Current interest in lone‐pair⋅⋅⋅π (lp⋅⋅⋅π) interactions is gaining momentum in biochemistry and (supramolecular) chemistry. However, the physicochemical origin of the exceptionally short (ca. 2.8 Å) oxygen‐to‐nucleobase plane distances observed in prototypical Z‐DNA CpG steps remains unclear. High‐level quantum mechanical calculations, including SAPT2+3 interaction energy decompositions, demonstrate that lp⋅⋅⋅π contacts do not result from n→π* orbital overlaps but from weak dispersion and electrostatic interactions combined with stereochemical effects imposed by the locally strained structural context. They also suggest that the carbon van der Waals (vdW) radii, originally derived for sp3 carbons, should not be used for smaller sp2 carbons attached to electron‐withdrawing groups. Using a more adapted carbon vdW radius results in these lp⋅⋅⋅π contacts being no longer of the sub‐vdW type. These findings challenge the whole lp⋅⋅⋅π concept that refers to elusive orbital interactions that fail to explain short interatomic contact distances.
Synthesis of RNA nucleobases from formamide is one of the recurring topics of prebiotic chemistry research. Earlier reports suggest that thymine, the substitute for uracil in DNA, may also be synthesized from formamide in the presence of catalysts enabling conversion of formamide to formaldehyde. In the current paper, we show that to a lesser extent conversion of uracil to thymine may occur even in the absence of catalysts. This is enabled by the presence of formic acid in the reaction mixture that forms as the hydrolysis product of formamide. Under the reaction conditions of our study, the disproportionation of formic acid may produce formaldehyde that hydroxymethylates uracil in the first step of the conversion process. The experiments are supplemented by quantum chemical modeling of the reaction pathway, supporting the plausibility of the mechanism suggested by Saladino and coworkers.
5 November 2021 in Journal of Chemical Information and Modeling
Phosphate…π, also called anion…π, contacts occur between nucleobases and phosphate OP oxygens in r(GNRA) and r(UNNN) U-turn motifs (N = A,G,C,U; R = A,G). We investigated these contacts in detail by using state-of-the-art quantum chemical methods (QM) to characterize some of their physico-chemical properties and to evaluate the ability of the AMBER force field (AFF) to describe these contacts. We found that AFF interaction energies of phosphate…π contacts calculated for model dimethyl phosphate…nucleobase systems are less stabilizing in comparison with double-hybrid DFT methods and that the minimum contact distances are stretched for all nucleobase systems. This distance stretch is also observed in large-scale AFF computations on several r(gcGNRAgc) tetraloop hairpins when compared to QM/MM. Further, classical molecular dynamics (MD) simulations of these tetraloop hairpins confirm this distance stretch and reveal shifted OP2/nucleobase positions when compared to experimental data extracted from high-resolution X ray/cryo EM structures (≤ 2.5 Å) of r(GNRA) tetraloops using the WebFR3D bioinformatic tool. We propose that discrepancies between QM and AFF are caused by a combination of missing polarization, too large AFF Lennard-Jones (LJ) radii of nucleobase carbon atoms and exaggerated short-range repulsion due to an approximate r−12 LJ repulsive term. We put these results in regard with those obtained in earlier investigations on lone pair…π contacts occurring in CpG Z-steps. Charge-transfer calculations do not support any significant n->π* donation effects and hence this label is inappropriate. We also investigated thiophosphate…π contacts for which we calculated less stabilizing interaction energies than for the phosphate…π contacts. We thus challenge suggestions that the experimentally observed enhanced thermodynamic stability of phosphorothioated r(GNRA) tetraloops can be straightforwardly explained by larger London dispersion.