Control of Long-Term Synaptic Potentiation and Learning by Alternative Splicing of the NMDA Receptor Subunit GluN1

NMDA receptors (NMDARs) are critical for physiological synaptic plasticity, learning, and memory and for pathological plasticity and neuronal death. The GluN1 subunit is encoded by a single gene, GRIN1, with 8 splice variants, but whether the diversity generated by this splicing has physiological consequences remains enigmatic. Here, we generate mice lacking from the GluN1 exon 5-encoded N1 cassette (GluN1a mice) or compulsorily expressing this exon (GluN1b mice). Despite no differences in basal synaptic transmission, long-term potentiation in the hippocampus is significantly enhanced in GluN1a mice compared with that in GluN1b mice. Furthermore, GluN1a mice learn more quickly and have significantly better spatial memory performance than do GluN1b mice. In addition, in human iPSC-derived neurons in autism spectrum disorder NMDARs show characteristics of N1-lacking GluN1. Our findings indicate that alternative splicing of GluN1 is a mechanism for controlling physiological long-lasting synaptic potentiation, learning, and memory.