Use of fluorescence-detected sedimentation velocity to study high-affinity protein interactions

Abstract

Sedimentation velocity (SV) analytical ultracentrifugation (AUC) is a classic technique for the real-time observation of free macromolecular migration in solution driven by centrifugal force. This enables the analysis of macromolecular mass, shape, size distribution, and interactions. Although traditionally limited to determination of the sedimentation coefficient and binding affinity of proteins in the micromolar range, the implementation of modern detection and data analysis techniques has resulted in marked improvements in detection sensitivity and size resolution during the past decades. Fluorescence optical detection now permits the detection of recombinant proteins with fluorescence excitation at 488 or 561 nm at low picomolar concentrations, allowing for the study of high-affinity protein self-association and hetero-association. Compared with other popular techniques for measuring high-affinity protein-protein interactions, such as biosensing or calorimetry, the high size resolution of complexes at picomolar concentrations obtained with SV offers a distinct advantage in sensitivity and flexibility of the application. Here, we present a basic protocol for carrying out fluorescence-detected SV experiments and the determination of the size distribution and affinity of protein-antibody complexes with picomolar K_D values. Using an EGFP-nanobody interaction as a model, this protocol describes sample preparation, ultracentrifugation, data acquisition, and data analysis. A variation of the protocol applying traditional absorbance or an interference optical system can be used for protein-protein interactions in the micromolar K_D value range. Sedimentation experiments typically take ∼3 h of preparation and 6-12 h of run time, followed by data analysis (typically taking 1-3 h).

Figure 1 |. Recorded sedimentation boundaries and fit.

(a) Screenshot of the SEDFIT window showing all scans. The upper plot shows a superposition the raw scan data for cell #1 sector A (0.1 nM EGFP + 0.02 nM Nanobody) in fluorescence counts vs. radius with higher color temperature for later scans, superimposed with the best-fit model. The vertical lines (emphasized by the arrows superimposed to the screenshot) are the locations of meniscus (red), bottom (blue), and fit limits (green), respectively. Residuals of the fit are represented as a grey bitmap and overlay below the raw data. The lower plot is the calculated sedimentation coefficient distribution c(s) associated with the fit. It shows a small buffer peak close to 0 S, a main peak of EGFP and EGFP-nanobody complex at ~2.7 S, and a shallow broad signal contribution extending to ~ 7 S from BSA oligomers. Close to the upper right corner of the distribution plot, a yellow arrow points to the integration button. The dashed yellow rectangle shows an example for a suitable integration range. (b) GUSSI overlay of raw scan data (points) and best-fit distributions (lines) from the same experiment as in (a), but showing only every 10^th data set.

Figure 2 |. GUSSI overlay of sedimentation coefficient distributions

GUSSI overlay of sedimentation coefficient distributions showing the EGFP control (black), and mixtures with 0.1 nM, 0.3 nM, and 20 nM nanobody.

Figure 3 |. Screenshot of the isotherm text-file created using Windows Notepad.

In the first column are the micromolar values of the EGFP concentration, the second the nanobody micromolar concentrations, and in the third are the adjusted signal weighted-average sedimentation coefficients from integration of the sedimentation coefficient distributions (s_w,adj). Values may be separated by space or tab characters.

Figure 4 |. SEDPHAT screenshots of experimental and global parameters

SEDPHAT screenshots of (a) the experimental parameters box, and (b) the global parameters box of the A + B AB hetero-association model.

Figure 5 |. GUSSI display of the signal-average sedimentation coefficients

Signal-average sedimentation coefficients of the EGFP-nanobody mixtures (circles) and the best-fit binding isotherm (solid line).