G protein coupled receptors (GPCRs) are a superfamily of proteins that are activated by a wide range of natural ligands, for instance neurotransmitters and hormones, and are targeted by many of the marketed and leisure drugs. Given their pharmaceutical relevance, there is great interest in the experimental elucidation or the modeling of their three-dimensional (3D) structures, which can be employed as platforms to discover or design new ligands that can modulate their activity. Through this proposal, we aim at furthering the field of virtual screening applied to ho ology models of G proteincoupledreceptors (GPCRs), thus enabling the scientific community to harness the knowledge deriving from the blossoming GPCR structural studies and expand it to the other superfamily members (1-4). In particular we are focusing on the largest class of GPCRs (class A, also known as family I or rhodopsin family), which comprises about 84% of the entire superfamily. Since class A GPCRs are proteins of very high pharmaceutical relevance, our research will ultimately provide tools that facilitate the rational discovery of drugs acting through this prominent class of targets. Though my career, I have been at the forefront of GPCR modeling. In 2008, with a single author article, I was the first to conclusively demonstrate that accurate GPCR homology models can be constructed (1). Later the same year, my structures resulted to be the most accurate of all those submitted to the first blind assessment of GPCR modeling and docking (2). Hence, I am in a very good position to conduct the proposed research. A substantial set of preliminary data for the research proposed here stem from a systematic study on GPCR homology modeling that I conducted at American University with a large cohort of research students. The study describes the construction of models of the ?2-adrenergic receptor in the inactive state, using as templates all the other class A GPCRs for which structures that reflect the inactive state have been solved. One of the key results of the study is the linear correlation that we found between the structural accuracy of the models and the sequence identity between modeled receptor and templates. This result suggests that, for a given receptor, the accuracy of the attainable models can be predicted on the basis of the available templates. As outlined below, we aim at furthering our investigation through two specific aims.
Specific aims of this proposal include: 1) Probing the applicability of Class A GPCR models to virtual screening for the identification of blockers and delineating best practices; 2) Probing the applicability of Class A GPCR models to virtual screening for the identification of agonists and delineating best practices. A further objective of this study is to strengthen the research environment at American University and give our students the opportunity to conduct research and receive training in the field of molecular modeling.
Through this research we aim at furthering the computational modeling framework for the discovery of drugs that act through G protein-coupled receptors (GPCRs). GPCRs are a superfamily of proteins that are found on the external membrane of cells and function like switches that can be turned on by specific chemicals. They are activated by a wide range of natural ligands, such as neurotransmitters and hormones, and are also targeted by many of the marketed and leisure drugs, which either turn on their signaling or prevent their activation by natural ligands. Because having knowledge of the three-dimensional structure of cellular targets greatly facilitates the discovery of drugs that act through them, GPCRs are object of many experimental biology studies. However, a thorough experimental characterization of all the receptors is not feasible due to the large size of the superfamily. Nevertheless, the available experimental structures can be used as templates for the construction of structural models of the other superfamily members through computational molecular modeling. The outcome of our research will provide the scientific community with the much-needed information to harness the currently available structural knowledge and maximize its applicability to drug discovery. Moreover, our research will support the training of students and prepare them for a career in biomedicine.
| Sagratini, Gianni; Buccioni, Michela; Marucci, Gabriella et al. (2018) Chiral analogues of (+)-cyclazosin as potent ?1B-adrenoceptor selective antagonist. Bioorg Med Chem 26:3502-3513 |
| Costanzi, Stefano; Skorski, Matthew; Deplano, Alessandro et al. (2016) Homology modeling of a Class A GPCR in the inactive conformation: A quantitative analysis of the correlation between model/template sequence identity and model accuracy. J Mol Graph Model 70:140-152 |
