| Abstract: | One- and two-dimensional NMR studies at 300 MHz and 500 MHz were carried out on the two oligonucleotides d(C-C-G-A-A-T-T-C-G-G) and d(C-C-G-A-m6A-T-T-C-G-G) in aqueous solution. NMR spectra were observed at 10 mM sample concentration over the temperature range 273-368 K. Assignments are given of the base, H1', H2', H2", H3' and of some H4' resonances, based upon a combination of two-dimensional correlation spectra (COSY) and two-dimensional nuclear Overhauser effect spectra (NOESY); imino-proton resonances were assigned with the aid of a two-dimensional NOE experiment. Chemical shift vs temperature profiles were constructed in order to gain insight into the influence of N6-methylation of residue A(5) on the temperature-dependent conformational behaviour of the decamer and to determine thermodynamic parameters for the duplex-to-coil transition. The NOESY spectra, the imino-proton spectra and the shift profiles of the two compounds, under conditions where each forms a B-DNA-type duplex, are very similar. This is taken to indicate that the influence of N6-methylation of residue A(5) on the local structure of the duplex must be small. However, the temperature dependence of the (non-)exchangeable proton resonances of the two compounds reveals that methylation slows down the duplex-single-strand exchange. Furthermore, a thermodynamic analysis of the two compounds indicates that N6-methylation slightly decreases the stability of the duplex relative to the monomeric forms (Tm is reduced from 332 K down to 325 K at 10 mM sample concentration). Proton-proton couplings were obtained by means of one-dimensional and two-dimensional NMR experiments and were used in a conformational analysis of the sugar ring of each residue of the two compounds in the duplex form. The analysis indicated that all sugar rings display conformational flexibility in the intact duplex: population S-type sugar conformation ranges from 70% to 100%. A more refined analysis of the sugar rings of the parent compound revealed a sequence-dependent variation of the sugar geometry. This variation does not follow well the trend predicted by the Calladine/Dickerson sigma 3-sum rule [Dickerson, R. E. (1983) J. Mol. Biol. 166, 419-441; Calladine, C. R. (1982) J. Mol. Biol. 161, 343-352]; moreover the actual variations appear to be smaller in solution than those expected on the basis of known X-ray structures. |
