The Journal of Neuroscience, November 10, 2004, ():
Molecular Determinants of Ligand Selectivity in a Vertebrate Odorant Receptor
J. Neurosci. Luu et al.
24: 10128
Supplemental data
Files in this Data Supplement:
- supplemental material -
Figure S1. Alignment of Receptor 5.24, Receptor ZO6, mGlu1 and LIVBP
For the purpose of homology modeling, the NTD sequences of goldfish receptor 5.24 (Speca et al., 1999) and zebrafish receptor ZO6 (the ortholog of receptor 5.24) were aligned to the corresponding sequence of the mGlu1 receptor and LIVBP. Letter coloration indicates secondary structure (?-helix: red; ?-sheet: green); the ?-helices or ?-sheets are identified above each row of sequence. The mGlu1 sequence region which is not resolved in 1ewk:A (residues 125 – 153) is colored in purple. The ribbons beneath the sequence indicate domain localization, (cyan: lobe 1; magenta: lobe 2; yellow: hinge region). Yellow highlighting shows residues that bind to the glycine amino acid moiety in the mGlu1 crystal structure. Blue highlighting indicates mGlu1 residues that coordinate side chain functions. Grey highlighting indicates loops at the receptor 5.24 binding site. Secondary structure predictions of receptors 5.24 and ZO6 were performed at http://cubic.bioc.columbia.edu/predictprotein. Receptor 5.24 and mGlu1 NTDs display 25% amino acid sequence identity and 48% sequence similarity. Receptors 5.24 and ZO6 exhibit 70.8% identity and 85% similarity in this region.
- supplemental material
-
Figure S2. Docking and Scoring of Amino Acid Ligands in the Model of Receptor 5.24 NTD Correlate with Observed Affinities
Using LigandFit (DS Modeling 1.2, Accelrys), all 20 L-amino acid ligands were docked in the model of receptor 5.24. Theoretical binding scores were obtained using the Jain scoring function, and were plotted against the -logKi of the respective amino acids (Speca et al., 1999). The binding of glycine, which has no side chain, reflects its interactions with the proximal binding pocket residues. Thus, lower Ki values (higher affinities) represent stabilizing ligand side chains, while higher values (lower affinities) could be due to destabilizing side chains. Linear hydrophobic side chains (e.g., methionine) appear to stabilize the active closed form of NTD as opposed to more bulky amino acids (e.g., histidine). Arginine and lysine – the longest ligands – are predicted to interact with the largest number of residues as described above. Other hydrophilic amino acids can be ranked according to the number of possible hydrogen bonds that can be established with the cluster of serines and N310 in the distal binding pocket (see above). In general, the longer the ligand’s side chain, the more of these interactions are possible. For example, glutamine is shorter than arginine or lysine but longer than asparagine, serine and threonine, and indeed a consistent ranking is found with the Jain scores. The remarkable correlation between theoretical binding scores and actual binding data shows that the Jain scoring function accurately predicts the rank order of the ligands in this structural model.
- supplemental material
-
Figure S3. Calcium Imaging Assay for Receptor Function
A) HEK 293 cells were transiently transfected with a receptor 5.24 expression construct. Receptor activation in these cells was monitored optically using the fluorescent calcium indicator dye, Fura-2AM (see Materials and Methods). Fura-2AM loaded cells were exposed to a continuous flow of CIB buffer and exposed sequentially to L-arginine at the following concentrations: 250nM (60-180s), 500nM (480-540s), 1µM (960-1020s), 50µM/saturation (1320-1380s). The figure shows averaged raw traces from 38 receptor 5.24 positive and responsive cells plotted as ?F/F0 versus time.
B) Dose response of receptor 5.24 expressing cells to L-arginine, plotted as %peak response vs. ligand concentration. Peak response was determined for each cell in an imaged field by exposure to a saturating concentration of arginine (50µM). Each point consists of measurements from a minimum of 30 cells. Error bars show the standard error for each point. The EC50 from this curve is 788 +/- 54.8 nM.
- supplemental material
-
Figure S4. Comparison of Peak Responses from Different Receptor Constructs and Agonists
Comparison of peak responses from different receptor 5.24 (panel A) and receptor ZO6 (panel B) constructs and agonists. Peak responses, expressed as ?F/F0 values, were determined on a per-cell basis in response to a saturating dose of the indicated ligand (100x EC50 for all ligands shown). Each box plot is composed of data from at least two runs, each containing 20 cells or more. The center of each box represents the mean peak response in the population of responding cells; the box height represents the 25-75% range for the data; the error bars represent the 5-25% and the 75-95% range. X marks represent the 1% and 99% range. The extremes are plotted as dashes. Individual plots are labeled with the construct name, followed by the ligand used for stimulation (e.g. S152AR denotes the peak responses for the mutant S152A with Arginine (R) as the ligand).
- supplemental material
-
Figure S5. Proposed Ionic Interactions of the Arginine Ligand’s Basic Side Chain with Distal Pocket Residues
“Top-down” view of the wild type receptor 5.24 binding pocket, highlighting selected distal pocket residues (E47, D388 and Y72) involved in contacting the guanidinium moiety of an arginine ligand in the wild-type receptor (panel A). The orientation of this view is from above lobe 1 looking down onto the bound ligand, with the proximal pocket toward the bottom and the distal pocket toward the top of the figure. Single amino acid substitutions were made at D388 (D388A, panel B) and E47 (E47L, panel C; E47K, panel D), and the arginine ligand was redocked to show the preferred ligand and side chain conformations. In the D388A and E47L mutants, the guanidinium is predicted to shift closer to the remaining carboxylate, as indicated; this deflection is not predicted with the E47K mutation. In the E47K mutant, the K47 side chain is predicted to turn away from the binding pocket, possibly forming a salt bridge with D50 or D112; the arginine is not predicted to change position. For all panels, receptor residues are shaded cyan, with oxygens shown in red. On the ligand, hydrogens are shown in white, carbons in green, nitrogens in blue and oxygens in red. Proposed intermolecular interactions and distances are indicated with dashed green lines.
