Quantifying binding of fluorescent VEGF isoforms at VEGFR2 using NanoBRET

C. J. Peach¹, L. E. Kilpatrick¹, R. Friedman Ohana², M. Robers², J. Woolard¹, S. J. Hill¹. ¹School of Life Sciences, University of Nottingham, Nottingham, United Kingdom, ²Promega Corporation, Madison,

Introduction: Vascular endothelial growth factor (VEGF) is a key mediator of angiogenesis, primarily signalling via VEGF receptor 2 (VEGFR2) (1). Alternative splicing of the Vegfa gene produces VEGF ligand isoforms with distinct signalling and physiological outcomes (2), such as the recently identified VEGF-Ax (3). Despite approved anti-cancer therapeutics targeting VEGF/VEGFR signalling (4), there are limited quantitative pharmacological studies of these endogenous isoforms at full-length VEGFR2 in living cells. Here we have used fluorescent variants of VEGF₁₆₅a, VEGF₁₆₅b and VEGF₁₂₁a to quantify ligand binding at NanoLuc-tagged VEGFR2 using bioluminescence resonance energy transfer (BRET).

Method: HEK293 cells expressing N-terminal NanoLuc-VEGFR2 were seeded 24 hours prior to experimentation in white 96-well plates. Increasing concentrations of VEGF₁₆₅a, VEGF₁₆₅b or VEGF₁₂₁a, labelled at a single site with the fluorophore tetramethylrhodamine (TMR), were added in the presence and absence of 100nM unlabelled VEGF in Hanks buffered saline solution/0.1% bovine serum albumin (pH 7.4). For competition experiments, cells were co-stimulated with fixed concentrations of VEGF₁₆₅a-TMR, VEGF₁₆₅b-TMR or VEGF₁₂₁a-TMR and increasing concentrations of unlabelled VEGF. Following 60min stimulation at 37°C, the NanoLuc substrate furimazine (10μM) was added and BRET ratios were recorded using a BMG Pherastar. NanoBRET kinetic experiments were performed at 37°C where VEGF₁₆₅a-TMR, VEGF₁₆₅b-TMR or VEGF₁₂₁a-TMR were added following furimazine addition and BRET ratios were calculated every 30 seconds for 20 minutes. Data are expressed as mean ± S.E.M.

Results: Saturable binding was observed at NanoLuc-VEGFR2 for VEGF₁₆₅a-TMR, VEGF₁₆₅b-TMR and VEGF₁₂₁a-TMR with nanomolar affinities (Table 1, N=3) and minimal non-specific binding. In competition experiments, all VEGF-TMR ligands were fully displaced by a panel of unlabelled VEGF isoforms. Derived pK_i values showed similar binding affinities for all isoforms, ranging between 0.2nM and 1.4nM, and were independent of VEGF-TMR probe used (N=5). Real-time binding kinetics fitted to a simple association exponential model showed comparable association and dissociation rates for all fluorescent VEGF variants (Table 1, N=5). Kinetic K_d values were similar to those derived from saturation experiments.

Conclusions: Fluorescently tagged VEGF₁₆₅a, VEGF₁₆₅b and VEGF₁₂₁a enabled ligand binding affinities and kinetics to be quantified in living cells expressing full-length VEGFR2 without endogenous co-receptors present. These data suggest distinct signalling between VEGF isoforms does not arise from binding VEGFR2 alone.

References:

(1) Koch S et al. (2011) Biochem. J. 437:169-183;

(2) Woolard J et al. (2009) Microcirculation 16:572-592;

(3) Eswarappa S et al. (2014) Cell 157:1605-1618;

(4) Ferrara N and Adamis A (2016) Nat. Rev. Drug Discov. 15:385-403.

Table 1. Ligand binding affinity and kinetics at NanoLuc-VEGFR2 measured using NanoBRET.