivity that was comparable in breadth, but distinct in neutralization specificity, to that of the other CD4 binding site-specific neutralizing mAb b12. A second mAb (HGN194) bound a conserved epitope in the V3 crown and neutralized all Tier-1 and a proportion of Tier-2 pseudoviruses tested, irrespective of clade. A third mAb (HK20) with broad neutralizing activity, particularly as a Fab fragment, recognized a highly conserved epitope in the HR-1 region of gp41, but showed striking assay-dependent selectivity in its activity.

Conclusions: This study reveals that by using appropriate screening methods, a large proportion of memory B cells can be isolated that produce mAbs with HIV-1 neutralizing activity. Three of these mAbs show unusual breadth of neutralization and therefore add to the current panel of HIV-1 neutralizing antibodies with potential for passive protection and template-based vaccine design.

Figure 1. Neutralization of 20 HIV-1 isolates in HOS.CD4-R5 cells by a panel of human mAbs.

(A) 46 mAbs (upper rows) were purified and tested in triplicates at a fixed concentration (100 µg/ml; *50 µg/ml and **25 µg/ml) for their capacity to neutralize 20 HIV-1 pseudoviruses (left columns) representing 6 different clades and both Tier-1 and Tier-2 isolates using HOS.CD4-R5 as target cells. Indicated is also the donor's HIV-1 clade. White, neutralization below 50%; yellow, 51–69%; orange, 70–89% and red, 90–100% neutralization. VSV-G pseudotyped HIV-1 was also tested as a negative control. (B) HK20, HGN194 and HJ16 were tested in parallel with b12, 2G12, 2F5 and 4E10 for their capacity to neutralize 20 HIV-1 pseudoviruses representing 6 different clades and both Tier-1 and Tier-2 isolates using HOS-CD4.R5 cells as in (A). Shown are IC50 values in µg/ml. -, indicates IC50 values >100 µg/ml.

Figure 2. Neutralization of 92 HIV-1 isolates in TZM-bl cells by HK20, HGN194, HJ16.

HK20, HGN194 and HJ16 were tested in parallel with b12, 2G12, 2F5, 4E10 and 447-52D for their capacity to neutralize 92 HIV-1 pseudoviruses representing 7 different clades and both Tier-1 and Tier-2 isolates using TZM-bl as target cells. Shown are IC50 values in µg/ml. -, indicates IC50 values >50 µg/ml, -* mAb tested starting from 25 µg/ml, nd, not determined. MuLV pseudotyped HIV-1 was also tested as a negative control.

Figure 3. HJ16 binds to a CD4bs epitope distinct from that recognized by b12.

(A) Binding of HJ16 and b12 to IIIB gp120 envelope protein. (B) Inhibition of IIIB gp120 binding to immobilized sCD4 by HJ16 and b12. (C–D) Inhibition of binding of HJ16 (C) or b12 (D) to immobilized gp120 by increasing concentrations of unlabeled HJ16 or b12. (E–F) Binding of b12 (E) or HJ16 (F) to YU2 wt gp120 and to the CD4i (I420R) or CD4bs (D368R) mutant YU2 proteins. Shown is mean ± SD of triplicates.

Figure 4. Epitope mapping of V3-specific mAbs by linear and circular peptide scanning.

(A) Eight gp120-specific mAbs were mapped with linear and cyclic peptides to the V3 region. Shown is the HIV-1 isolate used for the mapping, the minimal epitope, the binding breadth expressed as number of recombinant Env proteins out of the 16 tested, and the fraction of isolates neutralized in the HOS and TZMbl-based neutralization assays, respectively. (B) Alignment of the region corresponding to the epitope recognized by mAb HGN194 in 44 HIV-1 isolates. Co-receptor binding residues are bold. Highlighted in grey are the isolates neutralized by HGN194 either in the TZMbl-based or in the HOS-based assays. (C) Replacement analysis at positions R(307), S(308), V(309), R(310), I(311), G(312), Q(315), T(316 and F(317) of the epitope recognized by HGN194. Shown is the binding to 92 peptides carrying various amino acid substitutions at positions critical for maintenance of MAb binding. The bars represent ELISA values of HGN194 binding with the wt peptide (red) and the variant peptide (black). The amino acid replacements corresponding to V3 sequences from all isolates shown in (B) are highlighted with green bars. The binding of antibody to each peptide was tested in a PEPSCAN-based ELISA. Numbering according to HIV-1 HXB2.

Figure 5. HK20 binds to the HR-1 region within gp41.

(A) HK20 binding at 4 ug/ml to all overlapping linear peptides (15-mer peptides overlapping by 14 residues) spanning the gp41 sequence of the HXB2 isolate. Numbers at X-axis denote the first amino-terminal residue of the 15-mer gp41 peptide (numbering according to HIV-1 HXB2 sequence). Y- axis similar to Figure 4C (B) Immunoprecipitation of HR-1, 5HB and HR-1-FP constructs in the presence (+) or absence (−) of HK20 mAb. Proteins were separated in 10% polyacrylamide gels under reducing conditions and stained with Coomassie blue. C, HK20 mAb alone; MW, molecular weight. (C) HK20 binding to gp41 constructs by ELISA. (D–E) Neutralization of 96ZM651.2 (D) and CH064.20 (E) HIV-1 primary isolates by HK20 IgG and Fab fragments in a HOS-based assay.

Figure 6. Epitope specificity and neutralizing activity of the mAb panel.

The pie chart shows the specificity of the 58 mAbs isolated from the 21 interrogated individuals. The fraction of antibodies with neutralizing activity against at least one isolate is indicated in parentheses. Partial CD4bs, incomplete inhibition of CD4 binding; HR-1/FP, epitope located in between HR-1 and the fusion peptide as determined by 5F3 competition; HR-1/5HB, recognition of trimeric HR-1 and 5-helix bundle; 5HB, 5 helix bundle; C-C region, gp41 immunodominant region; HR-2, heptad-repeat 2; ND, not defined epitopes within gp120 or gp41.