GAT1 C-terminus controls cargo concentration and ER export

GABA transporter-1 (GAT1) terminates the action of GABA, the major inhibitory neurotransmitter in the brain. This makes GAT1 an interesting drug target, e.g. for epilepsy. In addition, GAT1 can be considered a model protein for a large class of membrane proteins that undergo quality control in the endoplasmic reticulum. We have recently uncovered an ER export motif in the proximal portion of the GAT1 carboxyl terminus. This motif consists of 3 hydrophobic amino acids (569VMI571). Mutation of this motif to serine residues (GAT1-SSS) led to intracellular retention of the transporter. Preliminary experiments showed that each of these hydrophobic residues is important for ER export, but that V569 seemed to play the most important role. We also found out that the proximal portion contains a binding motif for Sec24D. Nevertheless, this interaction with Sec24D is not mediated by the VMI-motif but rather by residues upstream (566RL567) of the VMI-motif. Mutation of this motif to alanine and serine (GAT1-RL/AS) led to loss of this interaction. GAT1-RL/AS was retained within the cells and exhibited a diffuse intracellular staining. Some staining was seen at the plasma membrane. In contrast GAT1-SSS exhibited a punctate staining and was actually never seen at the cell surface. From these data, a hypothetical model emerges where (i) GAT1-SSS is concentrated into ER exit sites (ERES) or another pre-Golgi compartment, but fails to be exported from these (ii) In contrast, GAT1-RL/AS fails to be concentrated into ERES. This latter assumption explains the lower levels of surface expression compared to the wild type transporter. Based on this model interpretation, the following questions need to be addressed: - Is Sec24D also important for ER export of GAT1 in vivo? - Is GAT1-RL/AS deficient in concentration into ERES? - What is the nature of the compartment where GAT1-SSS is concentrated? - Is oligomeric assembly of GAT1 per se required for the concentration process? - Is the rate of ER-to-Golgi transport of GAT1-RL/AS slower that that of the wild type transporter? - What is/are the protein(s) that interact(s) with the VMI-motif in the GAT1 carboxyl terminus? This information will be useful to generate a mechanistic model that accounts for the sequential steps of the export of GAT1 from the ER. It is conceivable that the model drawn for GAT1 will be applicable for other neurotransmitter transporters and possibly for other polytopic transmembrane proteins. The list of human diseases that are due to membrane proteins that fail to escape the ER is long and cystic fibrosis and the nephrogenic diabetes insipidus are just two among many others.