Open Access

Molecular dynamics simulations: from structure function relationships to drug discovery

In Silico Pharmacology20142:4

DOI: 10.1186/s40203-014-0004-8

Received: 23 October 2014

Accepted: 4 November 2014

Published: 21 November 2014


Molecular dynamics (MD) simulation is an emerging in silico technique with potential applications in diverse areas of pharmacology. Over the past three decades MD has evolved as an area of importance for understanding the atomic basis of complex phenomena such as molecular recognition, protein folding, and the transport of ions and small molecules across membranes. The application of MD simulations in isolation and in conjunction with experimental approaches have provided an increased understanding of protein structure-function relationships and demonstrated promise in drug discovery.


Molecular dynamics simulations Cytochrome P450 Drug-drug interactions Genetic polymorphism Drug design Allosteric binding sites Cryptic binding sites


This perspective highlights the importance of MD simulations in the area of structure function relationships (using the cytochromes P450 as an example) and drug discovery.

Main text


Molecular Dynamics (MD) simulations are computational approaches based on Newton?s equations of motion and statistical mechanics principles that relate the motions and distributions of atoms and molecules. The equations of motions are solved numerically for a system of interacting atoms to extract numerous chemical and biophysical properties from the atomic data (van Gunsteren and Berendsen [1990]). The key feature of this in silico technique is the possibility to mimic in vitro and in vivo conditions. For example, the protein may be simulated at varying pH, in the presence of water molecules and ions, with different salt or ionic concentrations, and even in the presence of lipid bilayer and other cellular components.

In MD simulations the forces between atoms and the potential energy of the system are defined by molecular mechanics biomolecular force fields. These biomolecular force fields are parameterized to fit quantum-mechanical calculations and experimental spectroscopic data. Parameterization involves definition of chemical bonding, atomic angles and dihedral angles, along with determination of partial atomic charges for calculation of the electrostatic-interaction energies, identification of appropriate van der Waals atomic radii, etc. A number of biomolecular force fields are available for simulations, the most common being AMBER (Cornell et al. [1995]), CHARMM (MacKerell et al. [1998]), and GROMOS (Oostenbrink et al. [2004]). The potential energy function used in the GROMOS96 (Oostenbrink et al. [2004]) force field is given in equation (1). Although these biomolecular force fields differ in the way they are parameterized, the results obtained from each are generally similar. Some of the commonly used softwares for MD simulations include GROMACS (Van Der Spoel et al. [2005]), NAMD (Phillips et al. [2005]), AMBER (Case et al. [2005]), and CHARMM (Brooks et al. [2009]).
V r 1 , r 2 , ? ? ? , r N = ? bonds 1 4 K b b 2 ? b 0 2 2 + ? angles 1 2 K ? cos? ? cos ? 0 2 + ? impropers 1 2 K ? ? ? ? 0 2 + ? dihedrals K ? 1 + cos ? cos m ? 2 + ? pairs C 12 i j r i j 12 ? C 6 i j r i j 6 + ? pairs q i q j 4 ? ? 0 ? 1 1 r i j

Eq. 1 In this function, V is the potential energy, r is the position of the particle (coordinate) at a particular time, b is the bond length, b 0 is the bond length at equilibrium, and k b is the bond force constant; ? is the bond angle, ? 0 is the bond angle at equilibrium, and k ? is the bond angle force constant; ? is the improper dihedral angle, ? 0 is the improper dihedral angle at equilibrium, k ? is the force constant for improper dihedral; ? is the dihedral, ? 0 is the dihedral angle at equilibrium, k ? is the force constant for torsional dihedral interactions; r ij is the distance between particles i and j, C12 and C6 are Lennard-Jones parameters; ?0 is the dielectric permittivity of vacuum, ?1 is the relative permittivity of the medium in which the particles are embedded, and q i and q j are the charges of particles i and j, respectively.

The spatiotemporal resolution attained at a particular timescale and the biomolecular processes that can be captured are important considerations in MD simulations. Simulating a system over the time scale 10?12?s - 10?9?s (picoseconds to nanoseconds) can identify motions such as atomic fluctuations, conformational changes in amino acid side chains and loop motions. However, to understand more complex biomolecular events such as large domain motions, protein folding, protein-ligand binding and the transport of molecules across membranes, simulation timescales of microseconds to milliseconds are required (Dror et al. [2012]). The first molecular dynamics simulation of a biomacromolecule of therapeutic interest was published almost four decades ago. This simulation was performed in a vacuum with simple molecular mechanics parameters for a short time (9.2?ps) (McCammon et al. [1977]). Since then, several advancements have taken place both in terms of computational power and algorithms. For instance, a recent two-millisecond time scale MD simulation of the ?2-adrenergic receptor using Google?s Exacycle cloud-computing platform has led to a detailed understanding of multiple activation pathways and differential interactions of agonists and inverse agonists (Kohlhoff et al. [2014]). This demonstrates the application of MD simulations for investigating complex biological phenomenon.

The importance of simulation techniques arises from the fact that biomacromolecules such as proteins exist in a dynamic state of motion. These dynamic motions are essential for the specific functions of biomacromolecules such as intermolecular protein-binding interactions or downstream signalling. Moreover, dynamic motions among the molecules are key driving forces for biomolecular events including molecular self-assembly, dimerization/oligomerization, and adaptive conformational changes on ligand binding or the transport of drugs and ions across channels and cellular membranes. Even though multiple experimental techniques aid the understanding of the structural features of biomolecules, they fail to characterize dynamic motions. MD simulations provide a means to model the flexibility and conformational changes in molecules at an atomic level and thus explore areas that are difficult to characterize experimentally.

Structure Function Relationships

X-ray crystallography is the most widely used experimental technique employed for elucidating the 3-dimensional (3D) structures of biomolecules. The underlying principle of this technique is that a periodic crystal with an ordered arrangement of the atoms can diffract x-ray beams. A diffraction pattern, which is characteristic to the particular arrangement of atoms in a crystal, is then generated using a series of mathematical calculations. Further calculations leads to the generation of electron densities, which are then used to build the 3D structure of the molecule based on atomic coordinates. However, structures derived by x-ray crystallography can only provide a snapshot of one particular conformational state of a biomolecule. For instance, in the case of human cytochromes P450 (CYP), which comprise a superfamily of enzymes responsible for the metabolism of drugs, non-drug xenobiotics and endogenous compounds (Miners et al. [1988], Miners and Birkett [1998]), x-ray crystallographic techniques have provided invaluable insights into the secondary and tertiary structures of CYP enzymes (Yano et al. [2004]). However, it is now evident from the approximately 150 human cytochrome P450 crystal structures that have been elucidated to date that CYP proteins are highly flexible and show dramatic structural adaptability in the presence of a ligand. For example, the experimental structure of CYP3A4 shows an active site volume of 950??3 in the ligand-free state, 1650??3 when bound to ketoconazole and 2,000??3 when bound to erythromycin (Ekroos and Sj?gren [2006]). To understand such dramatic changes in the active site architecture associated with the dynamic behavior of P450s, MD simulations provide a more practical alternative than the generation of x-ray crystal structures for every enzyme-ligand combination. In recent years, several applications have shown the potential of MD simulations for providing insights into structural aspects of human cytochromes P450, including demonstration of the plasticity of the substrate binding site (eg. Hendrychova et al. [2012]) and identification of solvent and ligand (access and egress) channels (eg. Schleinkofer et al. [2005]). Recent MD simulation studies have also demonstrated ligand cooperativity and allosteric binding sites associated with certain CYP enzymes (Li et al. [2011], Bren and Oostenbrink [2012]). This demonstrates the potential of MD simulation for understanding drug-drug interactions arising from CYP enzyme inhibition or activation at an atomic level. Moreover, MD simulations can be used to investigate altered conformational states of CYP proteins that result from genetic polymorphism.

Drug Discovery

In the recent years, the ability to design drugs rationally based on the structures of the protein targets has been a successful strategy in drug discovery. The so-called structure based drug design (SBDD) approach involves the design of molecules that can specifically bind to the key structural domains (binding pocket(s)) of the protein target (e.g. Varney et al. [1992], Dorsey et al. [1994]). The aim is to modulate the activity of the protein in order to alleviate disease progression. For instance, designing molecules that can maximize the complementary interaction at the active site of the target receptor can be a useful strategy in the structure guided inhibitor design (Varney et al. [1992], Dorsey et al. [1994]). However, recent studies have shown the importance of MD simulation to investigate the biomolecular flexibility associated with ligand recognition (Nair et al. [2011], Nair et al. [2012]). Studying the flexibility of the target receptor would thus permit the improved design of drugs over the simplistic lock and key conceptualization of the static receptor. Also, simulations allow the exploration of additional druggable sites (cryptic or allosteric) on the target receptor that is not evident from experimental (e.g. X-ray) structures (Frembgen-Kesner and Elcock [2006]). An example is the identification of a novel binding trench in HIV-integrase through dynamic simulation (Schames et al. [2004]). This led to the discovery of raltegravir, the first of a new class of anti-HIV (integrase inhibitor) drugs (Summa et al. [2008]). Similarly, a novel transiently open binding pocket was identified recently using MD simulation of the p53 protein, which has potential application for the design of novel anticancer drugs (Wassman et al. [2013]). Recent MD simulations have also shown the likelihood of identifying allosteric binding sites in the human ?1 and ?2 adrenergic receptors (Ivetac and McCammon [2010]).


There has been a dramatic increase in our understanding of disease states and therapeutic targets over the last two decades. With the current bioinformatics applications and sequencing data it is likely that the number of putative drug targets will continue to increase in the coming years, as in the case of G-protein coupled receptors. With increased computing power and continued developments in the efficiency of simulation codes and faster algorithms, the future of in silico approaches is promising. Molecular dynamics simulations are likely to play an increasingly important role for understanding the structure function relationships of pharmacological targets and in the development of novel therapeutics.

Authors? contributions

PCN and JOM wrote the manuscript. Both authors read and approved the final manuscript.



PCN acknowledges Flinders University for Research Fellowship.

Authors’ Affiliations

Department of Clinical Pharmacology, Flinders University School of Medicine


  1. Bren U, Oostenbrink C: Cytochrome P450 3A4 inhibition by ketoconazole. J Chem Inf Model 2012, 52: 1573–1582. 10.1021/ci300118xView ArticlePubMedGoogle Scholar
  2. Brooks BR, Brooks CL 3rd, Mackerell AD Jr, Nilsson L, Petrella RJ, Roux B, Won Y, Archontis G, Bartels C, Boresch S, Caflisch A, Caves L, Cui Q, Dinner AR, Feig M, Fischer S, Gao J, Hodoscek M, Im W, Kuczera K, Lazaridis T, Ma J, Ovchinnikov V, Paci E, Pastor RW, Post CB, Pu JZ, Schaefer M, Tidor B, Venable RM, et al.: CHARMM: the biomolecular simulation program. J Comput Chem 2009, 30: 1545–1614. 10.1002/jcc.21287View ArticlePubMedPubMed CentralGoogle Scholar
  3. Case DA, Cheatham TE III, Darden T, Gohlke H, Luo R, Merz KM Jr, Onufriev A, Simmerling C, Wang B, Woods RJ: The Amber biomolecular simulation programs. J Comput Chem 2005, 26: 1668–1688. 10.1002/jcc.20290View ArticlePubMedPubMed CentralGoogle Scholar
  4. Cornell WD, Cieplak P, Bayly CI, Gould IR, Merz KM Jr, Ferguson DM, Spellmeyer DC, Fox T, Caldwell JW, Kollman PA: A Second Generation Force Field for the Simulation of Proteins, Nucleic Acids, and Organic Molecules. J Am Chem Soc 1995, 117: 5179–5197. 10.1021/ja00124a002View ArticleGoogle Scholar
  5. Dorsey BD, Levin RB, McDaniel SL, Vacca JP, Guare JP, Darke PL, Zugay JA, Emini EA, Schleif WA, Quintero JC, Lin JH, Chen IW, Holloway MK, Fitzgerald PMD, Axel MG, Ostovic D, Anderson PS, Huf JR: L-735,524: The Design of a Potent and Orally Bioavailable HIV Protease Inhibitor. J Med Chem 1994, 37: 3443–3451. 10.1021/jm00047a001View ArticlePubMedGoogle Scholar
  6. Dror RO, Dirks RM, Grossman JP, Xu H, Shaw DE: Biomolecular simulation: a computational microscope for molecular biology. Annu Rev Biophys 2012, 41: 429–452. 10.1146/annurev-biophys-042910-155245View ArticlePubMedGoogle Scholar
  7. Ekroos M, Sj?gren T: Structural basis for ligand promiscuity in cytochrome P450 3A4. Proc Natl Acad Sci U S A 2006, 103: 13682–13687. 10.1073/pnas.0603236103View ArticlePubMedPubMed CentralGoogle Scholar
  8. Frembgen-Kesner T, Elcock AH: Computational sampling of a cryptic drug binding site in a protein receptor: explicit solvent molecular dynamics and inhibitor docking to p38 MAP kinase. J Mol Biol 2006, 359: 202–214. 10.1016/j.jmb.2006.03.021View ArticlePubMedGoogle Scholar
  9. Hendrychova T, Berka K, Navratilova V, Anzenbacher P, Otyepka M: Dynamics and hydration of the active sites of mammalian cytochromes P450 probed by molecular dynamics simulations. Curr Drug Metab 2012, 13: 177–189. 10.2174/138920012798918408View ArticlePubMedGoogle Scholar
  10. Ivetac A, McCammon JA: Mapping the druggable allosteric space of G-protein coupled receptors: a fragment-based molecular dynamics approach. Chem Biol Drug Des 2010, 76: 201–217.PubMedPubMed CentralGoogle Scholar
  11. Kohlhoff KJ, Shukla D, Lawrenz M, Bowman GR, Konerding DE, Belov D, Altman RB, Pande VS: Cloud-based simulations on Google Exacycle reveal ligand modulation of GPCR activation pathways. Nat Chem 2014, 6: 15–21. 10.1038/nchem.1821View ArticlePubMedPubMed CentralGoogle Scholar
  12. Li J, Wei DQ, Wang JF, Li YX: Negative cooperativity mechanism of human CYP2E1 inferred from molecular dynamics simulations. J Chem Inf Model 2011, 51: 3217–3225. 10.1021/ci2004016View ArticlePubMedGoogle Scholar
  13. MacKerell AD, Bashford D, Bellott M, Dunbrack RL, Evanseck JD, Field MJ, Fischer S, Gao J, Guo H, Ha S, Joseph-McCarthy D, Kuchnir L, Kuczera K, Lau FT, Mattos C, Michnick S, Ngo T, Nguyen DT, Prodhom B, Reiher WE, Roux B, Schlenkrich M, Smith JC, Stote R, Straub J, Watanabe M, Wi?rkiewicz-Kuczera J, Yin D, Karplus M: All-atom empirical potential for molecular modeling and dynamics studies of proteins. J Phys Chem B 1998, 102: 3586–3616. 10.1021/jp973084fView ArticlePubMedGoogle Scholar
  14. McCammon JA, Gelin BR, Karplus M: Dynamics of folded proteins. Nature 1977, 267: 585–590. 10.1038/267585a0View ArticlePubMedGoogle Scholar
  15. Miners JO, Birkett DJ: Cytochrome P4502C9: an enzyme of major importance in human drug metabolism. Br J Clin Pharmacol 1998, 45: 525–538. 10.1046/j.1365-2125.1998.00721.xView ArticlePubMedPubMed CentralGoogle Scholar
  16. Miners JO, Smith KJ, Robson RA, McManus ME, Veronese ME, Birkett DJ: Tolbutamide hydroxylation by human liver microsomes. Kinetic characterisation and relationship to other cytochrome P-450 dependent xenobiotic oxidations. Biochem Pharmacol 1988, 37: 1137–1144. 10.1016/0006-2952(88)90522-9View ArticlePubMedGoogle Scholar
  17. Nair PC, Malde AK, Mark AE: Using theory to reconcile experiment: the structural and thermodynamic basis of ligand recognition by Phenylethanolamine N-Methyl Transferase (PNMT). J Chem Theory Comput 2011, 7: 1458–1468. 10.1021/ct1007229View ArticlePubMedGoogle Scholar
  18. Nair PC, Malde AK, Drinkwater N, Mark AE: Missing fragments: detecting cooperative binding in fragment-based drug design. ACS Med Chem Lett 2012, 3: 322–326. 10.1021/ml300015uView ArticlePubMedPubMed CentralGoogle Scholar
  19. Oostenbrink C, Villa A, Mark AE, van Gunsteren WF: A biomolecular force field based on the free enthalpy of hydration and solvation: the GROMOS force-field parameter sets 53A5 and 53A6. J Comput Chem 2004, 25: 1656–1676. 10.1002/jcc.20090View ArticlePubMedGoogle Scholar
  20. Phillips JC, Braun R, Wang W, Gumbart J, Tajkhorshid E, Villa E, Chipot C, Skeel RD, Kal? L, Schulten K: Scalable molecular dynamics with NAMD. J Comput Chem 2005, 26: 1781–1802. 10.1002/jcc.20289View ArticlePubMedPubMed CentralGoogle Scholar
  21. Schames JR, Henchman RH, Siegel JS, Sotriffer CA, Ni H, McCammon JA: Discovery of a novel binding trench in HIV integrase. J Med Chem 2004, 47: 1879–1881. 10.1021/jm0341913View ArticlePubMedGoogle Scholar
  22. Schleinkofer K, Sudarko WPJ, L?demann SK, Wade RC: Do mammalian cytochrome P450s show multiple ligand access pathways and ligand channelling? EMBO Rep 2005, 6: 584–589. 10.1038/sj.embor.7400420View ArticlePubMedPubMed CentralGoogle Scholar
  23. Summa V, Petrocchi A, Bonelli F, Crescenzi B, Donghi M, Ferrara M, Fiore F, Gardelli C, Gonzalez Paz O, Hazuda DJ, Jones P, Kinzel O, Laufer R, Monteagudo E, Muraglia E, Nizi E, Orvieto F, Pace P, Pescatore G, Scarpelli R, Stillmock K, Witmer MV, Rowley M: Discovery of raltegravir, a potent selective HIV-integrase inhibitor for the treatment of HIV-AIDS infection. J Med Chem 2008, 51: 5843–5855. 10.1021/jm800245zView ArticlePubMedGoogle Scholar
  24. Van Der Spoel D, Lindahl E, Hess B, Groenhof G, Mark AE, Berendsen HJ: GROMACS: fast, flexible, and free. J Comput Chem 2005, 26: 1701–1718. 10.1002/jcc.20291View ArticlePubMedGoogle Scholar
  25. van Gunsteren WF, Berendsen HJC: Computer Simulation of Molecular Dynamics: Methodology, Applications and Perspectives in Chemistry Angew. Chem Int Ed Engl 1990, 29: 992–1023. 10.1002/anie.199009921View ArticleGoogle Scholar
  26. Varney MD, Marzoni GP, Palmer CL, Deal JG, Webber S, Welsh KM, Bacquet RJ, Bartlett CA, Morse CA, Booth CL, Herrmann SM, Howland EF, Ward RW, White J: Crystal-structure-based design and synthesis of benz[cd]indole-containing inhibitors of thymidylate synthase. J Med Chem 1992, 35: 663–676. 10.1021/jm00082a006View ArticlePubMedGoogle Scholar
  27. Wassman CD, Baronio R, Demir ?, Wallentine BD, Chen CK, Hall LV, Salehi F, Lin DW, Chung BP, Hatfield GW, Richard Chamberlin A, Luecke H, Lathrop RH, Kaiser P, Amaro RE: Computational identification of a transiently open L1/S3 pocket for reactivation of mutant p53. Nat Commun 2013, 4: 1407. 10.1038/ncomms2361View ArticlePubMedPubMed CentralGoogle Scholar
  28. Yano JK, Wester MR, Schoch GA, Griffin KJ, Stout CD, Johnson EF: The structure of human microsomal cytochrome P450 3A4 determined by X-ray crystallography to 2.05-A resolution. J Biol Chem 2004, 279: 38091–38094. 10.1074/jbc.C400293200View ArticlePubMedGoogle Scholar


© Nair and Miners; licensee Springer. 2014

This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited.