Limits...
Free fatty acid receptors: structural models and elucidation of ligand binding interactions.

Tikhonova IG, Poerio E - BMC Struct. Biol. (2015)

Bottom Line: This binding mode can explain mutagenesis results for residues at positions 4.56 and 5.42.The novel structural models of FFAs provide information on specific modes of ligand binding at FFA subtypes and new suggestions for mutagenesis and ligand modification, guiding the development of novel orthosteric and allosteric chemical probes to validate the importance of FFAs in metabolic and inflammatory conditions.Using our FFA homology modelling experience, a strategy to model a GPCR, which is phylogenetically distant from GPCRs with the available crystal structures, is discussed.

View Article: PubMed Central - PubMed

Affiliation: Molecular Therapeutics, School of Pharmacy, Medical Biology Centre, Queen's University Belfast, Belfast, BT9 7BL, Northern Ireland, UK. i.tikhonova@qub.ac.uk.

ABSTRACT

Background: The free fatty acid receptors (FFAs), including FFA1 (orphan name: GPR40), FFA2 (GPR43) and FFA3 (GPR41) are G protein-coupled receptors (GPCRs) involved in energy and metabolic homeostasis. Understanding the structural basis of ligand binding at FFAs is an essential step toward designing potent and selective small molecule modulators.

Results: We analyse earlier homology models of FFAs in light of the newly published FFA1 crystal structure co-crystallized with TAK-875, an ago-allosteric ligand, focusing on the architecture of the extracellular binding cavity and agonist-receptor interactions. The previous low-resolution homology models of FFAs were helpful in highlighting the location of the ligand binding site and the key residues for ligand anchoring. However, homology models were not accurate in establishing the nature of all ligand-receptor contacts and the precise ligand-binding mode. From analysis of structural models and mutagenesis, it appears that the position of helices 3, 4 and 5 is crucial in ligand docking. The FFA1-based homology models of FFA2 and FFA3 were constructed and used to compare the FFA subtypes. From docking studies we propose an alternative binding mode for orthosteric agonists at FFA1 and FFA2, involving the interhelical space between helices 4 and 5. This binding mode can explain mutagenesis results for residues at positions 4.56 and 5.42. The novel FFAs structural models highlight higher aromaticity of the FFA2 binding cavity and higher hydrophilicity of the FFA3 binding cavity. The role of the residues at the second extracellular loop used in mutagenesis is reanalysed. The third positively-charged residue in the binding cavity of FFAs, located in helix 2, is identified and predicted to coordinate allosteric modulators.

Conclusions: The novel structural models of FFAs provide information on specific modes of ligand binding at FFA subtypes and new suggestions for mutagenesis and ligand modification, guiding the development of novel orthosteric and allosteric chemical probes to validate the importance of FFAs in metabolic and inflammatory conditions. Using our FFA homology modelling experience, a strategy to model a GPCR, which is phylogenetically distant from GPCRs with the available crystal structures, is discussed.

Show MeSH

Related in: MedlinePlus

The ligand binding mode at the FFA1 crystal structure and homology models. a: the binding model of TAK-875, the ago-allosteric ligand in the FFA1 crystal structure, the ligand is pointed between helices 3 and 4 (mode 1 in the text). b: The binding mode of GW9508, the high potency agonist in the previous rhodopsin-based homology model of FFA1. The model was obtained from ref 12. c: The binding mode of TAK-875 in the PAR1-based model of FFA1. Hydrogen bonds are in black
© Copyright Policy - OpenAccess
Related In: Results  -  Collection

License 1 - License 2
getmorefigures.php?uid=PMC4561419&req=5

Fig3: The ligand binding mode at the FFA1 crystal structure and homology models. a: the binding model of TAK-875, the ago-allosteric ligand in the FFA1 crystal structure, the ligand is pointed between helices 3 and 4 (mode 1 in the text). b: The binding mode of GW9508, the high potency agonist in the previous rhodopsin-based homology model of FFA1. The model was obtained from ref 12. c: The binding mode of TAK-875 in the PAR1-based model of FFA1. Hydrogen bonds are in black

Mentions: In the crystal structure the carboxyl group of TAK-875 forms direct interactions with R1835.39, R2587.35, Y913.37 and Y2406.51 through H-bonding [16] (Fig. 3a). Although homology modelling in conjunction with mutagenesis predicted direct interactions with these residues for linoleic acid, GW9508 and TAK-875, interactions with Y913.37 and Y2406.51 were predicted through aromatic and hydrophobic contacts (Fig. 3b) [13, 14]. The deep cavity inside the helical bundle (Additional file 1: Figure 1S) in the earlier homology models locks agonists within the helical bundle allowing the hydrophobic tail of the ligand to interact with the tyrosines. In this position, GW9508 was predicted to form aromatic and amino-aromatic interactions with H1374.56 and hydrophobic interactions with L1865.42 (Fig. 3b). These interactions are not observed with TAK-875. H1374.56 and L1865.42 are at the distance of >6 Å from TAK-875 in the crystal structure. Instead, the hydrophobic part of TAK-875 interacts with F873.33, F1424.61, W174EL2 and L1384.57 and is pointed between helices 3 and 4 facing lipids in the experimental structure (Fig. 3a). This docking position could not have been predicted in the homology models as the space between helices 3 and 4 is 2 Å narrower and residues F873.33 and L1384.57 create an obstacle to the interhelical space. In contrast, docking is to some extent biased to the cavity formed by helices 4, 5 and 6 as the hydrophobic moiety of ligands in the crystal structures of rhodopsin and the β2 receptor is in this location. Docking to the FFA1 homology model based on the PAR1 receptor crystal structure predicts a different position for the hydrophobic part of an agonist compared to the rhodopsin- and β2 -based templates. Thus, the hydrophobic tail of the ligand is between helices 4 and 5 with the possibility of interacting with lipids (Fig. 3c). This orientation is also similar to the position of the ligand in the PAR1 receptor crystal structure. Clearly, the choice of a template for homology modelling is critical for establishing the size and shape of the binding cavity which, in turn, influences ligand docking solutions.Fig. 3


Free fatty acid receptors: structural models and elucidation of ligand binding interactions.

Tikhonova IG, Poerio E - BMC Struct. Biol. (2015)

The ligand binding mode at the FFA1 crystal structure and homology models. a: the binding model of TAK-875, the ago-allosteric ligand in the FFA1 crystal structure, the ligand is pointed between helices 3 and 4 (mode 1 in the text). b: The binding mode of GW9508, the high potency agonist in the previous rhodopsin-based homology model of FFA1. The model was obtained from ref 12. c: The binding mode of TAK-875 in the PAR1-based model of FFA1. Hydrogen bonds are in black
© Copyright Policy - OpenAccess
Related In: Results  -  Collection

License 1 - License 2
Show All Figures
getmorefigures.php?uid=PMC4561419&req=5

Fig3: The ligand binding mode at the FFA1 crystal structure and homology models. a: the binding model of TAK-875, the ago-allosteric ligand in the FFA1 crystal structure, the ligand is pointed between helices 3 and 4 (mode 1 in the text). b: The binding mode of GW9508, the high potency agonist in the previous rhodopsin-based homology model of FFA1. The model was obtained from ref 12. c: The binding mode of TAK-875 in the PAR1-based model of FFA1. Hydrogen bonds are in black
Mentions: In the crystal structure the carboxyl group of TAK-875 forms direct interactions with R1835.39, R2587.35, Y913.37 and Y2406.51 through H-bonding [16] (Fig. 3a). Although homology modelling in conjunction with mutagenesis predicted direct interactions with these residues for linoleic acid, GW9508 and TAK-875, interactions with Y913.37 and Y2406.51 were predicted through aromatic and hydrophobic contacts (Fig. 3b) [13, 14]. The deep cavity inside the helical bundle (Additional file 1: Figure 1S) in the earlier homology models locks agonists within the helical bundle allowing the hydrophobic tail of the ligand to interact with the tyrosines. In this position, GW9508 was predicted to form aromatic and amino-aromatic interactions with H1374.56 and hydrophobic interactions with L1865.42 (Fig. 3b). These interactions are not observed with TAK-875. H1374.56 and L1865.42 are at the distance of >6 Å from TAK-875 in the crystal structure. Instead, the hydrophobic part of TAK-875 interacts with F873.33, F1424.61, W174EL2 and L1384.57 and is pointed between helices 3 and 4 facing lipids in the experimental structure (Fig. 3a). This docking position could not have been predicted in the homology models as the space between helices 3 and 4 is 2 Å narrower and residues F873.33 and L1384.57 create an obstacle to the interhelical space. In contrast, docking is to some extent biased to the cavity formed by helices 4, 5 and 6 as the hydrophobic moiety of ligands in the crystal structures of rhodopsin and the β2 receptor is in this location. Docking to the FFA1 homology model based on the PAR1 receptor crystal structure predicts a different position for the hydrophobic part of an agonist compared to the rhodopsin- and β2 -based templates. Thus, the hydrophobic tail of the ligand is between helices 4 and 5 with the possibility of interacting with lipids (Fig. 3c). This orientation is also similar to the position of the ligand in the PAR1 receptor crystal structure. Clearly, the choice of a template for homology modelling is critical for establishing the size and shape of the binding cavity which, in turn, influences ligand docking solutions.Fig. 3

Bottom Line: This binding mode can explain mutagenesis results for residues at positions 4.56 and 5.42.The novel structural models of FFAs provide information on specific modes of ligand binding at FFA subtypes and new suggestions for mutagenesis and ligand modification, guiding the development of novel orthosteric and allosteric chemical probes to validate the importance of FFAs in metabolic and inflammatory conditions.Using our FFA homology modelling experience, a strategy to model a GPCR, which is phylogenetically distant from GPCRs with the available crystal structures, is discussed.

View Article: PubMed Central - PubMed

Affiliation: Molecular Therapeutics, School of Pharmacy, Medical Biology Centre, Queen's University Belfast, Belfast, BT9 7BL, Northern Ireland, UK. i.tikhonova@qub.ac.uk.

ABSTRACT

Background: The free fatty acid receptors (FFAs), including FFA1 (orphan name: GPR40), FFA2 (GPR43) and FFA3 (GPR41) are G protein-coupled receptors (GPCRs) involved in energy and metabolic homeostasis. Understanding the structural basis of ligand binding at FFAs is an essential step toward designing potent and selective small molecule modulators.

Results: We analyse earlier homology models of FFAs in light of the newly published FFA1 crystal structure co-crystallized with TAK-875, an ago-allosteric ligand, focusing on the architecture of the extracellular binding cavity and agonist-receptor interactions. The previous low-resolution homology models of FFAs were helpful in highlighting the location of the ligand binding site and the key residues for ligand anchoring. However, homology models were not accurate in establishing the nature of all ligand-receptor contacts and the precise ligand-binding mode. From analysis of structural models and mutagenesis, it appears that the position of helices 3, 4 and 5 is crucial in ligand docking. The FFA1-based homology models of FFA2 and FFA3 were constructed and used to compare the FFA subtypes. From docking studies we propose an alternative binding mode for orthosteric agonists at FFA1 and FFA2, involving the interhelical space between helices 4 and 5. This binding mode can explain mutagenesis results for residues at positions 4.56 and 5.42. The novel FFAs structural models highlight higher aromaticity of the FFA2 binding cavity and higher hydrophilicity of the FFA3 binding cavity. The role of the residues at the second extracellular loop used in mutagenesis is reanalysed. The third positively-charged residue in the binding cavity of FFAs, located in helix 2, is identified and predicted to coordinate allosteric modulators.

Conclusions: The novel structural models of FFAs provide information on specific modes of ligand binding at FFA subtypes and new suggestions for mutagenesis and ligand modification, guiding the development of novel orthosteric and allosteric chemical probes to validate the importance of FFAs in metabolic and inflammatory conditions. Using our FFA homology modelling experience, a strategy to model a GPCR, which is phylogenetically distant from GPCRs with the available crystal structures, is discussed.

Show MeSH
Related in: MedlinePlus