Solvent and linker influences on AQ(*-)/dA(*+) charge-transfer state energetics and dynamics in anthraquinonyl-linker-deoxyadenosine conjugates.

The goal of this work is to produce high yields of long-lived AQ(-)/dA(+) charge transfer (CT) excited states (or photoproducts). This goal fits within a larger context of trying generally to produce high yields of long-lived CT excited states within DNA nucleoside conjugates that can be incorporated into DNA duplexes. Depending upon the energetics of the anthraquinonyl (AQ) (3)(pi,pi) state as well as the reduction potentials of the subunits in particular anthraquinonyl-adenine conjugates, CT quenching of the AQ (3)(pi,pi*) state may or may not occur in polar organic solvents. In MeOH, bis(3',5'-O-acetyl)-N(6)-(anthraquinone-2-carbonyl)-2'-deoxyadenosine (AQCOdA) behaves as intended and forms a (3)(AQ(-)/dA(+)) CT state with a lifetime of 3 ns. However, in nonpolar THF the AQ(-)/dA(+) CT states of AQCOdA are too high in energy to be formed, and in DMSO a (1)(AQ(-)/dA(+)) CT state is formed but lives only 6 ps. Although the lowest energy excited state for AQCOdA in MeOH is a (3)(AQ(-)/dA(+)) CT state, for N(6)-(anthraquinone-2-methylenyl)-2'-deoxyadenosine (AQMedA) in the same solvent it is a (3)(pi,pi*) state. Changing the linking carbonyl in AQCOdA to methylene in AQMedA makes the anthraquinonyl subunit harder to reduce by 166 mV. This raises the energy of the (3)(AQ(-)/dA(+)) CT state above that of the (3)(pi,pi*) in AQMedA. The conclusion is that anthraquinonyl-dA conjugates will not have lowest energy AQ(-)/dA(+) CT states in polar organic solvents unless the anthraquinonyl subunit is also substituted with an electron-withdrawing group that raises the AQ-subunit's reduction potential above that of AQ. A key finding in this work is that the lifetime of the (3)(AQ(-)/dA(+)) CT excited state (ca. 3 ns) is ca. 500-times longer than that of the corresponding (1)(AQ(-)/dA(+)) CT excited state (ca. 6 ps).'