Supplementary Figure 1. LRRTM2 expression and diffusion during neuronal development. (a) Normalised expression of LRRTM2 in developing hippocampal neurons at different days in vitro (DIV), assessed by RT-qPCR using specific rat primers. Data are represented as mean ± SEM of at least 3 independent cultures per condition. (b) Number of trajectories obtained by uPAINT normalized to the cell area and (c) percentage of synaptic trajectories (> 50% track length inside synapses) for the different DIVs. (d-h) Frequency distribution of log(D) comparing synaptic and extrasynaptic trajectories at DIV7 (d), 9 (e) and 15 (f). Frequency distribution of log(D) of synaptic (g) and extrasynaptic (h) trajectories at different developmental stages. (i) Mean residency time of AP-LRRTM2 at synapses at different developmental stages. (DIV7, n = 20; DIV9, n = 15; DIV15, n = 12).

Supplementary Figure 2. LRRTM2 knockdown by shRNA disrupts excitatory synapse density. (a) Representative Western blot performed on protein extracts from COS-7 cells expressing AP-LRRTM2 or AP-LRRTM2r in the presence or absence of shLRRTM2. (b) Quantification of protein expression for LRRTM2 detected in the different conditions and normalized to β-actin levels, and expressed as a percentage of the control condition. Data are from 3 independent experiments. (c) Hippocampal neurons expressing soluble EGFP, or shLRRTM2 or shLRRTM2+AP-LRRTM2r+BirAER and immunostained for endogenous PSD-95 at DIV 14 (d) Quantifications of endogenous PSD-95 cluster density, average intensity, and size. Data are from 4 independent experiments, represented as mean ± SEM, and individual values (normalized for the average intensity).

Supplementary Figure 3. Expression levels of LRRTM2 mutants in COS cells. (a) Western blot performed on protein extracts from COS-7 cells expressing AP-LRRTM2 mutants. (b) Protein expression for LRRTM2 constructs detected in panel (a) normalized to actin and WTr expression levels. Data are from 2 independent experiments. (c) COS cells expressing AP-LRRTM2 mutants, fixed and labelled with anti-LRRTM2 antibody after permeabilization to detect total protein pool (d) Quantifications of the overall expression of the different mutants normalized to the mean of the WT for each experiment. Data are from two independent experiments. Grey dots in bars represent individual cell values. Data are represented as mean ± SEM.

Supplementary Figure 4. ECEV deletion but not YxxC mutation disrupt LRRTM2-PSD-95 interaction. COS-7 cells co-expressing mutants of AP-LRRTM2, BirAER and GFP-tagged PSD-95 and surface labeled with Alexa647-conjugated mSA. Notice that AP-LRRTM2-YACAr can still cluster PSD-95 like AP-LRRTM2-WTr, whereas AP-LRRTM2-∆CTDr and AP-LRRTM2-∆ECEVr cannot.

Supplementary Figure 5. Clustering of AP-LRRTM2 in the absence of shLRRTM2. Surface labelling of AP-LRRTM2 WTr (a) ∆Cr (b) ∆ECEVr (c) and YACAr (d) in the absence (GFP-) or presence (GFP+) of shLRRTM2. In the absence of shLRRTM2 all mutants appear more clustered. Images are scaled individually for each panel.

Supplementary Figure 6. Synaptic diffusion of AP-LRRTM2 mutants. (a) Semi-log plot of diffusion coefficient distributions for synaptic AP-LRRTM2-WTr, AP-LRRTM2-∆CTDr, AP-LRRTM2-∆ECEVr and AP-LRRTM2-YACAr (b) median diffusion coefficient showing an increase in diffusion in the ∆CTDr and YACAr mutant conditions (c) Percentage of synaptic immobile trajectories and (d) Mean squared displacement at t = 0.2sec. (e) Mean trajectory time spent at synapses for each LRRTM2 mutant (f) Semi-log distribution of diffusion coefficients for extra-synaptic AP-LRRTM2-WTr, AP-LRRTM2-∆CTDr, AP-LRRTM2-∆ECEVr and AP-LRRTM2-YACAr (g) Percentage of synaptic immobile trajectories for the different conditions.