QM/MM tutorial

QM/MM calculations on a Diels-Alder antibody catalyst.

IV. Optimization of product, reactant and transition state geometries in the fully solvated protein, using Linear Transit in gromacs


We have seen that in water the energy of the transition state is lower that in vacuo, with respect to the reactants. Now we will calculate the energy curve in the protein to see the effect of the protein environment on the reaction.

QM/MM subdivision

The fully solvated protein system is too large for even a semi-empirical QM calculation. Therefore, we resort to a QM/MM description of the system. The way we split up the system in a small QM part and a much bigger MM part, is shown in figure 7. The QM part consists of the same atoms as before and is again described at the semi-empirical PM3 level of theory. The remainder of the system, consisting of the tail part (-R, figure 1) of the reactants, the protein, the water molecules and the chloride ions, is modeled with the GROMOS96 forcefield.

Figure 7. Division of the system in a QM subsystem and an MM subsystem. The QM subsystem is described at the semi-empirical QM level,
while the remainder of the system, consisting of the reactants-aliphatic tail, protein, water molecules and ions is modeled with the GROMOS96

In this part of the tutorial we are going to perform a third Linear Transit calculation, but this time, the reactants are fully solvated. With the QM/MM subdivision shown in figure 7, we will perform the third and last Linear Transit computation. The details of how to perform such a calculation in gromacs can be reviewed here.

back to top

Finding product, reactant and transition state geometries in the fully solvated protein

The starting structure for this calculation is the x-ray model of the 1CE catalytic antibody by Xu et al.. Remember that in the x-ray model, the -R group of the analogue was not resolved. So we need to add it ourselves. We take the modified transition-state analogue of part II of this tutorial and fit it onto the analogue in the x-ray model. After the fit, we minimize the tail part, keeping the rest of the protein fixed.

Then we place this modified protein model in a periodic box, fill that box with water and equilibrate the water. Subsequently, we add 6 Cl- ions to compensate the overall net charge of -6 on the protein and equilibrate again. The procedure of preparing the system for the QM/MM geometry optimization is straightforward, but time-consuming. Therefore, we skip fitting and equilibrating and use the result (confin.gro) instead. An outline of the preparation is avalaible here.

And here are the scripts we use this time to perform the Linear Transit:

back to top


We have now found the geometries and energies of the transition state, the reactant state and the prodcut state of the Diels-Alder cycloaddition in the active site of the catalytic Diels-Alderase antibody. Table 4 lists the total QM/MM energies of these geomtries.

Table 4. Energies of the reactant,
transition state and product geom-
etries in the solvated protein at the
PM3/GROMOS96 QM/MM level.
The last column lists the energy
differences with respect to the
reactant state.

E (kJ/mol)ΔE (kJ/mol)
Trans. St.-1102.9113.5

Next:V. Conclusion, Discussion, and Outlook
Previous:III. Optimization of the product, reactant and transition state geometries in water, using Linear Transit in gromacs

back to top

updated 07/09/04