Generation of anti-GPC3/CD3 bispecific antibody

The anti-GPC3/CD3 bispecific antibody is an IgG-based heterodimeric antibody that simultaneously bind GPC3 expressed on tumor cells and CD3 epsilon chain on T cells (Figure 1a). The Fc region of bispecific provide extended half-life and markedly reduce Fc gamma receptors and complement component binding through amino acid substitution (P329 G/L234A/L235A). To facilitate heterodimerization of two chains, the knob (T366 W) and hole (T366 S/L368A/Y407 V) mutations were incorporated in the respective CH3 domain of each Fc. The bispecific antibody was expressed in HEK293 F cells and purified by Protein A affinity chromatography and cation exchange chromatography. Polymerase chain reaction (PCR) and sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) analysis showed a molecular weight of ~110 kDa of the bispecific antibody (Figure 1b,c) and the single peak of size exclusion HPLC (SE-HPLC) profile demonstrated the purified antibody is monomeric form (Figure 1d). Overall thermostability of the bispecific antibody was performed by differential scanning calorimetric with the onset of unfolding temperature (T_{m, onset}) of 54°C (Figure 1f), which had a similar T_{m, onset} with general IgG-like bispecific antibodies.¹⁶

A novel targeted GPC3/CD3 bispecific antibody for the treatment hepatocellular carcinoma

All authors

Lin Yu, Xi Yang, Nan Huang, Qiao-Li Lang, Qi-Lin He, Wang Jian-Hua & Ge Liang-Peng

https://doi.org/10.1080/15384047.2020.1743158

Published online:

02 April 2020

Figure 1. Characteristics of anti-GPC3/CD3 bispecific antibody. (a) Illustration structure of the bispecific antibody contains monovalent binding domains of GPC3 and CD3 antigens. The theoretical length of each chain was confirmed by PCR (b) and SDS-PAGE (c). (d) The aggregation was analyzed by size exclusion chromatography. (f) The onset of unfolding temperatures (T_{m, onset}) was measured by differential scanning calorimetry.

Display full size

Bispecific binding of anti-GPC3/CD3 antibody to antigens on cells

Three HCC cell lines had different expression level of GPC3 protein (Figure 2a), which were in correspondence with previous research.⁵ The bispecific binding ability of anti-GPC3/CD3 antibody was confirmed by flow cytometry using cells with overexpressing GPC3 or CD3 antigen. Results shown in Figure 2b demonstrated strong binding activity to both HepG2, Huh-7 cells and peripheral blood mononuclear cells (PBMCs), while no binding was observed on non-expressing GPC3 cells (SK-Hep-1). Subsequently, concentration-dependent binding to HepG2 and human PBMCs showed the avidity of the bispecific antibody of 4.8 and 5.5 nmol/L, respectively (Figure 2c)

A novel targeted GPC3/CD3 bispecific antibody for the treatment hepatocellular carcinoma

All authors

Lin Yu, Xi Yang, Nan Huang, Qiao-Li Lang, Qi-Lin He, Wang Jian-Hua & Ge Liang-Peng

https://doi.org/10.1080/15384047.2020.1743158

Published online:

02 April 2020

Figure 2. Binding ability of the anti-GPC3/CD3 bispecific antibody to human GPC3 and CD3 antigens. (a) Western blot result of three HCC cell lines had different expression level of GPC3 protein. (b) Flow cytometry analysis of antibody binding to different HCC cells (HepG2, Huh-7, SK-hep-1) and human PBMCs. The black lines indicate staining with an isotype control antibody, and the red lines indicate staining with anti-GPC3/CD3 bispecific antibody. (c) Median fluorescence intensity (MFI) of binding of bispecific antibody on HepG2 cells and PBMCs was measured by flow cytometry.

Display full size

Anti-GPC3/CD3 bispecific antibody mediated cytotoxicity and T-cell activation in vitro

The anti-GPC3/CD3 bispecific antibody mediated cytotoxicity of HCC cell lines HepG2, Huh-7 and SK-Hep-1 was evaluated in the presence of bispecific antibodies with freshly isolated human T cells. Concentration-dependent killing of bispecific antibodies was observed on GPC3-positive cells (HepG2, Huh-7) but no lysis to GPC3 negative SK-Hep-1 cells (Figure 3a). Previous study has proved the different surface expression level of human GPC3 on HepG2 and Huh-7 cells,⁵ which may led to the different potency of redirected T cells to lyse target cells by the bispecific antibodies.

A novel targeted GPC3/CD3 bispecific antibody for the treatment hepatocellular carcinoma

All authors

Lin Yu, Xi Yang, Nan Huang, Qiao-Li Lang, Qi-Lin He, Wang Jian-Hua & Ge Liang-Peng

https://doi.org/10.1080/15384047.2020.1743158

Published online:

02 April 2020

Figure 3. The anti-GPC3/CD3 bispecific antibody mediates GPC3-dependent tumor cells killing via activation of T cells. (a) Tumor cells and human T cells were incubated with increasing concentrations of bispecific antibodies at different ratios of effector to target cells (E:T) for 48 h. (b) Activation makers (CD25 and CD69) on CD4-positive and CD8-positive T cells were detected at 24, 48, and 72 h after incubation of HepG2 cells with bispecific antibodies and human T cells (E:T = 5:1). (c) Human T cells were labeled with CFSE dye and subsequently incubated in the presence of HepG2 cells (E:T = 5:1) and bispecific antibody for 4 days.

Display full size

CD69 and CD25 molecules relate to different activation pathways in various T-cell subsets.¹⁷The bispecific antibody induced CD8+ and CD4 + T cells activation was proved by determining the upregulation of CD25 and CD69 expression through flow cytometry (Figure 3b). The levels of CD69 and CD25 were reached rather high at 48 to 72 h on T cells. The proliferation of T cells subsequently was confirmed by incubating CFSE-labeled T cells and target cells with bispecific antibodies for 4 days to track cell division (Figure 3c).

Inhibition of tumor growth of murine xenografts by anti-GPC3/CD3 bispecific antibody

The ability of anti-GPC3/CD3 bispecific antibody to prevent tumor growth was investigated using a human HCC xenografts model with Huh-7 cancer cells subcutaneous injection in NOD/SCID mice. When the tumor volume reached 200 mm³, human PBMCs were injected intraperitoneally and treatment with the anti-GPC3/CD3 bispecific antibodies or vehicle the next day. Mean tumor volumes of 2 and 0.4 mg/kg antibody-treated group on day 45 were significantly lower than that of vehicle with T cells control group or 2 mg/kg (on T cells) group (P < 0.01), which demonstrated potent inhibition of GPC3-expressing tumor growth by the anti-GPC3/CD3 bispecific antibody (Figure 4a). In addition, the body weight of mice increased gently and no obvious difference among these groups (Figure 4a). To verify the antitumor effect of bispecific antibody, we removed the murine xenografts of vehicle and 2 mg/kg antibody treated groups on 3rd day after antibody administration to conduct immunohistochemical experiments. After incubation of second antibody (goat anti-human IgG -HRP antibody) and eikonogen, vehicle treated group showed negative staining while the antibody treated group was positive staining under microscopic examination (Figure 4b), which confirmed the bispecific antibody infiltrated the tumor tissue to exert its antitumor effect.

A novel targeted GPC3/CD3 bispecific antibody for the treatment hepatocellular carcinoma

All authors

Lin Yu, Xi Yang, Nan Huang, Qiao-Li Lang, Qi-Lin He, Wang Jian-Hua & Ge Liang-Peng

https://doi.org/10.1080/15384047.2020.1743158

Published online:

02 April 2020

Figure 4. Inhibition of cancer xenografts growth by the anti-GPC3/CD3 bispecific antibody. (a) The NOD/SCID mice were implanted Huh-7 cancer cells subcutaneously until tumor volumes reached approximately 200mm³. After received human PBMCs by intraperitoneal injection, the mice were treated with the indicated doses of antibodies or vehicle via intravenous administration the next day. Tumor volumes and body weight are plotted as means with SD (n = 6). (b) immunohistochemistry analysis of Huh-7 xenograft tissues 3^rd day after bispecific antibody administration. Black arrows indicate the timing of antibody administration, and red arrows indicate the positive staining (×200 magnification).

Display full size

Cell lines

The three human HCC cell lines (HepG2, Huh-7, SK-Hep-1) were purchased from National Infrastructure of Cell Line Resource (Beijing, China) and were cultured in DMEM (Gibco, Grand Island, NY, USA) supplemented with 1% penicillin/streptomycin (Gibco)and 10% fetal bovine serum (FBS, Biological Industries, Israel). PBMCs were isolated from the blood of healthy donors using Ficoll® Paque Plus (GE Healthcare, Piscataway, NJ, USA) according to the manufacturer’s instructions. PBMCs were cultured in RMPI-1640 medium (Gibco), supplemented with 10%FBS, 1% L-glutamine, and 1% penicillin/streptomycin. These cells incubated at 37°C with 5% CO₂. HEK293 F cell line was purchased (NANOGEN, Beijing, China) and cultured in NANO serum-free medium with 7% CO₂ humidified atmosphere at 37°C.

Bispecific antibody engineering

The VH-VL domain (ScFv) of anti-human CD3 antibody and VH domain (Fv) of anti-GPC3 antibody are derived from monoclonal antibody L2K¹² and HN3²⁶ separately. The bispecific molecules were designed in silico and chemically synthesized (GENEWIZ, Suzhou, China). Briefly, the ScFv fragment linked by 15 amino-acid long poly-glycine/serine linker consisting of (G₄ S)₃, was fused to the Fc of IgG1 via the hinge region. Similarly, the C-terminus of VH domain was fused to Fc domain. The regions of each Fc introduced P329 G/L234A/L235A (in EU numbering) mutations and the knob-in-hole mutations .³²

Bispecific antibody expression and purification

Chemically synthesized antibody genes were sub-cloned into pCDNA3.4 vector for transient transfection in mammalian expression system HEK293 F cells. Culture supernatants were harvested by centrifugation and filtered through a 0.22 micron filter before passage through a Protein A column (MabSelect SuRe LX, GE Healthcare) and Cation exchange column (CaptoTMS ImpAct, GE Healthcare). Monomeric fractions were pooled, quantified, concentrated and stored at 4°C. The purity of antibodies was confirmed by SDS-PAGE and SEC-HPLC. For SEC-HPLC, the samples (30 μg) were loaded on to a TSKgel super SW3000 column (Tosoh, Tokyo, Japan) and developed with an isocratic gradient of 0.1 M sodium phosphate and sodium sulfate pH6.7 at 0.3 mL/min. Continuous detection was by absorbance at 280 nm.

Antigen-binding measurements

Human HCC cells and PBMCs were harvested, wash, and resuspended in PBS buffer containing 2% FBS. Each cell type with 5 × 10⁵ cells, were incubated for 1 h at 4°C with 1 μg of bispecific antibodies or isotype control IgG1 antibody (Sigma-Aldrich, Munich, Germany). After the incubation, cells were washed and incubated with anti-human IgG-FITC antibody (Sigma) for 30 min at 4°C and subsequently washed again. The fluorescence associated with cells were analyzed using BD FACSVerse flow cytometer and BD FACSuite software (BD Bioscience, NJ, USA).

Thermostability assay

Thermostability was assessed by differential scanning calorimetry (DSC) using a Microcal VP-DSC scanning microcalorimeter (MicroCal, GE Healthcare UK, Little Chalfont, UK). Antibodies used for the studies were monomeric through analytical SEC-HPLC and filtered by a 0.22 micron filter before loading into the calorimeter. Bispecific antibodies with a concentration of 0.5 mg/ml (in pH 7.0 PBS buffer) were added to the sample well and DSC measurements were conducted at a 1°C/min scan rate from 25-95°C. Data analysis was carried out using the software Origin 7.0 provided by Microcal. The T_{m, onset} is defined as the qualitative temperature at which the thermogram appears to have a nonzero slope .¹⁶

Functional assays

Human T cells were isolated from the whole blood of healthy donors using human T-cell enrichment kit (Stem cell technologies, Vancouver, BC, Canada) according to the manufacturer’s instructions. Human T cells (effector cells, E) and HCC cells (target cells, T) were co-cultured with indicated E/T ratios and gradient diluted bispecific antibodies were added for incubation. Cytotoxicity was performed by WST-8 assay (Boster, Wuhan, China) after 48 h at 37°C and 5% CO₂. For detection of T-cell activation, the cells were stained with anti-CD4, anti-CD8, anti-CD25 and anti-CD69 antibodies (BD Bioscience). For analysis of T-cell proliferation, the T cells were labeled with carboxyfluorescein diacetate succinimidyl ester (CFSE) dye (MedChemExpress, Shanghai, China) and analyzed by flow cytometry.

Mouse xenograft studies

All protocols were approved by the Ethics Committee of Chongqing Academy of Animal Sciences, and the experiments were performed in accordance with the guidelines. Female NOD/SCID mice (Vital River Laboratories, Beijing, China) were received a suspension of 5 × 10⁶ of Huh-7 cells by subcutaneous injection. When the tumor volume reached approximately 200 mm³ (Tumor volumes were calculated using the formula: volume = 1/2 (length (width)²) and were allocated to treatment groups (n = 6). Mice were intraperitoneally injected with human PBMCs (1 × 10⁷ cells). Therapy administration started after 1 day of PBMCs transfer, followed by intravenously injection of bispecific antibodies or PBS (vehicle) with indicated doses. Tumor size was measured every 5 days.

Immunohistochemistry

Preparation of paraffin-embedded sections of mouse tumor tissues and principles for immunohistochemistry were described previously .³³ Bispecific antibody with a concentration of 2.5 µg/ml was applied for the detection.

Statistical analysis

Means, SDs, nonlinear regression analysis to determine EC₅₀ values, and were calculated using GraphPad Prism version 5.0 (GraphPad Software, La Jolla, CA, USA). Tumor volumes in mouse xenograft studies were compared by one-way ANOVA with Dunnett’s posttest. Values of P < 0.05 were considered to be statistically significant.