Revisiting DNA Sequence-Dependent Deformability in High-Resolution Structures: Effects of Flanking Base Pairs on Dinucleotide Morphology and Global Chain Configuration

Laplace approximation of J factors for rigid base and rigid basepair models of DNA cyclization

We apply the Laplace approximation to a mathematical formulation of DNA cyclization J factors, leading to a formula that involves energies of local minima of the DNA energy, factors coming from the Hessian of the energy near each minimum, and geometric factors arising from the orientational portion of J. The approximation is derived in a quite general setting that encompasses both rigid base and rigid basepair models common in the literature. The approximation is applied to several families of 200-400 bp DNA, some relatively straight (fragments of λ-phage) and others quite bent (constructs that include up to 10 A tracts). The accuracy of the approximation is assessed by comparing with (more time-consuming) Monte Carlo computations: Laplace is within 20% of Monte Carlo for most 200 bp molecules and undershoots Monte Carlo by about 30% for 300 bp and 50% for 400 bp. We explore length and sequence dependence, both for our overall approximation of J and for its energy and entropic components, and make comparisons to a different approximation of J proposed in the literature.

Predicting DNA structure using a deep learning method

Understanding the mechanisms of protein-DNA binding is critical in comprehending gene regulation. Three-dimensional DNA structure, also described as DNA shape, plays a key role in these mechanisms. In this study, we present a deep learning-based method, Deep DNAshape, that fundamentally changes the current k-mer based high-throughput prediction of DNA shape features by accurately accounting for the influence of extended flanking regions, without the need for extensive molecular simulations or structural biology experiments. By using the Deep DNAshape method, DNA structural features can be predicted for any length and number of DNA sequences in a high-throughput manner, providing an understanding of the effects of flanking regions on DNA structure in a target region of a sequence. The Deep DNAshape method provides access to the influence of distant flanking regions on a region of interest. Our findings reveal that DNA shape readout mechanisms of a core target are quantitatively affected by flanking regions, including extended flanking regions, providing valuable insights into the detailed structural readout mechanisms of protein-DNA binding. Furthermore, when incorporated in machine learning models, the features generated by Deep DNAshape improve the model prediction accuracy. Collectively, Deep DNAshape can serve as versatile and powerful tool for diverse DNA structure-related studies.

Deep DNAshape: Predicting DNA shape considering extended flanking regions using a deep learning method

Understanding the mechanisms of protein-DNA binding is critical in comprehending gene regulation. Three-dimensional DNA shape plays a key role in these mechanisms. In this study, we present a deep learning-based method, Deep DNAshape, that fundamentally changes the current k -mer based high-throughput prediction of DNA shape features by accurately accounting for the influence of extended flanking regions, without the need for extensive molecular simulations or structural biology experiments. By using the Deep DNAshape method, refined DNA shape features can be predicted for any length and number of DNA sequences in a high-throughput manner, providing a deeper understanding of the effects of flanking regions on DNA shape in a target region of a sequence. Deep DNAshape method provides access to the influence of distant flanking regions on a region of interest. Our findings reveal that DNA shape readout mechanisms of a core target are quantitatively affected by flanking regions, including extended flanking regions, providing valuable insights into the detailed structural readout mechanisms of protein-DNA binding. Furthermore, when incorporated in machine learning models, the features generated by Deep DNAshape improve the model prediction accuracy. Collectively, Deep DNAshape can serve as a versatile and powerful tool for diverse DNA structure-related studies.

Thermotoga maritima oriC involves a DNA unwinding element with distinct modules and a DnaA-oligomerizing region with a novel directional binding mode

Initiation of chromosomal replication requires dynamic nucleoprotein complexes. In most eubacteria, the origin oriC contains multiple DnaA box sequences to which the ubiquitous DnaA initiators bind. In Escherichia coli oriC, DnaA boxes sustain construction of higher-order complexes via DnaA-DnaA interactions, promoting the unwinding of the DNA unwinding element (DUE) within oriC and concomitantly binding the single-stranded DUE to install replication machinery. Despite the significant sequence homologies among DnaA proteins, bacterial oriC sequences are highly diverse. The present study investigated the design of oriC (tma-oriC) from Thermotoga maritima, an evolutionarily ancient eubacterium. The minimal tma-oriC sequence includes a DUE and a flanking region containing five DnaA boxes recognized by the cognate DnaA initiator (tmaDnaA). This DUE was comprised of two distinct functional modules, an unwinding module and a tmaDnaA-binding module. Three direct repeats of the trinucleotide TAG within DUE were essential for both unwinding and single-stranded DUE binding by tmaDnaA complexes constructed on the DnaA boxes. Its surrounding AT-rich sequences stimulated only duplex unwinding. Moreover, head-to-tail oligomers of ATP-bound tmaDnaA were constructed within tma-oriC, irrespective of the directions of the DnaA boxes. This binding mode was considered to be induced by flexible swiveling of DnaA domains III and IV, which were responsible for DnaA-DnaA interactions and DnaA box binding, respectively. Phasing of specific tmaDnaA boxes in tma-oriC DNA was also responsible for unwinding. These findings indicate that a single-stranded DUE recruitment mechanism was responsible for unwinding, and would enhance understanding of the fundamental molecular nature of the origin sequences present in evolutionarily divergent bacteria.

Modeling the Infrared Spectroscopy of Oligonucleotides with 13C Isotope Labels

The carbonyl stretching modes have been widely used in linear and two-dimensional infrared (IR) spectroscopy to probe the conformation, interaction, and biological functions of nucleic acids. However, due to their universal appearance in nucleobases, the IR absorption bands of nucleic acids are often highly congested in the 1600-1800 cm^-1 region. Following the fruitful applications in proteins, ¹³C isotope labels have been introduced to the IR measurements of oligonucleotides to reveal their site-specific structural fluctuations and hydrogen bonding conditions. In this work, we combine recently developed frequency and coupling maps to develop a theoretical strategy that models the IR spectra of oligonucleotides with ¹³C labels directly from molecular dynamics simulations. We apply the theoretical method to nucleoside 5'-monophosphates and DNA double helices and demonstrate how elements of the vibrational Hamiltonian determine the spectral features and their changes upon isotope labeling. Using the double helices as examples, we show that the calculated IR spectra are in good agreement with experiments and the ¹³C isotope labeling technique can potentially be applied to characterize the stacking configurations and secondary structures of nucleic acids.

Insights into DNA solvation found in protein-DNA structures

The proteins that bind double-helical DNA present various microenvironments that sense and/or induce signals in the genetic material. The high-resolution structures of protein-DNA complexes reveal the nature of both the microenvironments and the conformational responses in DNA and protein. Complex networks of interactions within the structures somehow tie the protein and DNA together and induce the observed spatial forms. Here we show how the cumulative buildup of amino acid atoms around the sugars, phosphates, and bases in different protein-DNA complexes produces a binding cloud around the double helix and how different types of atoms fill that cloud. Rather than focusing on the principles of molecular binding and recognition suggested by the arrangements of amino acids and nucleotides in the macromolecular complexes, we consider the proteins in contact with DNA as organized solvents. We describe differences in the mix of atoms that come in closest contact with DNA, subtle sequence-dependent features in the microenvironment of the sugar-phosphate backbone, a direct link between the localized buildup of ionic species and the electrostatic potential surfaces of the DNA bases, and sites of atomic buildup above and below the basepair planes that transmit the unique features of the base environments along the chain backbone. The inferences about solvation that can be drawn from the survey provide new stimuli for improvement of nucleic acid force fields and fresh ideas for exploration of the properties of DNA in solution.