Bioinformatics Advance Access originally published online on February 25, 2007
Bioinformatics 2007 23(8):1035-1037; doi:10.1093/bioinformatics/btm067
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
SwS: a solvation web service for nucleic acids
A.R.N., Université Louis Pasteur, IBMC-CNRS, 15 rue René Descartes, 67084 Strasbourg, France
*To whom correspondence should be addressed.
 |
ABSTRACT |
|---|
 TOP ABSTRACT ACKNOWLEDGEMENTSREFERENCES |
|---|
Summary: SwS, based on a statistical analysis of crystallographic structures deposited in the NDB, is designed to provide an exhaustive overview of the solvation of nucleic acid structural elements through the generation of 3D solvent density maps. A first version (v1.0) of this web service focuses on the interaction of DNA, RNA and hybrid base pairs linked by two or three hydrogen bonds with water, cations and/or anions. Data provided by SwS are updated on a weekly basis and can be used by: (i) those involved in molecular dynamics simulation studies for validation purposes; (ii) crystallographers for help in the interpretation of solvent density maps; and all those involved in (iii) drug design and, more generally, in (iv) nucleic acid structural studies. SwS provides also statistical data related to the frequency of occurrence of different types of base pairs in crystallographic structures and the conformation of the involved nucleotides. This web service has been designed to allow a maximum of flexibility in terms of queries and has also been developed with didactic considerations in mind.
Availability: http://www-ibmc.u-strasbg.fr/arn/sws.html
Contact: p.auffinger{at}ibmc.u-strasbg.fr
All information necessary for nucleic acid molecules to fold and associate into functional tertiary structures are present in their sequence of nucleotides. The folding/association process is mainly governed by the hierarchical formation of base pairs and stacking interactions. However, the solvent (water molecules, cations and anions) can influence to a large degree the structure of nucleic acids through specific and/or non-specific interactions. Consequently, nucleic acid sequences also integrate information linked to solvent-related environmental factors that are crucial for the formation of active 3D folds. Moreover, solvent molecules participate in all types of molecular recognition phenomena involving for example: (i) small natural ligands; (ii) therapeutic molecules; (iii) proteins and (iv) other nucleic acid components. Furthermore, neutral or charged chemical groups belonging to interacting molecules often occupy well-defined solvation sites. Hence, in order to acquire deeper insight into such fundamental molecular recognition processes, it is mandatory to get a detailed access to the structure of the first solvation shell surrounding DNA and RNA structural elements. This can be achieved by exploring the crystallographic structures deposited in the ‘Nucleic acid DataBase’ or NDB (Berman et al., 2002). However, such an undertaking is very tedious and time consuming if performed ‘manually’ (Auffinger and Westhof, 1998). Consequently, in order to automate this exploration process, we developed SwS (a Solvation web service for nucleic acids) that is, to the best of our knowledge, the first web service allowing to analyze, from a statistical and structural point of view, the solvation of nucleic acids.
SwS is based on more than 2200 crystallographic structures of nucleic acids, with a resolution better or equal to 3.0 Å, currently deposited in the NDB and allows to calculate, in its first version (v1.0), solvent ‘pseudo-electron’ density maps around the 33 DNA, RNA or hybrid (one DNA and one RNA nucleotide) base-pair types involving two or three interbase hydrogen bonds described by Leontis and Westhof (2002) that represent 
80% of all detected base pairs. From these solvent density maps that are calculated on the fly by using a procedure described in earlier articles (Auffinger and Westhof, 2000; Schneider and Berman, 1995), it is possible to infer the most probable solvent molecule positions (water, cations and anions) around the selected structural fragments.
The complete SwS procedure comprises four selections and one visualization step. The selections steps allow to: (i) choose a base-pair type (Fig. 1, left); (ii) define resolution, sugar pucker and syn/anti constraints (iii) apply backbone dihedral angle constraints and (iv) define a subset of solvent molecule types (for example, all water molecules, all monovalent cations, all Mg2+ cations or all anions). The first three selection steps are required for generating a structurally homogeneous ensemble of base pairs and backbone conformations from which a coherent view of their solvation can be inferred. These steps are based on the outputs provided by the FINDPAIR and ANALYZE programs of the 3DNA toolkit (Lu and Olson, 2003). The visualization step (Fig. 1, right), makes use of the AstexTM 2.0 Java graphical applet (Hartshorn, 2002). First, an average base pair, calculated from all selected base pairs is represented. Second, all the solvent molecules located in the first solvation shell of the selected base pairs are visualized. Third, solvent ‘pseudo-electron’ density maps are displayed. These maps are calculated by using the SFALL program of the CCP4 (Collaborative computational project No. 4, 1994). In short, the distribution of the solvent molecules proximate to the selected base pairs is transformed into ‘pseudo-electron’ densities through a Fourier transformation. Fourth, density maxima are calculated by using the PEAKMAX program of the CCP4. Fictitious solvent molecules associated with these density maxima and colored according to their height, from the highest (red) to the lowest (blue) can be visualized. Further technical information related to SwS can be accessed through the ‘Help’ link of the home page.
|
Besides, the ‘Gallery’ link provides some, non-restrictive examples related to the usage and limitations of this web service. For instance, SwS can be used to estimate the relation between the conformation adopted by the backbone of a given base pair and its first solvation shell and, consequently, provide some insight into solvation-dependent conformational changes.
SwS can also help to map ion-binding sites, a feature that could facilitate the interpretation of crystallographic solvent density maps. Moreover, since hydration and ion-binding sites correspond frequently to preferential binding sites for chemical groups belonging to drugs and proteins, SwS could be usefully included in drug design strategies targeting nucleic acids.
This web service has also been designed to become an important tool in the validation process of molecular dynamics simulations since it allows to perform efficient comparisons between calculated and experimental data. However, it must be noted that the data provided by SwS can only be as good as are the crystal structures on which they are based. Hence, the users are encouraged to always critically consider the data provided by this web service.
Finally, the SwS internal database is updated weekly by automatically retrieving the most recent crystal structures deposited in the NDB. Hence, the completeness of the data provided by SwS can only increase. In time, more features will be added to SwS such as: (i) the inclusion of the less frequent base pairs involving one regular and one C–H ... O hydrogen bond before (ii) addressing the problem of base pairs linked by a single hydrogen bond and (iii) taking into consideration neighboring base pairs for the calculation of sequence-dependent solvation patterns. When the number of structures deposited in the NDB will be sufficiently large, solvation patterns for base-pair steps and higher structural motifs will also become accessible.
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTACKNOWLEDGEMENTSREFERENCES |
|---|
The authors acknowledge the help of Dr Xiang-Jun Lu, Dr Michael J. Hartshorn, Prof. Neocles Leontis, Dr Fabrice Jossinet and Thomas Ludwig as well as Prof. Eric Westhof. Funding to pay the Open Access publication charges was provided by the Centre National de la Recherche Scientifique (CNRS).
Conflict of Interest: none declared.
|
FOOTNOTES |
|---|
Associate Editor: Keith Crandall
Received on January 30, 2007; accepted on February 20, 2007
|
REFERENCES |
|---|
|
TOPABSTRACTACKNOWLEDGEMENTSREFERENCES |
|---|
Auffinger P, Westhof E. Hydration of RNA base pairs. J. Biomol. Struct. Dyn, ( (1998) ) 16, : 693–707.[ISI][Medline].
Auffinger P, Westhof E. RNA solvation: a molecular dynamics simulation perspective. Biopolymers, ( (2000) ) 56, : 266–274.[CrossRef][Medline].
Berman HM, et al. The nucleic acid database. Acta Cryst., ( (2002) ) D58, : 889–898.[CrossRef][ISI].
Collaborative Computational Project No. 4. The CCP4 suite: programs for protein crystallography. Acta Cryst., ( (1994) ) D50, : 760–763.[ISI].
Hartshorn MJ. AstexViewer: a visualisation aid for structure-based drug design. J. Comput. Aided Mol. Des., ( (2002) ) 16, : 871–881.[CrossRef][ISI][Medline].
Leontis NB, Westhof. The non-Watson-Crick base pairs and their associated isostericity matrices. Nucleic Acids Res., ( (2002) ) 30, : 3497–3531.
Lu XJ, Olson WK. 3DNA: a software package for the analysis, rebuilding and visualization of three-dimensional nucleic acid structures. Nucleic Acids Res., ( (2003) ) 31, : 5108–5121.
Schneider B, Berman HM. Hydration of the DNA bases is local. Biophys. J, ( (1995) ) 69, : 2661–2669.[ISI][Medline].
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  W. Rypniewski, D. A. Adamiak, J. Milecki, and R. W. Adamiak Noncanonical G(syn)-G(anti) base pairs stabilized by sulphate anions in two X-ray structures of the (GUGGUCUGAUGAGGCC) RNA duplex RNA, September 1, 2008; 14(9): 1845 - 1851. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||