Fu et al B free D2 1790 D2-SPD 1780 1A 463 1.755 2000 8000 Simulation time(ps) Simulation time(ps) FiGuRE 8 Time-dependent Rg of the four models. (A)DI and DI-SPD.(B) D2 and D2-SPD and Tomic et al. (2)reached the same result on these two aromatic stacking with F-6.52 made it difficult for helix VI residues via S-542A and S-546A mutations and related and its neighbors to move. Such a move of interaction is in functional studies. The importance of w-7.40 and W-7. 43 in accord with the findings of Woodward et al. and others(49- the Di receptor can be attributed to their involvement in 51): H-6.55C and H-6.55L mutations caused a remarkable aromatic packing with rings A and B of SPD(Fig. 6). This is decrease in antagonist binding of the D2 receptor. Combined hy the W-740A and W-743A mutations performed by with these experimental results, it can be inferred that loth et al. (48)resulted in a decrease in binding affinity of residues such as D-3.32, S-5.42, S-5.46, S-3.36, H-6.55 the DI receptor with its agonists. Our SPD-D2 complex W-7.40, and W-7.43 should be structural determinants for model and its dynamic properties have illustrated why the the pharmacological specificity of the dual actions of SPD D2 receptor is antagonized by SPD, that is, because elec- against the DI/D2 receptors trostatic interactions with the protonated H-6.55 and the Proposed signal transduction model of the D1 receptor IMI TM4 Based on our structural modeling and md simulations. a stepwise signal transduction model of DI was proposed(Fig 10). Since the model of DI was built based on the bovine rhodopsin, which has low identity with DRs(-25%0), and TM7 the MD simulation is short compared with the real biological TM7 process of DI motion, we should say that this model is TM5 5 TM5 somewhat uncertain. However. our model and simulations B were supported by most known mutagenesis experiments. It provides some useful information/clues for researchers who TM4 investigate the activation process of the GPCR family. TMI Further biophysical experiments are needed to test and TM2 ImProv Typically, the DI receptor adopts the inactive state R through its intramolecular interaction and transits between several energy minima conformations by molecular thermo- TM TMs dynamic motion. As the protonated SPD diffuses to the extracellular mouth of the binding cavity of the DI receptor, D a kind of electrostatic signal is transmitted to the molecular surface of DI and promotes the disruption of the K-167-E nlarly) identified by essential dynamics analysis. (A)DI-SPD.(B) D2-SPD. 302 salt bridge. The electrostatic attraction between the For comparison, the minimum of helices is shown in green, and the negatively charged region of the DI receptor and positively maximum is colored in purple for both complexes charged SPD assists in their initial association(Rt, state) Biophysical Journal 93(5)1431-1441and Tomic et al. (2) reached the same result on these two residues via S-5.42A and S-5.46A mutations and related functional studies. The importance of W-7.40 and W-7.43 in the D1 receptor can be attributed to their involvement in aromatic packing with rings A and B of SPD (Fig. 6). This is why the W-7.40A and W-7.43A mutations performed by Roth et al. (48) resulted in a decrease in binding affinity of the D1 receptor with its agonists. Our SPD-D2 complex model and its dynamic properties have illustrated why the D2 receptor is antagonized by SPD, that is, because elec￾trostatic interactions with the protonated H-6.55 and the aromatic stacking with F-6.52 made it difficult for helix VI and its neighbors to move. Such a move of interaction is in accord with the findings of Woodward et al. and others (49– 51): H-6.55C and H-6.55L mutations caused a remarkable decrease in antagonist binding of the D2 receptor. Combined with these experimental results, it can be inferred that residues such as D-3.32, S-5.42, S-5.46, S-3.36, H-6.55, W-7.40, and W-7.43 should be structural determinants for the pharmacological specificity of the dual actions of SPD against the D1/D2 receptors. Proposed signal transduction model of the D1 receptor Based on our structural modeling and MD simulations, a stepwise signal transduction model of D1 was proposed (Fig. 10). Since the model of D1 was built based on the bovine rhodopsin, which has low identity with DRs (;25%), and the MD simulation is short compared with the real biological process of D1 motion, we should say that this model is somewhat uncertain. However, our model and simulations were supported by most known mutagenesis experiments. It provides some useful information/clues for researchers who investigate the activation process of the GPCR family. Further biophysical experiments are needed to test and improve this model. Typically, the D1 receptor adopts the inactive state R through its intramolecular interaction and transits between several energy minima conformations by molecular thermo￾dynamic motion. As the protonated SPD diffuses to the extracellular mouth of the binding cavity of the D1 receptor, a kind of electrostatic signal is transmitted to the molecular surface of D1 and promotes the disruption of the K-167-E- 302 salt bridge. The electrostatic attraction between the negatively charged region of the D1 receptor and positively charged SPD assists in their initial association (Rt1 state). FIGURE 8 Time-dependent Rg of the four models. (A) D1 and D1-SPD. (B) D2 and D2-SPD. FIGURE 9 Largest anharmonic motions of the TM helices (viewed intracel￾lularly) identified by essential dynamics analysis. (A) D1-SPD. (B) D2-SPD. For comparison, the minimum motion of helices is shown in green, and the maximum is colored in purple for both complexes. 1438 Fu et al. Biophysical Journal 93(5) 1431–1441