Subscriber access provided by University of Winnipeg Library
B: Biophysics; Physical Chemistry of Biological Systems and Biomolecules
DNA Base Pair Mismatches Induce Structural Changes and Alter the Free Energy Landscape of Base Flip Addie Kingsland, and Lutz Maibaum J. Phys. Chem. B, Just Accepted Manuscript • DOI: 10.1021/acs.jpcb.8b06007 • Publication Date (Web): 29 Nov 2018 Downloaded from http://pubs.acs.org on December 3, 2018
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
The Journal of Physical Chemistry
DNA Base Pair Mismatches Induce Structural Changes and Alter the Free Energy Landscape of Base Flip Addie Kingsland1 and Lutz Maibaum1* 1 *
Department of Chemistry, University of Washington, Seattle, WA 98195, USA To whom correspondence should be addressed: Email:
[email protected] Phone: (206) 221-3931
ACS Paragon Plus Environment
The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 2 of 24
Abstract Double-stranded DNA may contain mismatched base pairs beyond the Watson-Crick pairs guanine-cytosine and adenine-thymine. Such mismatches bear adverse consequences for human health. We utilize molecular dynamics and metadynamics computer simulations to study the structure and dynamics of both matched and mismatched base pairs. We discover significant differences between matched and mismatched pairs in structure and base flip work profiles. Mismatched pairs shift more in the plane normal to the DNA strand and exhibit more non-canonical structures, including the e-motif. We discuss potential implications on mismatch repair enzymes’ detection of DNA mismatches.
Introduction Bases in double-stranded DNA occur in complementary pairs: adenine (A) matches with thymine (T), and cytosine (C) matches with guanine (G). This exact pairing allows for DNA duplication during mitosis and cell growth. DNA replication, however, is susceptible to errors and mutates at a frequency of 1 per 107 base pairs per mitosis cycle; 1 this creates base pair mismatches, such as guanine-thymine (GT). DNA mismatches induce negative consequences on human health, including elevated mutation rates, 2 genetic defects, 3–5 and cancer. 5–8 We study the effects of base pair mismatches on DNA structure at the molecular level by employing molecular dynamics computer simulations. We quantify the differences in structure between DNA strands with matched and mismatched base pairs, focusing on geometric order parameters describing base pair structure, 9 hydrogen-bonding between opposing bases, and flipping DNA bases out of the double helix. For a subset of observables, matched and mismatched base pairs deviate significantly, especially those reporting on in-plane geometries. In 1994, Klimasauskas and coworkers 10 discovered that mismatch repair (MMR) proteins rotate entire DNA bases out of the duplex and into the protein active site. The proteins proofread the DNA, and upon encountering a mismatch, flip the offending nucleotide out of the double helix and into their active sites before proceeding with the repair pathway. 5,7,11–13 MMR proteins decrease the mutation rate 50-1000 times, to a rate of 1 per 109 -1010 base pairs per cell division. 1,14 This rotation is a crucial component of detecting and repairing post-replicative errors and is called base flip. Several experimental works have studied the energetics of base flip, 15–19 but the first reaction coordinate to quantify base flip in computation was defined in 1997 20 which was used to calculate free energy profile of an AT base opening. It was replaced in 2002 by a new coordinate 21 used to calculate the work required to rotate a single base out of the DNA double helix. This latter definition is still in wide use today, 22–25 though many other definitions also exist, including one defined more recently, 26 which has been shown to be more accurate. We employ this final definition to construct detailed free energy profiles of base flip for most every mismatched pair. Left alone, base flip is unlikely to occur on the timescale of typical molecular dynamics simulations, because it requires an activation energy far exceeding thermal fluctuations. We utilize the well-tempered metadynamics algorithm to drive the simulation away from the ground state in order to compile free energy profiles.
ACS Paragon Plus Environment
2
Page 3 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
The Journal of Physical Chemistry
The profiles reveal a metastable state in which both bases flip out of the DNA strand and nest in the minor groove. This may contribute to MMR proteins’ ability to recognize mismatches. Our results do not indicate a systematic trend in free energy difference between the flipped-in and flipped-out states as a quantitative predictor of mismatch recognition, largely due to this unusual minor-groove-nested structure. This arrangement does not predict MMR recognition, highlighting the importance of analyzing additional physical properties such as DNA bending. 13,27–30
Methods Model Building We construct atomistic models of B-DNA containing twelve base pairs by employing the 3DNA software. 31 The sequence is CTGA ACXA ATGT, where the placeholder X represents one of the four nucleotides. Bases 1-6 and 8-12 pair with matched nucleotides, whereas X pairs with either a matched or a mismatched base, denoted Y. We note our considered sequence with ’XY.’ We study a total of eleven systems, four matched (XY ∈ {AT, TA, CG, GC}) and seven with a single mismatch (XY ∈ {AA, AC, AG, CC, CT, GT, TT}). We do not include the GG mismatch because steric influences prevent a duplex equilibrium state. We minimize each structure for 200 steps of steepest descent by applying the CHARMM27 Force Field, 32 which we employed throughout. We solvated the structures with explicit TIP3P water 33 in a dodecahedral simulation box extending at least 1 nm beyond the DNA polymer in all directions. We randomly substituted 22 water molecules for Na+ counter-ions to neutralize the system. We minimized the solvent for 200 steps of steepest descent and equilibrated for 100 ps. We equilibrated each full system for 1 ns.
Molecular Dynamics Simulations We performed simulations at 300 K in the NVT ensemble with stochastic velocity rescaling, 34 periodic boundary conditions, and a 2 fs timestep employing the Gromacs simulation suite version 4.5.5 (unbiased) and 5.1.2 (metadynamics). 35 We utilized the SHAKE algorithm 36 to constrain covalent bonds that include a hydrogen atom, and treated long-range electrostatic interactions with the Particle Mesh Ewald method 37 with a non-bonded cutoff of 1.2 nm. After equilibration, we performed unbiased simulations for 100 ns and saved trajectory coordinates every 10 ps. To identify representative configurations, we applied the Biopython 38 module cluster.kmedoids.
Structure, Hydrogen Bonding, and Base Flip Roll, tilt, twist, slide, rise, shift, buckle, propeller, opening, stagger, shear, stretch, (pictured in Fig. S1) and inter-base hydrogen-bonding were calculated with 3DNA; 31,39 probability distributions and free energy profiles were determined by in-house computer code. We measure bend as proposed by Sharma et al.: 28 we section the DNA into three regions (bases 2-4/21-31, 5-8/17-20, and 9-11/14-16) and calculate the angle between the centers of
ACS Paragon Plus Environment
3
The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 4 of 24
mass of the heavy atoms. To measure base flip, we adopt a pseudo-dihedral angle described by Song et al., 26 defined as the dihedral angle between points P1, P2, P3, and P4 where P1 is the center of mass of the two neighboring base pairs, P2 and P3 are the centers of mass of the phosphates flanking the flipping base, and P4 is the center of mass of the pyrimidine ring, or the five-membered ring of the purine. (Fig. 1) Changes in base flip angle are positive if X crosses into the major groove or if Y crosses into the minor groove, and negative otherwise.
Metadynamics Simulations We employ well-tempered metadynamics simulations with two collective variables: the base flip angle of X and the base flip angle of Y. For comparison, we perform simulations in which we bias only one of two base flip angles as a collective variable. Using the PLUMED plug-in, 40 we add Gaussian biasing potentials to the collective variable(s) with a height of 3 kJ/mol, width of 0.08 radians, a deposition rate of 1 ps, and a well-tempered bias factor of 6, for a total simulation length of at least 700 ns per system. We reconstruct the 2D free energy surface as the inverse of the potential energy added to the system as described in Ref. 41 . We pinned the base pairs terminating the DNA strand and the base pairs neighboring the flipping bases by their inter-pair hydrogen bonds to prevent the system from artificially melting. We constrain each hydrogen bond to a distance of 0.2 nm with a harmonic constant of 40 kJ mol−1 nm−2 . Metadynamics deposits a lot of energy to the system, mangling the DNA. We implemented restraints strong enough to draw the DNA strands together if they separate, but weak enough to allow the pinned bases to separate during the simulation. Each run is thus capable of exploring their entire energy landscape, but also capable of returning to ground-state configurations, a key requirement for metadynamics to function properly. To ensure our simulations are converged and to estimate uncertainties, we require each simulation to satisfy the following conditions: 1. The difference in free energy between flipped-in and flipped-out states remains consistent over time. 2. The value of the reaction coordinate is no longer trapped in a local minimum. 3. The height of the Gaussian peaks added to the simulation energy become very small (
