Thus it would appear that the extracellular

domains of these neuroligins largely LY294002 ic50 account for the subtype differences in phenotype, while the intracellular domains are exchangeable. To narrow in on the specific region within the extracellular domain that might account for the unique properties of NLGN1, we constructed six additional chimeras with increasingly more of the NLGN3 extracellular domain and less of NLGN1. We found that chimeras containing at least 326 amino acids from the extreme N terminus of NLGN1 possessed the typical NLGN1 NMDAR enhancement, whereas chimeras that contained less than 254 amino acids of the NLGN1 N terminus instead displayed NLGN3 type NMDAR enhancement (Figures 3A and 3E). The difference between NLGN1 and NLGN3 in the region between amino acids 326 and 254 includes an alternatively spliced insertion in NLGN1 previously termed the site B (Ichtchenko et al., 1995; Figure 3B). Interestingly, inclusion of this B site has been shown to determine the specificity with which NLGN1 binds to specific splice variants of neurexin (Boucard et al., VX-809 concentration 2005). We tested an additional mutant of NLGN1 with a deletion of eight amino acids in

the B site and found that it indeed possessed a NLGN3-type NMDAR enhancement phenotype (Figure S3). We have demonstrated that NLGN1, but not NLGN3, is required for LTP in the adult dentate gyrus, but not adult CA1, and that at least some aspects of the phenotypic difference between expression of NLGN1 and NLGN3 are due to the B site insertion in the extracellular domain of NLGN1. What remains is to determine why NLGN1 is required for LTP in dentate gyrus and not CA1 and whether until the B site

has ramifications for LTP as well as the baseline synaptogenic phenotype of NLGN1. It has been shown that the dentate gyrus, one of two sites in the brain that incorporates substantial adult born neurons throughout life, remains more plastic into adulthood, perhaps accounting for the susceptibility to loss of a synaptogenic molecule (reviewed in Deng et al., 2010). Indeed, previous reports indicate that halting adult neurogenesis reduces the expression of LTP in the dentate gyrus (Massa et al., 2011; Singer et al., 2011). Perhaps then CA1 neurons would be susceptible to a knockdown of NLGN1 at an earlier developmental time point when the initial connections are still forming. To test this hypothesis we switched to in utero electroporations. By introducing the NLGN1 miR construct in utero we can check the basal state of synaptic currents and LTP in cells lacking NLGN1 at a very young age (Figure 4A).