Logo

Molecular Modeling & Drug Discovery

 

 13. BioActive Conformations:

Bioactive Conformations and
Pharmcophores

 

X-ray Structures of Complexes

Protein-ligand complex structures show bioactive conformations.

The ligands' bound state, or "binding mode," has biological activity.

Bioactive ligand conformations are utilized in ligand-based drug design to find more active molecules.

zxc

 

X-ray Structures of Complexes

To identify more active substances, bioactive ligand conformations are used.

Drug design based on ligands (LBVS)

  •  3D database search

  • Modeling/searching of pharmacophores

 

Pharmacophore

It is described as the spatial configuration of atoms or groups in a molecule that is essential for its bioactivity.

qwertyuiop

 

Pharmacophore Features

Consists of all inter-feature distances and three or four pharmacophore characteristics.

Six inter-feature distances and four characteristics define a 4-point pharmacophore.

Features are defined to take into consideration the chemical properties that various atoms and groups represent:

  • donors or acceptors of hydrogen bonds.

  • groupings that are negatively or positively charged.

  • hydrophobic molecules.

  • centers of aromatic systems or rings.

6 common pharmacophore characteristics

 

Structure-based Pharmacophore

Pharmacophore characteristics are chosen from X-ray structures to represent significant protein-ligand interactions.

Pharmacophore searching

Four features and six spacing between the features make up a four-point pharmacophore.

An encoded pharmacophore model searches 3D databases for more substances that share the same pharmacophore.

Potential candidates are database molecules with the same pharmacophore but distinct core structures (hits)

Inter-feature distances are encoded as short distance ranges when searching to prevent the search from being overdetermined (no hits)

 

Structure-based Drug Design

requires in-depth understanding of the three-dimensional ligand binding site.

attempts to represent ligands that are highly complementary in terms of shape, structure, and chemical makeup to the receptor site.

 

Pharmacophore-based Design

Find potential compounds by comparing them to a predetermined pharmacophore that "mimics" known actives.

The approach resembles 2D similarity searches based on molecular graphs.

While 2D pharmacophores can also be defined, the information content of bioactive conformations is higher.

 

Unknown Bioactive Conformation

Only a small portion of pharmacological targets have known bioactive conformations based on X-ray structures.

When using the pharmacophore idea, binding conformations of active chemicals must be modeled if they are unknown.

Pick at least two well-known active substances.

To sample low energy conformations, do a systematic conformational search with energy minimization.

 

Conformational Sampling

Look for low-energy conformations on a compound's potential energy surface.

Pick at least two well-known active substances.

Conduct a systematic search for conformations, sampling low-energy conformations.

Find the two active chemicals' sampled minimal energy conformations.

Choose every conformer that falls within a 5 kcal/mol energy range with respect to the minimum.

 

Strain Energy upon Binding

Global minimum conformations are often not those that are bioactive.

Strain energy is present in bound ligands as a result of (induced fit) binding.

cvbn

 

Pick at least two well-known active substances.

Sample low-energy conformations and conduct systematic conformational search.

Identify the conformation with the lowest energy.

Choose every conformer that is 5 kcal/mol or less from the minimum.

Comparing the chosen conformers for the two active substances methodically (superposition)

Pick the pair or pairs of conformations that are most comparable.

 

Comparison of Confomers

Comparison of the active compound A and B's low-energy conformations pairwise.

zse

The bioactive conformations are probably represented by the conformers that are the most comparable.

Similar pharmacophores and binding modalities

 

Modeling Bioactive Conformations

Pick at least two well-known active substances.

Sample low-energy conformations and conduct systematic conformational search.

Identify the conformation with the lowest energy.

Choose every conformer that is 5 kcal/mol or less from the minimum.

Comparing the selected conformers systematically.

Choose the most comparable pair(s) of conformations as the most likely bioactive conformations.

 

 

---- Summary ----

As of now you know all basics of all the laws of thermodynamics.

  • Structure Based Drug Design

  • Pharmacophore Based Drug Design

  • Bioactive conformations

  • etc..


________________________________________________________________________________________________________________________________
Footer
________________________________________________________________________________________________________________________________

Copyright © 2022-2023. Anoop Johny. All Rights Reserved.