A dual-targeting PDGFRbeta/VEGF-A molecule assembled from stable antibody fragments demonstrates anti-angiogenic activity in vitro and in vivo

Abstract

Targeting angiogenesis is a promising approach to the treatment of solid tumors and age-related macular degeneration (AMD). Inhibition of vascularization has been validated by the successful marketing of monoclonal antibodies (mAbs) that target specific growth factors or their receptors, but there is considerable room for improvement in existing therapies. Combination of mAbs targeting both the VEGF and PDGF pathways has the potential to increase the efficacy of anti-angiogenic therapy without the accompanying toxicities of tyrosine kinase inhibitors and the inability to combine efficiently with traditional chemotherapeutics. However, development costs and regulatory issues have limited the use of combinatorial approaches for the generation of more efficacious treatments. The concept of mediating disease pathology by targeting two antigens with one therapeutic was proposed over two decades ago. While mAbs are particularly suitable candidates for a dual-targeting approach, engineering bispecificity into one molecule can be difficult due to issues with expression and stability, which play a significant role in manufacturability. Here, we address these issues upstream in the process of developing a bispecific antibody (bsAb). Single-chain antibody fragments (scFvs) targeting PDGFRbeta and VEGF-A were selected for superior stability. The scFvs were fused to both termini of human Fc to generate a bispecific, tetravalent molecule. The resulting molecule displays potent activity, binds both targets simultaneously, and is stable in serum. The assembly of a bsAb using stable monomeric units allowed development of an anti-PDGFRB/VEGF-A antibody capable of attenuating angiogenesis through two distinct pathways and represents an efficient method for rapid engineering of dual-targeting molecules.

Figure 1

DSC analysis of the scFvs and scFv-Fc-scFv constructs. (A) Calculated tms for the scFvs: 74°C for the anti-pDGFRβ scFv, 78°C for the anti-VeGFA scFv. Both scFvs precipitated upon exposure to higher temperatures. (B) the Fc region and the scFv-Fc-scFv construct. Calculated tm for the bsAb, corresponding to the first unfolding event of the variable regions, is 72°C, very similar to the reported for the anti-scFv. An observed peak upstream at ∼62°C corresponds to the C_H2 region. All constructs were analyzed in standard buffer without any formulation enhancement.

Figure 2

SeC-MALS analysis of the scFvs and scFv-Fc-scFv constructs. the anti-pDGFRβ scFv and anti-VeGFA scFv run on Superdex-75 10 × 300 mm chromatography column for mass distribution analysis (molecular weight distribution in red, UV absorption in black). main peak observed for both the anti-and anti-VeGFA scFvs is monodisperse and has an average MW consistent with the expected monomeric form (A and B). Both scFvs exhibit some higher MW species eluting prior to the monomer, with the small peak eluting at approximately 19 minutes identified as a dimer. scFv-Fc-scFv bsAb was analyzed on a Superdex-200 10 × 300 mm column (C). main peak (at approximately 23 minutes) is monodisperse and has an average MW of 167 kDa, consistent with that expected for the glycosylated scFv-Fc-scFv. there is also a minor amount of higher MW material present (shown eluting between 18.5–21 minutes), a discrete oligomer with a MW corresponding to 2 × scFv-Fc-scFv, the homodimeric form of the construct.

Figure 3

DLS analysis of the scFvs and bsAb. Constructs were analyzed at both low concentration (1 mg/mL) and high concentration (25 mg/mL). Both anti-pDGFRβ (A) and anti-VeGFA (B) scFvs displayed a propensity to aggregate or form higher order oligomers at the higher concentration as indicated by the marked increase in hydrodynamic size (x-axis) compared to that observed at the lower concentration. the scFv-Fc-scFv also exhibited a predominant monodisperse population at the lower concentration with a radius of approximately 5.8 nm (C). However, upon concentration, the single population is only slightly larger at 6.8 nm, retains monodispersity and does not display the tendency to oligomerize or aggregate observed with the scFvs.

Figure 4

Sequence analysis of stable scFvs. (A) the light chain of the anti-pDGFRβ scFv is most closely related to V_κIV_B3 and a kappa J5 region. there are only two differences between the germline framework sequence and the derived light chain sequence; the A (alanine) at residue 1 is not found in other V_κIV sequences, but is found in VLI sequences, the same is true for the Q (glutamine) at position 3. heavy chain of the anti-scFv is derived from the Dyax FAB-310/410 libraries: based on V_H3_3-23 with a synthetic repertoire covering CDRs1 and 2, and an IgM repertoire from 33 autoimmune donors represented in CDR3; the heavy chain J region is J4. Numbering is per Kabat and Woo. CDRs, identified according to Kabat and Woo, are marked in red; germline framework residues possibly influencing solubility and folding efficiency are marked in blue. (B) the light chain of the anti-VeGF-A scFv is most closely related to V_LI_o12 and a kappa J2 region. there is only one difference between the germline framework sequence and the derived light chain sequence; the N (asparagine) at residue 1 is not found in other VLIV sequences, but is found in one VLI sequence. heavy chain of the scFv is derived from the Dyax FAB-310/410 libraries:20 based on V_H3_3-23 with a synthetic repertoire covering CDRs1 and 2, and an IgM repertoire from 33 autoimmune donors represented in CDR3; the heavy chain J region is J4. this derived heavy chain has one framework difference from the parent germline: the R (arginine) at residue 93 is found both in other V_H3 sequences, and other VH germline groups.

Figure 5

Dual-binding activity of the bsAb. Surface plasmon resonance analysis (Biacore t100) of bsAb for simultaneous (dual) binding to VeGF-A and PDGFRβ. the bsAb was captured onto a CM4 chip via hVeGF-A. either 0 nM (blue) or 500 nM (red) pDGFRβ-Fc was then allowed to bind to the bsAb.

Figure 6

Anti-pDGFRβ/VeGF-A ScFv-Fc-scFv is a potent inhibitor of VeGF-A and pDGFRβ-mediated activity in vitro. (A) primary HUVeCs were stimulated in SFM with 2 nM human VeGF-A for 48 hours in the presence of various concentrations of bevacizumab, anti-VeGF scFv or anti-pDGFRβ/VeGF-A scFv-Fc-scFv. 3H-thymidine was added to the culture for the last 18 hours. (B) primary human brain vascular pericytes (HBVp) were stimulated with 0.4 nM human pDGF-BB in the presence of various concentrations of anti-pDGFRβ mAb e9899, anti-scFv or anti-pDGFRβ/VeGF-A scFv-Fc-scFv for 18 hours. 3H-thymidine was added to the cultures after addition of reagents. Average 3H-thymidine-incorporation from triplicate wells are shown. Data is representative of four similar experiments.

Figure 7

Anti-pDGFRβ/VeGF-A scFv-Fc-scFv inhibits endothelial sprouting and pericyte/endothelial association in a coculture sprouting assay. Bevacizumab, anti-pDGFRβ mAb e9899 or the anti-pDGFR/VeGF-A scFv-Fc-scFv were used as antagonists in a endothelial:pericyte sprouting coculture assay as described in Methods. (A) Representative figures from each of the four groups (25 nM concentration). (B) Quantitative data using metamorph software for the four groups (25 nM concentration). (C) Dose response curve for bevacizumab alone or anti-VeGF-A scFv-Fc-scFv. the anti-mAb did not have any effect on endothelia sprouting in the assay (data not shown).

Figure 8

Prophylactic and therapeutic treatment with anti-pDGFRβ/VEGF-A scFv-Fc-scFv inhibits growth of A673 rhabdomyosarcoma tumors. (A) treatment—female SCID were injected with A673 tumor cells on Day 0. Starting Day 1, groups of mice were injected intraperitoneally (i.p.) with the indicated drugs (at specified concentrations) twice a week for a total of eight doses. (B) therapeutic treatment—female SCID were injected with A673 tumor cells on Day 0. Starting at a tumor volume of 150-200 mm3 groups of mice (n = 10/gp) were injected i.p. with the indicated drugs (at specified concentrations) twice a week for a total of five doses. Mean tumor volume change over time is shown in both figures.

Figure 9

Internalization of anti-pDGFRβ/VeGF-A scFv-Fc-scFv after binding to HBVps. Chamber slides containing HBVps were incubated with 1 µg/mL of anti-VEGF-A scFv-Fc-scFv for 1 hour on ice to allow binding. Following a pBS rinse, cells were cultured at 37°C for the indicated times, then fixed and permeabilized prior to staining with Alexafluor-488 labeled goat anti-human IgG antibody (green) and DApI nuclear stain (blue). the t = 0 time point was from cells cultured with scFv-Fc-scFv for 1 hour on ice.