Homology modeling is a computational approach very widely employed to recover theoretical structure of a macromolecule. Following the idea that structure is better conserved than sequence in a family of proteins sharing a common function, the main difficulty is to find a suitable template that is a protein of related activity and of properly similar sequence. The accepted limit in terms of sequence identity is about 30%, while even in cases of lower identity, the ratio of conserved residues is usually higher in the peptides involved in the function of the protein.

Despite targeting G-protein-coupled receptors (GPCRs) has become a challenge in drug design for many years, the structure of GPCRs is still hardly amenable to standard crystallographic or nuclear magnetic resonance (NMR) methods because their integral membrane protein nature renders them difficult to isolate and crystallize (White et al. 2001). However, homology modeling offers an alternative pathway through these difficulties by the construction of theoretical three-dimensional (3D) models based on the few crystallographic data yet available for GPCR structure. Even if the bovine rhodopsin has been the only structural template for 10 years (Palczewski et al. 2000), the homology modeling of monoamine GPCRs, particularly 5-HT2C, should greatly benefit with the very recent discovery of more homologous crystal templates.

The homology modeling of the 5-HT2C receptor is specially challenging for rational drug design since few discovered ligands, from the cloning of this receptor 15 years ago (Xie et al. 1996; Stam et al. 1994; Saltzman et al. 1991) up to now, are found to be very selective for 5-HT2C among other 5-HT receptor subtypes but also G-protein-coupled neurotransmitter receptors like dopaminergic or adrenergic receptors (adrenoceptors). According to its investigated pharmacological profile, antagonist or agonist ligands must target, respectively, inactive or active states of the

Laboratoire de Chimie Thérapeutique, Université de Lille 2, EA 1043, 3 Rue du Professeur Laguesse, B.P. 83, F-59006, Lille, France e-mail: [email protected]

G. Di Giovanni et al. (eds.), 5-HT2C Receptors in the Pathophysiology of CNS Disease, 97

The Receptors 22, DOI 10.1007/978-1-60761-941-3_6, © Springer Science+Business Media, LLC 2011

receptor. Computational molecular dynamics of the homology models associated with data from site-directed mutagenesis are able to simulate these different boundary states of activation while bound to known agonist or antagonist ligands in order to provide insights on activation mechanisms or new relevant sites to be targeted.

On the basis of the several published works of 3D rhodopsin-templated homol-ogy models and previously evoked novel crystallographic structures, this chapter will describe the different ways to generate models able to answer the real challenges associated with the 5-HT2C receptor. These range from the investigation of strictly 5-HT2C-dependent structural features in order to refine this actually detrimental ligand selectivity through an improved understanding of mechanisms leading to constitutive or ligand-induced activation of the receptor to the structural impact of genetic polymorphism or orthologous variations within mammalians.

6.2 Homology Modeling 6.2.1 Reference Structural Templates

Up to now the bovine rhodopsin (RHO) has been the only member of the eukaryotic GPCR family with an experimentally solved structure (Palczewski et al. 2000). As deduced from the first of the three classes of GPCRs, the rhodopsin-like family (class A), RHO is the representative protagonist of this largest protein family that includes hormone, neurotransmitter, and light receptors, all of which transducing extracellular signals through interaction with guanine nucleotide-binding (G) proteins. Although their activating ligands vary widely in structure and nature, the amino acid sequences of the receptors are very similar and are believed to adopt a common structural framework comprising seven transmembrane (TM) helices (Attwood and Findlay 1993; Birnbaumer 1990; Casey and Gilman 1988). However RHO remains unusual because it is highly abundant from natural sources and structurally stabilized by the covalently bound ligand H-ds-retinal, which maintains the receptor in a dark-adaptated, nonsig-naling conformation. In contrast, all other class A GPCRs are activated by diffusible ligands and are expressed at relatively low levels in native tissues.

Considerable progress has arisen from the recent crystallization of GPCRs nearer to therapeutically relevant monoamine receptor, successively the human 02-adrenoceptor (ADRB2) (Cherezov et al. 2007; Rosenbaum et al. 2007), the melea-gris b1-adrenoceptor (ADRB1) (Warne et al. 2008), and the human A2A adenosine receptor (A2AAR) (Jaakola et al. 2008). Their full-length sequence share 41%, 41%, and 36% homology, respectively, with the entire human 5-HT2C sequence versus 33% with the RHO sequence. Since they are also monoamine receptors, ADRB1 and ADRB2 get the highest score of sequence homology (41%) with the 5-HT2C receptor. Overall, these two crystal structures were reported in complex with the high-affinity antagonist cyanopindolol for ADRB1 and inverse agonist carazolol for ADRB2. This extensively improves the prediction of ligand binding sites and structural features of ligand accessibility in the other GPCR-type neu-rotransmitter receptors like the 5-HT2C receptor. Ligand-binding site accessibility is enabled by the second extracellular loop, which is held out of the binding cavity by two closely spaced disulfide bridges and a short helical segment within the loop. The ligand-binding pocket comprises 15 side chains from amino acid residues in four transmembrane alpha-helices and extracellular loop 2. Binding of either cyanopindolol to the pi-adrenergic receptor or carazolol to the p2-adrenergic receptor involves similar interactions mainly driven by one cluster of aromatic rings and one acidic side chain trapping the positive charge carried by the monoamine ligands like serotonin.

Additional insights are also gained from the examination of packing interactions and a network of hydrogen bonds, suggesting a conformational pathway from the ligand-, binding pocket to regions that interact with G proteins. Otherwise, a short well-defined helix in cytoplasmic loop 2 of ADRB1, not observed in either RHO or ADRB2, directly interacts by means of a tyrosine with the highly conserved DRY motif at the end of helix 3 that is essential for receptor activation. The A2AAR seems to be less interesting than the other structurally determined GPCRs with regard to modeling the structure of the 5-HT2C receptor. Indeed authors have found that the extracellular domain, combined with a subtle repacking of the transmembrane helices relative to the adrenergic and rhodopsin receptor structures, defines a pocket distinct from that of other relevant GPCR-type catecholamine receptors.

Superimposing the backbone of the TM regions (Fig. 6.1) shows that the four helix bundles of GPCR crystals are structurally very close, varying from a

Fig. 6.1 Superimposition of the four TM bundles of GPCR crystal structure. Helices are represented as cylinders for RHO (green), ADRB1 (red), ADRB2 (purple), and A2AAR (yellow) in (a) a transversal view and (b) a longitudinal view from the extracellular side to the cytoplasmic side on the right (For interpretation of the colors in this figure, the reader is referred to the web version of this chapter)

Fig. 6.1 Superimposition of the four TM bundles of GPCR crystal structure. Helices are represented as cylinders for RHO (green), ADRB1 (red), ADRB2 (purple), and A2AAR (yellow) in (a) a transversal view and (b) a longitudinal view from the extracellular side to the cytoplasmic side on the right (For interpretation of the colors in this figure, the reader is referred to the web version of this chapter)

2.44 A root-mean-square deviation (RMSD) between the more homologous sequences, ADRB1 and ADRB2, to 5.39 A between RHO and A2AAR (Table 6.1). Most structural variations occur in A2AAR, which shows a kink in the longitudinal axis of helices 2 and 3 in the top of the TM bundle at the extracellular side as well as a significantly longer helix 6 at the intracellular side. The second largest difference is a bend in the top of helix 1 of both adrenoceptors that is not found in RHO and A2AAR, which is comparatively straight. This structural difference may arise from the need for an accessible binding site in ADRB1 and ADRB2, which is provided in part by a lack of interactions between the N-terminus and extracellular loop segments. In contrast, the N-terminal region in RHO occludes the retinal-binding site through extensive interactions with the extracellular loops.

Consequently, ADRB1 and ADRB2 are preferential structural templates due to sequence homology and ligand-binding properties. However, modelers must take into account that these receptors have been modified in order to facilitate their crystallization, particularly cytoplasmic features. Thus, ADRB2 and A2AAR were engineered as fusion proteins in which T4 lysosyme replaces most of the third intracellular loop (IL3) of the GPCR. In the same manner, ADRB1 was modified in that six substitutions occur in TM 2, 5, and 7 and IL 1 and 3 as well as a 15-resi-dues deletion in IL3, resulting in an extensive improvement of the protein thermo-stability. Although crystallized proteins retain near-native pharmacological properties and could consequently be relevant structural template for study of ligand-binding sites, the analysis of structural events in signal transducing should be weighted, particularly in cytoplasmic location.

Defeat Drugs and Live Free

Defeat Drugs and Live Free

Being addicted to drugs is a complicated matter condition that's been specified as a disorder that evidences in the obsessional thinking about and utilization of drugs. It's a matter that might continue to get worse and become disastrous and deadly if left untreated.

Get My Free Ebook

Post a comment