The mechanism of DNA elongation in nanochannels was explored by Monte Carlo simulations as a function of the channel dimension D, DNA length, and stiffness. Simulations were based on the bead-spring model, representing double-stranded DNA chains of moderate length at a high salt concentration. As a rule, the channel-induced elongation profiles of R( parallel) vs D from the simulations were in qualitative agreement with those from microfluidic measurements of DNA. The longitudinal chain elongation in narrow channels was found to be correctly predicted by the Odijk relation for the deflection regime. The scaling relation of R( parallel) vs D(-1), based on the statistics of ideal-chain blobs, was used to explain the simulation data at the intermediate channel widths. Contrary to the blob-theory presumption, the nonlinear dependence of DNA elongation R( parallel) on the chain length N was observed in simulations at moderate confinement. It was suggested that discrepancies found between the simulations and the blob theory arose from the formation of various DNA hairpin structures within channels.

We compare in detail the channel confinement of flexible and semiflexible chains over a broad range of confinements and chain lengths using molecular simulations. Flexible and semiflexible chains differ over the regimes involved under confinement. For the stiff chain we confirm a transition between strong and weak confinement at the tube diameter approximately equal to the chain persistence length. Deviations from the predicted behavior for confined semiflexible macromolecules in the tubelike channel under the weak confinement are explained by a tendency of chains toward ideal chain behavior. There are several indicators of this ideality based on the shape of chain extension R-confinement D curves, on the scaling of R with the chain length, on the structure factor of coil in the range of individual blobs in the channel, on ambiguity in introduction of excluded volume for confined stiff chains in theoretical treatments as well as on the reported experimental observation.

No abstract