A unique sorting nexin regulates trafficking of potassium channels via a PDZ domain interaction.

G protein-gated potassium (Kir3) channels are important for controlling neuronal excitability in the brain. Using a proteomics approach, we have identified a unique rodent intracellular protein, sorting nexin 27 (SNX27), which regulates the trafficking of Kir3 channels. Like most sorting nexins, SNX27 possesses a functional PX domain that selectively binds the membrane phospholipid phosphatidylinositol-3-phosphate (PI3P) and is important for trafficking to the early endosome. SNX27, however, is the only sorting nexin to contain a PDZ domain. This PDZ domain discriminates between channels with similar class I PDZ-binding motifs, associating with the C-terminal end of Kir3.3 and Kir3.2c (-ESKV), but not with that of Kir2.1 (-ESEI) or Kv1.4 (-ETDV). SNX27 promotes the endosomal movement of Kir3 channels, leading to reduced surface expression, increased degradation and smaller Kir3 potassium currents. The regulation of endosomal trafficking via sorting nexins reveals a previously unknown mechanism for controlling potassium channel surface expression.

Bi-directional effects of GABA(B) receptor agonists on the mesolimbic dopamine system.

The rewarding effect of drugs of abuse is mediated by activation of the mesolimbic dopamine system, which is inhibited by putative anti-craving compounds. Interestingly, different GABA(B) receptor agonists can exert similarly opposing effects on the reward pathway, but the cellular mechanisms involved are unknown. Here we found that the coupling efficacy (EC(50)) of G-protein-gated inwardly rectifying potassium (GIRK, Kir3) channels to GABA(B) receptor was much lower in dopamine neurons than in GABA neurons of the ventral tegmental area (VTA), depending on the differential expression of GIRK subunits. Consequently, in rodent VTA slices, a low concentration of the canonical agonist baclofen caused increased activity, whereas higher doses eventually inhibited dopamine neurons. At behaviorally relevant dosages, baclofen activated GIRK channels in both cell types, but the drug of abuse gamma-hydroxy-butyric acid (GHB) activated GIRK channels only in GABAergic neurons. Thus GABA(B) receptor agonists exert parallel cellular and behavioral effects due to the cell-specific expression of GIRK subunits.

Three-dimensional structure of the ligand-binding core of GluR2 in complex with the agonist (S)-ATPA: implications for receptor subunit selectivity.

Two X-ray structures of the GluR2 ligand-binding core in complex with (S)-2-amino-3-(5-tert-butyl-3-hydroxy-4-isoxazolyl)propionic acid ((S)-ATPA) have been determined with and without Zn(2+) ions. (S)-ATPA induces a domain closure of ca. 21 degrees compared to the apo form. The tert-butyl moiety of (S)-ATPA is buried in a partially hydrophobic pocket and forces the ligand into the glutamate-like binding mode. The structures provide new insight into the molecular basis of agonist selectivity between AMPA and kainate receptors.

Competitive antagonism of AMPA receptors by ligands of different classes: crystal structure of ATPO bound to the GluR2 ligand-binding core, in comparison with DNQX.

Ionotropic glutamate receptors (iGluRs) constitute a family of ligand-gated ion channels that are essential for mediating fast synaptic transmission in the central nervous system. This study presents a high-resolution X-ray structure of the competitive antagonist (S)-2-amino-3-[5-tert-butyl-3-(phosphonomethoxy)-4-isoxazolyl]propionic acid (ATPO) in complex with the ligand-binding core of the receptor. Comparison with the only previous structure of the ligand-binding core in complex with an antagonist, 6,7-dinitro-2,3-quinoxalinedione (DNQX) (Armstrong, N.; Gouaux, E. Neuron 2000, 28, 165-181), reveals that ATPO and DNQX stabilize an open form of the ligand-binding core by different sets of interactions. Computational techniques are used to quantify the differences between these two ligands and to map the binding site. The isoxazole moiety of ATPO acts primarily as a spacer, and other scaffolds could potentially be used. Whereas agonists induce substantial domain closures compared to the apo structure, ATPO only induces minor conformational changes. These results are consistent with the hypothesis that domain closure is related to receptor activation. To facilitate the design of novel AMPA receptor antagonists, we present a modified model of the binding site that includes key residues involved in ligand recognition.

Structural determinants of agonist-specific kinetics at the ionotropic glutamate receptor 2.

Glutamate receptors (GluRs) are the most abundant mediators of the fast excitatory neurotransmission in the human brain. Agonists will, after activation of the receptors, induce different degrees of desensitization. The efficacy of agonists strongly correlates with the agonist-induced closure of the ligand-binding domain. However, the differences in desensitization properties are less well understood. By using high-resolution x-ray structure of the GluR2 flop (GluR2o) ligand-binding core protein in complex with the partial glutamate receptor agonist (S)-2-amino-3-(3-hydroxy-5-tert-butyl-4-isothiazolyl)propionic acid [(S)-thio-ATPA], we show that (S)-thio-ATPA induces an 18 degrees closure of the binding core similar to another partial agonist, (S)-2-amino-3-(4-bromo-3-hydroxy-5-isoxazolyl)propionic acid [(S)-Br-HIBO]. Despite the similar closure of the ligand-binding domain, we find in electrophysiological studies that (S)-thio-ATPA induced a 6.4-fold larger steady-state current than (RS)-Br-HIBO, and rapid agonist applications show that (S)-thio-ATPA induces a 3.6-fold higher steady-state/peak ratio and a 2.2-fold slower desensitization time constant than (RS)-Br-HIBO. Structural comparisons reveal that (S)-Br-HIBO, but not (S)-thio-ATPA, induces a twist of the ligand-binding core compared with the apostructure, and the agonist-specific conformation of Leu-650 correlates with the different kinetic profiles pointing at a key role in defining the desensitization kinetics. We conclude that, especially for intermediate efficacious agonists, the desensitization properties are influenced by additional ligand-induced factors beyond domain closure.