Nucleic acid simulations themed issue

Charles A. Laughton

Charlie Laughton is Associate Professor and Reader in Molecular Recognition in the School of Pharmacy, University of Nottingham. His doctorate was in synthetic organic chemistry, but for the last twenty years he has worked primarily on the development and application of simulation methodologies to problems in biomolecular structure, dynamics and recognition. Much of this work has been focussed on nucleic acids, and concerned with both their fundamental biological and biophysical properties, and with their importance as targets for anticancer drug development.

Modesto Orozco

Modesto Orozco received his PhD in Biochemistry in 1990 from the Universitat Autònoma de Barcelona. Since 2002 he has been a full professor of Biochemistry and Molecular Biology at the Departament de Bioquímica, Universitat de Barcelona, and since 2005 has also been the Director of the Computational Biology Programme, Barcelona Supercomputer Center. His wide-ranging research covers theoretical study of the mechanism of action of living organisms, using the basic rules of physics and chemistry.

---

While the first molecular dynamics (MD) simulations of proteins were reported by McCammon and Karplus between 1976 and 1977, the first reports of simulations of nucleic acids were not until seven years later: of double-helical DNA by the Levitt and Karplus groups and of tRNA by the McCammon and Harvey groups. In large part this gap between the fields of protein and nucleic acids MD reflected the observation that reliable simulations of nucleic acids were hard to achieve in the early days with limited computational resources, rough simulation protocols and approximate force-fields. Fortunately, the situation is now very different, and the last ten years have seen an explosion of interest in nucleic acid simulation, and an ever-widening range of aspects of the physics, chemistry, and biology of nucleic acids that have been very profitably explored through an increasingly diverse range of simulation methodologies. This themed issue of PCCP illustrates the wonderful diversity of interest that there is in these seemingly simple molecules, and the unique contributions that simulation techniques can make to understanding them better.

The deceptively simple structure of the classic DNA double helix masks intriguing physical properties. Sequence-dependent variations in structure and bendability are of profound biological relevance, but are still poorly understood. The article from the Maddocks group addresses issues in this area, while Barone’s team examines the, always complex, behaviour of single-stranded DNA.

Simulation studies continue to improve our understanding of how proteins have evolved to selectively recognise particular DNA sequences—a property without which life as we know it could never have developed. The article by Churchill’s group highlights a very direct mechanism of recognition—stacking interactions between the DNA and aromatic side chains in proteins—while Zakrzewska and coworkers describes how grid computing methods can be used to tackle questions about the role of induced fit in selective recognition, questions that could not be addressed without such powerful resources. The power of current computing facilities is also illustrated in Ponomarev’s work on nucleosome dynamics and recognition. Nevertheless, it remains the case that the length- and time-scale of many key processes that involve nucleic acids are not amenable to simple atomistic simulation methods. Fortunately a wide variety of mesoscale models and methods are available to overcome this limitation. The perspective article by Harvey illustrates this, as do those by Wocjan, Schlick, and Vologodskii. As examples of an alternative approach, enhanced sampling methods that maintain atomistic detail feature in the articles by Lin, Kannan, and de Marco.

New computers, either based on general-purpose or MD-optimized processors, are now opening up the possibility of moving from the nanosecond to the microsecond time scale; but the extension of trajectories into this regime is likely to make more evident any errors in algorithms or force-fields which will then need to be corrected. The critical evaluation and improvement of current simulation tools is therefore a crucial issue, a topic that is central in the contributions by Besseova and Noy.

The accurate prediction of ligand-target binding affinities is a major goal for simulation methodologies, and this applies as much to DNA–ligand recognition as to protein–ligand recognition. Issues such as induced fit, enthalpic versus entropic components of the recognition process, and potency versus selectivity, continue to be fertile fields for theoretical investigations, as the articles by Treesuwan and Wang illustrate.

Early on, the structure of the DNA double helix led to questions about its potential to conduct electricity, questions that still remain to be fully resolved and that may be as important for understanding mechanisms of DNA repair, as for fully exploiting the biotechnological properties of nucleic acids. Articles by D’Abramo and Voityuk throw more light on how the redox properties of DNA are influenced by the structure, and importantly the dynamics, of the molecule.

In summary, we believe this themed issue exemplifies the rich diversity of cutting-edge research in the field of nucleic acids simulation. It will provide a broad picture of the field for non-expert readers, and many details of interest to experts. We wish to thank the enthusiastic response of many friends who were so kind to find the time to contribute to this issue, and of course the superb work of the PCCP editorial team that has made it all possible.

Charles A. Laughton

Modesto Orozco

---