Long Time-Scale Atomistic Simulations of the Structure and Dynamics of Transcription Factor-DNA Recognition

Rugged fitness landscapes minimize promiscuity in the evolution of transcriptional repressors

How a protein's function influences the shape of its fitness landscape, smooth or rugged, is a fundamental question in evolutionary biochemistry. Smooth landscapes arise when incremental mutational steps lead to a progressive change in function, as commonly seen in enzymes and binding proteins. On the other hand, rugged landscapes are poorly understood because of the inherent unpredictability of how sequence changes affect function. Here, we experimentally characterize the entire sequence phylogeny, comprising 1,158 extant and ancestral sequences, of the DNA-binding domain (DBD) of the LacI/GalR transcriptional repressor family. Our analysis revealed an extremely rugged landscape with rapid switching of specificity, even between adjacent nodes. Further, the ruggedness arises due to the necessity of the repressor to simultaneously evolve specificity for asymmetric operators and disfavors potentially adverse regulatory crosstalk. Our study provides fundamental insight into evolutionary, molecular, and biophysical rules of genetic regulation through the lens of fitness landscapes.

Brewing COFFEE: A Sequence-Specific Coarse-Grained Energy Function for Simulations of DNA-Protein Complexes

DNA-protein interactions are pervasive in a number of biophysical processes ranging from transcription and gene expression to chromosome folding. To describe the structural and dynamic properties underlying these processes accurately, it is important to create transferable computational models. Toward this end, we introduce Coarse-grained Force Field for Energy Estimation, COFFEE, a robust framework for simulating DNA-protein complexes. To brew COFFEE, we integrated the energy function in the self-organized polymer model with side-chains for proteins and the three interaction site model for DNA in a modular fashion, without recalibrating any of the parameters in the original force-fields. A unique feature of COFFEE is that it describes sequence-specific DNA-protein interactions using a statistical potential (SP) derived from a data set of high-resolution crystal structures. The only parameter in COFFEE is the strength (λDNAPRO) of the DNA-protein contact potential. For an optimal choice of λDNAPRO, the crystallographic B-factors for DNA-protein complexes with varying sizes and topologies are quantitatively reproduced. Without any further readjustments to the force-field parameters, COFFEE predicts scattering profiles that are in quantitative agreement with small-angle X-ray scattering experiments, as well as chemical shifts that are consistent with NMR. We also show that COFFEE accurately describes the salt-induced unraveling of nucleosomes. Strikingly, our nucleosome simulations explain the destabilization effect of ARG to LYS mutations, which do not alter the balance of electrostatic interactions but affect chemical interactions in subtle ways. The range of applications attests to the transferability of COFFEE, and we anticipate that it would be a promising framework for simulating DNA-protein complexes at the molecular length-scale.

Simulated pressure changes in LacI suggest a link between hydration and functional conformational changes

The functions of many proteins are associated with interconversions among conformational substates. However, these substates can be difficult to measure experimentally, and determining contributions from hydration changes can be especially difficult. Here, we assessed the use of pressure perturbations to sample the substates accessible to the Escherichia coli lactose repressor protein (LacI) in various liganded forms. In the presence of DNA, the regulatory domain of LacI adopts an Open conformation that, in the absence of DNA, changes to a Closed conformation. Increasing the simulation pressure prevented the transition from an Open to a Closed conformation, in a similar manner to the binding of DNA and anti-inducer, ONPF. The results suggest the hydration of specific residues play a significant role in determining the population of different LacI substates and that simulating pressure perturbation could be useful for assessing the role of hydration changes that accompany functionally-relevant amino acid substitutions.

Nonspecific vs. specific DNA binding free energetics of a transcription factor domain protein

Transcription factor (TF) proteins regulate gene expression by binding to specific sites on the genome. In the facilitated diffusion model, an optimized search process is achieved by the TF alternating between 3D diffusion in the bulk and 1D diffusion along DNA. While undergoing 1D diffusion, the protein can switch from a search mode for fast diffusion along nonspecific DNA to a recognition mode for stable binding to specific DNA. It was recently noticed that, for a small TF domain protein, reorientations on DNA happen between the nonspecific and specific DNA binding. We here conducted all-atom molecular dynamics simulations with steering forces to reveal the protein-DNA binding free energetics, confirming that the search and recognition modes are distinguished primarily by protein orientations on the DNA. As the binding free energy difference between the specific and nonspecific DNA system slightly deviates from that being estimated directly from dissociation constants on 15-bp DNA constructs, we hypothesize that the discrepancy can come from DNA sequences flanking the 6-bp central binding sites that impact on the dissociation kinetics measurements. The hypothesis is supported by a simplified spherical protein-DNA model along with stochastic simulations and kinetic modeling.

Brewing COFFEE: A sequence-specific coarse-grained energy function for simulations of DNA-protein complexes

DNA-protein interactions are pervasive in a number of biophysical processes ranging from transcription, gene expression, to chromosome folding. To describe the structural and dynamic properties underlying these processes accurately, it is important to create transferable computational models. Toward this end, we introduce Coarse grained force field for energy estimation, COFFEE, a robust framework for simulating DNA-protein complexes. To brew COFFEE, we integrated the energy function in the Self-Organized Polymer model with Side Chains for proteins and the Three Interaction Site model for DNA in a modular fashion, without re-calibrating any of the parameters in the original force-fields. A unique feature of COFFEE is that it describes sequence-specific DNA-protein interactions using a statistical potential (SP) derived from a dataset of high-resolution crystal structures. The only parameter in COFFEE is the strength (λ D N A P R O ) of the DNA-protein contact potential. For an optimal choice of λ D N A P R O , the crystallographic B-factors for DNA-protein complexes, with varying sizes and topologies, are quantitatively reproduced. Without any further readjustments to the force-field parameters, COFFEE predicts the scattering profiles that are in quantitative agreement with SAXS experiments as well as chemical shifts that are consistent with NMR. We also show that COFFEE accurately describes the salt-induced unraveling of nucleosomes. Strikingly, our nucleosome simulations explain the destabilization effect of ARG to LYS mutations, which does not alter the balance of electrostatic interactions, but affects chemical interactions in subtle ways. The range of applications attests to the transferability of COFFEE, and we anticipate that it would be a promising framework for simulating DNA-protein complexes at the molecular length-scale.

Can molecular dynamics be used to simulate biomolecular recognition?

There are many problems in biochemistry that are difficult to study experimentally. Simulation methods are appealing due to direct availability of atomic coordinates as a function of time. However, direct molecular simulations are challenged by the size of systems and the time scales needed to describe relevant motions. In theory, enhanced sampling algorithms can help to overcome some of the limitations of molecular simulations. Here, we discuss a problem in biochemistry that offers a significant challenge for enhanced sampling methods and that could, therefore, serve as a benchmark for comparing approaches that use machine learning to find suitable collective variables. In particular, we study the transitions LacI undergoes upon moving between being non-specifically and specifically bound to DNA. Many degrees of freedom change during this transition and that the transition does not occur reversibly in simulations if only a subset of these degrees of freedom are biased. We also explain why this problem is so important to biologists and the transformative impact that a simulation of it would have on the understanding of DNA regulation.

C-B-A Test of DNA Force Fields

The DNA duplex may be locally strongly bent in complexes with proteins, for example, with polymerases or in a nucleosome. At such bends, the DNA helix is locally in the noncanonical forms A (with a narrow major groove and a large amount of north sugars) or C (with a narrow minor groove and a large share of BII phosphates). To model the formation of such complexes by molecular dynamics methods, the force field is required to reproduce these conformational transitions for a naked DNA. We analyzed the available experimental data on the B-C and B-A transitions under the conditions easily implemented in modeling: in an aqueous NaCl solution. We selected six DNA duplexes which conformations at different salt concentrations are known reliably enough. At low salt concentrations, poly(GC) and poly(A) are in the B-form, classical and slightly shifted to the A-form, respectively. The duplexes ATAT and GGTATACC have a strong and salt concentration dependent bias toward the A-form. The polymers poly(AC) and poly(G) take the C- and A-forms, respectively, at high salt concentrations. The reproduction of the behavior of these oligomers can serve as a test for the balance of interactions between the base stacking and the conformational flexibility of the sugar-phosphate backbone in a DNA force field. We tested the AMBER bsc1 and CHARMM36 force fields and their hybrids, and we failed to reproduce the experiment. In all the force fields, the salt concentration dependence is very weak. The known B-philicity of the AMBER force field proved to result from the B-philicity of its excessively strong base stacking. In the CHARMM force field, the B-form is a result of a fragile balance between the A-philic base stacking (especially for G:C pairs) and the C-philic backbone. Finally, we analyzed some recent simulations of the LacI-, SOX-4-, and Sac7d-DNA complex formation in the framework of the AMBER force field.

The development of nucleic acids force fields: From an unchallenged past to a competitive future

Molecular dynamics simulations have strongly matured as a method to study biomolecular processes. Their validity, however, is determined by the accuracy of the underlying force fields that describe the forces between all atoms. In this article, we review the development of nucleic acids force fields. We describe the early attempts in the 1990s and emphasize their strong influence on recent force fields. State-of-the-art force fields still use the same Lennard-Jones parameters derived 25 years ago in spite of the fact that these parameters were in general not fitted for nucleic acids. In addition, electrostatic parameters also are deprecated, which may explain some of the current force field deficiencies. We compare different force fields for various systems and discuss new tests of the recently developed Tumuc1 force field. The OL-force fields and Tumuc1 are arguably the best force fields to describe the DNA double helix. However, no force field is flawless. In particular, the description of sugar-puckering remains a problem for nucleic acids force fields. Future refinements are required, so we review methods for force field refinement and give an outlook to the future of force fields.

Conformational Change of Transcription Factors from Search to Specific Binding: A lac Repressor Case Study

In a process known as facilitated diffusion, DNA-binding proteins find their target sites by combining three-dimensional diffusion and one-dimensional scanning of the DNA. Following the trade-off between speed and stability, agile exploration of DNA requires loose binding, whereas, at the DNA target site, the searching protein needs to establish tight interactions with the DNA. To enable both efficient search and stable binding, DNA-binding proteins and DNA often switch conformations upon recognition. Here, we study the one-dimensional diffusion and DNA binding of the dimeric lac repressor (LacI), which was reported to adopt two different conformations when binding different conformations of DNA. Using coarse-grained molecular dynamic simulations, we studied the diffusion and the sequence-specific binding of these conformations of LacI, as well as their truncated or monomeric variants, with two DNA conformations: straight and bent. The simulations were compared to experimental observables. This study supports that linear diffusion along DNA combines tight rotation-coupled groove tracking and rotation-decoupled hopping, where the protein briefly dissociates and reassociates just a few base pairs away. Tight groove tracking is crucial for target-site recognition, while hopping speeds up the overall search process. We investigated the diffusion of different LacI conformations on DNA and show how the flexibility of LacI's hinge regions ensures agility on DNA as well as faithful groove tracking. If the hinge regions instead form α-helices at the protein-DNA interface, tight groove tracking is not possible. On the contrary, the helical hinge region is essential for tight binding to bent, specific DNA, for the formation of the specific complex. Based on our study of different encounter complexes, we argue that the conformational change in LacI and DNA bending are somewhat coupled. Our findings underline the importance of two distinct protein conformations for facilitated diffusion and specific binding, respectively.

Protein-Nucleic Acid Interactions for RNA Polymerase II Elongation Factors by Molecular Dynamics Simulations

RNA polymerase II (Pol II) forms a complex with elongation factors to proceed to the elongation stage of the transcription process. In this work, we studied the elongation factor SPT5 and explored the protein-nucleic acid interactions for the isolated systems of KOW1 and KOW4 domains of SPT5 with DNA and RNA, respectively. We performed molecular dynamics (MD) simulations using three commonly used force fields that are CHARMM c36m, AMBER ff14sb, and ff19sb. Simulations showed strong protein-nucleic acid interactions and low electrostatic binding free energies for all force fields used. RNA was found to be highly dynamic with all force fields, while DNA had relatively more stable conformations with the AMBER force fields compared to that with CHARMM. Furthermore, we performed MD simulations of the complete elongation complex using CHARMM c36m and AMBER ff19sb force fields to compare the dynamics and interactions with the isolated systems. Similarly, strong KOW1 and DNA interactions were observed in the complete elongation complex simulations and DNA was further stabilized by a network of interactions involving SPT5-KOW1, SPT4, and rpb2 of Pol II. Overall, our study showed that the differences between CHARMM and AMBER force fields strongly affect the dynamics of the nucleic acids. CHARMM provides highly flexible DNA, while AMBER largely stabilizes the DNA structure. Although the presence of the entire interaction network stabilized the DNA and decreased the differences in the results from the two force fields, the discrepancies of the force fields for smaller systems may reflect their problems in generating accurate dynamics of nucleic acids.

DNA-MBF1 study using molecular dynamics simulations : On the road to understanding the heat stress response in DNA-protein interactions in plants

Regulatory factor MBF1 is highly conserved between species and has been described as a cofactor and transcription factor. In plants, several reports associate MBF1 with heat stress response. Nevertheless, the specific physical processes involved in the MBF1-DNA interaction are still far from clearly understood. We thus performed extensive molecular dynamics simulations of DNA with a homology-based modethel of the MBF1 protein. Based on recent experimental data, we proposed two B-DNA sequences, analyzing their interaction with our model of the Arabidopsis MBF1c protein (AtMBF1c) at three different temperatures: 293, 300, and 320 K, maintaining a constant pressure of 1 bar. The simulations suggest that MBF1 binds directly to the DNA, supporting the idea of its role as a transcription factor. We identified two different conformations of the MBF1 protein when bound, and characterized the specific groups of amino acids involved in the formation of the DNA-MBF1 complex. These regions of amino acids are bound mostly to the minor groove of DNA by the attraction of positively charged residues and the negatively charged backbone, but subject to the compatibility of shapes, much in the sense of a lock-and-key mechanism. We found that only with a sequence rich in CTAGA motifs at 300 K does MBF1 bind to DNA in the DNA-binding domain Cro/C1-type HTH predicted. In the rest of the systems tested, we observed non-specific DNA-MBF1 interactions. This study complements findings previously reported by others on the role of CTAGA as a DNA-binding element for MBF1c at a heat stress temperature.

The N-terminal Helix-Turn-Helix Motif of Transcription Factors MarA and Rob Drives DNA Recognition

DNA-binding proteins play an important role in gene regulation and cellular function. The transcription factors MarA and Rob are two homologous members of the AraC/XylS family that regulate multidrug resistance. They share a common DNA-binding domain, and Rob possesses an additional C-terminal domain that permits binding of low-molecular weight effectors. Both proteins possess two helix-turn-helix (HTH) motifs capable of binding DNA; however, while MarA interacts with its promoter through both HTH-motifs, prior studies indicate that Rob binding to DNA via a single HTH-motif is sufficient for tight binding. In the present work, we perform microsecond time scale all-atom simulations of the binding of both transcription factors to different DNA sequences to understand the determinants of DNA recognition and binding. Our simulations characterize sequence-dependent changes in dynamical behavior upon DNA binding, showcasing the role of Arg40 of the N-terminal HTH-motif in allowing for specific tight binding. Finally, our simulations demonstrate that an acidic C-terminal loop of Rob can control the DNA binding mode, facilitating interconversion between the distinct DNA binding modes observed in MarA and Rob. In doing so, we provide detailed molecular insight into DNA binding and recognition by these proteins, which in turn is an important step toward the efficient design of antivirulence agents that target these proteins.

Kinetics of the Conformational Transformation between B- and A-Forms in the Drew-Dickerson Dodecamer

Some DNA sequences in crystals and in complexes with proteins can exist in the forms intermediate between the B- and A-DNA. Based on this, it was implied that the B-to-A transition for any DNA molecule should go through these intermediate forms also in kinetics. More precisely, the helix parameter Slide has to change first, and the molecule should take the E-form. After that, the Roll parameter changes. In the present work, we simulated the kinetics of the B-A transition in the Drew-Dickerson dodecamer, a known B-philic DNA oligomer. We used the "sugar" coarse-grained model that reproduces ribose flexibility, preserves sequence specificity, employs implicit water and explicit ions, and offers the possibility to vary friction. As the control parameter of the transition, we chose the volume available for a counterion and considered the change from a large to a small volume. In the described system, the B-to-A conformational transformation proved to correspond to a first-order phase transition. The molecule behaves like a small cluster in the region of such a transition, jumping between the A- and B-forms in a wide range of available volumes. The viscosity of the solvent does not affect the midpoint of the transition but only the overall mobility of the system. All helix parameters change synchronously on average, we have not observed the sequence "Slide first, Roll later" in kinetics, and the E-DNA is not a necessary step for the transition between the B- and A-forms in the studied system. So, the existence of the intermediate DNA forms requires specific conditions, shifting the common balance of interactions: certain nucleotide sequence in specific solution or/and the interaction with some protein.

Inchworm stepping of Myc-Max heterodimer protein diffusion along DNA

Oncogenic protein Myc serves as a transcription factor to control cell metabolisms. Myc dimerizes via leucine zipper with its associated partner protein Max to form a heterodimer structure, which then binds target DNA sequences to regulate gene transcription. The regulation depends on Myc-Max binding to DNA and searching for target sequences via diffusional motions along DNA. Here, we conduct structure-based molecular dynamics (MD) simulations to investigate the diffusion dynamics of the Myc-Max heterodimer along DNA. We found that the heterodimer protein slides on the DNA in a rotation-uncoupled manner in coarse-grained simulations, as its two helical DNA binding basic regions (BRs) alternate between open and closed conformations via inchworm stepping motions. In such motions, the two BRs of the heterodimer step across the DNA strand one by one, with step sizes reaching about half of a DNA helical pitch length. Atomic MD simulations of the Myc-Max heterodimer in complex with DNA have also been conducted. Hydrogen bond interactions are revealed between the two BRs and two complementary DNA strands, respectively. In the non-specific DNA binding, the BR from Myc shows an onset of stepping on one association DNA strand and starts detaching from the other strand. Overall, our simulation studies suggest that the inchworm stepping motions of the Myc-Max heterodimer can be achieved during the protein diffusion along DNA.

Molecular dynamics simulations of DNA-DNA and DNA-protein interactions

The all-atom molecular dynamics method can characterize the molecular-level interactions in DNA and DNA-protein systems with unprecedented resolution. Recent advances in computational technologies have allowed the method to reveal the unbiased behavior of such systems at the microseconds time scale, whereas enhanced sampling approaches have matured enough to characterize the interaction free energy with quantitative precision. Here, we describe recent progress toward increasing the realism of such simulations by refining the accuracy of the molecular dynamics force field, and we highlight recent application of the method to systems of outstanding biological interest.