MDCKpred: a web-tool to calculate MDCK permeability coefficient of small molecule using membrane-interaction chemical features

Abstract

Structure-based models to understand the transport of small molecules through biological membrane can be developed by enumerating intermolecular interactions of the small molecule with a biological membrane, usually a dimyristoylphosphatidylcholine (DMPC) monolayer. This ADME (absorption, distribution, metabolism, and excretion) property based on Madin–Darby Canine Kidney (MDCK) cell line demonstrates intestinal drug absorption of small molecules and correlated to human intestinal absorption which acts as a determining factor to forecast small-molecule prioritization in drug-discovery projects. We present here the development of MDCKpred web-tool which calculates MDCK permeability coefficient of small molecule based on the regression model, developed using membrane-interaction chemical features. The web-tool allows users to calculate the MDCK permeability coefficient (nm/s) instantly by providing simple descriptor inputs. The chemical-interaction features are derived from different parts of the DMPC molecule viz. head, middle, and tail regions and accounts overall intermolecular contacts of the small molecule when passively diffused through the phospholipid-rich biological membrane. The MDCKpred model is both internally (R² = .76; = .68; R_train = .87; R_test = .69) and externally (R_ext = .55) validated. Furthermore, we used natural molecules as application examples to demonstrate its utility in lead exploration and optimization projects. The MDCKpred web-tool can be accessed freely at http://www.mdckpred.in. This web-tool is designed to offer an intuitive way of prioritizing small molecules based on calculated MDCK permeabilities.

Keywords:

• MDCK
• Caco-2
• permeability
• web-tool
• permeability coefficient
• QSAR

Acknowledgements

The authors thank Rakesh M. Rawal and anonymous reviewer for their valuable suggestions.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

Rikin D. Patel acknowledges financial assistance from University Grants Commission (UGC), Govt. of India as Rajiv Gandhi National Fellowship. Sivakumar Prasanth Kumar acknowledges the National Post-Doctoral Fellowship [PDF/2016/000156] funded by Science & Engineering Research Board (SERB), Govt. of India.