Generation and functional characterization of a single-chain variable fragment (scFv) of the anti-FGF2 3F12E7 monoclonal antibody

Abstract

Single-chain variable fragments (scFvs) are small-sized artificial constructs composed of the immunoglobulin heavy and light chain variable regions connected by a peptide linker. We have previously described an anti-fibroblast growth factor 2 (FGF2) immunoglobulin G (IgG) monoclonal antibody (mAb), named 3F12E7, with notable antitumor potential revealed by preclinical assays. FGF2 is a known angiogenesis-associated molecule implicated in tumor progression. In this report, we describe a recombinant scFv format for the 3F12E7 mAb. The results demonstrate that the generated 3F12E7 scFv, although prone to aggregation, comprises an active anti-FGF2 product that contains monomers and small oligomers. Functionally, the 3F12E7 scFv preparations specifically recognize FGF2 and inhibit tumor growth similar to the corresponding full-length IgG counterpart in an experimental model. In silico molecular analysis provided insights into the aggregation propensity and the antigen-recognition by scFv units. Antigen-binding determinants were predicted outside the most aggregation-prone hotspots. Overall, our experimental and prediction dataset describes an scFv scaffold for the 3F12E7 mAb and also provides insights to further engineer non-aggregated anti-FGF2 scFv-based tools for therapeutic and research purposes.

Figure 1

Generation of 3F12E7 anti-FGF2 scFv. (a) Agarose gel electrophoresis of PCR amplified products of V_L and V_H genes from 3F12E7 hybridoma cells. (b) Design of the 3F12E7 scFv construct, formed by V_H and V_L amino acid sequences attached by a flexible peptide linker [(G₄S)₃]. (c) Schematic representation of the pET26-based vector used for the expression of 3F12E7 scFv in E. coli. The insert codes the scFv linked to a signal sequence and a C-terminal 6 × His tag. (d) SDS-PAGE analysis of the affinity-purified 3F12E7 scFv, under denaturing and reducing conditions. Gel protein content was detected by Coomassie blue staining. (e) Binding of 3F12E7 scFv to FGF2 was accessed by ELISA. 3F12E7 full-length IgG was used as a positive control.

Figure 2

In vitro and in vivo functional effects of 3F12E7 anti-FGF2 scFv. 3F12E7 scFv reduces in vitro endothelial cell proliferation (a) and migration (b). Cells were incubated with 10 µg/mL of the indicated mAbs. No difference was detected between 3F12E7 scFv and 3F12E7 full-length IgG groups. Cell proliferation and migration were accessed by trypan blue exclusion and scratch assays, respectively. Representative micrographs of the scratch assay are on (c). Dashed lines indicate original wound edges. Scale bar, 200 µm. *P < 0.05 compared to isotype and vehicle controls; one-way ANOVA/Bonferroni’s post-test. (d) Immunoblotting analyses of ERK1/2 phosphorylation in HUVEC after 48-h incubation with the indicated mAbs (50 µg/mL). β-actin was used as loading control. Graph shows the quantitative densitometry of the immunoblot results. Data are mean ± s.d. of the relative intensity of the bands, normalized to that of isotype ctrl IgG group, from two independent assays. The full-length image scans and the result of an additional independent assay are provided in Supplementary Fig. S4a. (e, f) 3F12E7 scFv reduces xenograft tumor growth similarly to 3F12E7 full-length IgG mAb. (e) Tumor growth curve. (f) Excised tumor mass on day 12. Treatment started four days after subcutaneous injection of B16-F10 cells. Result (mean ± s.d.) is representative of two independent experiments. Experimental groups: isotype control full-length IgG antibody, n = 6; 3F12E7 full-length IgG mAb, n = 6; 3F12E7 scFv, n = 6. *P < 0.05 compared with isotype control group; one-way ANOVA/Bonferroni’s post-test.

Figure 3

Analysis of the 3F12E7 scFv aggregation state and activity. (a) Evaluation of aggregation propensity of 3F12E7 scFv using Aggrescan3D web server and sequence-based predictors. Amino acid residues (numbered according to primary structure) that had positive scores are indicated in different colors for each predictor: Aggrescan (purple); Aggrescan3D (magenta); Waltz (blue); FoldAmyloid (yellow); Tango (orange). (b) Oligomeric profile of the 3F12E7 scFv, as assessed by blue native gel polyacrylamide electrophoresis (BN-PAGE). Gel was stained with silver nitrate. Graph on the right shows the corresponding densitometry analysis. (c) Dynamic light scattering (DLS) analysis for 3F12E7 scFv (left) and full-length IgG mAb (right). Average particle size results for each mAb are expressed by signal intensity. Each line denotes data obtained for the 3 independent samples. (d) Analysis of 3F12E7 scFv by size-exclusion chromatography (on a PD-10 column) and the binding of the obtained fractions to FGF2, as assessed by ELISA. Dashed line indicates ELISA background signal (BG). (e) Immunoblotting analysis of the 3F12E7 scFv reactivity to B16-F10 tumor protein extracts. For that, antibodies were labeled with biotin and detected with horseradish peroxidase (HRP)-streptavidin. These immunoblot images are also provided in Supplementary Fig. S4b. MW, molecular weight (kDa).

Figure 4

Prediction of 3F12E7 scFv binding to FGF2. (a) Metrics of the top-ranked 5 poses found by molecular docking between FGF2 (PDB ID: 1bfg) and 3F12E7 scFv with the Rosetta server. Based on CAPRI criteria, solution ID #2 was chosen for interaction analyses. (b) LIGPLOT⁺ diagram of the residues interacting across the 3F12E7 scFv-FGF2 interface. 3F12E7 scFv and FGF2 residues are labeled brown and magenta, respectively. Hydrophobic interactions are represented by arc with spokes and hydrogen bonds are indicated by dashed green lines. Hydrogen bonds were detected between R107 and T112 of FGF2 and N57 and N231 of the scFv entity, respectively. (c) 3D representation of the 3F12E7 scFv-FGF2 putative complex. FGF2 and 3F12E7 scFv residues found by LIGPLOT⁺ analyses are labeled magenta and orange, respectively. Hydrogen bonds are indicated by dashed lines. In dark blue, it is highlighted the FGF2 amino acid sequence (residues 104–146) used for the generation of the anti-FGF2 3F12E7 mAb, as reported. Within this segment, the residues predicted to interact with 3F12E7 scFv are schematically indicated on (d), along with the heparin and FGFR binding sites. (e) Localization of the putative FGF2-contacting residues and the Paratome CDRs in the 3F12E7 scFv sequence.