A critical evaluation of models for complex molecular dynamics: Application to NMR studies of double‐ and single‐stranded DNA

Abstract

SynopsisThe nature of internal and overall motions in native (double-stranded) and denatured (single-stranded) DNA fragments 120-160 base pairs (bp) long is examined by moleculardynamics modeling using 13C-nmr spin-relaxation data obtained over the frequency range of 37-125 MHz. The broad range of I3C frequencies is required to differentiate among various models. Relatively narrow linewidths, large nuclear Overhauser enhancements (NOES), and short T1 values all vary significantly with frequency and indicate the presence of rapid, restricted internal motions on the nanosecond time scale. For double-stranded DNA monomer fragments (147 bp, 24 8, diam a t 32"C), the overall motion is that of an axially symmetric cylinder ( r x = -lop6 s; 7, = -1.8 X s), which is in good agreement with values calculated from hydrodynamic theory ( T~ = -1.8 X s). The DNA internal motion can be modeled as restricted amplitude internal diffusion of individual C-H vectors of deoxyribose methine carbons Cl', C3', and C4', either with conic boundary conditions (7, = -4 X 10-9 s, Ocone = -21O) or as a bistable jump ( T A = T B = -2 X s, O = -15"). We discuss the critical role in molecular-dynamics modeling played by the angle (p) that individual C-H vectors make with the long axis of the DNA helix. Eeat denaturation brings about increases in both the rute and amplitude of the internal motion (described by the wobble model with rW = -0.2 X s, O,,,, = -50' 1, and overall motion is affected by becoming essentially isotropic ( T~ = T~ = -5 X s) for the single-stranded molecules. Since '3C-nmr data obtained at various DNA concentrations for C2' of the deoxyribose ring is not described weli by the above models, a new model incorporating an additional internal motion is proposed to take into account the rapid, extensive, and weakly coupled motion of C2'. s; T~ = -2.7 X

A computational approach to the mechanism of self-cleavage of hammerhead RNA.

“…In E the attacking 2'-oxygen is in line with the leaving 5'-oxygen. The energy barrier for the 3'-to 2'-endo change is small (17) and the equilibrium is rapid in solution [occurs in the nanosecond time scale for DNA (18,19)]. Consistently, the total energies of the two computer-generated hammerhead structures are comparable, with the 2'-endo structure favored by about 10 kcal/mol.…”

The amplitude of local angular motions of intercalated dyes and bases in DNA

“…Even these larger values of (8c2)l12 are much smaller than those estimated (incorrectly, in our view) for local polar (out of plane) motions of base pairs in DNA, which fall in the range of (8c2)l12 = 35°-500. [21][22][23][24][25][26][27][28][29][30][31] There is, of course, no reason a priori to expect intercalated ethidium to exhibit the same amplitudes of local internal motion as the base pairs. Indeed, it has been claimed that ethidium and its nearest-neighbor base pairs exhibit much smaller amplitudes of angular motion than the "normal base pairs.…”

31P-NMR analysis of the B to Z transition in double-stranded (dC-dG)3, and (dC-dG)4in high salt solution

“…This is in agreement with the results or temperature measurements of the circular dichroism of (dC-dG)3 in 2.8M NaCl (29), and is in contrast to the trend for poly(dG-dC) nucleotides, for which elevated temperature shifts the B-Z equilibrium toward the Z-form (17 We are currently expanding our relaxation measurements in an attempt to separate various relaxation mechanism contributions (CSA and DD) and to find optimal models to fit the relaxation data. One should be aware, however, of considerable complications in the analysis and the interpretation or 31P relaxation data of nucleic acids in solution (31)(32)(33)(34), especially where different mechanisms contribute to the relaxation (34). However, it appears that regardless of the details of the model for motions of DNA [which may be best probed by multi-field 13C measurements (31)], NMR relaxation experiments have produced a relatively homogenous picture ot the magnitudes of significant motional frequencies, if not a consistent picture of the forms of these motions.…”