Glutamate Class C

The Class C GPCR family comprises 22 human receptors [141], including such therapeutically important targets as the metabotropic glutamate receptors (GRM1. 8) [142-144] . GABA binding receptors (GABABRs), and the calcium-sensing receptor (CaR). For example, the GABAB agonist baclofen is used in the clinic as an anticonvulsant [122], while the allosteric CASR modulator cinacalcet is known to normalize calcium level in patients with hyper-parathyroidism [145] .

Class C receptors bind their endogenous ligands within a large (>500 amino acids) extracellular N-terminal domain. Crystal structures of this domain have recently been solved for GRM1, GRM3, and GRM7 [146-148]. This series of structures provides a high-resolution description of small molecule agonist and antagonist binding at an orthosteric pocket located between two subdomains (LB1 and LB2) connected by a flexible hinge. These structures provide an accurate template for structure-based design of small molecule ligands targeting the orthosteric site; however, an abundance of glutamate binding receptors in the CNS makes ligand specificity a serious concern. Nonetheless, homology modeling of the extracellular domain of other Class C receptors finds substantial diversity in the size and shape of the ligand binding pocket, suggesting that molecules substantially larger than glutamate can be used as selective Class C GPCR ligands [149, 150]. In one recent study, docking of tripeptides into the models based on crystal structure of the mGluR1 extracellular domain allowed identification of a specific orthosteric agonist of the CaR [151]. Docking in the orthosteric site of GRM1 has also been employed for rationalizing the agonistic or antagonistic action of a newly identified ligand series [152, 153]. Rational design strategies exploiting the new high-resolution structure of an extended glutamate orthosteric site [148] are likely to bring about other highly selective ligands with improved pharmacokinetic features and superior therapeutic potential.

While endogenous ligands bind the N-terminal domain of Class C GPCRs, many synthetic ligands identified in binding assays are noncompetitive. These compounds target an allosteric site between helices TM3-TM5-TM6-TM7, approximately corresponding to the binding cavity of rhodopsin-like GPCRs [142, 154-157]. Allosteric modulation has become the de facto approach to targeting Class C GPCRs, with cinacalcet (NPS-1493) being the first clini cally approved drug of its kind [158] . Though most previous applications to discovery of Class C allosteric modulators employed a ligand-based approach, 3D modeling and docking is critical for understanding ligand interaction patterns with the allosteric site and the design of more selective inhibitors [159, 160]. Encouraging results were obtained in studies using a "ligand supported" homology model of GRM5 and a fingerprint - based scoring function, where a reasonable enrichment factor (~25-fold at 1% cutoff) was achieved in retrospective screening of a diverse compound library [161]. In another study, selective retrieval of agonist and antagonists in a small benchmark test was reported using AutoDock (http://autodock.scripps.edu/) and ArgusLab (http://www.arguslab.com/) docking; models of active and inactive GRMs in this study were generated from the bRho structure, though details of homology modeling and binding pocket refinement were not reported [162] .

0 0

Post a comment