Anti-HIV-1 Nanobody-IgG1 Constructs With Improved Neutralization Potency and the Ability to Mediate Fc Effector Functions

Abstract

The most effective treatment for HIV-1, antiretroviral therapy, suppresses viral replication and averts the disease from progression. Nonetheless, there is a need for alternative treatments as it requires daily administration with the possibility of side effects and occurrence of drug resistance. Broadly neutralizing antibodies or nanobodies targeting the HIV-1 envelope glycoprotein are explored as alternative treatment, since they mediate viral suppression and contribute to the elimination of virus-infected cells. Besides neutralization potency and breadth, Fc-mediated effector functions of bNAbs also contribute to the in vivo efficacy. In this study multivalent J3, 2E7 and 1F10 anti-HIV-1 broadly neutralizing nanobodies were generated to improve neutralization potency and IgG1 Fc fusion was utilized to gain Fc-mediated effector functions. Bivalent and trivalent nanobodies, coupled using long glycine-serine linkers, showed increased binding to the HIV-1 Env and enhanced neutralization potency compared to the monovalent variant. Fusion of an IgG1 Fc domain to J3 improved neutralization potency compared to the J3-bihead and restored Fc-mediated effector functions such as antibody-dependent cellular phagocytosis and trogocytosis, and natural killer cell activation. Due to their neutralization breadth and potency and their ability to induce effector functions these nanobody-IgG1 constructs may prove to be valuable towards alternative HIV-1 therapies.

Keywords: Fc fusion; Fc-mediated effector functions; HIV-1; nanobodies; neutralization.

Figure 1

Visualization of nanobodies and their target epitope. (A) Schematic representation of a conventional immunoglobulin (IgG) antibody and a camelid heavy-chain only antibody with visualization of their variable domains. (B) Monovalent, bivalent and trivalent J3 targeting the CD4bs, monovalent and bivalent 1F10 targeting the V3 loop and monovalent and bivalent 2E7 targeting the gp41 heptad repeat-1 (HR1). (C) Schematic diagram of the HIV-1 trimeric envelope glycoprotein complex showing the epitopes that are recognized by the broadly neutralizing nanobodies.

Figure 2

Multivalent anti-HIV-1 nanobodies show increased binding towards HIV-1 Env trimers from global panel. (A) SDS-PAGE protein separation followed by coomassie blue staining of the purified anti-HIV-1 nanobody-constructs under reducing conditions. Molecular weight (MW) is indicated in kilodalton (kDa). (B) Average area under the curve (AUC) of binding curves of J3, 2E7 and 1F10 nanobody variants to SOSIP.v9.0 trimers from the global panel (CNE55, Ce1176_A3, 25710-2.43, CH119.10, BJOX002000.03.2, Ce7030 and 246-F3_C10) as determined by ELISA using OD₄₅₀. Average AUC of OD450 was determined by doing a definite integral between the start and end point using GraphPad. Friedman matched comparison (three groups) and Wilcoxon matched-pairs signed rank tests (two groups) was used to compare the groups with significance indicated as *p <0.05. Data shown are the average of three independent experiments. (C) Binding of nanobodies and nanobody-IgG1 constructs to Ce1176 as determined by ELISA. *The binding of the nanobody and nanobody-IgG1 constructs to the concerned epitopes was determined using an anti-VHH-HRP secondary antibody, starting concentration 500 nM.** The binding of the nanobody-IgG1 constructs to the concerned epitopes was determined using a Goat-anti-Human-HRP secondary antibody, starting concentration 50 nM. Kd values for BG505 SOSIP.664 as determined by BLI, based on the average association of different antibody concentrations. ND, not determined; NB, non-binding. Data are representative of at least two independent experiments.

Figure 3

Multivalent anti-HIV-1 nanobodies show enhanced neutralization capacity depending on target. (A) Neutralization IC₅₀ (µM) of J3 variants against 12 viruses from the global panel. (B) Neutralization capacity of bivalent (black dots indicate replicates) and trivalent J3 (red dots indicate replicates) as compared to monovalent J3 is depicted as fold change in IC_50. (C) Neutralization IC₅₀ (µM) of 1F10 variants against 12 viruses from the global panel. (D) Neutralization capacity of bivalent 1F10 as compared to monovalent 1F10 is depicted as fold change in IC₅₀. (E) Neutralization IC₅₀ (µM) of 2E7 variants against 12 viruses from the global panel. (F) Neutralization capacity of bivalent 2E7 as compared to monovalent 2E7 is depicted as fold change in IC_50. Blue dots indicate neutralization by bivalent nanobody but not by monovalent nanobody. Absence of neutralization (IC₅₀ > 1 or 1.5 µM) is indicated with a grey shade. Friedman matched comparison (three groups) and Wilcoxon matched-pairs signed rank tests (two group) was used to compare IC₅₀. Significance is indicated as *p < 0.05, **p < 0.005, ***p < 0.0005 and “ns” indicates not significant (p > 0.05). Data shown are the average of two or three independent experiments.

Figure 4

Creation of nanobody-IgG1 constructs and their ability to bind to their target epitope. (A) Representation of the different nanobody-IgG1 constructs. 2E7, 1F10 and J3 fused to an human IgG1 heavy chain, lacking the CH1 domain and the bispecific J3-1F10 and J3-2E7 antibodies containing the S354C, T366W, K409A, Y349C, T366S, L368A, Y407V, F405K knob in hole and electrostatic mutations. (B) Visualization of the anti-HIV-1 nanobody-IgG1 fusions and the conventional IgG1 VRC01 under reducing conditions. Molecular weight (MW) is indicated in kDa.

Figure 5

Nanobody-IgG1 constructs are able to induce Fc-mediated effector functions. Neutralization capacity of IgG1 fusion constructs compared to monovalent or bivalent versions of (A) J3 (B) 1F10 and (C) 2E7 and (D) J3-2E7 and (E) J3-1F10. (F) Neutralization capacity of J3-IgG1 compared to bispecific J3-2E7-IgG1 and J3-1F10-IgG1. Friedman matched comparison was used to compare neutralization IC₅₀ (µM) between three groups. Wilcoxon matched-pairs signed rank tests was used to compare neutralization IC₅₀ (µM) between two groups. Significance is indicated as *p <0.05, **p <0.005, ***p <0.0005 and not significant (ns). Data shown are the average of three independent experiments.

Figure 6

Neutralization is impacted by nanobody-IgG1 fusion. (A) The K_D of all nanobody-IgG1s to hFcRn, recorded at pH 6.0 using surface plasmon resonance. K_D was calculated as the average K_D from multiple antibody concentrations as seen in Supplementary Figure 5 . Binding of nanobody-IgG1s to (B) FcγRIIa and (C) FcγRIIIa as determined by FcgγR dimer ELISA using Ce1176 SOSIP.v9.0 as coating antigen, depicted as AUC values of the binding curves. (D) Binding of nanobody-IgG1s to CNE55 SOSIP.v9.0 as determined by ELISA and induction of ADCP using CNE55 SOSIP.v9.0 coated fluorescent beads, both depicted as AUC. (E) Binding of nanobody-IgG1s to BG505 gp160 expressing HEK-293T cells and induction of ADCT are quantified by a secondary PE-labeled anti-IgG antibody and uptake of membrane fragments by THP-1 cells using flow cytometry, both depicted as AUC. (F) The binding strength for all nanobody-IgG1s to Ce1176 SOSIP.v9.0 as determined by ELISA and subsequent activation of NK-cells plotted as the percentage of CD107+ NK-cells. 2G12-IgG1, specific for HIV-1 gp120, was used as a positive control and COVA2-15, specific for SARS-CoV-2, was used as a negative control. The dotted lines represent the background level that is observed in these assays in the absence of antibody. Effector function data is normalized to the positive control, 2G12. Data shown are the average of at least two independent experiments.