Computational Pharmacology for Identifying and Refining Novel Inhibitors of the Regulator of G protein Signaling type-12 (RGS12) Protein Target

Date

2023

Authors

Agogo-Mawuli, Percy
Siderovski, David

ORCID

0000-0001-6563-3376 (Agogo-Mawuli, Percy)
0000-0002-0688-8210 (Siderovski, David)

Journal Title

Journal ISSN

Volume Title

Publisher

Abstract

Purpose: Substance use disorders arise from persistent changes in CNS synaptic transmission, as caused by initial exposure to illicit substances that heighten dopamine levels in the brain’s reward circuitry (a key event in establishing long-term drug-seeking behavior) [1,2]. The Siderovski lab recently discovered that mice lacking Regulator of G protein Signaling-12 (RGS12) are attenuated in their normal hyperlocomotion elicited by acute cocaine, amphetamine, or methamphetamine [3,4]. RGS12-deficient mice have increased dopamine transporter (DAT) expression and increased dopamine uptake within the ventral striatum [3]. The target for RGS12’s action as a Galpha-directed GTPase-accelerating protein (GAP) is the presynaptic kappa opioid receptor (KOR) [4], as KOR activation is known to attenuate striatal dopaminergic tone [5]. Our hypothesis is that RGS12 directly modulates the output of dynorphin / KOR signaling to dopamine reuptake. Developing RGS12 inhibitors would provide complementary pharmacological means to test this hypothesis pre-clinically, including in rodent models. However, to date, there are no small-molecule inhibitors of the G-alpha: RGS domain protein-protein interaction that are not thiol-reactive covalent modifiers of the RGS protein (a highly undesirable chemical property anathematic to further drug development) [6].

Methods: Using the AtomNet® model, a deep convolutional neural network for structure-based drug discovery, we screened millions of compounds against the NMR structure of the RGS12 RGS domain (Protein Data Bank id 2EBZ; state 2) to identify 96 candidate compounds, followed by experimental testing of these candidate compounds using the Transcreener® GDP RGScreen™ developed in partnership with BellBrook Labs [7,8]. Follow-up computational chemistry is being performed with Schrödinger's suite of molecular dynamics software.

Results: Two hits out of the 96 candidate compounds were discovered to exhibit reproducible, double-digit micromolar IC50 values in the Transcreener® GDP RGScreen™ assay. We then tested 192 analogs of the two original hits and discovered 33 analogs with measurable IC50 values. Of the 33 congeneric compounds, all but one of the active congeners were structurally related to one of the original two hits, with a wide spread of IC50 values and many with improved potencies (IC50RGS12 = 0.84 – 153.2 microM). These hits do not inhibit the intrinsic GTPase activity of G-alpha.

Conclusions: To increase diversity of the chemical scaffolds capable of inhibiting RGS12 function in vitro, we are now performing computational modeling of this initial set of 33 congeneric compounds, including CPU-based molecular docking and GPU-based shape screening of new chemical libraries (including those sourced from MilliporeSigma, MolPort, and Enamine) using Schrödinger algorithms on a local HPC cluster. In vitro binding and structural studies, cell-based studies, and pre-clinical animal studies are also being planned to further characterize these congeneric compounds and future divergent chemotypes.

References:

[1] https://www.ncbi.nlm.nih.gov/books/NBK424849/

[2] https://nida.nih.gov/publications/drugs-brains-behavior-science-addiction/drugs-brain

[3] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5942192/

[4] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6785087/

[5] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7992303/

[6] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2084260/

[7] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2795102/

[8] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3306263/

Description

Keywords

Citation

Rights

License

Collections