Quantification, epitope mapping and genotype cross-reactivity of hepatitis B preS-specific antibodies in subjects vaccinated with different dosage regimens of BM32

Abstract

Background: Chronic hepatitis B virus (HBV) infections are a global health problem. There is a need for therapeutic strategies blocking continuous infection of liver cells. The grass pollen allergy vaccine BM32 containing the preS domain of the large HBV surface protein (LHBs) as immunogenic carrier induced IgG antibodies in human subjects inhibiting HBV infection in vitro. Aim of this study was the quantification, epitope mapping and investigation of HBV genotype cross-reactivity of preS-specific antibodies in subjects treated with different dosage regimens of BM32 METHODS: Hundred twenty eight grass pollen allergic patients received in a double-blind, placebo-controlled trial five monthly injections of placebo (aluminum hydroxide, n= 34) or different courses of BM32 (2 placebo + 3 BM32, n= 33; 1 placebo + 4 BM32, n= 30; 5 BM32, n= 31). Recombinant Escherichia coli-expressed preS was purified. Overlapping peptides spanning preS and the receptor-binding sites from consensus sequences of genotypes A-H were synthesized and purified. Isotype (IgM, IgG, IgA, IgE) and IgG subclass (IgG₁-IgG₄) responses to preS and peptides were determined by ELISA at baseline, one and four months after the last injection. IgG₁ and IgG₄ subclass concentrations specific for preS and the receptor-binding site were measured by quantitative ELISA.

Findings: Five monthly injections induced the highest levels of preS-specific IgG consisting mainly of IgG₁ and IgG₄, with a sum of median preS-specific IgG₁ and IgG₄ concentrations of >135 μg/ml reaching up to 1.8 mg/ml. More than 20% of preS-specific IgG was directed against the receptor-binding site. BM32-induced IgG cross-reacted with the receptor-binding domains from all eight HBV genotypes A-H.

Interpretation: BM32 induces high levels of IgG₁ and IgG₄ antibodies against the receptor binding sites of all eight HBV genotypes and hence might be suitable for therapeutic HBV vaccination.

Funding: This study was supported by the PhD program IAI (KPW01212FW), by Viravaxx AG and by the Danube-ARC funded by the Government of Lower Austria. Rudolf Valenta is a recipient of a Megagrant of the Government of the Russian Federation, grant No 14.W03.31.0024.

Keywords: Antibody response; Epitopes; Genotype cross-reactivity; Grass pollen allergy vaccine BM32; Hepatitis B; Vaccination; preS.

Fig. 1

Overview of the CS-BM32-004 study. Study subjects (n= 128) were divided into a group receiving five injections of placebo (n= 34), two placebo plus three injections of BM32 (n= 33), one placebo and four BM32 injections (n= 30) or five BM32 injections (n= 31) before the grass pollen season. Serum samples from visits 3, 8 and 11 were analyzed.

Fig 2

Schematic representation of the BM32 vaccine components. (a) Each of the four BM32 vaccine components (BM321, BM322, BM325, BM326) are recombinant fusion proteins consisting of allergen-derived peptides (pink) fused to the N- and C-terminus of preS, the N-terminal domain of LHBs protein, consisting of preS1 and preS2. In the LHBs the division between preS1 and preS2 does not exist, whereas preS2 domain forms the N-terminal domain of the non-essential middle MHBs protein. Certain HBV genotypes contain a deletion of 11 codons at the N-terminus of preS1. (b) Synthetic peptides (P1–P8) spanning preS1 and preS2 are indicated. The NTCP binding site is indicated in orange and the accessory domain involved in the inhibition of infection is marked in blue. (c) Sequences of three peptides A (green), B (violet) and C (red) used for mapping of antibody responses to the N-terminus are indicated in boxes.

Fig. 3

Identification of consensus sequences representing the preS1 region of HBV genotypes A–H containing the NTCP binding site and the accessory domain involved in the inhibition of infection. (a) Algorithm of the definition of the consensus sequences (AliView software). (b) Alignment of preS1-derived peptide sequences (aa 13–51) of genotypes A-H. Hydrophobic amino acids (A, I, L, M, F, W, V) are shown in blue, positively charged (K, R) in red, negatively charged (E, D) in magenta, polar (N, Q, S, T) in green, aromatic (H, Y) in cyan, cysteines (C) in pink, glycines (G) in orange, prolines (P) in yellow. (c) Percentages of identity to the consensus sequence of region aa 13–51 within each genotype are shown in the pie charts.

Fig. 4

Comparison of preS-specific isotype and IgG subclasses responses in the four groups of patients. Shown are optical density (OD) values (y-axes) corresponding to preS-specific IgG, IgA, IgM, IgE and IgG subclass (IgG₁–IgG₄) levels in patients immunized with 3, 4 or 5 injections of BM32 or with placebo (see inlay) at baseline (V3), after one (V8) or four (V11) months treatment (x-axes). Medians (horizontal bars) and significant differences are indicated: *p< 0.05, **p< 0.01, ***p< 0.001, ****p< 0.0001.

Fig. 5

Quantification of IgG₁ and IgG₄ subclass responses in patients treated with placebo or different numbers of BM32 injections. Shown are absolute increases of serum concentrations of IgG₁and IgG₄ (y-axes: μg/ml) specific (a) for preS, (b) for peptide A, (c) for peptide B and (d) for peptide C from baseline (V3) to visits 8 and 11 (x-axes) inpatients immunized with placebo, 3, 4 or 5 injections of BM32. Medians (horizontal bars) and significant differences are indicated: *p< 0.05, **p < 0.01, ***p< 0.001, ****p< 0.0001.

Fig. 6

Kinetics and fine specificities of anti-preS antibody responses in the individual subjects of the placebo group, and the groups having received 3, 4 or 5 injections. The heat maps depict the log₁₀ of the serum concentrations of preS- or peptide-specific (a) IgG₁ and (b) IgG₄. Missing values are indicated in gray.

Fig. 7

Vaccination with BM32 induces IgG antibodies recognizing HBV genotypes A–H. Shown are optical density (OD) values (y-axes) corresponding to IgG antibody levels specific for genotype A, B, C, D, E, F, G and H peptides in patients immunized with 5 injections of BM32 at baseline (V3) as well as one (V8) and four (V11) months after treatment (x-axes). Medians are indicated by red horizontal bars.