Chemistry Reference
In-Depth Information
During the time, a few mechanisms for protein-ligand and protein-protein
recognition and association have been proposed. The oldest and simplest is the “lock-
and-key” mechanism (Fischer
1894
) that is based on and emphasizes the chemical
and shape complementarity of the involved molecules. This mechanism could not
explain why a protein may bind more than one substrate and the scientists searched
for other mechanisms. The induced fit mechanism has been pro-posed by Koshland
(
1958
) and illustrated the importance of flexibility of both protein and ligand for
optimization of binding and interaction. For allosteric proteins the Monod-Wyman-
Changeux mechanism has been proposed (Monod et al.
1965
). This mechanism
considers that allosteric proteins present two conformational states in the unbound
form, tense and relaxed, and the ligand binding causes the transition to the relaxed
state. It also explains the positive cooperativity, but it cannot explain the negative
cooperativity. In order to account for both positive and negative cooperativity, the
Koshland-Nemethy-Filmer mechanism has been proposed (Koshland et al.
1966
)
considering that oligomeric proteins have some subunits in the weak binding state
and others in the strong binding state. Another proposed mechanism of cooperative
binding is the dynamic population shift mechanism (Freire
1999
). It considers that
protein exhibits a population of native conformations and upon ligand binding the
probability distribution of this ensemble is redistributed changing the stability of
certain residues and propagating a conformational change at specific residues.
The conformational selection mechanism (Ma et al.
1999
; Tsai et al.
1999
) is based
on the concept of the funnel-like free energy landscape of protein folding (Karplus
1997
) considering that the native state of a protein is an ensemble of co-existing
conformers one of them binding the ligand. This mechanism has been validated by
several experimental studies concerning the protein-ligand interactions (Berger et al.
1999
; Foote and Milstein
1994
; James et al.
2003
; Goh et al.
2004
;Levyetal.
2004
;
Tobi and Bahar
2005
; Kuzu et al.
2012
).
Complete deciphering and understanding of dynamic principles governing
protein-ligand and protein-protein interactions is a challenge both for description
and characterization of biological pathways and for designing of novel and efficient
therapeutic agents. Also, in protein science, there is a large gap between the number
of proteins with well characterized structures and biological functions (or malfunc-
tions) and the proteins with unknown structures and/or functions. From these points
of view, the computational approaches concerning the prediction and/or characteri-
zation of proteins structures, surfaces and interactions are increasingly needed and
developed in life sciences.
Protein structures are better conserved than protein sequences and also protein
structures contains more biological significant information. The biological functions
of proteins rely on their interactions with other biomolecules, the wide variety of
interactions being due to chemical and structural diversity observed at molecular
surfaces level. Thus, to decipher biological processes at molecular level is crucial
to accurately define the nature and shape of protein surfaces. It means to describe
the surface in terms of atoms, residues, and surface curvature, patches, clefts and
concavities or convexities and in order to do it the three-dimensional structure of
protein is needed. The atomic structures of proteins and their complexes are publicly
