Interpretation of intramolecular stacking effect on the fluorescence intensity decay of 3-methylbenzimidazolyl(5'-5')guanosine dinucleotides using a model of lifetime distribution.

Time-resolved fluorescence of 3-methylbenzimidazole (m(3)B) was used to study stacking interaction between base moieties in di-, tri- and tetra-phosphate analogues of 3-methylbenzimidazolyl(5'-5')guanosine (m(3)Bp( n )G, n = 2, 3, 4), using 5'-triphosphate of 3-methylbenzimidazole riboside (m(3)BTP) as reference. Fluorescence intensity decays of all compounds cannot be satisfactory fitted with single-exponential function. Although an increase of a number of exponents led to better fits, interpretation of the individual exponential terms, i.e. pre-exponential amplitudes and fluorescence lifetimes, cannot be adequately characterized. We show that these fluorescence decays are best fitted by power-like function derived from physically justified distribution of the fluorescence lifetimes, and characterized by the mean value of the excited-state lifetime and relative variance of lifetime fluctuations around the mean value. The latter led to the parameter of heterogeneity and number of decay paths, which depend on the factors responsible for non-radiative decay of the excited state, including base-base stacking interaction. This was studied by means of changes of temperature and the number of phosphate groups in dinucleotides. It was shown that the strongest effect of stacking interactions, characterized by lowest values of both fluorescence mean decay time and relative variance, occurs in the case of m(3)Bp(3)G containing the same number of phosphates as natural mRNA cap. The possible importance of these results for interpretation of the mechanism of function of the mRNA cap structure is discussed.