Using Cheminformatics in Drug Discovery.

Area of Science:

• Medicinal Chemistry
• Computational Chemistry
• Pharmacology

Background:

• Designing novel drug candidates requires balancing target activity with desirable pharmacokinetic properties.
• Cyclooxygenase (COX) enzymes are key targets for anti-inflammatory drugs, with selectivity for COX-2 over COX-1 being a major goal.

Purpose of the Study:

• To apply cheminformatics techniques for the design of novel cyclooxygenase inhibitors with enhanced COX-2 selectivity and favorable ADMET (Absorption, Distribution, Metabolism, Excretion, Toxicity) profiles.
• To demonstrate the utility of computational methods in identifying potent and selective drug candidates.

Main Methods:

• Construction of a knowledge database of cyclooxygenase inhibitors.
• Analysis of activity cliffs and matched molecular pair analysis to understand structure-activity relationships.
• Development of Quantitative Structure-Activity Relationship (QSAR) models for predicting COX potency and selectivity.
• In silico screening of virtual libraries for potency, selectivity, and ADMET properties using ADMET Risk™.
• Synthesis and in vitro testing of selected candidate compounds.

Main Results:

• QSAR models were developed to predict cyclooxygenase potency and selectivity.
• Virtual libraries were generated and screened, leading to the identification of eight synthetic candidates.
• Four compounds were synthesized and tested; SLP0020 exhibited a COX-1 IC50 of 770 nM and a COX-2 IC50 of 130 nM.
• The study successfully integrated cheminformatics for rational drug design, yielding a potent COX-2 inhibitor.

Conclusions:

• Cheminformatics is a powerful tool for accelerating the design of selective enzyme inhibitors.
• The integrated approach combining QSAR, virtual screening, and ADMET prediction is effective in identifying promising drug candidates.
• The compound SLP0020 represents a successful outcome of this cheminformatics-driven design process, showing potent COX-2 inhibition.