16 Molecular Modeling Method Applications …
321
16.3 Preparation of the Molecular Structure of EDC
Molecules
The molecular structure of EDCs can be downloaded from various software or
databases such as PubChem (https://pubchem.ncbi.nlm.nih.gov/) and ZINC (http://
blaster.docking.org/zinc). For the compounds not contained in those sources, their
molecular structures must be drawn directly. Then, the possible ionization states can
be generated at normal physiological or experimental pH conditions. Lastly, their
molecular structures can be optimized by various methods.
16.4 Obtaining the EDC—Biomacromolecule Complexes
16.4.1 Performing Molecular Docking
As shown in Fig. 16.2, a key prerequisite for successful application of molecular modeling to probe the binding mechanism lies in obtaining the toxicant–target complex.
The complex is a conformation demonstrating where (binding site location) and how
(binding orientation) the small molecule bound within the LBD of a given macromolecule. Undoubtedly, the experimental methods like NMR spectroscopy [59, 60]
are the most accurate methods to investigate toxicant–target interactions. The time
and cost limitations of NMR precipitate the need for alternative technologies.
Molecular docking is a powerful alternative technology for predicting the toxicant–target complex [61–64]. Generally speaking, molecular docking places a given
ligand into the binding site of a given target. The molecular docking protocol usually
contains two components: a search algorithm and a scoring function [65]. The search
algorithms determine the precise ligand-binding site location and binding orientation, which also referred to as a “pose.” The scoring function addresses the question,
“how well does the ligand bind to the protein?” by ranking the pose. In the past
thirty years, many search algorithms and scoring functions were developed. Detailed
discussion about the search algorithms and scoring functions are reviewed elsewhere
[66, 67] and are not considered herein.
To date, more than 60 software and Web servers for docking are available [68],
e.g., Autodock [69] and GOLD [70]. Previous studies document that the predictive
performance of the docking tools varies greatly between targets [62, 68]. It is therefore
difficult to readily conclude which tool is more appropriate than others for a given
modeling system. Thus, each given docking tools should be validated before use. In
practice, the target/ligand crystal structures are typically utilized for validation. After
extracting the 3D crystal ligand structure, it will be docked back into the binding site
of that target. Lastly, the root-mean-square deviation (RMSD) between the heavyatom positions of the crystal ligand and that of the predicted ligand is calculated [71].
If the RMSD value is <2 Å, the selected docking tool is considered reliable [61, 63].
321
16.3 Preparation of the Molecular Structure of EDC
Molecules
The molecular structure of EDCs can be downloaded from various software or
databases such as PubChem (https://pubchem.ncbi.nlm.nih.gov/) and ZINC (http://
blaster.docking.org/zinc). For the compounds not contained in those sources, their
molecular structures must be drawn directly. Then, the possible ionization states can
be generated at normal physiological or experimental pH conditions. Lastly, their
molecular structures can be optimized by various methods.
16.4 Obtaining the EDC—Biomacromolecule Complexes
16.4.1 Performing Molecular Docking
As shown in Fig. 16.2, a key prerequisite for successful application of molecular modeling to probe the binding mechanism lies in obtaining the toxicant–target complex.
The complex is a conformation demonstrating where (binding site location) and how
(binding orientation) the small molecule bound within the LBD of a given macromolecule. Undoubtedly, the experimental methods like NMR spectroscopy [59, 60]
are the most accurate methods to investigate toxicant–target interactions. The time
and cost limitations of NMR precipitate the need for alternative technologies.
Molecular docking is a powerful alternative technology for predicting the toxicant–target complex [61–64]. Generally speaking, molecular docking places a given
ligand into the binding site of a given target. The molecular docking protocol usually
contains two components: a search algorithm and a scoring function [65]. The search
algorithms determine the precise ligand-binding site location and binding orientation, which also referred to as a “pose.” The scoring function addresses the question,
“how well does the ligand bind to the protein?” by ranking the pose. In the past
thirty years, many search algorithms and scoring functions were developed. Detailed
discussion about the search algorithms and scoring functions are reviewed elsewhere
[66, 67] and are not considered herein.
To date, more than 60 software and Web servers for docking are available [68],
e.g., Autodock [69] and GOLD [70]. Previous studies document that the predictive
performance of the docking tools varies greatly between targets [62, 68]. It is therefore
difficult to readily conclude which tool is more appropriate than others for a given
modeling system. Thus, each given docking tools should be validated before use. In
practice, the target/ligand crystal structures are typically utilized for validation. After
extracting the 3D crystal ligand structure, it will be docked back into the binding site
of that target. Lastly, the root-mean-square deviation (RMSD) between the heavyatom positions of the crystal ligand and that of the predicted ligand is calculated [71].
If the RMSD value is <2 Å, the selected docking tool is considered reliable [61, 63].