G-Protein-Coupled Receptor Affinity Prediction Based on the Use of a Profiling Dataset:  QSAR Design, Synthesis, and Experimental Validation

Abstract

A QSAR model accounting for "average" G-protein-coupled receptor (GPCR) binding was built from a large set of experimental standardized binding data (1939 compounds systematically tested over 40 different GPCRs) and applied to the design of a library of "GPCR-predicted" compounds. Three hundred and sixty of these compounds were randomly selected and tested in 21 GPCR binding assays. Positives were defined by their ability to inhibit by more than 70% the binding of reference compounds at 10 microM. A 5.5-fold enrichment in positives was observed when comparing the "GPCR-predicted" compounds with 600 randomly selected compounds predicted as "non-GPCR" from a general collection. The model was efficient in predicting strongest binders, since enrichment was greater for higher cutoffs. Significant enrichment was also observed for peptidic GPCRs and receptors not included to develop the QSAR model, suggesting the usefulness of the model to design ligands binding with newly identified GPCRs, including orphan ones.

Computational Methods in Drug Discovery

“…These compounds, however, did not exhibit the desirable specificity, and the authors suggested that more specific pharmacophores may be necessary (Erickson et al, 2010). Rolland et al (2005) used a similar strategy to design a library that could be screened with GPCR targets. They collected binding profiles for 1939 compounds against 40 GPCR targets and used this information to train a global QSAR model.…”

Investigation of the Binding Site of CCR2 using 4-Azetidinyl-1-aryl-cyclohexane Derivatives: A Membrane Modeling and Molecular Dynamics Study

“…Several research groups have studies the SAR's of CCR2 antagonists experimentally 23,24,37 or computationally. 19,20,25,53 However, no previous study was conducted on the time dependent interactions between CCR2 and antagonists using MDS. This paucity of data prompted us to undertake the present study to determine the effects of active site residues in the vicinity of ligand.…”

“…The crystal structure of CCR2 has not yet been reported, and in the absence of firm structural data, ligand-based approaches have proven to be particularly useful for studying G proteincoupled receptors (GPCR). 20 However, few studies have been undertaken to examine CCR2 modeled structures.…”

On the Relationship between Molecular Hit Rates in High-Throughput Screening and Molecular Descriptors

“…An approximate way to model this is to assume that for a randomly chosen compound, the hit rate will vary accordingly to a beta distribution with a probability density function (pdf) depending on the molecular descriptor values according to equation (3). , where a(x) > 0, b(x) > 0, and 0 < h < 1.…”

“…Published computational models that investigate the relationship between promiscuity and adverse effects are based on data sets that contain compounds screened against panels of protein targets. [3][4][5] Several recent studies have shown that lipophilicity (ClogP) is a principal determinant of molecular promiscuity. Other molecular properties such as ionization state, basicity, and presence of certain functional groups have also been found to influence promiscuity.…”