Fig 2: EPTA stains all types of SCZs.(A) EPTA staining does not interfere with immunogold detection, as shown here for SyPhy immunoreactivity, also proving that SyPhy positive synapses (SyPhy+) are readily recognized in EPTA treated material; arrows point to SCZs from the postsynaptic side. Scale bar = 1 µm. (B) Unlike SyPhy immunoreactivity, EPTA does not fail to detect primary sensory endings (co.aff.) indicated by gold-detection of BDA transported into AVCN after intracochlear injection; p = presynapse, d = dendrite. Arrows point to SCZs from the postsynaptic side, arrowhead indicates presynaptic protrusions of an SCZ. Scale bar = 0.5 µm. (C) Glutamate immunogold staining (Glu+) demonstrated in conjunction with EPTA at POD7, indicating that not all glutamatergic synapses have disappeared from AVCN after cochlear nerve degeneration. Arrows point to SCZ from the dendritic side, arrowheads indicate presynaptic protrusions. Scale bar = 1 µm. (D) ChAT immunogold labeling (ChAT+) combined with EPTA staining at POD7; note that the ChAT-positive nerve ending carries an immature SCZ (asterisk), whereas a nearby ChAT-negative synaptic contact (white arrow) shows a fully differentiated EPTA staining including presynaptic protrusions (arrowheads). Scale bar = 1 µm. (E) Gephyrin immunogold labeling (black particles, Geph+) combined with EPTA staining (dark gray). Arrow points to SCZ from the dendritic (d) side, arrowheads indicate presynaptic (p) protrusions. Scale bar = 0.5 µm. (F) GAP-43 immunogold labeling (GAP-43+) combined with EPTA staining. Again, arrow points to SCZ from the dendritic (d) side, arrowhead indicates presynaptic (p) protrusions. Scale bar = 0.5 µm.

Fig 3: InSyn1 localization to the iPSD is DGC dependent.(A) Depletion of Gephyrin or aDG by CRISPR in neurons. Cas9 knock-in hippocampal neurons were transduced with AAV:Cre/(control)gRNA [control], AAV:Cre/(Gphn)gRNA [gephyrin] or AAV:Cre/(Dag1)gRNA [aDG] at DIV1 and stained with gephyrin or aDG at DIV13 (left panel). GFP fluorescence of the Cas9-2A-GFP (right panel). Graphs to the right show the normalized puncta density. Gphn vs control (two-tailed t-test, gephyrin n = 19, control n = 17, p<0.0001), Dag1 vs control (two-tailed t-test, aDG n = 16, control n = 20, p<0.0001). (B, C) InSyn1-HA localization after aDG or gephyrin CRISPR depletion. Neurons were depleted of aDG or gephyrin, followed by AAV:InSyn1-HA transduction 3 days before fixation. Exogenously expressed InSyn1-HA is shown in red. Endogenous gephyrin or aDG are shown in green. Bar graph showing the distribution index of InSyn1-HA as arbitrary units. Control (n = 36), Gphn (n = 36), Dag1 (n = 36). One-way ANOVA followed by Tukey's multiple comparisons test, F (2, 105)=25.49, ***p<0.001. Scale bars, 20 µm. (D). HITI labeling of endogenous InSyn1 with smFP-HA in Cas9 KI neurons. InSyn1 is shown in red and aDG is shown in green. Of note, InSyn1:smFP showed clear puncta staining colocalized with aDG. (E) InSyn1 puncta-positive cells were quantified in either Control, Gphn, or Dag1 depleted neurons (n = 3).

Fig 4: Hierarchical clustering of genes in the P21 hippocampus of quadruple knockout and wildtype mice. (a) The microarray analysis revealed that 438 genes are significantly altered in the quadruple knockout mouse. These are depicted in form of a volcano plot, which includes upregulated genes (red dots), downregulated genes (blue dots) and unaffected genes (grey). (b) The hierarchical clustering of genes in the P21 hippocampus uncovered a differential gene expression pattern in the quadruple knockout compared to the wildtype mouse using three dependent hippocampus samples of siblings (n = 3). The cluster heat map comprises significantly up- and downregulated genes of interest concerning the annotations “ECM” as well as “neurons/synapses” including Gpc3, Gabrq, Gad2, Wnt7, Syt9, Sod3, Dnm3, Cript, Sema4c, Calr, Cplx3, Col4a3, Col27a1, Adamts13, Grin2d, Col1a2, Spon2 and Rapsn. The colour shift of light blue to light red indicates the expression change of these genes, that is genes that appear in blue were down- whereas genes illustrated in red were upregulated in the hippocampus of the quadruple knockout mouse. (c) Linear fold change of exemplary ECM molecules (Spon2, Col1a2, Col4a3, Col27a1, Gpc3 and Calr) and (d) neuron related molecules (Adamts13, Cript, Sod3, Wnt7a) in the P21 quadruple knockout hippocampus. (e) Linear fold change of representative genes belonging to the field of neurons and their synapses (Rapsn, Cplx3, Grin2d, Dnm3, Sema4c, Syt9, Gad2, Cntnap4, Gabrq, SLC17A7, SLC32A1 and GPHN) in the quadruple knockout hippocampus.

Fig 5: Signaling Pathways Regulated by GluD1 at Inhibitory Synapses(A) Schematic of the experimental workflow used for mass spectrometry analysis. CoIP, co-immunoprecipitation; LC-MS/MS, liquid chromatography followed by tandem mass spectrometry.(B) Table displaying the 20 most abundant GluD1-interacting proteins. Exp#1–#3 represent independent coIP replicates with anti-GluD1 antibody; Ctrl#1–#3 corresponds to the control coIPs with total rabbit IgG. Only proteins enriched at least three times in GluD1 coIPs are indicated. GluD1 (encoded by the gene grid1) is highlighted in red.(C–G) Quantification of GPHN cluster density in oblique dendrites of layer 2/3 CPNs in utero electroporated in the indicated conditions (juvenile stage). MRCKa KinD, MRCKa kinase-dead mutant; MRCKa WT, wild-type MRCKa. nshControl (C) = 22, nshArhgef12 = 32, nRescue (C) = 24, nControl (D) = 41, n MRCKa WT = 20, n MRCKa KinD = 18, nshControl (E) = 17, nshPpp1r12a = 30, nRescue (E) = 21, nControl (F) = 15, nPpp1r12a-KO = 38, nArhgef12-KO = 19, nGrid1-KO (G) = 27, nGrid1-KO + Ppp1r12a-KO = 34, nGrid1-KO + Arhgef12-KO = 25.Statistics: bars indicate mean ± SEM, ns p > 0.05, ***p < 0.001 determined by one-way ANOVA test followed by Tukey’s post test in (D-G) and Kruskal-Wallis test followed by Dunn’s post test in (C).(H) Schematic illustrating GluD1 trans-synaptic signaling regulating inhibitory synapse development. See text for details.