Fluorescence anisotropy (FA) binding assays

To perform binding studies, RNA ligand concentrations were held constant and as low as possible while still maintaining good signal/noise (1–3 nM). Purified labeled RNA was snap cooled in 1× binding buffer (135 mM KCl, 15 mM NaCl, 50 mM Tris (pH 8.3 at 25°C), 10% glycerol) at 2× final concentration and the carrier molecule, tRNA^Leu, was added at 2× final concentration (3.6 μM). Protein dilutions were performed separately in 1× binding buffer at 2× final concentration. Protein and RNA were then mixed in a 1:1 volume ratio in a 20 μl reaction and allowed to come to equilibrium at room temperature in the dark for 40–60 min. Flat-bottom low-flange 384-well black polystyrene plates (Corning) were used. Perpendicular and parallel fluorescence intensities were measured using a ClarioStar Plus FP plate reader (BMG Labtech) and anisotropy values were calculated for each protein titration point where anisotropy = (I_‖ – I_⊥)/(I_‖ + 2*I_⊥) and correlates directly with fraction bound. Associated anisotropy was plotted as a function of the log of activity corrected protein concentration. To determine the K_D the data were fit to the simplified binding isotherm, anisotropy = O + (S*P)/(K_D + P) with KaleidaGraph where S and O are saturation and offset respectively, and P is the protein concentration. All binding reactions were performed in triplicate or more using different protein dilutions on separate days. Standard errors of the mean were calculated and reported. ΔG is calculated from the fitted K_D, ΔG = RT*ln(K_D). Statistical significance is determined for differences between averages of apparent dissociation constants using the two-tailed paired t-test. K_D determination for the interaction of full-length hnRNPK to the B motif RNA by FA was verified by EMSA (Supplemental Figure S1).