Differential chemoproteomic analysis of RRS-1 candidate molecule and molecules of several nonsteroidal anti-inflammatory drugs

Abstract

Background. To plan effective and safe pharmacotherapy for inflammation and pain, it is important to evaluate the mechanisms and spectrum of action of nonsteroidal anti-inflammatory drugs (NSAIDs), including their effects on human proteome.Objective: to identify and evaluate the most significant specific differences of candidate molecule RRS-1 (N-{(Z)-2-(1-methyl-1H-indol-3-yl)-1-[(propylamino)carbonyl]vinyl}benzamide) from other NSAIDs through differential chemoreactome analysis.Materials and methods. Chemoproteomic modeling of pharmacological effects of RRS-1 molecule and a number of well-known NSAIDs (diclofenac, nimesulide, ketorolac) on human proteome was carried out on the basis of numerical prediction algorithms over the space of heterogeneous feature descriptions, developed in the topological approach to recognition by Yu.I. Zhuravlev and K.V. Rudakov scientific school.Results. Significant differences in the effects of the studied molecules were found for 1232 proteins of human proteome. The features of assessing interactions of the studied molecules with 47 target proteins, which most distinguished the effects of RRS-1 molecule from all others were identified. RRS-1 could activate adenosine and dopamine receptors, cannabinoid receptor 2 and GABAA receptor to a greater extent than other molecules. Activation of these receptors corresponded to anti-inflammatory, anti-nociceptive and neuroprotective effects. RRS-1 could preferably inhibit a number of pro-inflammatory proteins, receptor bradykinin 1, metabotropic glutamate receptor 5, matrix metalloproteinases 8, 9, 12, and blood coagulation factor X. Additionally, RRS-1 molecule showed preferable inhibition of a number of kinases targeted in antitumor and anti-inflammatory therapy. RRS-1, less than other studied molecules, interacted with the receptors of vitamin D3, thyroid hormone, acetylcholine, cannabinoids and opioids, orexin, and various metabolic enzymes, which is important in assessment of the safety of using drugs based on this molecule. RRS-1 characteristically exhibited a moderate profile of antivitamin action: the total score of vitamin and mineral loss (7.4±3.7) was significantly less in comparison to diclofenac (11.7±4.5) and was actually on the same level as nimesulide (6.9±3.7) and ketorolac (6.7±3.6).Conclusion. Chemoreactomic and chemoproteomic profiling of RRS-1 candidate molecule provided pre-experimental assessments of its efficacy and safety through modeling interactions with the human proteome.

Keywords

• inflammation
• rational drug design
• rrs-1
• machine learning