Quantum Mechanical Insights into Biological Processes at the Electronic Level - Computational Modeling of Biological Systems - page 134

about the dynamics of the entire protein being a player in catalysis. Undoubtedly,
nature constructs large proteins for a number of possible reasons, including efficient
substrate binding and product release, chemical security and preservation of the
architecture of the active site, and the specific overall protein properties. However,
the motion of bulky protein parts is a process that occurs on a much slower time
scale than does the catalyzed reaction. Therefore, as long as the conformation of
the protein does not change dramatically from its crystal structure upon substrate
binding, consideration of the catalytic mechanism on a small model is not a bad
idea. As a payoff, ab initio calculations yield fairly accurate activation barriers
and molecular properties of the reacting system, which can be compared to the
experiment.
As an example, consider the mechanistic study of the enzyme urease performed
using UDFT [ 16 ]. Urease is a di-Ni enzyme that catalyzes the hydrolysis of urea,
and many things about the reaction mechanism, such as the role of the second Ni
center, the protonation state of the bridging water molecule, the group playing the
role of the nucleophile, etc. are still unclear. The model complex prepared by the
authors included the two Ni centers, truncated amino acids that bind them, i.e., the
immediate coordination shell of the metal ions, and the substrate (Fig. 2 ). First, the
geometry of the active site extracted from the crystal structure was optimized, and
the ground state multiplicity was determined. It appeared that the lowest energy
state is a quintet. The catalytic mechanism was then explicated using UB3LYP.
The initial complex between urea and the active site, and all intermediates on the
reaction path were optimized, and vibrational frequency calculations confirmed
that they are true minima on the PES. The transition states were also found via
geometry optimization to the saddle point between the two minima. All relevant
transition state structures were confirmed to be true saddle points, by calculating
their vibrational frequencies, and making sure that in each case there is only
one imaginary frequency corresponding to the displacements along the reaction
coordinate. It was found that there are two competing modes of binding of urea
to the active site: bidentate (Fig. 2 ) and monodentate (Fig. 3 ). In both complexes,
urea can get hydrolyzed by the active site, and the calculated energetics of the two
paths renders these mechanisms competitive.
In a similar spirit, Solomon and coworkers [ 17 ] performed a systematic and re-
markably exhaustive UDFT study of chemically possible peroxo-type intermediates
occurring in the nonheme di-iron enzyme class Ia ribonucleotide reductase (RNR).
This enzyme is responsible for the oxygen atom removal from RNA building
blocks, ribonucleoside diphosphates, to yield corresponding DNA building blocks,
deoxyribonucleoside diphosphates, by RNR. Class I RNRs contain two iron atoms
in one part of the protein, the R2 subunit, which is separated from the catalytic site
in the R1 subunit where the radical chemistry involving the ribonucleotide occurs.
The study was conducted on the R2 subunit using spectroscopically calibrated
density functional computations of equilibrium structures. Fe-O and O-O stretch
frequencies, Mossbauer isomer shifts, absorption spectra, J-coupling constants,
electron affinities, and free energies of O 2 and proton or water binding were
presented for a series of possible intermediates. The study explored how water or