Wright L A, Lerner L E
Department of Chemistry, University of Wisconsin, Madison 53706.
Biopolymers. 1994 Jun;34(6):691-700. doi: 10.1002/bip.360340602.
We have used 25Mg-nmr to investigate the binding of magnesium ions to double-stranded DNA. We have measured line shapes for 25Mg in the presence of monodisperse calf thymus DNA (160 base pairs; b.p.) (magnesium: phosphate = 2.0) at two different field strengths, 11.75 T and 7.05 T, and used the isotropic model of two-site exchange developed by Westlund and Wennerstrom to simultaneously fit the line shapes at both field strengths. This model does not reproduce the observed field dependence. This is in contrast to a previous study [E. Berggren, L. Nordenskiold, and W.H. Braunlin (1992), Biopolymers, Vol. 32, pp. 1339-1350] in which a similar model of isotropic two-site exchange qualitatively reproduced the temperature dependence of the line widths. Relaxation rates were also measured as a function of magnesium: phosphate ratio and coion type. These measurements were used to assess the sensitivity of magnesium relaxation measurements to small changes in DNA structure induced by changes in the solvent environment. The temperature dependence of the line shape varies with the type of coion (chloride or sulfate) present. This coion dependence of the line shape is consistent with the coion dependence of the aggregation midpoint temperature reported by Bloomfield and co-workers [O.A. Knoll, M.G. Fried, and V.A. Bloomfield (1988) in Structure and Expression, Vol. 2, R.H. Sarma and M. H. Sarma, Eds., Adenine Press, New York] and attributed to a lyotropic effect. These results suggest that even at low magnesium: phosphate ratios, relaxation parameters are specific to each magnesium-coion-DNA system.
