Defense Date

9-29-2020

Graduation Date

Fall 12-18-2020

Availability

One-year Embargo

Submission Type

dissertation

Degree Name

PhD

Department

Pharmacology

School

School of Pharmacy

Committee Chair

Paula A. Witt-Enderby

Committee Member

Rehana K. Leak

Committee Member

David A. Johnson

Committee Member

Kevin Tidgewell

Committee Member

Antonio Ferreira

Keywords

serotonin transporter, SERT, virtual screen, drug discovery, computer aided drug discovery, structure based drug discovery

Abstract

Depression is a mental health disorder affecting greater than 350 million people worldwide with roughly 7% of the United States population diagnosed as of 2017. The selective serotonin reuptake inhibitors (SSRIs) have been the mainstay of pharmacotherapies for depression for the last 40 years. The SSRIs target the serotonin transporter (SERT), a monoamine transporter (MAT) responsible for terminating serotonergic neurotransmission. The SSRIs are not perfect therapeutics and suffer from delayed response times, inconsistent efficacy among patients, and often produce intolerable side effects. Therefore, a strong need exists to develop new antidepressants that are more efficacious and have fewer adverse effects. The Surratt and Madura laboratories approached this problem through the application of computational chemistry and classical pharmacology to rationally identify novel MAT inhibitors and ligands. The work within this doctoral thesis encompasses a structure-based virtual screen targeting SERT and the pharmacological analysis of the compounds identified from the screen.

Previous virtual screens utilized SERT homology models based on a bacterial leucine transporter as the structural template (Manepalli et al., 2011; Kortagere et al., 2013; Gabrielsen et al., 2014; Nolan et al., 2014). More recently, the human SERT crystal structure was published by the Eric Gouaux laboratory (Coleman et al., 2016) and used as the template for the present study. The Molecular Operating Environment software was chosen to target the orthosteric binding pocket S1 due to performance during benchmarking evaluations of the scoring function parameters. The HitDiscoverer chemical library was screened with the SERT computational model, and SERT ligand candidates were evaluated by predicted binding affinity, the Lipinski Rule of 5, and chemical uniqueness. Nine compounds were purchased and subjected to pharmacological analysis for binding, inhibition efficacy, and release potential. One compound bound to SERT with reasonable affinity; two compounds inhibited serotonin transport in in vitro assays. None of the compounds promoted the release of internal serotonin (i.e., efflux). In conclusion, computational modeling was successfully used to identify novel inhibitors of the human SERT in a time and cost-efficient manner demonstrating the applicability to academic research.

Language

English

Download

Share

COinS