Binding mode of DiSC2(5) with duplex DNA determined by two-dimensional NMR

ORGN 723

Kristen C. Hannah, kch@andrew.cmu.edu, Roberto R. Gil, rgil@andrew.cmu.edu, and Bruce Armitage, army@andrew.cmu.edu. Department of Chemistry, Carnegie Mellon University, 4400 5th Ave., Pittsburgh, PA 15213
We have previously shown by optical spectroscopic methods that the dithiacarbocyanine dye DiSC2 (5) binds to the minor groove of DNA as a cofacial dimer. We have now used a combination of both one- and two-dimensional nuclear magnetic resonance experiments to confirm minor groove binding of DiSC2 (5) to the DNA duplex AT5. We have collected 1D Watergate, COSY and NOESY experiments for the DNA duplex both with and without dye. From these data we have unambiguously assigned the resonances of the DNA duplex, particularly the imino protons, except for the terminal residues which are fraying. In the presence of DiSC2 (5) , shifting, broadening or attenuation of some of the DNA imino protons identify the dye binding site within the minor groove. A reverse titration of DiSC2 (5) with AT5 was performed. After the first addition of AT5 (2:1, AT5:Dye), the dye proton signals broaden immediately. This supports a fast exchange equilibrium on the NMR time scale between the monomer and dimer bound states. Our data supports a central binding site in the 5 A/T base pairs that is slightly shifted to the 5'G, as indicated by perturbations of the G 3 imino proton. Widening of the minor groove when the dimer binds likely causes the perurbations of G 3 signals. We have designed short synthetic DNA duplexes to investigate the shifted binding site. A similar decrease in DiSC2 (5) dimerization is observed when either G 3-C 9 is replaced with an A-T base pair or the binding site is reduced to 4 A/T pairs. However replacement of G 3-C 9 by an I/C does not alter binding. This seems to indicate that G 3-C 9 does not actively participate in the binding site, but is rather affected indirectly by the dimer binding nearby.