Formaldehyde as a probe of DNA structure. I. Reaction with exocyclic amino groups of DNA bases.

TLDR: The reaction product is proven to be a hydroxymethyl group and a reaction mechanism is proposed, and some implications of these results forpolynucleotide studies are discussed.

Abstract: A comprehensive description is given of both the equilibrium and the kinetic aspects of the reaction of formaldehyde with the exocyclic amino groups of derivatives of adenine, cytosine, and guanine; the results extend previous data in the literature to the point where formaldehyde can now be used as a quantitative probe of DNA structure and dynamic behavior. The main results are: (i) the reaction product is proven (by isolation followed by nuclear magnetic resonance (NMR) spectroscopy) to be a hydroxymethyl group; (ii) a dihydroxymethyl adduct is shown to exist at high formaldehyde concentrations; (iii) equilibrium constants at 25 degrees for forming the monoadduct with adenine and cytosine compounds are about 12 (M-1), while those for forming the dihydroxymethyl adduct are about 0.4 (M-1); (iv) the standard enthalpies for forming the monoadducts with adenine and cytosine compounds are about minus 4 to minus 6 kcal/mol; (v) indirect evidence is presented suggesting that a monohydroxymethyl group on adenine or cytosine derivatives exists preferentially as that rotational isomer which blocks Watson-Crick hydrogen bonding; (vi) in derivatives of guanine, it is shown that the N-1 endocyclic imino group can react with formaldehyde, as well as the amino group, the overall equilibrium constant being about 6 (M-1); (vii) all rate constants are reported, as well as their response to temperature, pH, and various solvent additives known to perturb DNA structure; (viii) using a series of substituted anilines, a linear free energy relation is obtained between the logarithm of both the forward and the reverse rate constant for the formaldehyde reaction and the amine pK, over a range of 10-8 change in amie basicity; (ix) using this relation, the pK's for protonating the nucleoside amino groups are estimated to lie in the range of minus 2 to minus 4; (x) a reaction mechanism is proposed; and (xi) some implications of these results forpolynucleotide studies are discussed.