Structure-based virtual screening of perfluoroalkyl and polyfluoroalkyl substances (PFASs) as endocrine disruptors of androgen receptor activity using molecular docking and machine learning

Ettayapuram Ramaprasad Azhagiya Singam; Phum Tachachartvanich; Denis Fourches; Anatoly Soshilov; Jennifer C Y Hsieh; Michele A La Merrill; Martyn T Smith; Kathleen A Durkin

doi:10.1016/j.envres.2020.109920

Structure-based virtual screening of perfluoroalkyl and polyfluoroalkyl substances (PFASs) as endocrine disruptors of androgen receptor activity using molecular docking and machine learning

Environ Res. 2020 Nov:190:109920. doi: 10.1016/j.envres.2020.109920. Epub 2020 Jul 28.

Authors

Ettayapuram Ramaprasad Azhagiya Singam¹, Phum Tachachartvanich², Denis Fourches³, Anatoly Soshilov⁴, Jennifer C Y Hsieh⁵, Michele A La Merrill², Martyn T Smith⁶, Kathleen A Durkin⁷

Affiliations

¹ Molecular Graphics and Computation Facility, College of Chemistry, University of California, Berkeley, CA, USA.
² Department of Environmental Toxicology, University of California, Davis, CA, USA.
³ Department of Chemistry, North Carolina State University, Raleigh, NC, USA.
⁴ Division of Scientific Programs, Pesticide and Environmental Toxicology Branch, Water Toxicology Section, Office of Environmental Health Hazard Assessment, California Environmental Protection Agency, USA.
⁵ Division of Scientific Programs, Reproductive and Cancer Hazard Assessment Branch, Cancer Toxicology and Epidemiology Section, Office of Environmental Health Hazard Assessment, California Environmental Protection Agency, USA.
⁶ Division of Environmental Health Sciences, School of Public Health, University of California, Berkeley, CA, USA. Electronic address: martynts@berkeley.edu.
⁷ Molecular Graphics and Computation Facility, College of Chemistry, University of California, Berkeley, CA, USA. Electronic address: durkin@berkeley.edu.

PMID: 32795691
DOI: 10.1016/j.envres.2020.109920

Abstract

Perfluoroalkyl and polyfluoroalkyl substances (PFASs) pose a substantial threat as endocrine disruptors, and thus early identification of those that may interact with steroid hormone receptors, such as the androgen receptor (AR), is critical. In this study we screened 5,206 PFASs from the CompTox database against the different binding sites on the AR using both molecular docking and machine learning techniques. We developed support vector machine models trained on Tox21 data to classify the active and inactive PFASs for AR using different chemical fingerprints as features. The maximum accuracy was 95.01% and Matthew's correlation coefficient (MCC) was 0.76 respectively, based on MACCS fingerprints (MACCSFP). The combination of docking-based screening and machine learning models identified 29 PFASs that have strong potential for activity against the AR and should be considered priority chemicals for biological toxicity testing.

Keywords: Modeling; Steroid hormones; Tox21.

Publication types

Research Support, N.I.H., Extramural
Research Support, U.S. Gov't, Non-P.H.S.

MeSH terms

Endocrine Disruptors* / analysis
Endocrine Disruptors* / toxicity
Fluorocarbons* / toxicity
Machine Learning
Mass Screening
Molecular Docking Simulation
Receptors, Androgen

Substances

Endocrine Disruptors
Fluorocarbons
Receptors, Androgen

Grants and funding

S10 OD023532/OD/NIH HHS/United States