Triplex intermediates in folding of human telomeric quadruplexes probed by microsecond-scale molecular dynamics simulations

Abstract

We have carried out extended set of μs-scale explicit solvent MD simulations of all possible G-triplexes which can participate in folding pathways of the human telomeric quadruplex. Our study accumulates almost 60 μs of simulation data, which is by about three orders of magnitude larger sampling compared to the earlier simulations of human telomeric G-DNA triplexes. Starting structures were obtained from experimental quadruplex structures by deleting either the first or the last strand. The life-times of antiparallel triplexes with lateral and diagonal loops are at least on μs-scale, which should be sufficient to contribute to the folding pathways. However, the triplex states may involve structures with various local deviations from the ideal triplexes, such as strand tilting and various alternative and incomplete triads. The simulations reveal easy rearrangements between lateral and diagonal loop triplex topologies. Propeller loops of antiparallel triplexes may to certain extent interfere with the G-triplexes but these structures are still viable candidates to participate in the folding. In contrast, all-parallel all-anti triplexes are very unstable and are unlikely to contribute to the folding. Although our simulations demonstrate that antiparallel G-triplexes, if folded, would have life-times sufficient to participate in the quadruplex folding, the results do not rule out the possibility that the G-triplexes are out-competed by other structures not included in our study. Among them, numerous possible misfolded structures containing guanine quartets can act as off-path intermediates with longer life-times than the triplexes. Besides analyzing the structural dynamics of a diverse set of G-DNA triplexes, we also provide a brief discussion of the limitations of the simulation methodology, which is necessary for proper understanding of the simulation data.

Simulative and Experimental Characterization of a pH-Dependent Clamp-like DNA Triple-Helix Nanoswitch

“…At pH 5.0 the triple helix is stable, being characterized by small deviations from the starting structure in line with previous simulative data. 56,57 Conversely, at pH 8.0, deviations larger than 1 nm are present ( Figure 3A), indicating that the triple helix loses its conformation when the N3 atoms are deprotonated. The RMSD of only the two strands forming the double helix (i.e., the DNA target and the 3′ region of the clamp-switch) gives a similar result.…”

Dominant Driving Forces in Human Telomere Quadruplex Binding-Induced Structural Alterations

“…Detailed computational and experimental (thermodynamic, kinetic) investigation of folding and ligand binding pathways of quadruplex DNA has begun very recently (17)(18)(19)(20)(21)(22). These studies leave no doubt that in solution, the Tel22 folding/unfolding process may consist of several structural steps, meaning that such solutions should be considered as equilibrium mixtures of Tel22 molecules in their folded (Q), unfolded (U), and intermediate (I) states.…”

A pilgrim's guide to G-quadruplex nucleic acid folding

“…In a previous work, Stadlbauer et al investigated strand-slipped structures as latestage folding intermediates [4]. In a feature paper of the present issue [5], the authors report the longest to date unbiased sampling MD simulations of G-triplexes derived from the human telomeric sequence, with three tracts of three guanines each.…”

“…North of the map: some possible stopovers explored by unbiased molecular dynamics: slipped structures were found plausible points of interest for the fully parallel, all-anti structure A whereas they were found impossible for hybrid structures due to steric clash between bases (see for example strand slippage from structure C) [4]. In contrast, triplex structures were found plausible for structures containing both syn and anti bases (see for example the triplex of C), but unstable for the parallel, all-anti structure derived from structure A [5]. The map of the Way of St James and the shell-shaped logo were reproduced from Wikimedia Commons [12].…”

“…And in current state of the art and hardware, without thoughtful speculations simulations would wander about the landscape with no sense of direction. Mutual understanding between specialists of both communities is therefore crucial, and gaining understanding in the advantages and limitations of state-of-the-art MD simulations is another very good reason for experimentalists to read the present feature article from Stadlbauer et al [5].…”