The neurotransmitter serotonin (5-hydroxytryptamine or 5-HT) mediates rapid excitatory responses through ligand-gated channels (5-HT3 receptors). Recombinant expression of the only identified receptor subunit (5-HT3A) yields functional 5-HT3 receptors. However, the conductance of these homomeric receptors (sub-picosiemens) is too small to be resolved directly, and contrasts with a robust channel conductance displayed by neuronal 5-HT3 receptors (9-17 pS). Neuronal 5-HT3 receptors also display a permeability to calcium ions and a current-voltage relationship that differ from those of homomeric receptors. Here we describe a new class of 5-HT3-receptor subunit (5-HT3B). Transcripts of this subunit are co-expressed with the 5-HT3A subunit in the amygdala, caudate and hippocampus. Heteromeric assemblies of 5-HT3A and 5-HT3B subunits display a large single-channel conductance (16 pS), low permeability to calcium ions, and a current-voltage relationship which resembles that of characterized neuronal 5-HT3 channels. The heteromeric receptors also display distinctive pharmacological properties. Surprisingly, the M2 region of the 5-HT3B subunit lacks any of the structural features that are known to promote the conductance of related receptors. In addition to providing a new target for therapeutic agents, the 5-HT3B subunit will be a valuable resource for defining the molecular mechanisms of ion-channel function.

KCC2 is a neuron-specific K+-Cl− cotransporter that maintains a low intracellular Cl− concentration essential for hyperpolarizing inhibition mediated by GABAA receptors. Deficits in KCC2 activity occur in disease states associated with pathophysiological glutamate release. However, the mechanisms by which elevated glutamate levels alter KCC2 function are unknown. The phosphorylation of KCC2 residue S940 is known to regulate its surface activity. Here we demonstrated in dissociated rat neurons that NMDA receptor activity and Ca2+ influx caused the dephosphorylation of S940 leading to a loss of KCC2 function that lasted greater than 20 minutes. PP1 mediated the dephosphorylation events of S940 that coincided with a deficit in hyperpolarizing GABAergic inhibition due to the loss of KCC2 activity. Blocking dephosphorylation of S940 reduced the glutamate-induced downregulation of KCC2 and significantly improved the maintenance of hyperpolarizing GABAergic inhibition. Reducing the downregulation of KCC2 thus has therapeutic potential in the treatment of neurological disorders.

A common feature of general anaesthetic agents is their ability to potentiate neuronal inhibition through GABA(A) (gamma-aminobutyric acid) receptors. At concentrations relevant to clinical anaesthesia, these agents cause a dramatic stimulation of the chloride currents that are evoked by the binding of the natural ligand, GABA. Although there is widespread evidence that the sensitivity of GABA(A) receptors to anaesthetic agents is heterogeneous, the structural basis of these differences is largely unknown. Variations in subunit composition can have profound effects on the sensitivity of GABA(A) receptors to modulatory agents such as benzodiazepines. However, strict subunit specificity has not been demonstrated for the potentiating effects of anaesthetic agents. Here we describe a new class of human GABA(A) receptor subunit (epsilon) that can assemble with alpha- and beta-subunits and confer an insensitivity to the potentiating effects of intravenous anaesthetic agents. The epsilon-subunit also abolishes the normal outward rectification of recombinant receptors in which it assembles. The expression pattern of this subunit in the brain suggests a new target for manipulation of neuronal pathways within the basal ganglia.