MD in various solvents. The Fortunate summary
Dear Netters,
After the somewhat provocative summary I posted a week or so ago, the
good ol' ccl traditions emerged.
I received useful informations, this is the summary.
Many thanks for your help
Best regards,
luigi
--------------------------------------------------------------------------
ORIGINAL MESSAGE
>> I would like to receive informations on comparisons of MD simulations
>> in vacuo with simulations done by using explicit solvents.
>> In particular, how compare in vacuo simulations with those performed
>> in CHCl3 ?
>>
>> Any reference is appreciated.
>
>
>Well, I received several requests for a summary, it means that some
>peoples around the world are interested to the subject, but...
>
> !!! NO INFORMATIONS AT ALL !!!
>
>Like nobody has ever done such comparisons. Unbelivable.
--------------------------------------------------------------------------
From: vkitzing # - at - # sunny.mpimf-heidelberg.mpg.de (Eberhard von Kitzing)
Andrea Amadei, Antonius B.M. Linssen and Herman J.C. Berendsen (1993).
"Essential dynamics of proteins." Proteins-Structure Function and
Genetics
17(4) 412-425.
======================================================================
From: konrad.koehler # - at - # karobio.se (Konrad Koehler)
Please find below an excerpt from our chapter:
1) Koehler, K. F.; Rao, S. N.; Snyder, J. P. Modeling Drug?Receptor
Interactions.; in Guidebook on Molecular Modeling in Drug Design ;
Cohen, N. C., ed.; Academic Press, Inc.: San Diego, 1996, pp 234-336.
The references all deal with a comparision of vacuum vs. water.
As you may know, Clark Still's group at Columbia has done some
comparisons's with their GB/SA solvent continuum model for chloroform
vs. vacuum. I am not aware of any comparisons between explicit
chloroform vs. vacuum. If you are interested references of the use of
GB/SA chloroform calculations, I can send you these as well.
C. Macromolecular Conformation and Ligand Binding
Just as with small molecules, accounting for solvent effects is
crucial to the prediction of protein conformation and dynamics. Vacuum
molecular mechanics and dynamics calculations produce structures which
by a variety of measures are unrealistic [1]. Polar side-chains on the
surface of proteins are often extended to maximize their interaction
with water. In contrast, in vacuo minimization cause these side-chains
to fold-back onto the protein. This is partially a consequence of
electrostatic attractions that are not being shielded by solvent.
Another contributing factor is "van der Waals collapse" of the protein
structure during in vacuo optimization. In the gas phase, surface
atoms are only in contact with atoms on one face and therefore are
"pulled" toward the center of the protein. This causes protein
structure as a whole to be denser than observed experimentally.
In order to produce satisfactory protein structures in molecular
dynamics simulations, either an explicit solvent bath [2] or a solvent
continuum model [3,4] must be used. When using atomic solvation
parameters, the balance between solvation free energy and molecular
mechanics energy is critical [5,6]. If these two terms are carefully
balanced, then the simulations have been found to not seriously perturb
the structure of the protein from its crystallographic starting point
[7]. One drawback to this approach is that it is assumed that if a
protein atom is buried within a protein, it is in a hydrophobic
environment. However there is of course a great deal of variability in
the internal environments of proteins. Hence more accurate ASP's which
take into account the local environment of buried protein atoms have
been developed [8].
Consideration of solvent is also crucial for evaluating the energetic
differences between various conformation of macromolecules. For
example molecular mechanics energy evaluations which include a
solvation correction term are able to distinguish between native and
incorrectly folded protein structures [5,9]. The corresponding in
vacuo calculations were unable to make this distinction. Similarly the
crystallographic conformation of protein loops could only be
successfully predicted if solvation effects were included in the energy
evaluation [10].
Successful prediction of ligand binding affinities requires accounting
for solvation effects. In free energy perturbation predictions of
relative ligand affinities, explicit solvent is generally used (section
II.D) whereas empirical approaches generally rely on solvent continuum
models (section II.C.3). In an example of the latter strategy, 11
treat the proteinligand system as a three-dimensional grid (with
approximately 2 Å resolution) and at each grid point, the steric (van
der Waals interactions) and electrostatic components of the potential
are calculated using the force field (CHARMm). Solvent effects are
accounted for by (a) the use of a high dielectric constant (e.g., 80
for water) and by (b) calculating electrostatic interactions by the
finite difference Poisson-Boltzmann (FDPB) method which incorporates
the effects of solvent ionic strength and the differing
polarizabilities of protein and solvent. At grid points within the
protein, a constant dielectric of 2 is employed. This technique was
applied to the study of diffusion of superoxide into the electric field
of superoxide dismutase and calculation of association constants as a
func-tion of ionic strength and amino acid modifications in the enzyme
active site [11]. The protein was treated as rigid with no
conforma-tional mobility relative to its X-ray crystallographic
structure. The study demonstrated that the electric field of the
enzyme enhanced association rates of the superoxide by factors
exceeding 30 as evidenced by the lower association constants for the
mutant enzymes in which the catalytically important arginine and lysine
residues were modeled in their neutral forms. These cationic residues
were thought to lower the magnitude of the negative electrostatic
potential barrier around the protein which carries an overall negative
charge. This observation was substantiated through simulations on
mutants in which two glutamates in the vicinity of the copper site were
altered to lysines resulting in higher association constant for the
superoxide anion.
This implicit solvation method has been successfully applied to the
determination of relative binding energies of substrate/protein
interac-tions for a series of ligands for the arabinose and sulfate
binding proteins [12] in a manner analogous to the superoxide-SOD
calculations. The success of this study is attributable in part to the
small structural perturbations in the ligands or proteins and that
electrostatics dominate the differences in interactions between the
various ligands and mutant proteins. It is not clear however how
problems associated with conformational mobility likely to accompany
larger structural variations of the protein and of the ligand will
affect the calculated electrostatic and steric potentials at grid
points close to the proteinsolvent interface and hence the calculated
binding energies.
1) Stouten, P. F. W.; Frömmel, C.; Nakamura, H.; Sander, C.; "An
Effective solvation term based on atomic occupancies for use in protein
simulations"; Molec. Simulation 1993, 10, 97120.
2) Levitt, M.; Sharon, R.; "Accurate simulation of protein dynamics in
solution"; Proc. Natl. Acad. Sci. U.S.A. 1988, 85, 7557-7561.
3) Solmajer, T.; Mehler, E. L.; "Electrostatic screening in molecular
dynamics simulations"; Protein Eng. 1991, 4, 911-917.
4) Arnold, G. E.; Ornstein, R. L.; "An evaluation of implicit and
explicit solvent model systems for the molecular dynamics simulation of
bacteriophage T4 lysozyme"; Proteins 1994, 18, 19-33.
5) Cregut, D.; Liautard, J.-P.; Chiche, L.; "Homology modelling of
annexin I: implicit solvation improves side-chain prediction and
combination of evaluation criteria allows recognition of different
types of conformational error."; Prot. Engin. 1994, 7, 1333-1344.
6) Schiffer, C. A.; Caldwell, J. W.; Stroud, R. M.; Kollman, P. A.;
"Inclusion of solvation free energy with molecular mechanics energy:
alanyl dipeptide as a test case"; Protein Sci. 1992, 1, 396-400.
7) Schiffer, C. A.; Caldwell, J. W.; Stroud, R. M.; Kollman, P. A.;
"Protein structure prediction with a combined solvation free
energymolecular mechanics force field."; Mol. Simulations 1993, 10,
121-149.
8) Delarue, M.; Koehl, P.; "Atomic environment energies in proteins
defined from statistics of accessible and contact surface areas"; J.
Mol. Biol. 1995, 249, 675-690.
9) Novotny, J.; Rashin, A. A.; Bruccoleri, R. E.; "Criteria that
discriminate between native proteins and incorrectly folded models";
Proteins 1988, 4, 19-30.
10) Smith, K. C.; Honig, B.; "Evaluation of the conformational free
energies of loops in proteins."; Proteins 1994, 18, 119-132.
11) Sharp, K.; Fine, R.; Honig, B.; "Computer simulations of the
diffusion of a substrate to an active site of an enzyme"; Science 1987,
236, 1460-1463.
12) Shen, J.; Quiocho, F. A.; "Calculations of binding energy
differences for receptorligand systems using the Poisson-Boltzmann
method."; J. Comput. Chem. 1995, 16, 445-448.
--------------------------------------------------------------------------
From: Maria Turner <turnerm # - at - # CRHSC.UMontreal.CA>
I know it's not exactly what you're looking for but you might want to
look at "Computer Modeling Studies of G Protein Coupled Receptors"
Kontoyianni and Lybrand, Med. Chem. Res. (1993) 3:407-418 for a
comparison of MD in vacuo, in vacuo plus lipid bilayer, in vacuo with
lipid bilayer and continuum model for solvent effects.
--------------------------------------------------------------------------
From: Pieter Stouten <stoutepf # - at - # carbon.dmpc.com>
This is the only work I know of: S. Yun-yu, W. Lu & W.F. Van Gunsteren,
"On
the approximation of solvent effects on the conformation and dynamics of
cyclosporin A by stochastic dynamics simulation techniques", Molec.
Simulation 1988, 1, 369-388. It compares water MD, CCl4 MD and vacuum SD.
--------------------------------------------------------------------------
From: "Vijayakumar, Sundararajan" <sundararajan_vijayakumar # - at
- # merck.com>
I would like to add that there are a number of papers from Prof. David
Beveridge's lab at Wesleyan University examining the effects of
Implicit/Explicit Solvation and in vacuo on protein and DNA systems,
using the Gromos Force Field. The group has also carried out studies
comparing different solvation models and counter ion treatment for
oligonucleotides employing AMBER and GROMOS force fields. I don't have
the references handy right now but a simple literature search should
find some of the recent papers!
--------------------------------------------------------------------------
Let me add
Lauterbach M. Wippf G. In: "Physical Supramolecular Chemistry"
NATO ASI series (Eds. Echegoyen L., Kaifer A.) Kluwer Academic Publisher
Dordecht 1996, 65-102
Fraternali F. Van Gunsteren W.F. An efficient mean solvation force model
for use in molecular synamics simulations of proteins in aqueous solution
J. Mol. Biol. 1996, 256, 939
-------------------------------------------------------------------------
--------------------------------------------------------------------------------
| Dr. Luigi Cavallo |
| Department Of Chemistry Fax : ++39-81-5527771 |
| University Of Naples Ph : ++39-81-5476535 |
| Via Mezzocannone 4 Email : cavallo # - at - #
chemna.dichi.unina.it |
| I-80134 Naples, ITALY |
--------------------------------------------------------------------------------