Fluorescence depletion measurements in various experimental geometries provide true emission and absorption anisotropies for the study of protein rotation.

Use of fluorescence depletion methods for measuring slow protein rotational diffusion has been limited by failure to obtain, from depletion data, well-defined anisotropy functions dependent on the distribution of either fluorophore emission or absorption transition dipoles, but not both. Such anisotropies would be directly comparable to those obtained from phosphorescence emission or triplet absorption measurements. We now describe such procedures applicable to cuvet and microscope experimental geometries, together with supporting experimental results. In cuvet measurements, the pump and probe beams are colinear and fluorescence is collected at 90 degrees to this axis. The data analysis procedure for this geometry has been suggested by Wegener (Biophys. J., 46 (1984) 795) and permits calculation of the absorption and emission anisotropies and the interdipole angle. In microscope experiments, fluorescence emission is collected along the pump/probe beam axis. For microscope measurements, a new experimental procedure permits evaluation of absorption and emission anisotropies when the interdipole angle is independently known. In either case multiple depletion measurements are required, each with different relative orientations of the probe beam polarization, pump beam polarization and emission polarizer axis. We have used these methods to calculate the time-dependent anisotropies for eosin-derivatized BSA rotation in glycerol solutions in both experimental geometries. These data correspond well with those obtained from time-resolved phosphorescence anisotropy measurements.