, 2009) and CL1 (Boucard et al., 2012). While our lentiviral-expressed targeting sequences against each neuroligin were quite effective in a mixed hippocampal cell culture, it is possible that knockdown efficiency would differ in vivo, which we were unable to assess directly. Finally, stable adult CA1
selleck chemicals synapses may be less susceptible to the loss of neuroligin than the newly created or rapidly remodeling synapses found in young CA1 or the dentate gyrus. In the present study, we found that loss of neuroligin in adulthood led to a reduction in the number of synapses rather than a reduction in the number of AMPA or NMDA receptors per synapse. This is consistent with our previous finding, showing a loss of whole synapses upon knockdown of NLGNs1–3 in organotypic hippocampal slice culture (Shipman et al., 2011). However, other studies have Selleck GSK126 reported changes in the AMPA/NMDA ratio in the NLGN1 knockout which is at odds with these results (Chubykin et al., 2007; Soler-Llavina et al., 2011). This difference could
be the result of differences in methodology, particularly the difference between whole brain germline knockouts and sparsely expressed RNAi or the use of paired recording to individually measure changes in AMPAR- and NMDAR-mediated currents versus the use of AMPA/NMDA ratios. Others have reported impairment and of LTP following NLGN1 manipulations. Blundell et al. (2010) reported diminished LTP in a NLGN1 knockout mouse using field potential recordings in CA1, while another group found a loss of LTP in the amygdala following knockdown of NLGN1 (Jung et al., 2010; Kim et al., 2008). In each of these cases, however, unlike the present study, the manipulation caused apparent changes in NMDAR functioning and therefore the LTP effects were attributed to the loss of the NMDA-mediated inductive Ca2+ influx. It was quite unforeseen that the major difference in phenotype between overexpressed NLGN1 and NLGN3 would reside in the extracellular domain. This domain is known to mediate both cis and trans interactions. Specifically,
homo- and heterodimerization have been described as well as binding to the presynaptic neurexins ( Araç et al., 2007; Fabrichny et al., 2007). Based on our chimeric analysis and in vivo molecular replacement experiments, it is likely that the alternatively spliced insertion at site B in the extracellular domain of NLGN1 is responsible for its unique functions. Of the neuroligins, only the NLGN1 gene contains the possibility of an insertion at the B splice site, which affects the specificity of neurexin binding. Specifically, NLGN1 containing the B insertion binds preferentially to β-neurexins lacking an insertion at splice site 4 and does not bind the longer form α-neurexins ( Boucard et al., 2005).