Our research group develops and applies quantum mechanics/ molecular
mechanics (QM/MM) hybrid methods to investigate structure-function
relations and biochemical reactivity of biomacromolecules. Our goal is
to obtain a first principles understanding of structure and activity in
proteins and enzymes at the detailed molecular level. We base our
approach in that a large number of biological events, having a
localized molecular origin, require quantum detail and are therefore
computable, only within the framework of Quantum Mechanics. In the
long-term, this research aims to produce computational techniques
applicable to a large variety of biomacromolecules and biomaterials,
and to advance the connection between the detailed atomistic level and
macroscopic level.
Hybrid Methods
We are interested in understanding how proteins and enzymes work from
an atomic level. For that, we use quantum mechanics/molecular mechanics
(QM/MM) hybrid methods to investigate structure-function relations and
biochemical reactivity. QM/MM methods treat a reduced region of the
protein quantum mechanically (see figure on the below), and the rest of
the protein using molecular mechanics. One aspect of particular
interest to us is the incorporation of large scale polarization effects
in protein electrostatic potentials. For this we have developed a
computational protocol, Moving-Domain QM/MM, to obtain a
self-consistent polarization of electrostatic properties, such as
atomic charges, dipoles, polarizabilities, etc. Specifically, the
method provides a rigorous and efficient way to compute electrostatic
potentials in biomacromolecules.
The
basic strategy is to exploit the fact that the wave function in a QM/MM
calculation, provided that electronic embedding is used, is polarized
by the protein, and can be used to generate updated electrostatic
properties, such as atomic charges. By implementing a self-consistent
iteration scheme, these electrostatic properties, which are quantum
mechanically derived, can be updated for the entire protein. Moreover,
the protocol scales linearly with the size of the protein, making the
method specially attractive for large systems such as ribosomes.
Structure,
Reactivity and Spectroscopy of active sites
The computational tools that we use and develop are
ultimately ment to give a rigorous interpretation of experimental data
aimed to elucidate the relation between structure, reactivity and
function. Currently, we are interested in two systems, rhodopsin
proteins and vanadium haloperoxidases (VHPOs). Rhodopsin is a protein
found in the rod cells of the retina, and is where the first events in
vision occur. These early events in the vision process are yet to be
understood from a molecular level. Past effort has been focused in
modeling the very first event, in which the retinal chromophore
isomerizes after absorbing a photon. Our current interests focus in
modeling what happens after that initial step. In fact, the changes of
conformation and protonation around the active site that occurs after
isomerization remains an open problem.
The
structure fo the active site of visual rhodopsin after isomerization.
What happens next, in terms of
protonation states is not fully understood.
VHPOs
form a class of ensymes that catalyse the oxidation of halides (Cl-,
Br-, I-) in the presence of hydrogen peroxide. The active site of VHPOs
is paradigm of a hydrogen bound co-factor. Using
first principle calculations, our study aims to determine
the structure, energetics and spectroscopic properties of different
states during the catalytic cycle as well as the role of the different
residues within the active site of the enzyme. A full understanding of
the molecular mechanisms of these enzyme reactions requires quantum
detail. QM/MM methods are an excellent way to capture these details by
describing the active site very accurately, while including the
influence of the whole enzyme, and explicit solvents with a reasonable
computational cost. These computational studies will correlate a wealth
of experimental observations with structural and mechanistic aspects
governing the activity of these enzymes.
The
active site of vanadium chloroperoxidase.