Abstract
Abstract Analysis of the X-ray fibre diffraction pattern obtained from synthetic poly-2-thiouridylic acid (poly(s 2 U)), a 2-thiosubstituted homologue of polyuridylic acid, reveals that the poly(s 2 U) homopolymer exists in a structure similar to that of A -DNA: two polynucleotide chains are complexed in an antiparallel manner to form a helix with a pitch of 28.8 A and 11 base pairs per turn. The radius of the helix is 0.5 A less than that of A -DNA due to the exclusive occurrence of pyrimidine bases. A molecular model of poly(s 2 U) was derived with unsymmetric N (3) H · · · O (4) and S (2) · · · HN (3) hydrogen bonding between bases, and base pair tilt of 18 ° relative to the helix axis. The peculiar base pairing scheme implies that no symmetry relation exists between the two glycosidic C (1′) N (1) bonds or between the attached ribose-phosphate chains of a base pair. Thus two chains, α and β, are obtained which significantly differ from each other in geometry: the α-chain appears more compressed than an A -DNA polynucleotide chain, the β-chain more extended. Further, the sulphur atoms of the nucleobases within the β-chain are in contact with the glycosidic bonds of the adjacent nucleotides. The bases of the α-chain are stacked such that the base sulphur atoms are located near the N (1) atoms of adjacent bases, giving rise to continuous S · · · N, S · · · N, S · · · N interactions. Similar S · · · N stacking interactions have also been observed in a number of crystal structures of thiopyrimidine nucleosides and seem to indicate a general packing feature. These interactions might be responsible for the increased helix stability of poly(s 2 U) (melting point 68.5 °C) relative to poly(U) (melting point 8 °C under identical conditions) and for the relatively high stability of double helical polynucleotides containing 2-thiopyrimidine bases in contrast to those with 4-thiopyrimidine bases. Further, yeast tRNA 3 Glu contains a 2-thiouridine derivative in the 3′-position of its anticodon and recognizes GAA but not GAG as the codon for glutamic acid. This specificity can be explained when considering the preferred S · · · N stacking interactions which would restrict “wobbling” of the 3′-nucleotide of the anticodon.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have