The decrease in total,

but not synaptic, surface GluA1 after glycine treatment in the DKD cells suggests that the extrasynaptic AMPARs in the DKD cells may be relatively unstable and more susceptible to endocytosis. Consistent with this hypothesis, the constitutive endocytosis of GluA1-containing AMPARs increased after LRRTM DKD and returned to basal levels with expression of LRRTM2 click here (Figure S7). Our results suggest that LRRTMs are required for LTP at synapses on early postnatal CA1 pyramidal neurons in vivo and on cultured neurons in vitro. However, the effects of LRRTM DKD on basal AMPAR-mediated synaptic responses in vivo depend on the maturational state of the synapses (Soler-Llavina et al., 2011). Furthermore, NL1 KD was reported to impair LTP at early postnatal but not at mature synapses on CA1 pyramidal neurons (Shipman and Nicoll, 2012), although the NL1 JQ1 concentration KO does not cause a major impairment in LTP (Blundell et al., 2010), suggesting that NL1 is not required during development to render synapses competent for LTP. These findings raise the possibility that LRRTMs may not play a critical role in mediating LTP at mature synapses but instead that the in vivo LRRTM DKD

at P0 may prevent synapses from reaching a maturational state necessary to support LTP. To address this possibility, we injected the LRRTM DKD lentivirus into the CA1 region of P21 mice, a time point at which synapses have largely matured, and then performed recordings in slices prepared 14–18 days later (Figures 4A and 4B). P35–P39 control neurons expressed robust LTP (Figure 4C), whereas LTP was dramatically reduced in DKD neurons (Figures 4D and 4E; control = 2.1 ± 0.18, n = 13 cells; DKD = 1.26 ± 0.11, n = 12 cells). Furthermore, expression of LRRTM2 rescued LTP (Figures 4F and 4G; DKD-LRR2 = 2.0 ± 0.30, n = 7 cells) as did expression of LRR2Ex (Figures 4H and 4I; control = 2.14 ± 0.41, n = 5 cells; DKD-LRR2Ex = 2.08 ± 0.33, PD184352 (CI-1040) n = 6 cells). Despite decades of effort,

the molecular mechanisms underlying classic NMDAR-dependent LTP at excitatory synapses on hippocampal CA1 pyramidal neurons remain poorly understood. Indeed, recent work points out the need to re-examine current hypotheses about LTP mechanisms (Granger et al., 2013, Lee et al., 2013 and Volk et al., 2013) and the importance of testing the role of novel proteins. Here we investigated the role of LRRTMs (Laurén et al., 2003 and Linhoff et al., 2009) in standard LTP because, like NLs, LRRTMs form an adhesion complex with Nrxs (de Wit et al., 2009, Ko et al., 2009, Ko et al., 2011 and Siddiqui et al., 2010), their in vivo KD during early postnatal development affects AMPAR-mediated, but not NMDAR-mediated, synaptic responses (Soler-Llavina et al., 2011), and they may directly bind to AMPAR subunits (de Wit et al., 2009 and Schwenk et al.