Enabling the Application of Virtual Screening to GPCRs:
Identifying Ligands of the GPR101 Receptor
PI: Dr. Stefano Costanzi, Department of Chemistry, American University, Washington, DC
Collaborator: Dr. Constantine Stratakis, NICHD, NIH
Project summary/abstract
We aim at identifying new ligands of GPR101, an orphan Class A G protein-coupled receptor (GPCR) involved
in X-linked acrogigantism (X-LAG). Our rational approach is based on the construction of validated homology
models of the receptor and their use as platforms for virtual screening for the identification of new ligands. We
also aim at providing access to biomedical research and training to a diverse cohort of undergraduate
students at American University. The proposed research is significant because: 1) It will provide GPR101
ligands, possibly inverse agonists (any ligand of this orphan receptor will be a valuable tool to illuminate its
pathophysiological role; in particular, inverse agonists have the potential to serve as tools to probe the feasibility
of blocking GPR101 activity to treat X-LAG); 2) It will provide 3D models of GPR101 as well as data on their
applicability to virtual screening (these models will be made available to the scientific community and, in the
absence of experimental GPR101 structures, can serve as useful ligand-discovery tools); 3) it will advance our
understanding of the applicability of homology models to virtual screening, particularly with respect to the
correlation between model/template sequence identity and virtual screening performance. The premises of our
proposed research are very solid given that: 1) as shown in preliminary results, we have identified 8 GPR101
inverse agonists through the experimental screening of a library of 3,000 compounds by the Stratakis Lab in
collaboration with NCATS (this data enables us to run the proposed controlled virtual screenings); 2) we have
gathered preliminary results that show the effectiveness in controlled virtual screening campaigns of a GPR101
model in complex with Compound 1 (the most potent of the inverse agonists presented in the preliminary results)
– more in general, we and others have amply demonstrated the applicability of homology models to virtual
screening, especially when optimized in complex with known ligands; 3) the Stratakis Lab has demonstrated the
implication of GPR101 in X-LAG. The innovation of our proposed research rests on the facts that: 1) ligands of
the GPR101 receptor are currently not known, with the exception of the 8 inverse agonists presented in the
preliminary results and two more putative ligands reported in the literature; 2) validated GPR101 models
applicable to virtual screening are currently unavailable; 3) targets for the potential treatment of X-LAG are
understudied and there is a lack of effective options to manage this condition. We seek to achieve our objective
by: 1) building GPR101 models in complex with known inverse agonists and subjecting them to controlled virtual
screening campaigns to identify those that are more suited to prioritize inverse agonists of the receptor (Specific
Aim 1); 2) using the most promising models to conduct prospective virtual screening campaigns followed by
experimental tests, to identify novel GPR101 ligands, possibly inverse agonists (Specific Aim 2). To ensure
rigor and reproducibility, we will conduct a thorough and systematic study, following highly reproducible
modeling procedures, and subjecting experimental results to a thorough validation through secondary tests.
