G-protein coupled receptor (GPCR) activation
The ultimate goal of the proposed research is to better understand the molecular mechanism(s) for G-protein coupled receptor (GPCR) activation induced by a ligand initially bound to the inactive state of the receptor. It is essential to construct a realistic atomic model of the GPCR-G-protein complex in consideration of the lipid environment. The current objectives are: (1) to construct a molecular model of the human brain cannabinoid (CB1) receptor, including the transmembrane (TM) helices, the extracellular/intracellular loops, and the shortened N- and C-terminals; and (2) to construct a molecular model of the CB1-G-protein complex embedded in a lipid bi-layer.
Simulation system of the CB1 receptor in a POPC bilayer is shown Figure 1. The palmitoyl moiety (PAL) that is covalently bonded to C415 was included. A total of ~68,000 atoms, including 14,625 water molecules, 140 lipids, 62 Na+, and 77 Cl-, resulted in a system of 80 Å × 98 Å × 98 Å, was subjected to a MD simulation of > 100 ns. The lateral area of the simulated system was ~ 63 Å2, which appeared to be in agreement with the experimentally measured values in the range of 63 - 68 Å2 of the liquid-crystalline phase, which is the most biologically relevant phase. Density profiles for the CB1 receptor in the POPC bilayer along the membrane normal is shown in Figure 2. It is revealed that several coordinated water molecules are located close to the highly conserved residues of the S(N)LAxAD motif, the D(E)RY motif, the CWxP motif and the NPxxY motif, forming water-mediated indirect H-bond networks. In Figure 3, a large, complex H-bond network near the S(N)LAxAD motif at the middle of the interhelical region is shown. It appears that hydrophobic residues inside the helical core would play an important role in stabilizing the helical bundle, especially by excluding water molecules coming from the intracellular side and interfering with the stability of the helical bundle. In Figure 4, a hydrophobic belt above the S(N)LAxAD motif is represented. K3.28(192) is located more than two helical turns down from the extracellular helical boundary of H3. It appears that K3.28(192), a critical residue for cannabinoid ligand binding would play a important role in stabilizing the protein by keeping H2, H3 and E2 in place through not only salt bridge formation but also hydrophobic interaction in the absence of the ligand (Figure 5). It is expected that such stabilization is interfered upon the binding of the K3.28(192) mutation-sensitive ligands, such as D9-THC and AEA. It is also shown that positively charged residues and negatively charged lipid head groups are particularly important for the protein-lipid interactions (Figure 6). It is noted that the Arg residues are predominant over any other residues in the protein interaction with the lipid head groups. Thus, it appears that the Arg residues are crucial in determining the structure of the CB1 receptor by interacting with the lipids and water at the interfacial region.
Click the Figures to enlarge

Figure 1. Simulation system of the CB1 receptor in a POPC bilayer. The CB1 receptor structure (in ribbon). The residues are colored by residue type: hydrophobic residues in white; hydrophilic residues in green; positively charged residues in blue; and negatively charged residues in red. The palmitoyl moiety (PAL) that is covalently bonded to C415 is represented in stick. Some of the identified water molecules inside the helical core region are also represented by balls (in red).

Figure 2. Lipid bilayer. Density profiles for the CB1 receptor in the POPC bilayer along the membrane normal. Density of the lipid tails and water are shown to a scale of 0.1. Profile color scheme: the sn-1 tail (in black, dotted), the sn-2 (in black, straight), the head group oxygen atoms (in red, dotted), the head group phosphate P atoms (in magenta), the head group N atoms (in blue), ions (in yellow), water (in red, straight), and the protein (in green).

Figure 3. H-bonding networknear the middle of the helical core of H1/H2/H7. N1.50(134) forms the water-mediated H-bond to D2.50(163), both known as the highly conserved residues among GPCRs. N1.50(134), D2.50(163), and N7.45(389), N7.49(393). N1.50(134) forms the water-mediated H-bond to D2.50(163) which possibly forms H-bond to N7.49(393) which, in turn, possibly forms H-bond to N7.45(389).

Figure 4. Hydrophobic belt composed of V1.53(137), I2.43(156), L2.46(159), A2.47(160), L3.43(207), I5.54(290), I6.40(348), L6.41(351), P7.50(394), I7.51(395), and I7.52(396) are represented in the space filling mode. All water molecules are represented in the space filling mode. Color coding for the TM helices (in ribbons): red for H1; orange for H2; yellow for H3; green for H4; cyan for H5; blue for H6; and purple for H7.

Figure 5. K3.28(192) of the inactive state of the CB1 receptor. K3.28(192) interacts not only with the negatively charged residues D83 of E1 and D266 of E2 but also with the surrounding hydrophobic residues, including I2.56(169), V3.24(188), F3.25(189), L3.26(190), F3.27(191), L3.29(193), V3.32(196), and I267 of E2, which are colored in white. The structure at 50 ns of the simulation was used.

Figure 6. Protein-lipid polar interactions between the protein R6.32(340)/K6.35(343)/R405/R409 residues and the lipid head groups. Lipids (in line) and water molecules (in line) within 3.5 Å of the interested Arg/Lys residues (in stick) are represented. The structure at 50 ns of the simulation was used.