Characterizing protein motions from structure

https://doi.org/10.1016/j.jmgm.2011.08.004Get rights and content

Abstract

To clarify the extent structure plays in determining protein dynamics, a comparative study is made using three models that characterize native state dynamics of single domain proteins starting from known structures taken from four distinct SCOP classifications. A geometrical simulation using the framework rigidity optimized dynamics algorithm (FRODA) based on rigid cluster decomposition is compared to the commonly employed elastic network model (specifically the Anisotropic Network Model ANM) and molecular dynamics (MD) simulation. The essential dynamics are quantified by a mode subspace constructed from ANM and a principal component analysis (PCA) on FRODA and MD trajectories. Aggregate conformational ensembles are constructed to provide a basis for quantitative comparisons between FRODA runs using different parameter settings to critically assess how the predictions of essential dynamics depend on a priori arbitrary user-defined distance constraint rules. We established a range of physicality for these parameters. Surprisingly, FRODA maintains greater intra-consistent results than obtained from MD trajectories, comparable to ANM. Additionally, a mode subspace is constructed from PCA on an exemplar set of myoglobin structures from the Protein Data Bank. Significant overlap across the three model subspaces and the experimentally derived subspace is found. While FRODA provides the most robust sampling and characterization of the native basin, all three models give similar dynamical information of a native state, further demonstrating that structure is the key determinant of dynamics.

Highlights

► Aggregate conformational ensembles provide a basis for quantitative comparisons. ► Geometric simulation using FRODA efficiently characterizes native state dynamics. ► Robust sampling is obtained in ANM and FRODA, but not using MD. ► Underlying structure plays the dominant role in determining essential dynamics.

Section snippets

Introduction:

The protein data bank [1], [2] (PDB) (www.pdb.org) is a repository of protein structures that continues to grow on a daily basis containing tens of thousands of structures derived from X-ray and nuclear diffraction and NMR. A key component of protein science is to generate an ensemble of conformations when provided a static structure in order to identify essential motions important to function. Molecular Dynamics (MD) [3] is a model that implements a force field to represent interactions of a

Anisotropic Network Model (ANM)

ANM calculations were done using the Anisotropic Network Model web server.

All runs were performed online at http://ignmtest.ccbb.pitt.edu/cgi-bin/anm/.

Each analysis used a distance cutoff of 15 Å and a weighted C–C distance of 2.5 Å.

Molecular dynamics (MD)

MD trajectories were downloaded [16] from www.Dynameomics.org.

The methodology used to generate the trajectories is available at the same URL.

Geometrical simulation (FIRST/FRODA)

FRODA trajectories were created using FIRST/FRODA version 6.2. Software downloaded [13] from http://flexweb.asu.edu/.

For each

The dynamical models and essential dynamics

FIRST uses a set of parameters that determine how constraints are identified, which is ultimately responsible for outcomes in determining the number of iDOF and the predicted rigid and flexible regions of a protein. Based on the RCD, a geometric simulation using FRODA is very efficient. The advantage of FIRST/FRODA is that the generation of output structures is by some comparisons four orders of magnitude faster than MD. However, this tremendous gain in speed comes at the price of

Conclusions

The existence for a range of physicality using FRODA has been demonstrated in this work for the first time. For the four proteins studied here: We established that the default settings for hydrophobic tethers (rule H3) combined with a H-bond energy cutoff between −1 kcal/mol to −3 kcal/mol is robust. For much larger proteins, the H-bond energy cutoff range may shift slightly lower because the decrease in surface to volume in larger proteins gives slightly higher density of H-bonds. More

Acknowledgements

We wish to thank Dan Farrell and Mike Thorpe for their support of the FRODA software, especially in regards to keeping us current with new features as they are added. This work has been supported in part by NIH (GM073082) and a subcontract from Pennsylvania State University through NIH (HL093531).

References (41)

  • S. Fong et al.

    Structure and stability of an immunoglobulin superfamily domain from twitchin, a muscle protein of the nematode Caenorhabditis elegans

    J. Mol. Biol.

    (1996)
  • S. Vijay-Kumar et al.

    Structure of ubiquitin refined at 1.8 A resolution

    J. Mol. Biol.

    (1987)
  • D.R. Livesay et al.

    A flexible approach for understanding protein stability

    FEBS Lett.

    (2004)
  • D.J. Jacobs et al.

    Elucidating quantitative stability-flexibility relationships within thioredoxin and its fragments using a distance constraint model

    J. Mol. Biol.

    (2006)
  • P.A. Absil et al.

    On the largest principal angle between random subspaces

    Lin. Algeb. Appl.

    (2006)
  • D.J. Jacobs et al.

    Kinetics and thermodynamics of the rate-limiting conformational change in the actomyosin V mechanochemical cycle

    J. Mol. Biol.

    (2011 Apr 15)
  • H.M. Berman et al.

    The protein data bank

    Nucl. Acids Res.

    (2000)
  • J.A. McCammon et al.

    Dynamics of folded proteins

    Nature

    (1997)
  • M.A. Balsera et al.

    Principal component analysis and long time protein dynamics

    J. Phys. Chem.

    (1996)
  • M.M. Tirion

    Large amplitude elastic motions in proteins from a single-parameter, atomic analysis

    Phys. Rev. Lett.

    (1996)
  • Cited by (21)

    • Dynamic rotation featured translocations of human serum albumin with a conical glass nanopore

      2022, Journal of Electroanalytical Chemistry
      Citation Excerpt :

      In the natural state, proteins are in a tertiary structure and participate in life activities. Therefore, to study the mechanism of action of proteins in life activities, it is necessary to characterize or detect the structure of protein molecules to reveal the relationship between protein structure and function [1–6]. Due to the complexity and diversity of structures, higher requirements are placed on the characterization or detection of single-molecule proteins.

    • Constrained geometric simulation of the nicotinic acetylcholine receptor

      2014, Journal of Molecular Graphics and Modelling
      Citation Excerpt :

      The principal component subspaces spanned by froda simulations have been shown to be very robust with respect to the chosen value of Ecut [22]. For froda dynamics, a step length of 0.1 Å [22] was employed. Sixteen simulations, each sampling 275,000 configurations, were run starting from the open-channel structure, and sixteen were run starting from the closed-channel structure.

    • Switch II mutants reveal coupling between the nucleotide- and actin-binding regions in myosin v

      2012, Biophysical Journal
      Citation Excerpt :

      We generated all-atom trajectories using the Floppy Inclusion and Rigid Substructure Topology (FIRST)/Framework Rigidity Optimized Dynamic Algorithm (FRODA) software (35), version 6.2. The geometrical simulation paradigm (36,37) represents the molecular structure as a distance constraint network (36–38), which is used to explore accessible motions at the all-atom level of resolution. The FRODA simulations performed in this work mirror those of our previous study (22), with the exception of the starting structures.

    View all citing articles on Scopus
    View full text