The computational-based structure of Dwarf14 provides evidence for its role as potential strigolactone receptor in plants
© Gaiji et al.; licensee BioMed Central Ltd 2012
Received: 6 December 2011
Accepted: 19 June 2012
Published: 19 June 2012
Strigolactones (SLs) are recently identified plant hormones modulating root and shoot branching. Besides their endogenous role within the producing organism, SLs are also key molecules in the communication of plants with arbuscular mycorrhizal (AM) fungi and parasitic weeds. In fact SLs are exuded into the rhizosphere where they act as a host-derived signal, stimulating the germination of the seeds of parasitic plants which would not survive in the absence of a host root to colonize. Similarly, their perception by AM fungi causes extensive hyphal branching; this is a prerequisite for effective root colonization, since it increases the number of potential contact points with the host surface. In spite of the crucial and multifaceted biological role of SLs, there is no information on the receptor(s) which bind(s) such active molecules, neither in the producing plants, or in parasitic weeds or AM fungi.
In this work, we applied homology modelling techniques to investigate the structure of the protein encoded by the gene Dwarf14, which was first identified in rice as conferring SLs insensitivity when mutated. The best sequence identity was with bacterial RsbQ. Both proteins belong to the superfamily of alpha/beta-fold hydrolases, some members of which play a role in the metabolism or signalling of plant hormones. The Dwarf14 (D14) structure was refined by means of molecular dynamics simulations. In order to support the hypothesis that D14 could be an endogenous SLs receptor, we performed docking experiments with a natural ligand.
It is suggested that D14 interacts with and thereby may act as a receptor for SLs in plants. This hypothesis offers a starting point to experimentally study the mechanism of its activity in vivo by means of structural, molecular and genetic approaches. Lastly, knowledge of the putative receptor structure will boost the research on analogues of the natural substrates as required for agricultural applications.
Strigolactones (SLs) are a group of plant-produced carotenoid-derived terpenoid lactones that have been recently implicated in the regulation of shoot and root branching [1–3]. Already long before then, SLs were known as root-exuded molecules capable to provoke the germination of seeds from parasitic plants, like Striga and Orobanche and, more recently, their role was extended to the induction of hyphal branching and of a burst of mitochondrial activity in arbuscular mycorrhizal fungi (AMF) [5, 6]. In undisturbed ecosystems, most plants are colonized by AMF , a group of soil-borne fungal endophytes belonging to the ancient Glomeromycota phylum. The association is the result of co-evolution events dating back to the early Devonian times : its success in time and space is mostly due to the nutritional benefits both partners gain. How plants and AMF establish a molecular dialogue which eventually allows the symbiosis, is a crucial question in plant biology. The release of soluble signals in the rhizosphere was suggested as an easy solution for both partners to be timely informed of the presence of each other, even before physical contact . AMF produce several bioactive compounds with a chitin-based structure called “Myc factors” [10, 11]. At the same time they respond with profuse branching to root exudates from compatible hosts [12, 13], whose bioactive molecules are SLs . The ability to perceive SLs has been very recently expanded also to non-AMF, phytopathogenic filamentous fungi . The ability to synthesize SLs is widespread along the plant taxa, also including non-host plants for AMF like Arabidopsis .
The characterization of these versatile molecules, the identification of their biosynthetic genes in plants and of their receptors in plants and fungi is at the moment a hot spot in plant biology [16–18]. In spite of the increasing knowledge on SLs synthesis and mechanism of action, the proteins that mediate their perception within the plant are still poorly known. Only two genes, encoding an F-Box protein and a predicted α/β-fold hydrolase, have been identified so far as potentially involved in the perception and/or transduction mechanisms of the SLs signal, since their mutants are SL-insensitive [16, 19]. As far as the proteins that are expected to perceive SLs in parasitic plants, even less is known . These studies open the question whether similar proteins with analogous functions may be present in the AMF, but on this respect, no genetic or biochemical information is available as yet. Nevertheless, Akiyama and co-workers  have highlighted the molecular structural requirements that specifically correlate with the activity of SLs in stimulating AMF branching vs. seed germination in parasitic plants. The incomplete overlap between such requirements makes it likely that the nature of SLs receptors is different in organisms belonging to different kingdoms, whereas it can be envisaged that SLs receptors in parasitic plants may share similarity with the receptor(s) for endogenously produced SLs.
To identify the structure of a potential SLs receptor in plants we used multiple bioinformatics and computational approaches. These allowed us to propose a model compatible with the protein described in rice as D14, and to verify whether its docking features fit to a natural reference SLs molecule (Strigol). A similar approach was applied to GID1  and SABP2 , two plant proteins belonging to the same super-family as D14 and broadly involved in the metabolism and perception of two important classes of plant hormones, i.e. gibberellins and salicylic acid with its derivatives.
Results and discussion
Sequence analysis and homology modelling of D14
RsbQ belongs to the α/β-fold hydrolase super-family and is involved in the stress response of Bacillus subtilis[26, 27]. The 3D structure of the RsbQ protein was solved by X-ray crystallography and two 3D structures are reported in PDB: the native [PDB: 1WOM] and the inhibitor-bound one [PDB: 1WPR], the inhibitor being phenylmethanesulfonic acid (PMSF). Both were used as templates for the generation of the 3D model of D14 with three different methods, in order to prevent methodological biases (see Methods). The catalytic nucleophile, Ser96, has the same position in both the native and the PMSF-bound structure. This position of the nucleophile is shared among the α/β hydrolases and is due to the formation of a sharp turn, called the “nucleophile elbow” [28, 29]. The catalytic triad is buried inside the molecule and the active site is a hydrophobic cavity that is nearly isolated from the solvent. It is inferred from this feature that the catalytic site of RsbQ has specificity for a hydrophobic, small compound, rather than a macromolecule such as RsbP (a protein phosphatase physically interacting with RsbQ and involved in its signalling pathway). Instead, structural comparison with other α/β hydrolases demonstrates that a unique loop region of RsbQ is a likely candidate for the interaction site with RsbP, and that this interaction might be responsible for the product release by operating the hydrophobic gate between the cavity and the solvent.
Structural refinement and stability evaluation by Molecular Dynamics of the D14 model
Having built the 3D structure of D14 by homology modelling, we noticed that the catalytic triad of RsbQ (Ser95, Asp219, His250) almost overlaps in the active site of the modelled D14. The homologous structure is then optimized without restraints by means of molecular mechanics (MM), in explicit solvent.
The binding site of D14 is elucidated by means of the MOE site-finder module  and by using the structure of RsbQ bound to the inhibitor PMSF [PDB: 1WPR] as target.
The docking, both flexible and rigid, of Strigol into the site of the protein structure averaged over the last 2 ns of MD of the complex D14-solvent was performed. The putative ligand is well arranged in the site, as already evidenced by MD, and the binding energy is very favourable (−11.3 Kcal/mol). This supports the idea that, similarly to what suggested by the authors in the case of RsbQ, the cognate ligand of D14 could be a small hydrophobic molecule such as SLs.
D14 and the Hydrolases mechanism
Docking energies of substrate and product for the DLH-like reaction
Both structural and catalytic features are consistently in agreement with the hypothesis put forward for D14, that is, D14 can bind to SLs and modify their structure by opening the lactone ring.
Comparison between D14 and other proteins playing a role in plant defence, development and metabolism
We hence compared D14 and the enzyme showing the highest sequence similarity to GID1A: this is tobacco SABP2 [PDB: 1Y7H], a methylsalicylate (MeSA) esterase. The product, salicylic acid (SA), is a critical signal for the activation of plant defence responses against pathogen infections. SABP2 is thought to convert MeSA to SA as part of the signal transduction pathways that activate systemic acquired resistance and perhaps local defence responses as well .
The structural comparison between SABP2, DLH and D14 supports an analogy between their enzymatic activity as esterases: the figures evidence the good fit among the α/β fold and the diversity in the αF-chain and loop that may in fact recognize the substrates or act as lids in the signal transduction pathway.
The aim of this work was to describe the structure of the protein D14, which is encoded by the gene Dwarf14 and was suggested to be a potential SLs receptor in plants . From the structural comparison of D14 with a bacterial protein sharing 38% of sequence identity (RsbQ), it is inferred that SL signalling may involve a step with a hydrolasic-like catalytic mechanism: this is consistent with the structural requirements for SLs molecules active in parasitic plants.
Indeed, the detailed comparison of D14 with known structures shows that the active site is highly similar to that of DLHs. However in the case of D14, the pocket is not exposed to solvent, but protected by a helical domain that appears to be flexible: its hinge movement could be related to the recognition of the substrate and be part of the signal transmission by means of a conformational movement. This mechanism has been already postulated for the recognition of gibberellins by GID1 : however the binding sites of D14 and GID1 are different, and both their sequences and 3D structures have a low degree of similarity. Instead, high similarity is found between D14 and a known plant esterase, SABP2. This is involved in the processing of MeSA to obtain SA, a molecule controlling a subset of stress responses in plants. Once again, 3D structures of D14 and SABP2 are well correlated, and the demonstrated (SABP2) and postulated (D14, this work) enzymatic activities show intriguing analogies.
In conclusion, our results on one hand confirm previous findings suggesting that D14 can be regarded as a receptor for SLs in plants. On the other hand, they provide the first computational reconstruction of the 3D structure of D14 offering a model to be tested for experimental studies of in vitro and in vivo activity by means of structural, molecular and genetic approaches. The knowledge of the putative receptor structure will boost the research on analogues of the natural substrates as required for agricultural applications.
The aminoacidic sequence of Dwarf14 (D14) was taken from the article of Arite et al. . Multiple sequences alignment was performed using the ClustalW program accessible on-line through the European Bioinformatics Institute  and BLAST . A model of the D14 protein was generated using SWISS MODEL  and EsyPred3D . The model structure was based on the files PBD: 1WP0 and 1WPR; these template proteins, belonging to the family of α/β hydrolases, were chosen because of a significant sequence similarity with D14, in addition to their satisfactory crystallographic resolution. The model was subsequently verified using MOE-ProEval, an implementation of the PROCHECK suite of stereochemical measurements, and Ramachandran’s maps .
Molecular Dynamics simulations
The 3-D molecule was locally minimized in vacuo by constraining the backbone to the template molecule in order to give a first optimization of the rough geometry derived from homology modeling, particularly for the side chains and the added polar hydrogen atoms.
GROMACS [30, 38] was then used for MD simulations. The structure of D14 was inserted into a cubic box maintaining a minimum of 9 Å between the box edges and the protein surface. The resulting system was solvated with Simple Point Charge (SPC) water molecules provided in the GROMACS package and then minimized with the GROMOS96 force field using the steepest descent method in order to lead the system to a more favourable energetic condition before starting the MD simulation. The temperature of the bath was set to 300 K and the coupling time constant was set to 0.1 ps. The box pressure was maintained at 1 bar using 1ps time constant and a water compressibility of 4.5 × 10–5 bar−1. Coulombic interactions were treated with the PME (Particle Mesh Ewald algorithm) model with a cutoff of 1.6 nm. Configurations were saved every 100 fs for analysis. After equilibrating the system, a 5 ns production simulation was conducted with a 1 fs time-step at a pressure of 1 bar and a temperature of 300 K. At this stage no constraints or restraints to the template structure 1WOM were added. The only constraint applied was to the α-helices and β -sheets H bonds, using the LINCS  algorithm: this is an algorithm that resets bonds to their correct lengths after an unconstrained update. The following parameters were used: lincs-order of 4, lincs-warn angle of 30 and unconstrained start. Computer simulations describe protein dynamics, and under the limit of their accuracy and extension, they should contain information on functional motion and ability to address the relationship that motion has with structure. Essential dynamics (ED)  has been a fairly applied method to extract useful information from protein simulation. In particular the ED analysis reveals high-amplitude concerted motions in the equilibrated portion of the trajectories, based on the diagonalization of the Cα covariance matrix of the atomic positional fluctuations. The collection of the selected eigenvectors describing the collective motions is termed “essential subspace” and can describe protein motions at a reasonable level of accuracy. Correlation plots were obtained by first computing Cα correlation matrices C(i,j), where C(i,j) is the covariance matrix of protein fluctuations between residues i and j.
Binding-site identification and analysis
The Site Finder module of MOE 2008.10  was used to identify the putative binding pockets and protein ligand-binding sites in the energy-minimized 3D structure of D14. The Site Finder module of MOE 2008.10 generates hydrophobic and hydrophilic alpha spheres serving as probes denoting zones of tight atom packing. These alpha spheres are then used as centroids for the creation of dummy atoms used to define potential binding sites .
The average structure of D14 resulting from the last 2 ns of molecular dynamics with the Strigol molecule in the binding pocket was used to carry out ligand-receptor simulations. The molecular docking simulations were performed with MOE dock package  and Delos package (N. Gaiji, F. Archetti, P.C. Fantucci, E.L. Zimolo, L. Roggia DELOS: Method of construction and selection of virtual libraries in combinatorial chemistry. European Patent Application EP1628234, holder: Università Milano Bicocca). The ligand explores the conformational space to locate the most favourable binding orientation and conformation by aligning and matching all triangles of the template points with compatible geometry, while the protein atoms remain fixed. An affinity scoring function, ΔG, was employed to rank candidate poses. This simulation is divided into three stages: 1. Conformational analysis, during which ligand is treated in a flexible manner by rotating rotatable bonds. 2. Placement, during which a collection of orientations is generated from the pool of ligand conformations. In this case, the alpha-triangle placement method was used, which generates orientations by superposition of ligand atom triplets and triplet points in the receptor site. The receptor site points are alpha sphere centres which represent locations of tight packing. At each iteration, a random conformation is selected; a random triplet of ligand atoms and a random triplet of alpha sphere centres are used to determine the orientation. 3. Scoring, during which each orientation generated by the placement methodology is subjected to scoring in an effort to identify the most favourable orientations.
NG and GR are researcher and head of scientific projects, respectively, at Geol Sas – via Monte Bo, 2 – 13100 Vercelli
FC is staff researcher in Plant Physiology at the Dept. of Arboriculture, Università di Torino – via L. da Vinci, 44 – 10095 Grugliasco (TO) Italy
CP is associate professor of Organic Chemistry at the Dept. of Chemistry, Università di Torino – via P. Giuria, 7 – 10125 Turin
PB is full professor of Plant Biology at the Dept. of Life Science and Systems Biology, Università di Torino – viale P. Mattioli, 25–10025 Turin, Italy.
Arbuscular mycorrhizal fungi
Gibberellin insensitive 1
Particle mesh ewald algorithm
Phenylmethane sulfonic acid
Root mean square deviation
Regulator of sigma (b) P
Regulator of sigma (b) Q
Salicylic acid-binding protein 2
Simple point charge.
This work was funded by the Piedmont Region, Converging Technologies call 2007, BioBITs Project.
- Gomez-Roldan V, Fermas S, Brewer PB, Puech-Pagès V, Dun EA, Pillot JP, Letisse F, Matusova R, Danoun S, Portais JC, et al: Strigolactone inhibition of shoot branching. Nature. 2008, 455: 189-194. 10.1038/nature07271.PubMedView ArticleGoogle Scholar
- Umehara M, Hanada A, Yoshida S, Akiyama K, Arite T, Takeda-Kamiya N, Magome H, Kamiya Y, Shirasu K, Yoneyama K, et al: Inhibition of shoot branching by new terpenoid plant hormones. Nature. 2008, 455: 195-200. 10.1038/nature07272.PubMedView ArticleGoogle Scholar
- Arbuscular Mycorrhizas: Physiology and Function. Edited by: Koltai H, Kapulnik Y. 2010, Springer Science + Business Media BV, Dordrecht Heidelberg London New York, 2Google Scholar
- Bouwmeester HJ, Roux C, Lopez-Raez JA, Bécard G: Rhizosphere communication of plants, parasitic plants and AM fungi. Trends Plant Sci. 2007, 12: 224-230. 10.1016/j.tplants.2007.03.009.PubMedView ArticleGoogle Scholar
- Besserer A, Puech-Pagès V, Kiefer P, Gomez-Roldan V, Jauneau A, Roy S, Portais JC, Roux C, Bécard G, Sejalon-Delmas N: Strigolactones stimulate arbuscular mycorrhizal fungi by activating mitochondria. PLoS Biol. 2006, 4: e226-10.1371/journal.pbio.0040226.PubMedPubMed CentralView ArticleGoogle Scholar
- Besserer A, Bécard G, Jauneau A, Roux C, Sejalon-Delmas N: GR24, a synthetic analog of strigolactones, stimulates the mitosis and growth of the arbuscular mycorrhizal fungus Gigaspora rosea by boosting its energy metabolism. Plant Physiol. 2008, 148: 402-413. 10.1104/pp.108.121400.PubMedPubMed CentralView ArticleGoogle Scholar
- Smith SE, Read DJ: Mycorrhizal Symbiosis. 2008, Academic, Amsterdam, Boston, 3Google Scholar
- Redecker D, Hijri I, Wiemken A: Molecular identification of AM fungi in roots: perspectives and problems. Folia Geobot Phytotaxon. 2003, 38: 113-124. 10.1007/BF02803144.View ArticleGoogle Scholar
- Bonfante P, Genre A: Mechanisms underlying beneficial plant-fungus interactions in mycorrhizal symbiosis. Nat Commun. 2010, 1: 48-PubMedView ArticleGoogle Scholar
- Bonfante P, Requena N: Dating in the dark: how roots respond to fungal signals to establish arbuscular mycorrhizal symbiosis. Curr Opin Plant Biol. 2011, 14: 451-457. 10.1016/j.pbi.2011.03.014.PubMedView ArticleGoogle Scholar
- Maillet F, Poinsot V, Andre O, Puech-Pagès V, Haouy A, Gueunier M, Cromer L, Giraudet D, Formey D, Niebel A, et al: Fungal lipochitooligosaccharide symbiotic signals in arbuscular mycorrhiza. Nature. 2011, 469: 58-63. 10.1038/nature09622.PubMedView ArticleGoogle Scholar
- Giovannetti M, Sbrana C, Citernesi AS, Avio L: Analysis of factors involved in fungal recognition responses to host-derived signals by arbuscular mycorrhizal fungi. New Phytol. 1996, 133: 65-71. 10.1111/j.1469-8137.1996.tb04342.x.View ArticleGoogle Scholar
- Bécard G, Taylor LP, Douds DD, Pfeffer PE, Doner LW: Flavonoids are not necessary plant signal compounds in arbuscular mycorrhizal symbiosis. Mol Plant Microbe Interact. 1995, 8: 252-258. 10.1094/MPMI-8-0252.View ArticleGoogle Scholar
- Akiyama K, Matsuzaki K, Hayashi H: Plant sesquiterpenes induce hyphal branching in arbuscular mycorrhizal fungi. Nature. 2005, 435: 824-827. 10.1038/nature03608.PubMedView ArticleGoogle Scholar
- Dor E, Joel DM, Kapulnik Y, Koltai H, Hershenhorn J: The synthetic strigolactone GR24 influences the growth pattern of phytopathogenic fungi. Planta. 2011, 234: 419-427. 10.1007/s00425-011-1452-6.PubMedView ArticleGoogle Scholar
- Arite T, Umehara M, Ishikawa S, Hanada A, Maekawa M, Yamaguchi S, Kyozuka J: d14, a strigolactone-insensitive mutant of rice, shows an accelerated outgrowth of tillers. Plant Cell Physiol. 2009, 50: 1416-1424. 10.1093/pcp/pcp091.PubMedView ArticleGoogle Scholar
- Zwanenburg B, Mwakaboko A, Reizelman A, Anilkumar G, Sethumadhavan D: Structure and function of natural and synthetic signaling molecules in parasitic weed germination. Pest Manag Sci. 2009, 65: 478-491. 10.1002/ps.1706.PubMedView ArticleGoogle Scholar
- Koltai H: Strigolactones are regulators of root development. New Phytol. 2011, 190: 545-549. 10.1111/j.1469-8137.2011.03678.x.PubMedView ArticleGoogle Scholar
- Ishikawa S, Maekawa M, Arite T, Onishi K, Takamure I, Kyozuka J: Suppression of tiller bud activity in tillering dwarf mutants of rice. Plant Cell Physiol. 2005, 46: 79-86. 10.1093/pcp/pci022.PubMedView ArticleGoogle Scholar
- Akiyama K, Ogasawara S, Ito S, Hayashi H: Structural requirements of strigolactones for hyphal branching in AM fungi. Plant Cell Physiol. 2010, 51: 1104-1117. 10.1093/pcp/pcq058.PubMedPubMed CentralView ArticleGoogle Scholar
- Murase K, Hirano Y, Sun TP, Hakoshima T: Gibberellin-induced DELLA recognition by the gibberellin receptor GID1. Nature. 2008, 456: 459-463. 10.1038/nature07519.PubMedView ArticleGoogle Scholar
- Forouhar F, Yang Y, Kumar D, Chen Y, Fridman E, Park SW, Chiang Y, Acton TB, Montelione GT, Pichersky E, et al: Structural and biochemical studies identify tobacco SABP2 as a methyl salicylate esterase and implicate it in plant innate immunity. Proc Natl Acad Sci U S A. 2005, 102: 1773-1778. 10.1073/pnas.0409227102.PubMedPubMed CentralView ArticleGoogle Scholar
- BLAST: [http://blast.ncbi.nlm.nih.gov/Blast.cgi]
- Thompson JD, Higgins DG, Gibson TJ: CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 1994, 22: 4673-4680. 10.1093/nar/22.22.4673.PubMedPubMed CentralView ArticleGoogle Scholar
- RCSB Protein Data Bank: [http://www.pdb.org]
- Pane-Farre J, Lewis RJ, Stulke J: The RsbRST stress module in bacteria: a signalling system that may interact with different output modules. J Mol Microbiol Biotechnol. 2005, 9: 65-76. 10.1159/000088837.PubMedView ArticleGoogle Scholar
- Brody MS, Vijay K, Price CW: Catalytic function of an alpha/beta hydrolase is required for energy stress activation of the sigma(B) transcription factor in Bacillus subtilis. J Bacteriol. 2001, 183: 6422-6428. 10.1128/JB.183.21.6422-6428.2001.PubMedPubMed CentralView ArticleGoogle Scholar
- Nardini M, Dijkstra BW: Alpha/beta hydrolase fold enzymes: the family keeps growing. Curr Opin Struct Biol. 1999, 9: 732-737. 10.1016/S0959-440X(99)00037-8.PubMedView ArticleGoogle Scholar
- Molecular Operating Environment (MOE 2008.10). [http://www.chemcomp.com]
- Lindahl E, Hess B, Van Der Spoel D: GROMACS 3.0: a package for molecular simulation and trajectory analysis. J Mol Model. 2001, 7: 306-317.Google Scholar
- Holmquist M: Alpha/Beta-hydrolase fold enzymes: structures, functions and mechanisms. Curr Protein Pept Sci. 2000, 1: 209-235. 10.2174/1389203003381405.PubMedView ArticleGoogle Scholar
- Pathak D, Ollis D: Refined structure of dienelactone hydrolase at 1.8 Å. J Mol Biol. 1990, 214: 497-525. 10.1016/0022-2836(90)90196-S.PubMedView ArticleGoogle Scholar
- Brückmann M, Blasco R, Timmis KN, Pieper DH: Detoxification of protoanemonin by dienelactone hydrolase. J Bacteriol. 1998, 180: 400-402.PubMedPubMed CentralGoogle Scholar
- Ueguchi-Tanaka M, Matsuoka M: The perception of gibberellins: clues from receptor structure. Curr Opin Plant Biol. 2010, 13: 503-508. 10.1016/j.pbi.2010.08.004.PubMedView ArticleGoogle Scholar
- Park SW, Kaimoyo E, Kumar D, Mosher S, Klessig DF: Methyl salicylate is a critical mobile signal for plant systemic acquired resistance. Science. 2007, 318: 113-116. 10.1126/science.1147113.PubMedView ArticleGoogle Scholar
- Kiefer F, Arnold K, Kunzli M, Bordoli L, Schwede T: The SWISS-MODEL Repository and associated resources. Nucleic Acids Res. 2009, 37: D387-D392. 10.1093/nar/gkn750.PubMedPubMed CentralView ArticleGoogle Scholar
- Lambert C, Leonard N, De Bolle X, Depiereux E: ESyPred3D: prediction of proteins 3D structures. Bioinformatics. 2002, 18: 1250-1256. 10.1093/bioinformatics/18.9.1250.PubMedView ArticleGoogle Scholar
- Berendsen HJC, van der Spoel D, van Drunen R: GROMACS: a message-passing parallel molecular dynamics implementation. Comput Phys Commun. 1995, 91: 43-56. 10.1016/0010-4655(95)00042-E.View ArticleGoogle Scholar
- Hess B, Bekker H, Berendsen HJC, Fraaije JGEM: LINCS: a linear constraint solver for molecular simulations. J Comput Chem. 1997, 18: 1463-1472. 10.1002/(SICI)1096-987X(199709)18:12<1463::AID-JCC4>3.0.CO;2-H.View ArticleGoogle Scholar
- Amadei A, Linssen ABM, Berendsen HJC: Essential dynamics of proteins. Proteins. 1993, 17: 412-425. 10.1002/prot.340170408.PubMedView ArticleGoogle Scholar
- Hunenberger PH, Mark AE, van Gunsteren WF: Fluctuation and cross-correlation analysis of protein motions observed in nanosecond molecular dynamics simulations. J Mol Biol. 1995, 252: 492-503. 10.1006/jmbi.1995.0514.PubMedView ArticleGoogle Scholar