A systematically coarse-grained model for DNA and its predictions for persistence length, stacking, twist, and chirality

Abstract

We introduce a coarse-grained model of DNA with bases modeled as rigid-body ellipsoids to capture their anisotropic stereochemistry. Interaction potentials are all physicochemical and generated from all-atom simulation/parameterization with minimal phenomenology. Persistence length, degree of stacking, and twist are studied by molecular dynamics simulation as functions of temperature, salt concentration, sequence, interaction potential strength, and local position along the chain, for both single-and double-stranded DNA where appropriate. The model of DNA shows several phase transitions and crossover regimes in addition to dehybridization, including unstacking, untwisting, and collapse which affect mechanical properties such as rigidity and persistence length. The model also exhibits chirality with a stable right-handed and metastable left-handed helix.

Moving beyond Watson–Crick models of coarse grained DNA dynamics

“…Of 50 independent simulations conducted with the 10 random dsDNA sequences listed in Table IV, we found that right-handed helices are formed in 62% of the structures. This result is reasonable since our model has no built-in handedness, torsional constraints that favor B-DNA, 7, 13-16, 24 bottom up parameterization from an all atom B-DNA model, [2][3][4][5][6][7] or a method to remove the stacking interactions in left-handed twist. 20 Indeed, other models of this type lead to equilibrated structures that are sometimes left-handed.…”

“…As will be apparent shortly, the force fields for the two models are quite different. The most notable difference is our use of non-spherical bonding potentials, which allows us to avoid the need for dihedral potentials 7,[13][14][15][16]24 that introduce an unphysical bias towards the B-form of dsDNA when the DNA is single-stranded. 13 Rather, we can smoothly move between dsDNA and ssDNA.…”

“…To enforce the directionality of the hydrogen bonding and stacking/cross-stacking interactions, we have modified their spherical potential functions with a smoothly varying prefactor. 7,15,16,21 Since the bonds have directionality, all of the base-base interactions (including excluded volume) are enforced at all times in the simulation.…”

“…Some coarse-grained models are parameterized from the bottom up, [2][3][4][5][6][7] where the functional form and strength of the potentials are tuned to match trajectories from all atom simulations. We chose a top down approach 9,10,[12][13][14][15][16][17][18][19][20][21] for the reasons expounded by Ouldridge et al 21 (Naturally, one can also employ a mixture of top down and bottom up approaches.…”

Molecular determinants of the interactions between proteins and ssDNA

“…The phosphate bead in the model bears a negative charge. The model potential for ssDNA used in our study, which follows other models (30)(31)(32), is given by E ssDNA = E Bond ssDNA + E Angle ssDNA + E Dihedral ssDNA + E Base pairing…”