Figure 1.
In vitro characterization of anti-VEGF RabMAbs.
(A) Dose-dependent inhibition of VEGF/VEGFR-2 interaction by 4 representative anti-VEGF RabMAbs. Bevacizumab and a non-relevant anti-human Factor VIII RabMAb were included as controls. Points represent means of three replications and error bars represent standard deviations. The IC50 values are shown in the table below. (B) Inhibition of VEGF-stimulated receptor tyrosine phosphorylation in 293/KDR cells in the presence of neutralizing antibodies against VEGF. Four representative anti-VEGF RabMAbs inhibited VEGF stimulated VEGFR-2 phosphorylation in 293/KDR cells. Bevacizumab and a non-relevant anti-human Factor VIII RabMAb were used as positive and negative controls, respectively. Total VEGFR-2 staining was performed as sample quantitative controls. (C) Specificity and cross-reactivity of anti-VEGF RabMAbs. A panel of seven representative anti-human VEGF antibodies exhibited no reactivity to human VEGF-B, -C, -D. Four of the seven antibodies (EBV302, EBV307, EBV320 and EBV321) were cross-reactive with mouse VEGF.
Table 1.
In vitro characterization of anti-VEGF RabMAbs.
Figure 2.
Mutational lineage guided humanization of anti-VEGF RabMAb EBV321.
(A–B) Phylogenetic analysis of VK (A) and VH (B) amino acid sequences of 15 neutralizing anti-VEGF RabMAbs by Clustal X. The human antigen-specific clones are underlined. RabMAbs of the same lineage group are boxed and labeled as 1, 2 or 3. (C–D) Alignments of VH (C) and VK (D) protein sequence of the EBV321 lineage group 2 (EBV302, EBV307, EBV320 and EBV321) with human germline and humanized EBV321 sequences. ‘–’ denotes residues that are identical at the corresponding positions. ‘*’ denotes the residues in RabMAb framework regions potentially involved in CDR contacts or inter-chain contacts. ‘$’ denotes the residues considered not critical to the structural activity. ‘#’ denotes the residues humanized in the CDR region. Chothia numbering scheme and Kabat CDR loop definition were used [55], [56].
Table 2.
Antigen binding activity of random pairs of 15 anti-VEGF antibody H and L chains.
Figure 3.
Comparison of hEBV321 and Bevacizumab binding properties to human VEGF.
A. VEGF competition ELISA measuring the binding of hEBV321 to VEGF coated on ELISA plate in the presence of increasing concentration of competitor (hEBV321, Bevacizumab or an irrelevant antibody Humira). B-E. Direct binding of hEBV321 and Bevacizumab to various forms of human VEGF captured on ELISA plate. B. VEGF wild type; C. VEGF I46A; D. VEGF G89A; E. VEGF G88A/Q89A.
Figure 4.
Humanized EBV321 retains full biological activities in vitro and in vivo.
(A) Comparison of hEBV321 and its parental RabMAb EBV321 in the receptor-ligand interaction ELISA assay. Bevacizumab was included as a positive control. Points represent the means of three replicates and error bars represent standard deviations. The IC50 values are shown in the table below. (B) Dose-dependent inhibition of HUVEC proliferation by RabMAb EBV321, hEBV321, and Bevacizumab. Points represent the means of three replicates and error bars represent standard deviations. The IC50 values are shown in the table below. (C) Inhibition of tumor growth in vivo by anti-VEGF RabMAbs EBV321 and hEBV321 in NCI-H460 xenograft model. Bevacizumab, EBV321, and hEBV321 were dosed at 5 mg/kg 3 times per week for a total of 9 doses (8 mice per group). Error bars represent the standard deviation. (D) Photomicrographs of immunohistochemistry staining with an anti-CD34 mAb (x400). Representative areas of tumor sections from cohorts receiving saline as control (i), Bevacizumab (ii), EBV321 (iii) and hEBV321 (iv). (E) MVD scoring of CD34 staining in NCI-H460 xenograft tumors. (F) Inhibition of tumor growth by hEBV321 in A673 xenograft model. Animals were i.p. injected with normal saline, 5 mg/kg hEBV321 and Bevacizumab, 3 times per week for a total of 9 doses (8 to 11 mice per group). Symbols and bars, mean + standard deviation. Significant difference when compared to the control group by ANOVA was denoted with *: P<0.05 or **: P<0.01. †, significant difference compared with the bevacizumab group (P<0.05, student's t test).
Table 3.
Binding affinities of anti-VEGF-A RabMAb EBV321 and its humanized version.