Analysis of steady-state Förster resonance energy transfer data by avoiding pitfalls: Interaction of JAK2 tyrosine kinase with N-methylanthraniloyl nucleotides

Förster resonance energy transfer (FRET) between the fluorescent ATP analogue 2′/3′-(N-methyl-anthraniloyl)-adenosine-5′-triphosphate (MANT–ATP) and enzymes is widely used to determine affinities for ATP–protein binding. However, in analysis of FRET fluorescence data, several important parameters are often ignored, resulting in poor accuracy of the calculated dissociation constant (K_d). In this study, we systematically analyze factors that interfere with K_d determination and describe methods for correction of primary and secondary inner filter effects that extend the use of the FRET method to higher MANT nucleotide concentrations. The interactions of the fluorescent nucleotide analogues MANT–ATP, MANT–ADP 2′/3′-O-(N-methylanthraniloyl) adenosine diphosphate], and MANT–AMP 2′/3′-O-(N-methylanthraniloyl) adenosine monophosphate] with the JAK2 tyrosine kinase domain are characterized. Taking all interfering factors into consideration, we found that JAK2 binds MANT–ATP tightly with a K_d of 15 to 25 nM and excluded the presence of a second binding site. The affinity for MANT–ADP is also tight with a K_d of 50 to 80 nM, whereas MANT–AMP does not bind. Titrations of JAK2 JH1 with nonhydrolyzable ATP analogue MANT–ATP-γ-S 2′/3′-O-(N-methylanthraniloyl) adenosine-5′-(thio)- triphosphate] yielded a K_d of 30 to 50 nM. The methods demonstrated here are applicable to other enzyme–fluorophore combinations and are expected to help improve the analysis of steady-state FRET data in MANT nucleotide binding studies and to obtain more accurate results for the affinities of nucleotide binding proteins.