Interpretation of fluorescence decays in proteins using continuous lifetime distributions.

The decay of the tryptophanyl emission in proteins is often complex due to the sensitivity of the tryptophan excited state to its surroundings. The traditional analysis of the decay curve using exponential components is based on the identification of each component with a particular protein conformation. An alternative approach assumes that proteins can exhibit a large number of conformations and that, at room temperature, the interconversion rate between conformations can be of the same order of magnitude as the excited-state decay rate. Following this assumption, the analysis of the protein emission was performed using continuous distributions of lifetime values. The number of average protein conformations, the range of mobility around each conformation, and the rate of interconversion between conformations determines the characteristics of the lifetime distribution. The fluorescence decay from some single tryptophan proteins was measured using multifrequency phase fluorometry and analyzed using a sum of exponentials, unimodal and bimodal probability-density functions, and the analytical form for lifetime distribution obtained for a model in which the tryptophan residue can move in a single potential well. For ribonuclease T1 and neurotoxin variant 3, the sum of two exponentials and bimodal probability-density functions gave comparable results, whereas for phospholipase A2, the description of the decay required three exponentials or bimodal probability-density functions. Also the temperature dependence of the fluorescence decay was investigated. It was found that the lifetime distribution was broader and shifted toward longer lifetime values at lower temperature. The analysis of the decay of tryptophan in buffer and of some tryptophan derivatives gave single-exponential decays. The single-potential well lifetime distribution, which has only three adjustable parameters, gave good fits for all cases investigated, but in the case of phopholipase A2, the temperature dependence of the parameters that describe the single-potential well distribution indicated the inadequacy of this model at lower temperature, suggesting that multiple potential wells can describe better the decay for this protein.