Fluorescence resonance energy transfer in dye-labeled DNA

We present the results of molecular modeling of dye-labeled, double-stranded DNA. The structural information obtained from the simulations are used as input to an analysis of energy transfer in this system. The simulations reveal the nature of the interaction between a pair of fluorophores and DNA. The donor, tetramethylrhodamine, TMR, attached to the 5'-end of DNA with a six-carbon tether, interacts primarily with DNA's minor groove, but occasionally stacks against the DNA base pairs. The acceptor, Cy5, attached to the opposite strand at positions n (n = 7, 12, 14, 16, 19, 24, 27), binds in the major groove in two distinct locations on the upper and lower part of the groove. We analyzed in detail the dye-to-dye distances, dipole orientation factors and fluorescence resonance energy transfer (FRET) rates. Tests of the validity of the Forster model were conducted using the transition density cube (TDC) method, which provides the exact Coulombic interaction within a certain model chemistry. Our studies show that the use of long tethers does not guarantee rotational freedom of the dyes, as intended in the experiments. Instead, the tethers allow Cy5 to bind in two different geometries, which causes a large uncertainty in the dye-to-dye distances. Our results also show significant fluctuation in the orientation factor, kappa(2,) which, together with uncertainty in dye-to-dye distances, cause considerable uncertainty in interpreting FRET measurements. We suggest that molecular modeling, combined with the TDC method, provides a useful tool in designing and interpreting FRET experiments. (c) 2007 Published by Elsevier B.V.