Helix rigidity of DNA: the meroduplex as an experimental paradigm.

The intrinsic rigidity of the DNA helix is generally believed to arise primarily from vertical base-stacking interactions; however, relatively little experimental information exists regarding the relationship between the thermodynamic stability of base-stacking interactions and the mechanical rigidity imparted by such interactions. To address this issue, the solution conformations of complexes formed between adenine (A) or N-6-methyladenine (meA) monomer and deoxythymidylate (dTn) polymers of varying length (n = 40, 60, 81, and 110) have been examined. Such complexes are known to exist as extended, chiral structures in which the purine monomers exist as extensively stacked arrays. Thus, one can in principle examine the structural consequences of base-pair stack formation in the absence of any change in stoichiometric (phosphate) charge. The current approach has utilized the method of transient electric birefringence (TEB), which is highly sensitive to changes in nucleic acid conformation. Addition of millimolar concentrations of either A or meA to the dTn species leads to the formation of relatively rigid, chiral complexes whose dimensions are strictly limited by the length of the polymer strand. For adenine, the principal species appears to be [A] approximately n/2-dTn in which the polymer strand doubles back to form the two continuous strands of the complex (merotriplex). The addition of a methyl group to the N-6 position of adenine (meA) results in a shift to a meroduplex form, [meA] approximately n-dTn, with an intrinsic rigidity that is nearly identical to the rigidity of the corresponding duplex, dAn-dTn, despite the fact that the stoichiometric charge of the meroduplex is only one-half of that of the full duplex. The current results thus support a model in which helix rigidity is primarily due to the intrinsic resistance to deformation of base-stacking interactions; the deformation energies, as with the stacking energies themselves, are expected to be quite sequence-dependent. Phosphate-phosphate (repulsive) interactions, whose contributions are both salt-dependent and relatively sequence-independent, appear to play a secondary role in establishing helix rigidity. In particular, the DNA helix is likely to possess substantial rigidity in the absence of phosphate interactions. Thus, proteins whose interactions with DNA lead to substantial bending of the helix axis may facilitate such distortions through solvation of bases in addition to partial charge neutralization.