Conformational Dynamics of Ligand Binding Domains of GluR2

离子型谷氨酸受体(IGluRs)是中枢神经系统中介导兴奋性信号的配体门控离子通道。当配基结合到iGluRs的胞外区时，局部构象发生变化，这种运动被翻译到跨膜区，诱导通道开放。我们使用一个分离的配体结合域(LBD)GluR2-S1S2J(GluR2)作为模型系统来研究蛋白质-配体复合体。利用时间分辨荧光和各向异性测量，我们表征了分离的LBD，GluR2中Trp残基的激发态性质和局部迁移率。 具体地说，我们确定了广泛使用和结构表征的拮抗剂6，7-二硝基-2，3-二酮(DNQX)是Trp的有效荧光能量转移(FET)受体。与结晶学数据一致，我们的结果表明这四个天然色氨酸位于结合配体的Forster半径(33埃)内。此外，我们通过确定GluR2的原始FET配体3-硝基酪氨酸(3NY)(24埃，表观解离常数，Kd约为170微摩尔)展示了这项技术的更广泛的价值。从色氨酸激发态衰变中估计的平均供体-受体(Trp-to-Ligand)距离对于两种配体来说(24埃)是相似的，这表明3NY结合在结构特征的配体结合裂隙中。 有趣的是，我们观察到3NY的多个速率分量，这表明配体-蛋白质复合体结构不均一。然而，由于存在多个Trp供体，我们只能估计到Trp残基的平均DNQX/3NY距离。虽然从我们对DNQX-Trp GluR2的分析中提取的距离(24埃)与结晶学确定的距离(12-20埃)大致相同，但它们略长一些。这可能是由于目前可分辨速率的极限(tau为0.5 ns)，这对应于供体-受体距离比Trp-DNQX估计的短，也可能是由于折叠蛋白质中Trp和DNQX过渡偶极的不利取向所致。或者，我们的较长距离可能只是反映了蛋白质在溶液中的动力学性质。 虽然DNQX是一个众所周知的配合物，但它的光谱性质和作为配体与GluR2结合的灵敏报告的潜在应用一直被忽视。我们已经描述了一种DNQX竞争分析，它比目前使用的荧光分析方法有了显著的改进，因为这是一个开启的传感器，消除了由于光漂白而导致的猝灭事件固有的假阳性观察。当使用DNQX或结构类似物(即CNQX和其他硝基芳香族化合物)时，FET竞争配体和本征Trp荧光的使用将允许未来不仅为GluR2识别新的配体，而且还为其他蛋白质的新配体进行识别。

Ionotropic glutamate receptors (iGluRs) are ligand-gated ion channels that mediate excitatory signaling in the central nervous system. Upon ligand binding to the extracellular domain of iGluRs local conformational changes ensue and this motion is translated to the transmembrane domain inducing channel opening. We have used an isolated ligand binding domain (LBD), GluR2-S1S2J (GluR2), as a model system to study the protein-ligand complex. Using time-resolved fluorescence and anisotropy measurements, we characterized the excited state properties and local mobility of Trp residues in the isolated LBD, GluR2. Specifically, we determined that the widely used and structurally characterized antagonist, 6,7-dinitroquinoxaline-2,3-dione (DNQX) acts as an efficient fluorescence energy transfer (FET) acceptor for Trp. Consistent with crystallographic data, our results indicate that the four native tryptophans are within Forsters radius (33 angstroms ) of the bound ligand. Additionally, we demonstrate the broader value of this technique by identifying an original FET ligand, 3-nitrotyrosine (3NY) for GluR2 (24 angstroms , apparent dissociation constant, Kd of approximately 170 micromolar). Estimated average donor-acceptor (Trp-to-ligand) distances extracted from tryptophan excited-state decays are similar for both ligands (24 angstroms) suggesting that 3NY binds in the structurally characterized ligand-binding cleft. Interestingly, we observe multiple rate components for 3NY suggestive of ligand-protein complex structural heterogeneity. However, due to the presence of multiple Trp donors we are only able to estimate average DNQX/3NY distances to the Trp residues. While the distance extracted (24 angstroms) from our analysis on DNQX-Trp GluR2 are on the order of the crystallographically determined distances (12-20 angstroms), they are somewhat longer. This may be due to the current limit of resolvable rates (tau of 0.5 ns) which would correspond to donor-acceptor distances shorter than those estimated for Trp-DNQX or it is possible that it is attributable to unfavorable orientations of the Trp and DNQX transition dipoles in the folded protein. Alternatively, our longer distances may simply reflect dynamical properties of the protein in solution. While DNQX is a well-known complex, its spectroscopic properties and potential application as a sensitive reporter for ligand binding to GluR2 have been overlooked previously. We have described a DNQX competition assay that represents a significant improvement over the fluorescence assays currently in use in that this is a turn-on sensor, eliminating false positive observations inherent to quenching events resulting from photobleaching. The use of FET competitor ligands and intrinsic Trp fluorescence should allow for the future identification of novel ligands for not only for GluR2 but also for other proteins when DNQX or structural analogs (i.e. CNQX and other nitrated aromatic compounds) are employed.