Abstract

Fibroblast Growth Factor Receptor 1 (FGFR1) plays a direct role in tumour growth and metastasis and is a key therapeutic target for the treatment of multiple cancers [1]. A major obstacle to therapeutic success is the prediction of drug efficacy in vivo. Identification of slow compound dissociation kinetics (koff), or increased residence time (1/koff), however, can be beneficial for the selection of small molecule inhibitors with improved in vivo efficacy [2]. Here, we demonstrate how Cisbio’s HTRF® technology can be applied – in conjunction with Motulsky and Mahan (M-M) analysis [3] – to derive the binding kinetics of unlabelled inhibitors at the FGFR1 receptor. To provide accurate input parameters for M-M analysis, the association kinetics of varying concentrations of fluorescent tracer (staurosporine-red) was assessed upon injection of donor complex (purified, biotinylated-FGFR1 kinase and streptavidin-Eu3+-cryptate) using BMG’s PHERAstar® FSX system. Fitting global association kinetics provided staurosporine-red association- and dissociation-rate constants (kon 7.2 ± 0.7 x107 M-1 min-1, koff 1.6 ± 0.08 min-1), alongside a kinetically derived dissociation constant (Kd 22.9 ± 1.4 nM). Next, competition association experiments were performed using a fixed concentration of staurosporine-red in the presence of three concentrations of unlabelled inhibitor. However, at Kd concentration of staurosporine-red, inhibition of FGFR1 with the broad-spectrum and low-affinity kinase inhibitor dasatinib was subject to a high degree of M-M curve-fitting error. This was overcome by increasing tracer concentration, which increased the definition required for accurate curve fitting during the initial stages of ligand association. Reasonable kinetic estimates were subsequently derived for dasatinib (kon 211 ± 86 x107 M-1 min-1, koff 18 ± 7 min-1, Kd 852 ± 37 nM). Additionally, the confidence interval associated with slower-on/off inhibitors staurosporine (kon 3.6 ± 0.4 x107 M-1 min-1, koff 2.1 ± 0.2 min-1, Kd 60 ± 6 nM) and PD173074 (kon 2.2 ± 0.4 x108 M-1 min-1, koff 0.18 ± 0.02 min-1, Kd 0.89 ± 0.09 nM), also improved. In summary, application of the M-M method in conjunction with Cisbio’s HTRF® technology – which provides coverage for up to 80% of the kinome – provides a robust and valuable format for the profiling of inhibitor binding in a medium-throughput assay format [4]. Particularly for low-affinity ligands however, careful assay optimisation (such as tracer characteristics and concentration) is important for the acquisition of accurate data from the M-M model [5]. 1. Katoh, M. (2019) Fibroblast growth factor receptors as treatment targets in clinical oncology. Nat. Rev. Clin. Oncol. 16, 105-122. 2. Lee, K.S.S., et al., (2019) Drug-target residence time affects in vivo target occupancy through multiple pathways. ACS Cent. Sci. 5, 1614-1624. 3. Motulsky, H.J. and Mahan, L.C. (1984) The kinetics of