Fig. 2: Spine loss and excitatory synapse aberrations in mPFC in absence of LRRTM1 and SynCAM 1.

a Diagram of dendritic spine classification. b Left, diagram marking analyzed PFC in red. Right, DiI-labeled secondary and tertiary dendrites in mPFC layer II/III of WT, LRRTM1 KO, SynCAM 1 KO, and DKO mice. Asterisks, color code as in a. Scale bar, 2 μm. c Fewer thin and mushroom spines in mPFC of LRRTM1 KO and SynCAM 1 KO mice compared to WT quantified from images as in b. DKO mice exhibited a larger, more than additive decrease in thin spines. (One-way ANOVA with post-hoc Tukey’s test; n = 29 dendritic segments from 3 WT, 29 from 3 LRRTM1 KO, 28 from 3 SynCAM 1 KO, 41 from 3 DKO). d Left, diagram marking analyzed mPFC. Right, representative immunostainings for excitatory pre- and post-synaptic markers vGlut1 and Homer1 in mPFC layer II/III (red and white, respectively, in algorithm-traced merge). Single optical 1.0 μm confocal sections were obtained. Scale bar, 5 μm. e Lower densities of excitatory postsynaptic Homer (left) and presynaptic vGlut1 (right) in single LRRTM 1 and LRRTM1/SynCAM 1 DKO mPFC compared to WT. Fewer Homer but no change in vGlut1 sites was observed in single SynCAM 1 KO mPFC. Images as in d were quantified. (Homer, WT vs. LRRTM1 KO, p < 0.0001; WT vs SynCAM 1 KO, p = 0.006; WT vs. DKO, p < 0.0001; LRRTM1 KO vs SynCAM 1 KO, p = 0.022; vGlut1, WT vs. LRRTM1 KO, p < 0.0001; WT vs. DKO, p < 0.0001) (One-way ANOVA with post-hoc Tukey’s test; n = 35 ROI from 4 WT, 64 from 4 LRRTM1 KO, 35 from 4 SynCAM 1 KO, and 64 from 4 DKO). f The density of excitatory synapses, defined as sites where vGlut1 and Homer puncta co-localize, is reduced in single and double KO mice. Images as in d were quantified. WT control and LRRTM1 KO data replicate the results in Fig. 1b. (One-way ANOVA with post-hoc Tukey’s test; WT vs. all, p < 0.0001) Number of analyzed ROI and mice as in e. g Single or combined loss of LRRTM1 and SynCAM 1 reduces the extent to which vGlut1 puncta co-localize with Homer (left) and the co-localization of Homer with vGlut1 (right). DKO mice exhibit the strongest reduction of Homer co-localization with vGlut1. Images as in d were quantified. (One-way ANOVA with post-hoc Tukey’s test; vGlut 1 with Homer, WT vs. all, p < 0.0001; Homer with vGlut1, SynCAM 1 KO vs DKO, p < 0.0001; WT vs. all others, p < 0.0001) Number of analyzed ROI and mice as in e. h Dendritic spine density in the hippocampal CA1 stratum radiatum. Left, diagram with hippocampus marked in red. Right, representative images of DiI-labeled dendritic shafts and spines in CA1 of WT, LRRTM1 KO, SynCAM1 KO, and DKO mice. Asterisks, spine color code as in a. Scale bar, 2 μm. i Spine density losses in CA1 stratum radiatum of DKO mice was indistinguishable from single KOs. Images as in h were quantified. (One-way ANOVA with post-hoc Tukey’s test; n = 26 dendritic segments from 3 WT mice, 26 from 3 LRRTM1 KO, 28 from 3 SynCAM 1 KO, 19 from 3 DKO). *p < 0.05, **p < 0.01, ***p < 0.001, n.s., not significant. Error bars in bar graphs show Standard Error of the Mean. Source data for panels c, e–g and i are provided as a Source Data file.