Progress in elucidating the structural and dynamic character of G Protein-Coupled Receptor oligomers for use in drug discovery.

G Protein-Coupled Receptors (GPCRs) are the most targeted group of proteins for the development of therapeutic drugs. Until the last decade, structural information about this family of membrane proteins was relatively scarce, and their mechanisms of ligand binding and signal transduction were modeled on the assumption that GPCRs existed and functioned as monomeric entities. New crystal structures of native and engineered GPCRs, together with important biochemical and biophysical data that reveal structural details of the activation mechanism(s) of this receptor family, provide a valuable framework to improve dynamic molecular models of GPCRs with the ultimate goal of elucidating their allostery and functional selectivity. Since the dynamic movements of single GPCR protomers are likely to be affected by the presence of neighboring interacting subunits, oligomeric arrangements should be taken into account to improve the predictive ability of computer-assisted structural models of GPCRs for effective use in drug design.

Differentiation of delta, mu, and kappa opioid receptor agonists based on pharmacophore development and computed physicochemical properties.

Compounds that bind with significant affinity to the opioid receptor types, delta, mu, and kappa, with different combinations of activation and inhibition at these three receptors could be promising behaviorally selective agents. Working on this hypothesis, the chemical moieties common to three different sets of opioid receptor agonists with significant affinity for each of the three receptor types delta, mu, or kappa were identified. Using a distance analysis approach, common geometric arrangements of these chemical moieties were found for selected delta, mu, or kappa opioid agonists. The chemical and geometric commonalities among agonists at each opioid receptor type were then compared with a non-specific opioid recognition pharmacophore recently developed. The comparison provided identification of the additional requirements for activation of delta, mu, and kappa opioid receptors. The distance analysis approach was able to clearly discriminate kappa-agonists, while global molecular properties for all compounds were calculated to identify additional requirements for activation of delta and mu receptors. Comparisons of the combined geometric and physicochemical properties calculated for each of the three sets of agonists allowed the determination of unique requirements for activation of each of the three opioid receptors. These results can be used to improve the activation selectivity of known opioid agonists and as a guide for the identification of novel selective opioid ligands with potential therapeutic usefulness.

Molecular determinants of non-specific recognition of delta, mu, and kappa opioid receptors.

Identification of the molecular determinants of recognition common to all three opioid receptors embedded in a single three-dimensional (3D) non-specific recognition pharmacophore has been carried out. The working hypothesis that underlies the computational study reported here is that ligands that bind with significant affinity to all three cloned opioid receptors, delta, mu, and kappa, but with different combinations of activation and inhibition properties at these receptors, could be promising behaviorally selective analgesics with diminished side effects. The study presented here represents the first step towards the rational design of such therapeutic agents. The common 3D pharmacophore developed for recognition of delta, mu, and kappa opioid receptors was based on the receptor affinities determined for 23 different opioid ligands that display no specificity for any of the receptor subtypes. The pharmacophore centers identified are a protonated amine, two hydrophobic groups, and the centroid of an aromatic group in a geometric arrangement common to all 23, non-specific, opioid ligands studied. Using this three-dimensional pharmacophore as a query for searching 3D structural databases, novel compounds potentially involved in non-specific recognition of delta, mu, and kappa opioid receptors were retrieved. These compounds can be valuable candidates for novel behaviorally selective analgesics with diminished or no side effects, and thus with potential therapeutic usefulness.

Development of a 3D pharmacophore for nonspecific ligand recognition of alpha1, alpha2, alpha3, alpha5, and alpha6 containing GABA(A)/benzodiazepine receptors.

Transfected cells containing GABA(A) benzodiazepine receptors (BDZRs) have been utilized to systematically determine the affinity of ligands at alpha1, alpha2, alpha3, alpha5 and alpha6 subtypes in combination with beta2 and gamma2. All but a few of the ligands thus far studied have relatively high affinities for each of these alpha subtype receptors. Thus, these ligands must contain common stereochemical properties favorable for recognition by each of the subtype combinations. In the present work, such a common three-dimensional (3D) pharmacophore for recognition of alpha1, alpha2, alpha3, alpha5 and alpha6 containing GABA(A)/BDZRs types of receptors has been developed and assessed, using as a database receptor affinities measured in transfected cells for 27 diverse compounds. The 3D-recognition pharmacophore developed consists of three proton accepting groups, a hydrophobic group, and the centroid of an aromatic ring found in a common geometric arrangement in the 19 nonselective ligands used. Three tests were made to assess this pharmacophore: (i) Four low affinity compounds were used as negative controls, (ii) Four high affinity compounds, excluded from the pharmacophore development, were used as compounds for pharmacophore validation, (iii) The 3D pharmacophore was used to search 3D databases. The results of each of these types of assessments provided robust validation of the 3D pharmacophore. This 3D pharmacophore can now be used to discover novel nonselective ligands that could be activation selective at different behavioral end points. Additionally, it may serve as a guide in the design of more selective ligands, by determining if candidate ligands proposed for synthesis conform to this pharmacophore and selecting those that do not for further experimental assessment.

Benzodiazepine-induced hyperphagia: development and assessment of a 3D pharmacophore by computational methods.

Benzodiazepine receptor (BDZR) ligands are structurally diverse compounds that bind to specific binding sites on GABA(A) receptors and allosterically modulate the effect of GABA on chloride ion flux. The binding of BDZR ligands to this receptor system results in activity at multiple behavioral endpoints, including anxiolytic, sedative, anticonvulsant, and hyperphagic effects. In the work presented here, a computational procedure developed in our laboratory has been used to obtain a 3D pharmacophore for ligand recognition of the GABA(A)/BDZRs initiating the hyperphagic response. To accomplish this goal, 17 structurally diverse compounds, previously assessed in our laboratory for activity at the hyperphagic endpoint, were used. The result is a four-component 3D pharmacophore. It consists of two proton acceptor atoms, the centroid of an aromatic ring and the centroid of a hydrophobic moiety in a common geometric arrangement in all compounds with activity at this endpoint. This 3D pharmacophore was then assessed and successfully validated using three different tests. First, two BDZR ligands, which were included as negative controls in the set of seventeen compounds used for the pharmacophore development, did not fit the pharmacophore. Second, some benzodiazepine ligands known to have activity at the hyperphagia endpoint, but not included in the pharmacophore development, were used as positive controls and were found to fit the pharmacophore. Finally, using the 3D pharmacophore developed in the present work to search 3D databases, over 50 classical benzodiazepines were found. Among them, were benzodiazepine ligands known to have an effect at the hyperphagic endpoint. In addition, the novel compounds also found in this search are promising therapeutic agents that could beneficially affect feeding behavior.

Docking of peptide-T onto the D1 domain of the CD4 receptor.

Peptide T (pepT) is a segment of the human immunodeficiency virus (HIV) envelope protein gp120. The peptide competitively binds to the CD4 receptor of a subset of peripheral T lymphocytes and inhibits binding of gp120. Previous studies of this laboratory allowed the assessment of a bioactive form of the peptide and a pharmacophore for the peptide-receptor interaction. In the present study the proposed bioactive form of pepT and its (4-8) segment, the smallest pepT fragment shown to retain full activity, were docked onto the D1 domain of the CD4 receptor. The bioactive conformation of the peptides complements well a cleft on the surface of the CD4 receptor, shown to be the attachment site of gp120 from site directed mutagenesis experiments. These studies provide an improved description of the ligand-receptor pharmacophore.

3D modeling, ligand binding and activation studies of the cloned mouse delta, mu; and kappa opioid receptors.

Refined 3D models of the transmembrane domains of the cloned delta, mu and kappa opioid receptors belonging to the superfamily of G-protein coupled receptors (GPCRs) were constructed from a multiple sequence alignment using the alpha carbon template of rhodopsin recently reported. Other key steps in the procedure were relaxation of the 3D helix bundle by unconstrained energy optimization and assessment of the stability of the structure by performing unconstrained molecular dynamics simulations of the energy optimized structure. The results were stable ligand-free models of the TM domains of the three opioid receptors. The ligand-free delta receptor was then used to develop a systematic and reliable procedure to identify and assess putative binding sites that would be suitable for similar investigation of the other two receptors and GPCRs in general. To this end, a non-selective, 'universal' antagonist, naltrexone, and agonist, etorphine, were used as probes. These ligands were first docked in all sites of the model delta opioid receptor which were sterically accessible and to which the protonated amine of the ligands could be anchored to a complementary proton-accepting residue. Using these criteria, nine ligand-receptor complexes with different binding pockets were identified and refined by energy minimization. The properties of all these possible ligand-substrate complexes were then examined for consistency with known experimental results of mutations in both opioid and other GPCRs. Using this procedure, the lowest energy agonist-receptor and antagonist-receptor complexes consistent with these experimental results were identified. These complexes were then used to probe the mechanism of receptor activation by identifying differences in receptor conformation between the agonist and the antagonist complex during unconstrained dynamics simulation. The results lent support to a possible activation mechanism of the mouse delta opioid receptor similar to that recently proposed for several other GPCRs. They also allowed the selection of candidate sites for future mutagenesis experiments.

Molecular modeling study of the differential ligand-receptor interaction at the mu, delta and kappa opioid receptors.

3D models of the opioid receptors mu, delta and kappa were constructed using BUNDLE, an in-house program to build de novo models of G-protein coupled receptors at the atomic level. Once the three opioid receptors were constructed and before any energy refinement, models were assessed for their compatibility with the results available from point-site mutations carried out on these receptors. In a subsequent step, three selective antagonists to each of three receptors (naltrindole, naltrexone and nor-binaltorphamine) were docked onto each of the three receptors and subsequently energy minimized. The nine resulting complexes were checked for their ability to explain known results of structure-activity studies. Once the models were validated, analysis of the distances between different residues of the receptors and the ligands were computed. This analysis permitted us to identify key residues tentatively involved in direct interaction with the ligand.

Assessment of the bioactive conformation of the farnesyltransferase protein binding recognition motif by computational methods.

Ras farnesyltransferase catalyzes the carboxyl-terminal farnesylation of Ras as well as other proteins involved in signal transduction processes. Previous studies demonstrated that its inhibition suppresses the activity of Ras transformed phenotypes in cultured cells, causing tumor regression in animal models. This observation led to the consideration of farnesyltransferase as a target for cancer therapy. In the present work we report the results of a computational study aimed at assessing the bioactive conformation of the peptide Cys-Val-Phe-Met, known to be the minimum peptide sequence that inhibits farnesyltransferase. For this purpose the conformational preferences of four analogs of the peptide were assessed by means of thorough searches of their respective conformational spaces, using a simulated annealing protocol as sampling technique. Specifically, two active analogs: Cys-Val-Tic-Met and Cys-Val-psi(CH2NH)Tic-Met and two inactive analogs: Cys-Val-Tic-psi(CH2NH)Met and Cys-Val-Aic-Met were selected for the present study. Low energy conformations of the four analogs were classified according to their structural motifs. The putative bioactive conformation of the minimum farnesyltransferase recognition motif was assessed by cross-comparison of the different classes of conformations obtained for the two active and the two inactive analogs. The putative bioactive conformation is characterized by two structural motifs: i) a C14 pseudo-ring stabilized by a hydrogen bond between the amino group of Cys1 and the carboxylate group of Met4 and a C11 pseudo-ring involving the residues Cys1 and Tic3. In addition, the thiol group of Cys1 side chain of the bioactive conformation points to the carboxylate moiety of Met4.

New insights into the conformational requirements of B2 bradykinin antagonism.

The conformational profiles of a selected group of a new series of small linear and cyclic penta- and hexapeptides, inspired on the C-terminal segment of second-generation bradykinin (BK) antagonists, were independently computed in order to assess the chemical and geometrical requirements necessary for BK antagonism. Specifically, four cyclic peptides: cyclo-(Gly-Thi-D-Tic-Oic-Arg), cyclo-(Gly-Ala-D-Tic-Oic-Arg), cyclo-(Abu-Ala-Ser-D-Tic-Oic-Arg), cyclo-(Abu-D-Phe-Ala-D-Tic-Oic-Arg), and a linear peptide: Thi-Ser-D-Tic-Oic-Arg were selected for the present study. The first three BK analogs are capable to antagonize kinin-induced rabbit jugular vein and rabbit aorta smooth muscle contraction, while last two show no detectable affinity for the BK B2 receptor. The conformational space of the five peptides was thoroughly explored using simulated annealing (SA) in an iterative fashion as sampling technique. The bioactive conformation was assessed by pairwise cross comparisons between each of the unique low energy conformations found for each of the different peptides studied within a 5 kcal/mol threshold in respect to the global minimum. The conformational profile of the highly potent BK antagonist HOE-140, computed in an independent study, was also used in conjunction with the bioactive form assessed in the present study, to propose a pharmacophore that includes the stereochemical requirements for B2 BK antagonism.

BUNDLE: a program for building the transmembrane domains of G-protein-coupled receptors.

The only information available at present about the structural features of G-protein-coupled receptors (GPCRs) comes from low resolution electron density maps of rhodopsin obtained from electron microscopy studies on 2D crystals. Despite their low resolution, maps can be used to extract information about transmembrane helix relative positions and their tilt. This information, together with a reliable algorithm to assess the residues involved in each of the membrane spanning regions, can be used to construct a 3D model of the transmembrane domains of rhodopsin at atomic resolution. In the present work, we describe an automated procedure applicable to generate such a model and, in general, to construct a 3D model of any given GPCR with the only assumption that it adopts the same helix arrangement as in rhodopsin. The present approach avoids uncertainties associated with other procedures available for constructing models of GPCRs based on a template, since sequence identity among GPCRs of different families in most of the cases is not significant. The steps involved in the construction of the model are: (i) locate the centers of the helices according to the low-resolution electron density map; (ii) compute the tilt of each helix based on the elliptical shape observed by each helix in the map; (iii) define a local coordinate system for each of the helices; (iv) bring them together in an antiparallel orientation; (v) rotate each helix through the helical axis in such a way that its hydrophobic moment points in the same direction of the bisector formed between three consecutive helices in the bundle; (vi) rotate each helix through an axis perpendicular to the helical one to assign a proper tilt; and (vii) translate each helix to its center deduced from the projection map.

Conformational analysis of the highly potent bradykinin antagonist Hoe-140 by means of two different computational methods.

The AMBER 4.0 force field was used to perform the characterization of the conformational profile of the highly potent bradykinin antagonist Hoe-140 (D-Arg0-Arg1-Pro2-Hyp3-Gly4-Thi5-Ser6-D-++ +Tic7-Oic8-Arg9). The structural features of the peptide were assessed using two different computational methods, both capable to provide a good sampling of the low-energy conformations of the molecule. Specifically, the conformational space of the peptide was explored: i) computing molecular dynamics trajectories in cycles of high (900 K) and low (300 K) temperature and ii) using simulated annealing (SA) in an iterative fashion. Analysis of the structures characterized indicates that most of the low-energy conformations of the peptide exhibit a betaII'-turn motif at its C-terminus, in agreement with previous experimental and theoretical studies. On the other hand, about a 50% of the low-energy conformations characterized also exhibit different beta-turn type motifs at the N-terminus, whereas the rest of the conformations can be described as bends. Finally, in order to get new insights into the structural requirements necessary to design more potent and selective antagonists of bradykinin, present results were compared with those previously reported by this laboratory on the conformational preferences of the native nonapeptide and its DPhe7 analog.

Amygdalin binds to the CD4 receptor as suggested from molecular modeling studies.

The geometrical features of the proposed bioactive conformation of peptide T assessed by computational methods in a previous study, together with available structure-activity studies on peptide T, led us to propose a pharmacophore for the CD4-peptide T interaction. Subsequent, data base searching permitted us to identify amygdalin as a peptide T peptidomimetic.

Conformational study of vasoactive intestinal peptide by computational methods.

The conformational profile of vasoactive intestinal peptide (VIP) was characterized using computational methods. The strategy devised included a close examination of the conformational profile of the first 11 residues fragment followed by a study that considered the compatibility of the different conformations found with a continuation of the polypeptide chain in a alpha-helical conformation. Accordingly, a detailed analysis of the conformational preferences of the N-terminal fragment of VIP(1-11) was carried out within the framework of the molecular mechanics approach, using simulated annealing in an iterative fashion as the sampling technique. In a second step, low-energy structures of the fragment were fused to the remainder of the VIP chain in the form of two noninteracting alpha-helices, according to a model of the structure of the peptide proposed from NMR studies. After investigation for compatibility of each of the low-energy structures of VIP(1-11) with the two helical regions by energy minimization, only 5 of 35 structures were discarded. Analysis of the structures characterized indicates that most of the conformations of VIP(1-11), including the global minimum, can be described as bent conformations. Conformations exhibiting alpha-turns and beta-turns, previously proposed by NMR studies were also characterized. The conformational analysis also suggests that the common structural features found in VIP(1-11) should also be present in VIP. Finally, because of the sequence homology between VIP and Peptide T, and the fact that both are ligands of the CD4 receptor, both sets of low-energy conformations were compared for similarity. The relevance of these results as guidance of the design of new peptide analogs targeted to the CD4 receptor is also discussed.

Computational study of the conformational domains of peptide T.

The conformational preferences of peptide T (ASTTTNYT) were analysed by means of computational methods. A thorough exploration of the conformational space was carried out within the framework of the molecular mechanics approach, using simulated annealing as a searching strategy. Specifically, in order to obtain a subset of low-energy conformations with energies close to the global minimum as complete as possible, a simulated annealing protocol was repeated several times in a recursive fashion. The results of the search indicate that the peptide exhibits a alpha-helical character although most of the conformations characterized, including the global minimum, can be described as bent conformations. Conformations exhibiting beta-turn motives previously proposed from NMR studies were also characterized, although they are not very predominant in the set of low-energy conformations.

Comparative molecular modeling study of the three-dimensional structures of prostaglandin endoperoxide H2 synthase 1 and 2 (COX-1 and COX-2).

To understand the structural features that dictate the selectivity of diverse nonsteroidal antiinflammatory drugs for the two isoforms of the human prostaglandin H2 synthase (PGHS), the three-dimensional (3D) structure of human COX-2 was assessed by means of sequence homology modeling. The ovine COX-1 structure, solved by X-ray diffraction methods and sharing a 61% sequence identity with human COX-2, was used as template. Both structures were energy minimized using the AMBER 4.0 force field with a dielectric constant of 4r. (S)-Flurbiprofen, a nonselective COX inhibitor, and SC-558, a COX-2-selective ligand, were docked at the cyclooxygenase binding site in both isozymes, evidencing the role of different residues in the ligand-protein interaction. The 3D structures of the constructed four ligand-enzyme complexes were refined by energy minimization. Molecular dynamics simulations were also carried out, to understand more deeply the structural origins of the selectivity. Distances calculated during the dynamics process between the different ligands and the interacting residues of the two PGHS isozymes provided evidence of the flexible nature of the cyclooxygenase active site, permitting the identification of different conserved and nonconserved residues as responsible for ligand selectivity.

Bioactive peptides: conformational studies of [Tyr4] cyclolinopeptide A.

The solid state conformational analysis of [Tyr4] cyclolinopeptide A has been carried out by x-ray diffraction studies. The crystal structure of the monoclinic form, grown from a dioxane-water mixture [alpha = 9.849 (5) A, b = 20.752 (4) A, c = 16.728 (5) A, beta = 98.83 (3) degrees, space group P21, Z = 2], shows the presence of five intramolecular N-H...O = C hydrogen bonds, with formation of one C17 ring structure, one alpha-turn (C13), one inverse gamma-turn (C7), and two beta-turns (C10, one of type III and one of type I). The Pro1-Pro2 peptide unit is cis (omega = 5 degrees), all others are trans. The structure is almost superimposable with that of cyclolinopeptide A. The rms deviation for the atoms of the backbones is on the average 0.33 A.

[Aib5,6-D-Ala8]-cyclolinopeptide A, grown from a benzene/acetonitrile mixture.

cyclo-(Prolyl-prolyl-phenylalanyl-phenylalanyl- alpha-aminoisobutyryl-alpha-aminoisobutyryl-isoleucyl-D-alan yl-valyl) ([Aib5,6-D-Ala8]-cyclolinopeptide A), grown from benzene/acetonitrile mixture, crystallizes with one acetonitrile and two water molecules. The molecular structure is almost identical to that obtained from methanol/water. The dimension of the solvent channels found in these structures is reduced in the present one, but the intramolecular hydrogen-bond pattern is preserved. The Pro1-Pro2 peptide unit is cis (omega = 8 degrees); all others are trans.