![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
Materials and Process Simulation Center, Beckman Institute, California Institute of Technology, Pasadena, CA 91125
* To whom correspondence should be addressed. E-mail: wag{at}wag.caltech.edu.
 |
ABSTRACT |
|---|
 TOP ABSTRACT ResultsDiscussionLITERATURE CITED |
|---|
340 nonolfactory receptors organized in 5 families: glutamate, rhodopsin, adhesion, frizzled, and secretin (1). A variety of bioactive molecules (including biogenic amines, peptides, lipids, nucleotides, and proteins) modulate GPCR activity to affect regulation of essential physiological processes (e.g. neurotransmission, cellular metabolism, secretion, cell growth, immune defense, and differentiation). Thus, many important cell recognition and communication processes involve GPCR (2,3). Due to their mediation of numerous critical physiological functions, GPCR are involved in all major disease areas, including cardiovascular, metabolic, neurodegenerative, psychiatric, cancer, and infectious diseases (4,5). Indeed, GPCR represent 30–50% of the current drug targets for activation (by agonist drugs) or inhibition (by antagonists) (5,6). It is estimated that the 80 GPCR-targeting drugs currently marketed account for $50 billion in annual sales, and many have annual sales >$2 billion. Target evaluation, lead identification, and optimization of GPCR have accelerated progress in identifying subtypes with specific cell and tissue functions. However, progress in developing new drugs with reduced toxicity and side effects has been hampered by the lack of a 3-dimensional (3D) structure for any human GPCR or the common animal surrogate targets. Despite intensive efforts for decades by many protein crystallography groups, the only experimental 3D structure for any GPCR of any form of life is bovine rhodopsin (7). This is due to poor expression levels, difficulties in large-scale receptor purification, the insolubility in media lacking phospholipids, and the difficulty in crystallization. Unfortunately, the homology of bovine rhodopsin with most drug targets is too low (<20%) for use in homology methods to predict structures.
To provide the 3D structures needed to understand the function of GPCR and to help design new ligands, we developed the MembStruk method (8) to predict the 3D structure (without using homology to known 3D structures), which we have applied to predict the 3D structures for 10 GPCR families. We will summarize here the results for ß2 adrenergic receptor (AR), D2 dopamine receptor (DR), CCR1 chemokine receptor, M1 muscarinic acetylcholine receptor (MAR), and the Mas-related gene (Mrg)C11 receptor. These GPCR families have been the focus of many experimental studies, providing a great deal of binding and mutation experimental data for the validation of our predictions.
After predicting the best structure for each GPCR, we validated this structure by using the HierDock procedure (9) or MSCDock (10) to predict the binding site, binding configuration, and binding energy of known strongly binding ligands. Then we used the predicted structure for the ligand-GPCR complex to predict which mutations would dramatically decrease or increase binding. We have applied HierDock/MSCDock to successfully predict the binding site structure and binding energy of 5–15 ligands to human D2 DR (11), human ß2 AR (12), human M1 MAR (13), human CCR1 chemokine receptor (14), and mouse MrgC11 receptors (J. Heo, S.-K. Han, N. Vaidehi, J. Wendel, P. Kekenes-Huskey, and W. A. Goddard III, unpublished results). These predictions have compared well to available experimental binding and mutation data even for larger ligands such as haloperidol and domperidone. Indeed, in some cases, the predictions preceded the experiments.
Our approach was to predict the structures for GPCR without using any experimental data (no constraints). We did, of course, compare our predictions to any available experimental data; however, we did not modify our predicted structures to fit experimental data. Rather, we used any apparent disagreements with experiments to warn us to reexamine our methods and assumptions. This was useful for improving our methods, which now generally give the correct answer for all known cases.
These atomic level 3-D structures are providing a detailed understanding of the structural conformational space of these receptors and their ligand-binding properties, which should allow these structures to be related to functional activities of these important receptors. Given the importance of GPCR physiologically and as pharmacologic targets, the results may provide useful new insights of broad biological and medical importance.
|
Results |
|---|
|
TOPABSTRACTResultsDiscussionLITERATURE CITED |
|---|
Structure and function of ß2-AR
The first extensive validation of the MembStruk and HierDock methods was for the 3D structure of agonists and antagonists bound to ß2 AR (12). Figure 1 shows the overall structure, whereas Figure 2A shows the details of the predicted structure for epinephrine (EPI) bound to human ß2 AR. We predicted that: 1) the amine group of EPI makes a 2.9-Å salt bridge with aspartic acid (Asp)-113 (TM3); 2) the 2 catechol hydroxy groups make hydrogen bonds to the 3 serine (Ser) in TM5 [the 4 hydrogen bond distances are 2.94 Å (Ser-204-meta OH), 3.05 Å (Ser-207-para OH), 3.14 Å (Ser-203-meta OH), and 3.00 Å (Ser-203-para OH)]; 3) asparagine (Asn)-293 make a hydrogen bond to the alkyl OH of EPI; and 4) phenylalanine (Phe)-290 makes a hydrophobic contact with the aromatic ring of EPI.
|
|
|
Dynamic behavior of fully solvated ß2-AR, embedded in the membrane with bound agonist or antagonist
To optimize all possible positions of the ligand and all possible packings of the helices while simultaneously describing the structures of the lipid, water, and ions was impractical. To find the correct structures, it was essential that MembStruk and HierDock sample all reasonable configurations of the ligand and protein and, hence, we severely restricted the description of the lipid and solvent so important to the stability of the GPCR-ligand complex. Thus, the calculations used implicit solvent and only 50 lipid molecules (2 layers around the TM domain). Then, after predicting the ligand-GPCR complex, we removed these restrictions by embedding the optimum protein-ligand complex in an infinite (periodic) lipid bilayer, embedding the membrane-protein-ligand complex in an infinite (periodic) box of equilibrated water, and adding the appropriate counter ions or salt. We then minimized the full structure (40,000–100,000 atoms), heated gradually to 300 K, and equilibrated the structure for 10 ns at 300 K.
For the 6 HierDock-MembStruk structures we have tested in this way (several ligands with ß2 AR (17), 1 ligand with CCR1 chemokine receptor (14), 2 ligands with M1 MAR (13), we found that the predicted structures remained stable with modest changes in structures and interactions. However, in several systems, a couple of water molecules invaded the active site, often helping to stabilize hydrogen bond interactions. We illustrated this observation with 1 case, ß2 AR (17).
Spijker et al. (17) considered 3 systems: human apo-ß2 AR, the endogenous agonist EPI bound to ß2 AR, and the ß2 AR selective antagonist butoxamine bound to ß2 AR. A visual indication of the stability of the structure is shown in Figure 4, which compares the original structure from the MembStruk-HierDock predictions of EPI-ß2 AR (left half of Fig. 4) to the structure after 4 ns of molecular dynamics (MD) simulation in the presence of a full membrane and water (100,000 atoms). Slight changes occurred in the conformation of the protein (7° change in tilt for 2 helices, bringing them closer). The EPI moved by 1.2 Å (accommodating a water near the catechol OH and TM5 Ser), but it retained all of the previous strong interactions with the crucial residues. The biggest changes occurred in the intracellular (IC) and extracellular (EC) loops. These underwent large deformations (as shown in Fig. 4), with C
-atom root mean squared differences from 3–5 Å.
|
Figure 5 shows the variation over 6 ns of various nonbond distances of EPI (upper) or butoxamine (lower) to the residues of ß2 AR. They showed that the salt bridge between the Asp-113 (TM3) and the protonated amine was preserved throughout the dynamics for both ligands (arrows 1 and 2), leading to an average length of 2.8 Å for EPI-ß2 AR and 3.0 Å for butoxamine-ß2 AR. The ligand-protein hydrogen bonds, however, fluctuated during the dynamics. The hydrogen bond between the meta-catechol OH group of EPI with Ser-203 (TM5) was maintained as a direct hydrogen bond for 0.8 ns (breaking at arrow 3) and as a water-mediated hydrogen bond for another 0.8 ns (breaking at arrow 4). The hydrogen bond between Asn-293 and the alkyl OH group of EPI was lost near the beginning of the simulation (arrow 3) but formed again after 2 ns (arrow 5). Most hydrogen bonds between TM4 (Ser-165) and TM5 (Ser-204) were lost almost immediately (arrow 3). During the MD simulations, butoxamine continued to exhibit hydrophobic interactions with histidine (His)-172 (TM4), Trp-286 (TM5), and Ile-294 (TM6). Figure 5 shows that these interactions remained tight through the entire MD run. We found that the hydrophobic interactions with Val-119, Ile-169, and Phe-290 remained rather constant (4.5 Å) throughout the MD run.
|
This MD validation of the structure predictions for GPCR in explicit lipid and water suggests that these methods lead to predicted 3D structures sufficiently accurate for predicting the binding sites of agonists and antagonists. Thus, it should be possible to use these structures to design subtype-specific ligands.
The changes in the interhelical hydrogen bonding between TM 4 and 5 as the agonist binds were most interesting, remaining fairly stable throughout the simulation. In addition, the incorporation of water-mediated hydrogen bonds showed that water can play a critical role in ligand binding. The differences in the simulations among the antagonist-protein, the agonist-protein, and the apo-protein structures indicated that 6 ns of simulation was already sufficient to show differences between agonist and antagonist. Such differences should be testable experimentally.
Although we found that the agonist-ß2 AR dynamics differed substantially from the dynamics of the apo-protein and antagonist-protein of systems, we did not expect such simulations to be sufficiently long to follow the activation events. We are now extending our procedures to generate an ensemble of low-lying structures with agonist bound with the hope that some of these will correspond to intermediates in the process of activation, providing structural information that could be used to design experimental tests.
Prediction of structure and function for human D2 DR
There are 5 human DR, usually partitioned into 2 subfamilies. DR1 (consisting of the D1 and D5 receptors, which have an 82% sequence identity) and DR2 (consisting of D2–D4, which also have high sequence identity), with a sequence identity between D1 and D2 of only 44%. The DR2 class has a long 3rd IC loop and a short carboxyl terminus, whereas the DR1 class has a short 3rd IC loop and a long carboxyl terminus. Such differences are likely to cause functional differences between these classes of the receptors. Certainly, the differences in the length of the 3rd IC loop could affect the interaction with the G-protein.
Kalani et al. (11) used MembStruk to predict the 3D structure of the human D2 DR from the primary sequence and used HierDock to predict the binding site for dopamine and 10 other ligands. The predicted binding site for dopamine (see Fig. 6) and for other agonists is located between TM helices 3–6. The residues in contact with dopamine in the predicted binding site (shown in Fig. 7) are in excellent agreement with the results from the mutation studies.
|
|
Other residues in the binding pocket provided a mostly hydrophobic packing for the ligand. These residues included TM3 (Phe-110, methionine-117, and cysteine-118), TM4 (Phe-164), TM5 (Phe-189 and Val-190), and TM6 (Trp-386, Phe-390, and Trp-393). Although Trp-386 and Phe-390 of the well-conserved WXXFF motif on TM6 were within 5 Å of the dopamine-binding site, Phe-389 was slightly (>5.5 Å) away from dopamine (20,21). Kalani et al. (11) found that His-393 in TM6 provided a very strong component of the binding energy of dopamine to the receptor. Because Asp-114 can form either a bidentate salt bridge or a salt bridge and a hydrogen bond to the amino group of dopamine, the interaction with His-393 fulfills the ability of the amino group to form favorable interactions with any other residues.
Agonists- vs. antagonists-binding site. To validate further the predicted structure for D2, Kalani et al. (11) used HierDock to predict the binding configuration and the relative binding affinities of 3 known dopamine agonists (antiparkinsonian) and 7 known antagonists (antipsychotic) in the D2 receptor (Fig. 8). For each of these ligands, the entire molecular surface of D2 receptor was scanned using the ScanBindSite procedure (9). For these agonists, the predicted binding site is located between TM–6. From the binding mode of agonists, it was speculated that the agonist binding pulls TM3 and TM5 closer together while allowing flexibility for the motion of TM6.
|
Class II antagonists are long molecules with aromatic moieties at each end connected by a flexible linker containing a protonated amine (e.g. haloperidol). Thus, one aromatic moiety occupies the hydrophobic pocket formed by TM3–6, whereas the other occupies the hydrophobic pocket formed by TM2, 3, and 7. The relative location of the agonist- vs. antagonist-binding sites is shown in Figure 9. The residues within 5.5 Å of the clozapine-binding site are shown in Figure 10A and that of class II antagonist haloperidol are shown in Figure 10B. Also listed are the residues implicated by mutation experiments to be involved in binding of agonists and antagonists. Such residues found within 5 Å of the predicted structure are colored black in the list in Figure 10B. The residues in red are residues that experiment (18) found to affect binding affinity of the antagonist but were not found by theory. The residues listed in blue are predicted within 5 Å of the binding site but have not yet been studied experimentally. Thus, the predicted binding site agrees well with the mutation experiments (18,22–25).
|
|
Comparison of antagonist-binding sites. Class I antagonists The clozapine-like antagonists form a salt bridge to the TM3 Asp-114 with their protonated amino group. They have just 1 aromatic moiety and therefore utilize the aromatic microdomain between TM4 and TM6 (24). They form only 1 weak interaction with the TM6 Ser network, as also for both class I and class II antagonists. The Kalani results indicate that the interaction may be with Ser-193 or Ser-197. These calculations suggest that both the number and strength of the interactions with the TM5 Ser are important for activation. However, the situation must be more complicated, because Strange et al. (26) identified a nonhydroxylated member of the DPAT series as a weak agonist.
The critical distinguishing feature of an agonist vs. an antagonist appears to be its relative position to TM6. Class I antagonists bury their aliphatic domain deep into the conserved TM6 WXXFF motif. Experimental studies on rhodopsin suggest that motion of TM3 and TM6 is necessary for activation (27). The binding of antagonists at the TM6 WXXFF would prevent such motion, specifically the hinge motion between TM3 and TM6 that is likely necessary for activation. The class I antagonists are further stabilized by Phe-110 in TM3.
Class II antagonists The protonated amino group of haloperidol-like antagonists forms a salt bridge to TM3 Asp-114. This leaves aromatic ring moieties at both ends. In most cases, only 1 of the aromatic rings is halogenated and this ring binds to the first TM46 aromatic microdomain, with the second ring binding in the TM27 aromatic microdomain. The halogenated ring binds effectively into the cavity between TM4 and 6 to form weak interactions with either Ser-193 or Ser-197 in TM5. It gains stability from the presence of TM4 Trp-160 and Phe-164 and TM6 Trp-386 and Phe-390. In some cases, His-393 (TM6) may also stabilize the class II antagonists. The nonhalogenated aromatic domain located in the cavity between TM2 and TM7 gains stabilization from Trp-90 (TM2), Phe-110 (TM3), Trp-413 (TM7), and tyrosine (Tyr)-416 (TM7). As for class I, the class II antagonists prevent motion between TM3 and TM6 by burying their aliphatic portion between TM3 and TM6, thereby preventing interaction of these helices. There is little difference between the binding sites of the class II antagonists, although some more fully utilize the cavity.
Figure 11 compares the calculated binding energies with experimental lnKd values. It is difficult to validate the accuracy of the predictions, because the experimental Kd values for a given ligand and receptor often vary by factors of 10 or 20. Generally, the trend of calculated binding energies agrees with the trend of experimental Kd values and usually ligands with binding constants that differ by a factor of 10 are properly ordered. It may seem surprising that the theory does so well at predicting the binding site but only marginally at predicted binding energy. There are 2 problems here. One is that the binding energy is the difference between the energy of the ligand-protein complex and that of the separate ligand and protein complexes. We generally do little to optimize these separate species, usually just removing the ligand, minimizing (to remove strain), and solvating. The 2nd problem is that the calculated binding energies are for minimized structures. That is, we ignore dynamical effects, including entropy. We plan to use the full solvent dynamics at 300 K to extract entropy and enthalpy information, but long time scales may be required to obtain significant results.
|
The 18 human chemokine GPCR are important drug targets for such autoimmune diseases as multiple sclerosis and for blockading HIV-1 entry. CCR1 responds to a number of ligands, particularly those associated with a wide variety of autoimmune and pro-inflammatory diseases. This made the CCR1 GPCR an attractive therapeutic target. In particular, Berlex developed BX 471, a potent, orally available specific antagonist to CCR1, currently in phase II clinical trials. To resolve some puzzles in the structure-activity relations and to explain significant differences between the effect of BX471 on human and an animal model, we used MembStruk to predict the 3D structure of the human CCR1 receptor and HierDock to predict the binding site of BX 471 (14). These results explained a variety of structure-activity relations data and some puzzles regarding differential binding in animal models.
The predicted structure for BX 471 bound to the predicted structure of the CCR1 receptor is shown in Figure 12. The binding site is located between TM3 and TM7. The strongest protein-ligand interactions are of hydrophobic character, arising from Y113 (TM3) and I259 (TM6), which anchor the piperazine ring, and Y114 (TM3), which provides pi-stacking interactions to the adjacent phenyl ring with the urea group. Neither Tyr forms strong hydrogen bonds to BX 471. The fluorine on BX 471 interacts favorably with 1 of the methyl groups of leucine-87 and with the phenyl ring of Phe-83. The urea group of BX 471 does not make tight polar interactions with the protein (one might have expected it to form a hydrogen bond to the carbonyl group of the glycine-168 backbone or the OH-group of Tyr-114). MD simulations of the CCR1/BX 471 complex in explicit lipid bilayer and water indicate that this group is highly flexible and forms hydrogen bonds with water molecules that enter the binding region. The possible contribution of residues in the loop to binding was not included in these calculations.
|
|
Prediction of the 3D structure for FMRF-amide peptides bound to mouse MrgC11 GPCR and experimental validation
Mouse MrgC11 belongs to the Mrg receptor family of GPCR and is localized mainly to isolectin B4+, the small-diameter nociceptors in the dorsal root ganglia likely associated with pain perception. It is not known which endogenous ligand activates mMrgC11 or what function it has, but Dong et al. (28) showed that mMrgC11 is activated by FMRF-amides (Phe-Met-Arg-Phe) and suggested that it is involved in pain sensation or modulation. In particular, Dong et al. (28) showed that mMrgC11 is activated by the F-M-R-F-NH2, (D)F-M-R-F-NH2, and F-(D)M-R-F-NH2 tetrapeptides at 10–100 nmol concentration, while F-M-(D)R-F-NH2 and F-M-R-(D)F-NH2 are inactive (>10 µmol).
Using MembStruk and HierDock, Heo et al. (J. Heo, S-K Han, N. Vaidehi, J. Wendel, P. Kekenes-Huskey, and W. A. Goddard III, unpublished results) predicted the 3D structure for mMrgC11 and the binding site for the 5 chiral FMRF-NH2 ligands, one of which is shown in Figure 13. Table 2 shows that Heo predicted correctly the relative binding observed for the 5 FMRF-NH2 ligands. Thus, Heo predicted that the binding is in the sequence F-(D)M-R-F-NH2 > F-M-R-F-NH2 > (D)F-M-R-F-NH2, as observed experimentally and predicted that F-M-(D)R-F-NH2 and F-M-R-(D)F-NH2 bind far more weakly. The reason for this dramatic dependence on chirality is that first, the RFa (Arg-Phe amide) terminus is buried, interacting strongly with Tyr-110 (TM3), Asp-161 (TM4), and Asp-179 (TM5), then the M comes out of the pocket with little selective binding, and finally, the terminal F sticks back into the pocket, interacting with some Y and W residues. The 2 bad binders have the wrong chirality for the terminal RFa to stick into the primary part of the binding site.
|
|
These results showing that mutations to alanine of Tyr-110, Asp-161, or Asp-179 eliminate or dramatically reduce binding to mMrgC11 validate both the predicted GPCR structure and the predicted binding site. This validation indicates that it is now practical to predict GPCR structures sufficiently accurately to identify binding sites for selective ligands.
Structure and function of human M1 MAR
All 5 subtypes of MAR have been implicated in the various pathogenetical processes of a wide range of diseases, including cognitive dysfunctions (Alzheimer's disease, Parkinson's disease, and schizophrenia) and smooth muscle disorders (irritable bowel syndrome, overactive bladder, and chronic obstructive pulmonary disease). These receptors have attracted much attention for the design of drug molecules (e.g. agonists for M1 and antagonists for M2) to stimulate the cholinergic system for potential treatment of Alzheimer's disease. The search for M1 receptor agonists has led to a number of drug candidate molecules, many of which had to be discontinued due to undesirable side effects resulting from cross reactivity to other receptor subtypes. Such cross reactivity would likely be greatly reduced by using a structure-based approach for the rational design of drugs coupled with a database of 3D structures to use as antitargets.
Peng et al. (13) predicted the structure of the human MAR 1 (M1) using the MembStruk method and validated this structure by using HierDock to predict the ligand-binding sites for the 7 agonists and antagonists shown in Figure 14. The interhelical contacts that stabilize the receptor conformation, the predicted binding sites for agonists and antagonists, and the results of in silico mutagenesis all agree very well with available experimental mutagenesis data. The calculated relative binding energies also reproduce well the order of measured binding affinities.
|
|
|
|
interactions (with a geometry similar to that in the nicotinic acetylcholine receptor crystal structure complex). Peng's (13) results also suggest that upon binding of ligands with a charged tertiary amine head group, there might be proton transfer from the ligand to Asp-105 (TM3). In addition, an extensive aromatic network involving Tyr-106 (TM3), Trp-157 (TM4), Phe-197 (TM5), Trp-378 (TM6), and Tyr-381 (TM6) is important in stabilizing the antagonist binding sites (Fig. 16 for QNB antagonist). For antagonists with 2 terminal aromatic rings, this aromatic network extends to Trp-164 (TM4), Tyr-179 (EC loop 2), and possibly Phe-391 (TM6). Asn-382 (TM6) forms hydrogen bonds with selected antagonists. Tyr-381 (TM6) and Ser-109 (TM3) form hydrogen bonds with the ester moiety of acetylcholine, which binds in the gauche conformation.
The predicted structures show the importance of aromatic residues in the binding site, illustrating the cation- interactions and aromatic-aromatic interactions critical to ligand binding in M1. This study provides an improved structural understanding of the M1 receptor and its binding sites. In addition, it sets the stage for applying these procedures to predict the structures and function of the 4 other muscarinic receptor subtypes, which should enable the development of subtype-specific agonists and antagonists with reduced side effects.
|
Discussion |
|---|
|
TOPABSTRACTResultsDiscussionLITERATURE CITED |
|---|
Most of the current successes using MembStruk and HierDock have been with the A family of GPCR; however, the methodologies used here should also apply to the other families. Thus, with sufficient funding it would be practical to obtain similar quality structures for all 350 nonsensory receptors of humans over 5 y. The advantage of having all these structures is that in designing a drug to target 1 of these GPCR, we could use the other 349 as antitargets, minimizing the binding to them. This might provide a means to dramatically reduce cross reactivity, thereby reducing toxic side effects.
We have also reported structures and ligand bonding for 10 olfactory receptors (9,29–32) and a taste receptor (33). Almost no experimental data on mutations or binding of antagonists exists. In addition, there may be much less selectivity. It would be practical to obtain good quality structures for all 500 human sensory receptors. A special advantage of doing this might be to compare binding of various odorants and tastes to perceptions by humans of these same odorants and tastes. This might lead to new formulations for perfumes and foods. It might also be useful in training humans or animals to detect odorants associated with diseases or the vapors associated with explosives and nerve gases. Once the structures of 500 odorant receptors and their binding characteristics for a library of ligands is known, it will be especially interesting to compare the binding strengths (from theory) with the actual perceptions of people along with functional NMR to follow which structures are associated with the perceptions.
It will also be interesting to compare the 3D structures and binding sites of GPCR in mammals with those in other forms of life to better understand the evolution of neural and sensory systems.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Supported in part by NIH (R21-MH073910-01-A1) and by the Materials and Process Simulation Center. The original funding for this work was through an ARO-MURI (Dr. R. Campbell) and then by Aventis (now Sanofi-Aventis) and Berlex (now part of Bayer-Schering).
3 Author disclosures: Travel expenses for W. A. Goddard III to attend meeting paid by The Ajinomoto Company, Inc.; R. Abrol, no conflicts of interest.
4 Color versions of Figures 1, 2, 3, 6, 7, 9, 12, 16, and 17 are available with the online posting of this paper at jn.nutrition.org.
5 Abbreviations used: AR, adrenergic receptor; Asn, asparagine; Asp, aspartic acid; 3D, 3-dimensional; DR, dopamine receptor; EC, extracellular; EPI, epinephrine; GPCR, G protein-coupled receptor; His, histidine; IC, intracellular; Ile, isoleucine; MAR, muscarinic acetylcholine receptor; MD, molecular dynamics; Mrg, mas-related gene; Phe, phenylalanine; QNB, quinuclidinyl benzilate; Ser, serine; TM, transmembrane; Trp, tryptophan; Tyr, tyrosine; Val, valine.
|
LITERATURE CITED |
|---|
|
TOPABSTRACTResultsDiscussionLITERATURE CITED |
|---|
1. Fredriksson R, Lagerstrom MC, Lundin LG, Schioth HB. The G-protein-coupled receptors in the human genome form five main families. Phylogenetic analysis, paralogon groups, and fingerprints. Mol Pharmacol. 2003;63:1256–72.
2. Lefkowitz RJ, Pierce KL, Luttrell LM. Dancing with different partners: protein kinase A phosphorylation of seven membrane-spanning receptors regulates their G protein-coupling specificity. Mol Pharmacol. 2002;62:971–4.
3. Bartfai T, Benovic JL, Bockaert J, Bond RA, Bouvier M, Christopoulos A, Civelli O, Devi LA, George SR, et al. The state of GPCR research in 2004. Nat Rev Drug Discov. 2004;3:575, 577–626.
4. Tang CM, Insel PA. Genetic variation in G-protein-coupled receptors-consequences for G-protein-coupled receptors as drug targets. Expert Opin Ther Targets. 2005;9:12472–65.
5. Lundstrom K. Latest development in drug discovery on G-protein coupled receptors. Curr Protein Pept Sci. 2006;7:465–70.[Medline]
6. Hopkins AL, Groom CR. The druggable genome. Nat Rev Drug Discov. 2002;1:727–30.[Medline]
7. Palczewski K, Kumasaka T, Hori T, Behnke CA, Motoshima H, Fox BA, Le Trong I, Teller DC, Okada T, et al. Crystal structure of rhodopsin: a G protein-coupled receptor. Science. 2000;289:739–45.
8. Vaidehi N, Floriano WB, Trabanino R, Hall SE, Freddolino P, Choi EJ, Goddard WA III. Structure and function of GPCR. Proc Natl Acad Sci USA. 2002;99:12622–7.
9. Floriano WB, Vaidehi N, Singer M, Shepherd G, Goddard WA III. Molecular mechanisms underlying differential odor responses of a mouse olfactory receptor. Proc Natl Acad Sci USA. 2000;97:10712–6.
10. Cho A, Wendel JA, Vaidehi N, Kekenes-Huskey PM, Floriano WB, Maiti PK, Goddard WA III. The MPSim-Dock Hierarchical Docking Algorithm: application to the eight trypsin inhibitor co-crystals. J Comput Chem. 2005;26:48–71.[Medline]
11. Kalani MY, Vaidehi N, Hall SE, Floriano WB, Trabanino RJ, Freddolino PL, Kam V, Goddard WA III. The predicted 3D structure of the human D2 dopamine receptor and the binding site and binding affinities for agonists and antagonists. Proc Natl Acad Sci USA. 2004;101:3815–20.
12. Freddolino PL, Kalani MY, Vaidehi N, Floriano WB, Trabanino RJ, Freddolino PL, Kam V, Goddard WA III. Structure and function prediction for human ß2-adrenergic receptor. Proc Natl Acad Sci USA. 2004;101:2736–41.
13. Peng JY, Vaidehi N, Hall SE, Goddard WA III. The predicted 3D structures of the human M1 muscarinic acetylcholine receptor with agonist or antagonist bound. ChemMedChem. 2006;1:878–90.[Medline]
14. Vaidehi N, Schlyer S, Trabanino RJ, Floriano WB, Abrol R, Sharma S, Kochanny M, Koovakat S, Dunning L, et al. Predictions of CCR1 chemokine receptor structure and BX 471 antagonist binding followed by experimental validation. J Biol Chem. 2006;281:27613–20.
15. Strader CD, Fong TM, Tota MR, Underwood D, Dixon RAF. Structure and function of G-protein coupled receptors. Annu Rev Biochem. 1994;63:101–32.[Medline]
16. Liapakis G, Ballesteros JA, Papachristou S, Chan WC, Chen X, Javitch JA. The forgotten serine: a critical role for Ser-203(5.42) in ligand binding to and activation of the beta(2)-adrenergic receptor. J Biol Chem. 2000;275:37779–88.
17. Spijker P, Vaidehi N, Freddolino PL, Hilbers PA, Goddard WA III. Dynamic behavior of fully solvated ß2-adrenergic receptor, embedded in the membrane with bound agonist or antagonist. Proc Natl Acad Sci USA. 2006;103:4882–7.
18. Shi L, Javitch JA. The binding site of aminergic G protein-coupled receptors: the transmembrane segments and second extracellular loop. Annu Rev Pharmacol Toxicol. 2002;42:437–67.[Medline]
19. Teeter MM, Froimowitz M, Stec B, DuRand CJ. Homology modeling of the dopamine D2 receptor and its testing by docking of agonists and tricyclic antagonists. J Med Chem. 1994;37:2874–88.[Medline]
20. Cho W, Taylor LP, Mansour A, Akil H. Hydrophobic residues of the D2 dopamine receptor are important for binding and signal transduction. J Neurochem. 1995;65:2105–15.[Medline]
21. Javitch JA, Ballesteros JA, Weinstein H, Chen J. A cluster of aromatic residues in the sixth membrane-spanning segment of the dopamine D2 receptor is accessible in the binding-site crevice. Biochemistry. 1998;37:998–1006.[Medline]
22. Fu D, Ballesteros JA, Weinstein H, Chen J, Javitch JA. Residues in the seventh membrane-spanning segment of the dopamine D2 receptor accessible in the binding-site crevice. Biochemistry. 1996;35:11278–85.[Medline]
23. Shi L, Simpson MM, Ballesteros JA, Javitch JA. The first transmembrane segment of the dopamine D2 receptor: accessibility in the binding-site crevice and position in the transmembrane bundle. Biochemistry. 2001;40:12339–48.[Medline]
24. Simpson MM, Ballesteros JA, Chiappa V, Chen J, Suehiro M, Hartman DS, Godel T, Snyder L, Sakmar TP, et al. Dopamine D4/D2 receptor selectivity is determined by a divergent aromatic microdomain contained within the second, third, and seventh membrane-spanning segments. Mol Pharmacol. 1999;56:1116–26.
25. Dijkstra D, Rodenhuis N, Vermeulen ES, Pugsley TA, Wise LD, Wikstrom HV. Further characterization of structural requirements for ligands at the dopamine D2 and D3 receptor: exploring the thiophene moiety. J Med Chem. 2002;45:3022–31.[Medline]
26. Strange PG. Dopamine receptors: studies on structure and function. Adv Drug Res. 1996;28:313–51.
27. Hubbell WL, Altenbach C, Hubbell CM, Khorana HG. Rhodopsin structure, dynamics, and activation: a perspective from crystallography, site-directed spin labeling, sulfhydryl reactivity and disulfide cross-linking. Adv Protein Chem. 2003;63:243–90.[Medline]
28. Dong X, Han S, Zylka MJ, Simon MI, Anderson DJ. A diverse family of GPCR expressed in specific subsets of nociceptive sensory neurons. Cell. 2001;106:619–32.[Medline]
29. Floriano WB, Vaidehi N, Goddard WA III. Making sense of olfaction through predictions of the 3-D structure and function of olfactory receptor. Chem Senses. 2004;29:269–90.
30. Hall SE, Floriano WB, Vaidehi N, Goddard WA III. Predicted 3-D structures for mouse I7 and rat I7 olfactory receptors and comparison of predicted odor recognition profiles with experiment. Chem Senses. 2004;29:595–616.
31. Hummel P, Vaidehi N, Floriano WB, Hall SE, Goddard WA III. Test of the binding threshold hypothesis for olfactory receptors: explanation of the differential binding of ketones to the mouse and human orthologs of olfactory receptor 912–93. Protein Sci. 2005;14:703–10.[Medline]
32. Trabanino R, Hall SE, Vaidehi N, Floriano WB, Goddard WA III. First principles prediction of the structure and function of G protein-coupled receptors: validation for bovine rhodopsin. Biophys J. 2004;86:1904–21.[Medline]
33. Floriano WB, Hall S, Vaidehi N, Kim U, Drayna D, Goddard WA III. Modeling the human PTC bitter-taste receptor interactions with bitter tastants. J Mol Model. 2006;12:931–41.[Medline]
This article has been cited by other articles:
|
|
 |
|
  J. L. Klotz, B. H. Kirch, G. E. Aiken, L. P. Bush, and J. R. Strickland Contractile response of fescue-naive bovine lateral saphenous veins to increasing concentrations of tall fescue alkaloids J Anim Sci, January 1, 2010; 88(1): 408 - 415. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Grosmaitre, S. H. Fuss, A. C. Lee, K. A. Adipietro, H. Matsunami, P. Mombaerts, and M. Ma SR1, a Mouse Odorant Receptor with an Unusually Broad Response Profile J. Neurosci., November 18, 2009; 29(46): 14545 - 14552. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Takai Introduction to the Transdisciplinary International Conference on Aromatic Amino Acids and Related Substances: Chemistry, Biology, Medicine, and Application J. Nutr., June 1, 2007; 137(6): 1501S - 1503S. [Full Text] [PDF]
|
|
|
|
|
|
Y. Shimomura and S. M. Morris Jr. Biology and Biochemistry: Discussion of Session 2 J. Nutr., June 1, 2007; 137(6): 1548S - 1548S. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
