In silico exploration of the fingerprints triggering modulation of glutaminyl cyclase inhibition for the treatment of Alzheimer’s disease using SMILES based attributes in Monte Carlo optimization

Abstract

Alzheimer’s disease is the most common neurodegenerative disorder and being a social burden Alzheimer’s has become an economic liability on developing countries. With limited understanding regarding the cause of disease, it is commonly identified by extracellular deposit of amyloid β (Aβ) peptides as senile plaques. Pyroglutamated Aβ is identified from the brain of AD patients and constituted the majority of total Aβ present. The formation of Pyroglutamated Aβ could be hindered by the use of Glutaminyl cyclase inhibitors and could efficiently improve the symptoms of Alzheimer’s. The literature revealed the competence of quantitative structure activity/property relationship studies in drug discovery. The present work explores the efficiency of Monte Carlo based QSAR modelling studies on a dataset of 125 Glutaminyl cyclase inhibitors with pKi taken as the endpoint for QSAR analysis. The dataset is divided into training, subtraining, calibration and validation sets resulting in the generation of five random splits. The validation is performed in accordance with the Organization of Economic Corporation and Development principles. The values of R², Q², index of ideality of correlation, concordance correlation coefficient, av. $r_{\mathrm{m}}^2$ and delta $r_{\mathrm{m}}^2$ of calibration set of the best split are found to be 0.9012, 0.8775, 0.9479, 0.9435, 0.8347 and 0.0847, respectively. The structural features responsible for increasing the inhibitory activity are identified. These structural features are added to a base compound from the dataset to design six novel molecules. These new molecules possess improved inhibitory activity as compare to the base compound. The results are further supported by docking studies.

Communicated by Vsevolod Makeev

Keywords:

• Alzheimer’s disease
• amyloid β
• glutaminyl cyclase inhibitors
• quantitative structure activity relationship
• SMILES
• Monte Carlo

Acknowledgements

The authors are grateful to Dr Andrey A. Toropov and Dr Alla P. Toropov for providing CORAL software. The authors are also thankful to the respective universities for providing necessary facilities.

Disclosure statement

There is no conflict of interest.