Limits...
Parameterization of highly charged metal ions using the 12-6-4 LJ-type nonbonded model in explicit water.

Li P, Song LF, Merz KM - J Phys Chem B (2014)

Bottom Line: In the present study, by treating the experimental hydration free energies and ion-oxygen distances of the first solvation shell as targets for our parametrization, we evaluated 12-6 LJ parameters for 18 M(III) and 6 M(IV) metal ions for three widely used water models (TIP3P, SPC/E, and TIP4PEW).The final parameters reproduced the target values with good accuracy, which is consistent with our previous experience using this potential.Finally, tests were performed on a protein system, and the obtained results validate the transferability of these nonbonded model parameters.

View Article: PubMed Central - PubMed

Affiliation: Department of Chemistry, Department of Biochemistry and Molecular Biology, Michigan State University , 578 S. Shaw Lane, East Lansing, Michigan 48824-1322, United States.

ABSTRACT
Highly charged metal ions act as catalytic centers and structural elements in a broad range of chemical complexes. The nonbonded model for metal ions is extensively used in molecular simulations due to its simple form, computational speed, and transferability. We have proposed and parametrized a 12-6-4 LJ (Lennard-Jones)-type nonbonded model for divalent metal ions in previous work, which showed a marked improvement over the 12-6 LJ nonbonded model. In the present study, by treating the experimental hydration free energies and ion-oxygen distances of the first solvation shell as targets for our parametrization, we evaluated 12-6 LJ parameters for 18 M(III) and 6 M(IV) metal ions for three widely used water models (TIP3P, SPC/E, and TIP4PEW). As expected, the interaction energy underestimation of the 12-6 LJ nonbonded model increases dramatically for the highly charged metal ions. We then parametrized the 12-6-4 LJ-type nonbonded model for these metal ions with the three water models. The final parameters reproduced the target values with good accuracy, which is consistent with our previous experience using this potential. Finally, tests were performed on a protein system, and the obtained results validate the transferability of these nonbonded model parameters.

Show MeSH
(a) HFE errors for the12-6 IOD and 12-6-4 parameter sets for M(II),M(III), and M(IV) metal ions. (b) IOD errors for the 12-6 HFE and12-6-4 parameter sets for the M(II), M(III), and M(IV) metal ions.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4306492&req=5

fig2a: (a) HFE errors for the12-6 IOD and 12-6-4 parameter sets for M(II),M(III), and M(IV) metal ions. (b) IOD errors for the 12-6 HFE and12-6-4 parameter sets for the M(II), M(III), and M(IV) metal ions.

Mentions: After initial parameterselection and subsequent fine-tuning, the final 12-6-4 parameterswere determined. The final optimized 12-6-4 parameters are given inTable 6 while the simulated HFE, IOD, and CNvalues are shown in Table SI.4. These parametersreproduce the experimental HFE values by ±1 kcal/mol and theIOD values by ±0.01 Å for the M(III) ions, while they reproducethe HFE values by ±2 kcal/mol and the IOD values by ±0.01Å for the M(IV) ions. Just as in the 12-6-4 parameter sets fordivalent metal ions, the Rmin/2 termsare similar between the three water models, while the C4 term for TIP4PEW water is generally largerthan for the other two water models for the same metal ion. This maydue to the smaller dipole of the TIP4PEW water model (2.32D) relative to the TIP3P (2.35 D) and SPC/E (2.35 D) water models.Figure 2 shows the accuracy comparison betweenthe 12-6-4 parameter set and the 12-6 parameter sets for divalent,trivalent, and tetravalent metal ions. We can see that there is significantimprovement in the accuracy using the 12-6-4 parameter set, whichis able to reproduce the experimental HFE and IOD values simultaneously.While for the 12-6 LJ nonbonded model, if you want to reproduce theexperimental HFE values, the error in the simulated IOD values wouldincrease along with the formal charge of the metal ions. Vice versa,if you simulate the IOD values using the 12-6 LJ nonbonded model,the error of the calculated HFE would increase markedly with an increasein the oxidation state of the metal ion in question.


Parameterization of highly charged metal ions using the 12-6-4 LJ-type nonbonded model in explicit water.

Li P, Song LF, Merz KM - J Phys Chem B (2014)

(a) HFE errors for the12-6 IOD and 12-6-4 parameter sets for M(II),M(III), and M(IV) metal ions. (b) IOD errors for the 12-6 HFE and12-6-4 parameter sets for the M(II), M(III), and M(IV) metal ions.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4306492&req=5

fig2a: (a) HFE errors for the12-6 IOD and 12-6-4 parameter sets for M(II),M(III), and M(IV) metal ions. (b) IOD errors for the 12-6 HFE and12-6-4 parameter sets for the M(II), M(III), and M(IV) metal ions.
Mentions: After initial parameterselection and subsequent fine-tuning, the final 12-6-4 parameterswere determined. The final optimized 12-6-4 parameters are given inTable 6 while the simulated HFE, IOD, and CNvalues are shown in Table SI.4. These parametersreproduce the experimental HFE values by ±1 kcal/mol and theIOD values by ±0.01 Å for the M(III) ions, while they reproducethe HFE values by ±2 kcal/mol and the IOD values by ±0.01Å for the M(IV) ions. Just as in the 12-6-4 parameter sets fordivalent metal ions, the Rmin/2 termsare similar between the three water models, while the C4 term for TIP4PEW water is generally largerthan for the other two water models for the same metal ion. This maydue to the smaller dipole of the TIP4PEW water model (2.32D) relative to the TIP3P (2.35 D) and SPC/E (2.35 D) water models.Figure 2 shows the accuracy comparison betweenthe 12-6-4 parameter set and the 12-6 parameter sets for divalent,trivalent, and tetravalent metal ions. We can see that there is significantimprovement in the accuracy using the 12-6-4 parameter set, whichis able to reproduce the experimental HFE and IOD values simultaneously.While for the 12-6 LJ nonbonded model, if you want to reproduce theexperimental HFE values, the error in the simulated IOD values wouldincrease along with the formal charge of the metal ions. Vice versa,if you simulate the IOD values using the 12-6 LJ nonbonded model,the error of the calculated HFE would increase markedly with an increasein the oxidation state of the metal ion in question.

Bottom Line: In the present study, by treating the experimental hydration free energies and ion-oxygen distances of the first solvation shell as targets for our parametrization, we evaluated 12-6 LJ parameters for 18 M(III) and 6 M(IV) metal ions for three widely used water models (TIP3P, SPC/E, and TIP4PEW).The final parameters reproduced the target values with good accuracy, which is consistent with our previous experience using this potential.Finally, tests were performed on a protein system, and the obtained results validate the transferability of these nonbonded model parameters.

View Article: PubMed Central - PubMed

Affiliation: Department of Chemistry, Department of Biochemistry and Molecular Biology, Michigan State University , 578 S. Shaw Lane, East Lansing, Michigan 48824-1322, United States.

ABSTRACT
Highly charged metal ions act as catalytic centers and structural elements in a broad range of chemical complexes. The nonbonded model for metal ions is extensively used in molecular simulations due to its simple form, computational speed, and transferability. We have proposed and parametrized a 12-6-4 LJ (Lennard-Jones)-type nonbonded model for divalent metal ions in previous work, which showed a marked improvement over the 12-6 LJ nonbonded model. In the present study, by treating the experimental hydration free energies and ion-oxygen distances of the first solvation shell as targets for our parametrization, we evaluated 12-6 LJ parameters for 18 M(III) and 6 M(IV) metal ions for three widely used water models (TIP3P, SPC/E, and TIP4PEW). As expected, the interaction energy underestimation of the 12-6 LJ nonbonded model increases dramatically for the highly charged metal ions. We then parametrized the 12-6-4 LJ-type nonbonded model for these metal ions with the three water models. The final parameters reproduced the target values with good accuracy, which is consistent with our previous experience using this potential. Finally, tests were performed on a protein system, and the obtained results validate the transferability of these nonbonded model parameters.

Show MeSH