Search

The intrinsic curvature of seven 98 bp DNA molecules containing up to four centrally located A6-tracts has been measured by gel and capillary electrophoresis as a function of the number and arrangement of the A-tracts. At low cation concentrations, the electrophoretic mobility observed in polyacrylamide gels and in free solution decreases progressively with the increasing number of phased A-tracts, as expected for DNA molecules with increasingly curved backbone structures. Anomalously slow electrophoretic mobilities are also observed for DNA molecules containing two pairs of phased A-tracts that are out of phase with each other, suggesting that out-of-phase distortions of the helix backbone do not cancel each other out. The mobility decreases observed for the A-tract samples are due to curvature, not cation binding in the A-tract minor groove, because identical free solution mobilities are observed for a molecule with four out-of-phase A-tracts and one with no A-tracts. Surprisingly, the curvature of DNA A-tracts is gradually lost when the monovalent cation concentration is increased to ∼200 mM, regardless of whether the cation is a hydrophilic ion like Na+, NH4+, or Tris+ or a hydrophobic ion like tetrabutylammonium. The decrease in A-tract curvature with increasing ionic strength, along with the known decrease in A-tract curvature with increasing temperature, suggests that DNA A-tracts are not significantly curved under physiological conditions.

The effect of monovalent cations on the thermal stability of a small model DNA hairpin has been measured by capillary electrophoresis, using an oligomer with 16 thymine residues as an unstructured control. The melting temperature of the model hairpin increases approximately linearly with the logarithm of increasing cation concentration in solutions containing Na(+), K(+), Li(+), NH(4)(+), Tris(+), tetramethylammonium (TMA(+)), or tetraethylammonium (TEA(+)) ions, is approximately independent of cation concentration in solutions containing tetrapropylammonium (TPA(+)) ions, and decreases with the logarithm of increasing cation concentration in solutions containing tetrabutylammonium (TBA(+)) ions. At constant cation concentration, the melting temperature of the DNA model hairpin decreases in the order Li(+) ∼ Na(+) ∼ K(+) > NH(4)(+) > TMA(+) > Tris(+) > TEA(+) > TPA(+) > TBA(+). Isothermal studies indicate that the decrease in the hairpin melting temperature with increasing cation hydrophobicity is not due to saturable, site-specific binding of the cation to the random coil conformation, but to the concomitant increase in cation size with increasing hydrophobicity. Larger cations are less effective at shielding the charged phosphate residues in B-form DNA because they cannot approach the DNA backbone as closely as smaller cations. By contrast, larger cations are relatively more effective at shielding the phosphate charges in the random coil conformation, where the phosphate-phosphate distance more closely matches cation size. Hydrophobic interactions between alkylammonium ions interacting electrostatically with the phosphate residues in the coil may amplify the effect of cation size on DNA thermal stability.

The binding of five different monovalent cations to DNA oligomers containing A-tracts, runs of four or more contiguous adenine residues, has been assessed by capillary electrophoresis, using the Replacement Ion method. In this method, a nonbinding cation in the background electrolyte is gradually replaced by a binding cation, keeping the ionic strength of the solution constant. Monovalent cation binding reduces the effective charge of an A-tract-containing oligomer, decreasing its free solution mobility. The cations bind in the A-tract minor groove, because the binding site can be blocked by the minor groove binding drug netropsin. Li(+), NH(4)(+), and Tris(+) ions bind to A-tracts with similar affinities; the binding of Na(+) ions is weaker, and K(+) ion binding is highly variable. Each A-tract appears to bind one monovalent cation upon saturation of the binding site(s). For a given cation, the apparent dissociation constants depend on A-tract sequence and orientation, but not on the phasing of the A-tracts with respect to the helix repeat. Differences in the cooperativity of binding of the various cations to A-tracts with different sequences suggest that monovalent cation binding may be coupled with a conformational transition leading to the formation of the characteristic narrow minor groove A-tract structure.

The free solution mobilities of the adenosine nucleotides 5'-adenosine triphosphate (ATP), 5'-adenosine diphosphate (ADP), 5'-adenosine monophosphate (AMP), and 3'-5'-cyclic AMP (cAMP) have been measured in diethylmalonate buffers containing a wide variety of monovalent cations. The mobilities of all nucleotides increase gradually with the increase in intrinsic conductivity of the cation in the BGE. However, at a given conductivity, the mobilities observed for ATP, ADP, and AMP in BGEs containing alkali metal ions and other cations are lower than these observed in BGEs containing tetraalkylammonium ions. Since the mobility of cAMP is independent of the cation in the BGE, the results suggest that the relatively low mobilities observed for ATP, ADP, and AMP in BGEs containing cations other than a tetraalkylammonium ion are due to cation binding, reducing the effective net charge of the nucleotide and thereby reducing the observed mobility. To measure the binding quantitatively, the mobilities of the nucleotides were measured as a function of ionic strength. The mobilities of ATP, ADP, and AMP decrease nonlinearly with the square root of ionic strength (I(1/2)) in BGEs containing an alkali metal ion or Tris(+). By contrast, the mobilities decrease linearly with I(1/2) in BGEs containing a nonbinding quaternary ammonium ion, as expected from Debye-Hückel-Onsager (DHO) theory. The mobility of cAMP, a nonbinding analyte, decreases linearly with I(1/2), regardless of the cation in the BGE. Hence, a nonlinear decrease of the mobility of an analyte with I(1/2) appears to be a hallmark of counterion binding. The curved mobility profiles observed for ATP, ADP, and AMP in BGEs containing an alkali metal ion or Tris(+) were analyzed by nonlinear curve fitting, using difference mobility profiles to correct for the effect of the physical properties of BGE on the observed mobilities. The calculated apparent dissociation constants range from 22 to 344 mM, depending on the particular cation-nucleotide pair. Similar values have been obtained by other investigators, using different methods. Interestingly, Tris(+) and Li(+) bind to the adenosine nucleotides with approximately equal affinities, suggesting that positively charged Tris(+) buffer ions can compete with alkali metal ions in Tris-buffered solutions.

A variation of affinity capillary electrophoresis, called the replacement ion (RI) method, has been developed to measure the binding of monovalent cations to random sequence, double-stranded (ds) DNA. In this method, the ionic strength is kept constant by gradually replacing a non-binding ion in the solution with a binding ion and measuring the mobility of binding and non-binding analytes as a function of binding ion concentration. The method was validated by measuring the binding of Li+ ions to adenosine nucleotides; the apparent dissociation constants obtained by the RI method are comparable to literature values obtained by other methods. The binding of Tris+, NH4+, Li+, Na+, and K+ to dsDNA was then investigated. The apparent dissociation constants observed for counterion binding to a random-sequence 26-base pair (bp) oligomer ranged from 71 mM for Tris+ to 173 mM for Na+ and K+. Hence, positively charged Tris buffer ions will compete with other monovalent cations in Tris-buffered solutions. The bound cations identified in this study may correspond to the strongly correlated, tightly bound ions recently postulated to exist as a class of ions near the surface of dsDNA (Tan, Z.-J., and Chen, S.-J. (2006) Biophys. J. 91, 518-536). Monovalent cation binding to random-sequence dsDNA would be expected to occur in addition to any site-specific binding of cations to A-tracts or other DNA sequence motifs. Single-stranded DNA oligomers do not bind the five tested cations under the conditions investigated here.

The free solution mobilities of single- and double-stranded DNA molecules of various molecular weights have been measured by capillary electrophoresis in solutions of constant ionic strength containing a common anion and fifteen different monovalent cations. In solutions with the same ionic composition, the mobilities of different DNA molecules can vary by up to 20%, depending on molecular weight, the number of strands, and the presence or absence of A-tracts, runs of four or more contiguous adenine residues. Importantly, the mobilities observed for the same DNA sample can vary by up to 40% in solutions containing different cations. The mobility differences observed for the same DNA in solutions containing different cations cannot be rationalized by differences in the anhydrous radii or intrinsic conductivities of the various cations, or by the sequence-dependent binding of certain cations to A-tracts. Instead, the observed mobilities are linearly correlated with the average number of water-water hydrogen bonds that are present in solutions containing different cations. The mobilities are also correlated with the viscosity B coefficients of the various cations and with the rotational correlation times frictional coefficients observed for water molecules in solutions containing different cations. Hence, monovalent cations modify the free solution mobility of DNA primarily by perturbing the hydrogen-bonded structure of water, affecting the friction experienced by the migrating DNA molecules during electrophoresis., (Copyright 2005 Wiley Periodicals, Inc.)