7 Applications of Pharmacophore-Based Approaches
In this section, we discuss the diverse applications of the pharmacophore approaches under different scenarios.
7.1 Pharmacophore Approaches for Virtual Screening
Pharmacophore models being very simple by their definition can be used in a variety
of ways depending on the research problem. This simplicity makes ‘pharmacophore
based search’ a tool of choice for drug discovery scientists in the last decade [93].
When the structure of a set of molecules with similar or different scaffolds active on a
particular target are known, then ligand-based pharmacophore models can be
developed using their structures as described in Sect. 5.1. If the structures of some
inactive derivatives are also known, then contribution of each feature towards the
bioactivity can be compared between the positive and negative datasets to distinguish the wanted and unwanted features. The allowable steric arrangement of the
ligands can also be mapped. When only the structure of the receptor or a receptor–
ligand complex is available, then pharmacophore models are generated as described
in Sects. 5.2 and 5.3 and can be utilized as queries to screen a database not only to
screen compounds satisfying certain geometric and chemical restraints, but also to
filter molecules with undesirable properties. For example, Voet and co-workers
identified specific antagonists of human androgen receptor by applying two pharmacophoric filters back to back. One model is being generated from the available
receptor-agonist complexes, while the other filter applied was a pharmacophore
model generated from the receptor-antagonist complex. This approach enabled the
authors to screen the compound that matches the antagonist-specific feature [94].
7.1.1 Dynamic E-pharmacophore Models: A Case Study
with Mycobacterial CmaA1
We present here the summary of our recent work (Choudhury et al. [11, 17, 18]) on
generation and application of dynamic structure and ligand-based pharmacophore
models for screening a certain library against a mycobacterial target cyclopropane
synthase (CmaA1). Mycolic acids are the characteristic constituents of Mtb cell
wall which contribute towards the drug resistance, pathogenicity and persistence of
the parasite. CmaA1 enzyme catalyses the cis-cyclopropanation of unsaturated
mycolic acid chains at the distal position, which is an indispensable step in mycolic
acid biosynthesis and maturation, thus making CmaA1 an important Mtb drug
target. Five model systems of CmaA1 corresponding to different stages of cyclopropanation were studied using molecular dynamics (MD) simulations. A detailed
picture of the structural changes in the two distinct binding sites, i.e. cofactor and
40
C. Choudhury and G. Narahari Sastry
In this section, we discuss the diverse applications of the pharmacophore approaches under different scenarios.
7.1 Pharmacophore Approaches for Virtual Screening
Pharmacophore models being very simple by their definition can be used in a variety
of ways depending on the research problem. This simplicity makes ‘pharmacophore
based search’ a tool of choice for drug discovery scientists in the last decade [93].
When the structure of a set of molecules with similar or different scaffolds active on a
particular target are known, then ligand-based pharmacophore models can be
developed using their structures as described in Sect. 5.1. If the structures of some
inactive derivatives are also known, then contribution of each feature towards the
bioactivity can be compared between the positive and negative datasets to distinguish the wanted and unwanted features. The allowable steric arrangement of the
ligands can also be mapped. When only the structure of the receptor or a receptor–
ligand complex is available, then pharmacophore models are generated as described
in Sects. 5.2 and 5.3 and can be utilized as queries to screen a database not only to
screen compounds satisfying certain geometric and chemical restraints, but also to
filter molecules with undesirable properties. For example, Voet and co-workers
identified specific antagonists of human androgen receptor by applying two pharmacophoric filters back to back. One model is being generated from the available
receptor-agonist complexes, while the other filter applied was a pharmacophore
model generated from the receptor-antagonist complex. This approach enabled the
authors to screen the compound that matches the antagonist-specific feature [94].
7.1.1 Dynamic E-pharmacophore Models: A Case Study
with Mycobacterial CmaA1
We present here the summary of our recent work (Choudhury et al. [11, 17, 18]) on
generation and application of dynamic structure and ligand-based pharmacophore
models for screening a certain library against a mycobacterial target cyclopropane
synthase (CmaA1). Mycolic acids are the characteristic constituents of Mtb cell
wall which contribute towards the drug resistance, pathogenicity and persistence of
the parasite. CmaA1 enzyme catalyses the cis-cyclopropanation of unsaturated
mycolic acid chains at the distal position, which is an indispensable step in mycolic
acid biosynthesis and maturation, thus making CmaA1 an important Mtb drug
target. Five model systems of CmaA1 corresponding to different stages of cyclopropanation were studied using molecular dynamics (MD) simulations. A detailed
picture of the structural changes in the two distinct binding sites, i.e. cofactor and
40
C. Choudhury and G. Narahari Sastry