The regulated trafficking of GluR1 contributes significantly to synaptic plasticity, but studies addressing the function of the GluR1 C-terminal PDZ-ligand domain in this process have produced conflicting results. Here, we resolve this conflict by showing that apparently similar C-terminal mutations of the GluR1 PDZ-ligand domain result in opposite physiological phenotypes during activity-and CamKII-induced synaptic plasticity.The modification of synapses is a key feature underlying plasticity that occurs during development and learning. Mechanistic studies have shown that synaptic incorporation of additional AMPA-type receptors (AMPA-Rs) contributes to activity-induced enhancement at glutamatergic synapses during long-term potentiation (LTP) and plasticity in vivo (Malinow and Malenka 2002;Bredt and Nicoll 2003;Takahashi et al. 2003;Rumpel et al. 2005). However, a clear understanding of the molecular interactions that lead to stable synaptic incorporation of AMPA-Rs is missing.Here, we investigate the role played by the last seven amino acids in the GluR1 C terminus, which ends in a PDZ type I consensus sequence (S/T-X-) (Songyang et al. 1997). PDZ domains play critical roles in protein-protein interactions for many cellular processes (Garner et al. 2000;Kim and Sheng 2004). The GluR1 C-terminal PDZ-ligand domain is conserved among humans (TGL), rodents (TGL), birds (TGL), fish (SGM), and worms (TLF), suggesting that this region is important for GluR1 function (Chang and Rongo 2005). Early studies showed that a point mutation in this PDZ-ligand of GluR1 (GluR1T887A) blocked interactions with PDZ proteins (Cai et al. 2002) and prevented longterm potentiation (LTP) and CaMKII-induced potentiation in organotypic hippocampal slices . Subsequent studies with this GluR1 (T887A) mutant in dissociated cultured neurons (Passafaro et al. 2001), hippocampal slice cultures (Piccini and Malinow 2002), or Caenorhabditis elegans in vivo (Chang and Rongo 2005) also found receptor trafficking deficits. In contrast, a recent study using a mouse knock-in model showed that removal of the last seven amino acids of GluR1 (GluR1⌬7) does not affect receptor incorporation into synapses or LTP . Although both mutations, T887A and ⌬7, prevent interactions between the GluR1 cytoplasmic tail and PDZ proteins (Cai et al. 2002;Kim et al. 2005), they appear to have opposite effects on synaptic receptor incorporation and plasticity. It is unclear if these results are due to experimental approaches, or if a mechanistic difference exists between the point mutant (GluR1[T887A]) and the deletion mutant (GluR1⌬7).To resolve this discrepancy, we compared the physiological effects of GluR1(wildtype[wt]), GluR1(T887A), and GluR1⌬7 in the same experimental system, using viral overexpression (Sindbis virus, Invitrogen) of recombinant receptors in organotypic slice cultures. To visualize neurons that express receptors, we tagged GluR1 and its mutants with GFP. GFP-GluR1(wt) and GFPGluR1(T887A), with or without coexpressing tCaMKII (using an internal ribo...