The pEGFP-N1-TrkB plasmid for expression of rat TrkB [a gift from Rosalind Segal (Addgene plasmid # 32500; RRID: Addgene_32500)]¹ was mutated by site-directed mutagenesis (QuikChange Kit, Agilent, Santa Clara, CA) to give yield to mutants deficient in Shc or PLCγ interaction (Minichiello et al., 2002) or in ATP-binding (McCarty and Feinstein, 1998). Respective primers (Supplementary Table S1) introduced point mutations in base triplets encoding Y515 or Y816 to generate TrkB Y515F (TrkB SHC-), TrkB Y816F (TrkB PLC-) or the kinase-dead mutant (TrkB K571A, TrkB KD). All mutations were subsequently verified by Sanger sequencing.

Wildtype (WT) TrkB or TrkB mutants were transfected into HEK293 and 3 T3 cells cultured in 6-well plates to 70% confluency alongside a control plasmid (pEGFP-N1, Takara Bio USA, Inc., San Jose, CA) using Lipofectamine 2000 (Thermo Fisher Scientific, Waltham, MA). After further incubation for 24 h, cells were subjected to FBS starvation for 24 h for the purpose of cell cycle synchronization and to minimize the activation of signaling pathways unrelated to Bdnf stimulation, e.g., through growth factors, hormones and nutrients contained in FBS. Prior to lysis, the cells were either stimulated with 100 ng/mL Bdnf in 0.1% BSA/PBS (PeproTec, Thermo Fisher Scientific, Waltham, MA) for 30 min or only with 0.1% BSA/PBS as a control. After a single wash with PBS, the cells were harvested and lysed in RIPA buffer (Thermo Fisher Scientific, Waltham, MA) supplemented with cOmplete™ mini protease inhibitors (Roche, Rotkreuz, Switzerland) and Pierce™ Phosphatase Inhibitors (Thermo Fisher Scientific, Waltham, MA) for 30 min on ice. Protein concentrations were determined using the Pierce™ BCA Protein Assay Kit (Thermo Fisher Scientific, Waltham, MA) according to the manufacturer’s protocol.

Cell lysates (30 μg of protein per lane) were separated by SDS-PAGE (NuPAGE 4–12% Bis-Tris gels, Thermo Fisher Scientific, Waltham, MA). After transfer, PVDF membranes were incubated with the following primary antibodies: rabbit anti-phospho-TrkA (Tyr674/675)/TrkB (Tyr706/707) (#4621), mouse anti-phospho-p44/42 MAPK (Thr202/Tyr204) (#9106), rabbit anti-p44/42 MAPK (#9102), rabbit anti-phospho-PLCγ1 (#2821) and rabbit anti-PLCγ1 (#2822) from Cell Signaling Technology, Danvers, MA and goat anti-TrkB (AF1494) from R&D Systems, Minneapolis, MN.

Detection was achieved by secondary antibodies which were either fluorescently labeled (for phospho-MAPK and MAPK detection) or HRP-labeled (for phospho-TrkB, TrkB, phospho-PLCγ1 and PLCγ1). Fluorescent signals were acquired using a LI-COR Imager (LI-COR Biosciences, Lincoln, NE), while ECL signals were detected with SignalFire™ Plus ECL Reagent (Cell Signal Technology, Danvers, MA) subsequently acquired with an ImageQuant™ LAS-400 mini (GE Healthcare, Freiburg, Germany). To quantify the relative protein bands, densiometric analysis was performed with Image Lab (Bio-Rad Laboratories, Feldkirchen, Germany).

miRNAs for the knockdown of TrkB (mi1973, mi162, or miTrkB), were transferred into pLenti04C/SEW (Kriebel et al., 2020) through Gateway^®-recombination (Life Technologies, Waltham, MA). The vector expresses EGFP under the control of a synapsin protein and miRNA under the control of a CaMKII promoter. Sequences targeted by miRNAs can be found in Supplementary Table S2.

Based on a previously described lentiviral vector [pLenti4CaMKII_s/V5-DEST (Kriebel et al., 2020)], TrkB expression vectors were constructed. TrkB WT, TrkB Shc-, TrkB PLC-, and TrkB KD fused to EGFP are driven by a functional mouse α-CaMKII promotor fragment. For vector details see Supplementary Figure S3. Lentiviral suspensions were produced in HEK239FT cells under serum-free conditions, purified as previously described (Kriebel et al., 2020) and titred with the Lenti-X qRT-PCR titration Kit (Takara Bio USA, Inc., San Jose, CA).

Adult pregnant female Sprague–Dawley rats were supplied by Janvier Labs, Le Genest-Saint-Isle, France. The housing of animals and submission to surgical procedures were performed in accordance with the European Union recommendations for the care and use of laboratory animals (2010/63/EU) and were approved by the regional authority (Regierungspräsidium Tübingen). Primary hippocampal neurons were isolated from E18 rat embryos and cultured as described earlier (Kriebel et al., 2020). Cells were seeded on polyethyleneimine (PEI)-coated 96-well μCLEAR^® plates (Greiner Bio-One, Frickenhausen, Germany) at a density of 4 × 10⁴ cells/well. On day 3 in vitro (DIV3), the cells were lentivirally transduced with 2 × 10⁶ viral copies (VCs)/well.

For chemical induction of inhibitory long-term potentiation (chem-iLTP), experimental procedures were adapted from (Petrini et al., 2014). At DIV14, the cells were washed and treated with 145 mM NaCl, 2 mM KCl, 2 mM CaCl, 2 mM MgCl, 10 mM Glucose, 10 mM HEPES (Pennacchietti et al., 2017) supplemented with 20 μM NMDA and 10 μM CNQX (MedChemExpress, Monmouth Junction, NJ) for 2 min, followed by a recovery period in the treatment solution for 18 min. To inhibit MEK1, CaMKII or TrkB, neurons were treated with either 50 μM PD98059, 3 μM KN-62, 1 μM Cyclotraxin B (all from MedChemExpress, Monmouth Junction, NJ) or DMSO (CTR) for 30 min in culture medium before induction of chem iLTP as described. All inhibitors and DMSO were present during the stimulation and recovery period.

Adult female Sprague–Dawley rats (250 g at the time of surgery) were used for stereotaxic injection of lentiviral suspensions. The animals were deeply anesthetized with 2–5% isoflurane/oxygen, followed by s.c. injection of metamizole (50 mg/kg) for intraoperative analgesia at least 30 min before the start of surgical procedures. Bilateral injections of 2.5 μL of lentiviral suspensions into the dorsal dentate gyrus (AP: −2.9 mm, ML: ±2.5 mm, DV: −4.3 mm; all coordinates relative to Bregma) were conducted using a Lab Standard™ Stereotaxic Instrument (Stoelting, Wood Dale, IL) connected to a 701 RN Hamilton syringe (10 μL, 30 gauge, pst 4; CS-Chromatographie Service, Langerwehe, Germany). Injection speed was set to 0.2 μL/min. To assure sufficient postoperative analgesia, 2 mg/kg meloxicam was administered by s.c. injection at the end of surgical procedures and during surgical follow-up care.

For fixation of brain tissue, animals were deeply anesthetized by ketamine (100 mg/kg i.p.) and xylazine (10 mg/kg i.p.) application and transcardially perfused with 100 mL of PBS followed by 250 mL of freshly prepared 4% paraformaldehyde/PBS 14 days after injection. Brain tissue was collected and additionally fixed in 4% paraformaldehyde/PBS at 4°C for 60 min.

Perfusion-fixated brains were washed in PBS and cut into 50 μm slices (Vibratome VT1000S, Leica Microsystems, Wetzlar Germany). After permeabilization (0.5% Triton X-100), slices were blocked by 1x BMB blocking reagent (Roche, Rotkreuz, Switzerland) for 2 h at room temperature. Further processing of brain slices and immunocytochemical staining was performed as described, previously (Beuter et al., 2016). Specimens were mounted on microscopic slides using Dako Fluorescent Mounting medium (Agilent, Santa Clara, CA).

Untreated or treated primary rat neurons were fixed at DIV14 in 4% PFA/PBS for 15 min at room temperature. After blocking and permeabilization for 30 min at room temperature in 0.1% Triton X-100/PBS containing 1 X BMB blocking reagent (Roche, Rotkreuz, Switzerland), cells were incubated at 4°C with primary antibodies diluted in the blocking solution overnight. Subsequently, secondary antibodies diluted in blocking solution were added for 2 h at room temperature. Nuclei were stained using Hoechst Dye 33,258 (Merck, Rahway, NJ). The following primary antibodies were used: polyclonal rabbit anti-TrkB (Ab6180, 1:50) from Abcam, Cambridge, UK, polyclonal chicken anti-MAP2 antibody (PA1-10005, 1:2500) and polyclonal rabbit anti-GABA (A2052, 1:500) from ThermoFisher, Waltham, MA, mouse anti-VGSC antibody (S8809, 1:100) from Sigma-Aldrich, St. Louis, MO and monoclonal mouse anti-gephyrin (#147021, 1:500), polyclonal rabbit anti-VGat (#131003, 1:500), monoclonal rabbit anti-Psd95 (#124011, 1:500) and monoclonal mouse anti-VGlut1 (#135511, 1:400), all from Synaptic Systems, Göttingen, Germany.

Confocal fluorescence images of tissue samples from in vivo experiments were acquired using a Zeiss LSM510 Meta confocal microscope equipped with a 63 × Plan-Apochromat oil immersion objective (NA 1.4; Carl Zeiss Microscopy, Oberkochen, Germany). Images of cultured primary neurons were retrieved with a spinning disc confocal microscope (Cell Observer SD, Carl Zeiss Microscopy, Oberkochen, Germany) equipped with a plan-apochromat 20x air objective and a plan-apochromat 63x oil immersion objective. Z-stacks were obtained from EGFP-positive neuronal somata.

For all recordings, settings including exposure time, laser intensity, and gain were kept constant across samples from different experimental groups. Each representative image is a maximum intensity projection of a z-stack.

To evaluate the specificity of CaMKII promoter-driven transgene expression, transduced primary hippocampal neurons were stained for GABA and MAP2 to identify inhibitory neurons. The proportion of GABA-positive neurons among EGFP-positive cells was quantified by calculating the percentage of GABA/EGFP double-positive neurons relative to the total EGFP-positive population.

To assess protein expression levels of TrkB, and phosphorylated CaMKII (pCaMKII), image stacks were loaded into IMARIS 10 (Bitplane, Oxford Instruments, Abingdon, UK). Following background subtraction, three-dimensional ROIs of EGFP-positive neuronal cell somata were generated using IMARIS’ surface creation functionality. These ROIs were then used to determine the mean gray values of TrkB and pCaMKII immunoreactivity. To identify Cleaved Caspase 3 (Cl.Csp3)-positive neurons, cells were analyzed based on the immunofluorescence intensity of the somatic Cl.Csp3 signal. A threshold was set using the mean fluorescence intensity to define Cl.Csp3-positive neurons, and this threshold was consistently applied across all groups. The ratio of apoptotic cells was calculated by dividing the number of Cl.Csp3/EGFP double-positive cells by the total number of EGFP-positive neurons within each microscopic section.

Quantification of synaptic marker protein clusters in vivo was performed as previously described (Beuter et al., 2016). Proximal dendritic segments (20 μm length) were defined as EGFP/MAP2-positive dendritic segments within 20 μm distance to the corresponding soma, while distal dendritic segments (20 μm length) were defined as those located 80 μm away (Supplementary Figure S1I). The axon initial segment of EGFP-positive neurons was defined by staining for voltage gated sodium channels (VGscs) (Supplementary Figures S1J). Pre-and postsynaptic markers were analyzed individually as described previously (Kriebel et al., 2011) To quantify synaptic marker protein clusters in vitro, image stacks were loaded into IMARIS 10 (Bitplane, Oxford Instruments, Abingdon, UK). Three-dimensional ROIs of the EGFP-positive neuronal cell somata were generated using IMARIS’ surface creation functionality. These ROIs were used to mask EGFP fluorescent signals and subsequently generate new image channels for the respective synaptic marker channels. Synaptic marker structures were quantified using IMARIS’ spot detection functionality, with a threshold setting maintained constant across samples from different experimental groups. The number of structures was normalized to the volume of the corresponding EGFP-based ROI to determine marker density. To determine mean cluster size of individual markers, the mean cluster volume for each neuron was analyzed after three-dimensional rendering of the respective image channels by using IMARIS’ surface creation tool. To identify colocalized pre- and postsynaptic markers in vitro, the IMARIS XTention MATLAB-based plugin “Spots Colocalize” was used. A maximum distance threshold of 0.7 μm between local intensity maxima was set.

To specifically measure calcium transients in EGFP-positive neurons, DIV14 cells were stained by replacing 50% of the neuronal medium with BrainPhys + SM1 (STEMCELL Technologies, Vancouver, Canada) containing the red-fluorescent calcium indicator 2 μM CalBryte™^* 590 AM (AAT Bioquest, Pleasanton, CA) Cells were incubated for 30 min at 37°C, 5% CO₂. After exchange of culture medium, culture plates were mounted on a Cell Observer SD (Carl Zeiss Microscopy, Oberkochen, Germany). Calcium transients were recorded in imaging optimized BrainPhys + SM1 (STEMCELL Technologies, Vancouver, Canada). Recordings of 3 min of multiple cells within a microscopic field were retrieved, with two microscopic fields per well. To identify EGFP-positive neurons and enable cell specific analysis, snapshots of the CalBryte and EGFP signals were taken for each microscopic field. Somatic calcium transients were measured from individual neurons by manually drawing ROIs with Fiji software (Schindelin et al., 2012) and extracting the mean fluorescence intensity over time. A minimum of three spontaneously active EGFP-positive neurons (at least two calcium transients within a measurement) were analyzed per recording. Extracted traces were further analyzed with the Origin 2015G Peak Analyzer module (OriginLab Corporation, Northampton, MA). Traces were normalized to baseline, and peak detection threshold and minimum peak distance were kept constant across all replicates. Peak parameters [area under the curve (AUC), full width half maximum (FHWM), amplitude] were averaged across all peaks of a single neuron. Data points in the graphs are normalized to the WT mean and represent the mean of multiple cells per well.

Statistical analysis was performed with GraphPad Prism 10 (GraphPad Software, Boston, MA). All data were submitted to outlier detection and tested for Gaussian distribution. The statistical tests used for assessment of significance are stated in the respective figure legends. All values are represented as mean ± S.E.M and values are stated in the respective figure legends. The p-values were assigned as follows: *p < 0.05; **p < 0.01; ***p < 0.001, ****p < 0.0001. Sample size is described in the respective figure legends. An independent experiment describes an individual dissection and culture of neurons. Detailed statistical values for all comparisons can be found in Supplementary Table S3.

In a previous report, we have demonstrated a role for Bdnf–TrkB signaling for gephyrin clustering on proximal dendritic segments of primary hippocampal neurons in vitro (Wuchter et al., 2012). Extending this work, we prepared recombinant lentiviral vectors for the expression of miRNA targeting rat Ntrk2 (TrkB) mRNA in vivo. TrkB-specific miRNAs mi162 and mi1973 were expressed under the control of a CaMKII promoter which is active in principal neurons, exclusively. Additionally, an EGFP reporter, driven by a synapsin promoter, was included to localize injections sites in vivo. Lentiviral vector maps and efficient knockdown of TrkB after lentiviral transduction at the mRNA and protein level has been shown recently (Beuter, 2017).

Lentiviral vectors for the knockdown of TrkB (mi162 and mi1973) as well as an ineffective control (miCTR) were injected into the dorsal dentate gyrus of anaesthetized juvenile rats. Application of two independent TrkB-specific miRNAs ruled out potential off-target effects (Colon et al., 2024). After 14 days, animals were sacrificed, brains fixed and stained for synaptic marker proteins. EGFP fluorescence analysis by confocal microscopy confirmed successful transduction of granule cells of the dentate gyrus (for representative low-magnification images see Supplementary Figure S1A,B). Likewise, GABAergic synapse markers VGat (for presynapses), gephyrin (Gphn, for postsynapses) as well as a glutamatergic postsynaptic marker (Psd95) were detected on EGFP-positive somata and neurites (Figures 1A-F; see Supplementary Figure S1C–H for maximum intensity projections of confocal z-stacks). Quantification of gephyrin cluster densities revealed that TrkB knockdown by both mi162 or mi1973 significantly reduced gephyrin cluster densities on proximal dendritic segments and somata of granule cells by at least 31 and 37%, respectively, as compared to miCTR (Figure 1G). In contrast, gephyrin cluster densities on distal segments and on the axon initial segment remained unaffected. Likewise, presynaptic GABAergic marker VGat densities on proximal segments were reduced while postsynaptic glutamatergic marker Psd95 densities remained unimpaired. These findings indicate that TrkB receptor expression is required for gephyrin clustering specifically at proximal dendritic and somatic compartments of granule cells in vivo.

For a detailed analysis of signaling pathways contributing to the TrkB-dependent regulation of gephyrin clustering, we generated a set of point mutations in the rat Ntrk2 coding sequence for overexpression of signaling deficient TrkB mutants in dissociated primary hippocampal neurons. To this end, we mutated the rat Ntrk2 sequence at amino acid residue 515 (Y515F) to remove the Shc binding site (further referred as TrkB SHC-), at amino acid residue 816 (Y816F) to interfere with PLCγ interaction (further referred as TrkB PLC-), and at amino acid residue 571 (K571A) to create a kinase-dead (further referred as TrkB KD) mutant. Successful mutation was confirmed by sequencing (Supplementary Figure S2A). Wild type TrkB (WT) as well as the TrkB mutants fused to an EGFP reporter downstream of a CMV promoter were expressed in NIH3T3 or HEK293 cells (Supplementary Figure S2B–D) (Minichiello et al., 2002; McCarty and Feinstein, 1998; Minichiello et al., 1998). Western Blot analysis with an antibody specific for TrkB indicated that all constructs expressed TrkB-EGFP fusion proteins at 120 kDa (Supplementary Figure S2B). A specific polyclonal TrkB antibody specific for phosphorylation of the catalytic domain showed no autophosphorylation of the TrkB KD mutant while TrkB WT, TrkB SHC- and TrkB PLC- were active in the presence or absence of Bdnf stimulation. TrkB is suggested to induce Erk1/2 activation as a downstream effector via interaction with Shc (Minichiello et al., 2002; Atwal et al., 2000). Analysis of Erk1/2 phosphorylation after overexpression of TrkB WT and TrkB mutants indicated a trend for reduced activation of Erk1 after expression of the TrkB SHC- by 54% and the TrkB KD mutant by 24% compared to TrkB WT and TrkB PLC- mutant (Supplementary Figure S2C). Moreover, PLCγ1 activation after expression of TrkB PLC- or TrkB KD mutants was reduced 73 and > 99%, respectively. While TrkB SHC- remained ineffective (Supplementary Figure S2D). In conclusion, the mutants showed the selective impairment of either TrkB activation (TrkB KD), reduced Shc-dependent signaling (TrkB SHC-) or PLCγ activation (TrkB PLC-).

Subsequently, TrkB WT and mutants were transferred into lentiviral backbones for expression under the control of a CaMKII promoter (Supplementary Figure S3A). All mutants were expressed in primary hippocampal neurons as detected by EGFP fluorescence (Supplementary Figures S3B–E). To evaluate cell specificity of the CaMKII promoter, we quantified the percentage of GABA-positive neurons among EGFP-transduced neurons. A total 10.96 ± 2.04% of all EGFP-positive cells expressed GABA indicating no specificity for excitatory neurons in vitro (Supplementary Figures S3F–G). Quantitative immunofluorescence analysis with an antibody specific for TrkB revealed an equal increase in TrkB expression in the presence of all constructs in comparison to neurons transduced with an EGFP expressing control vector (Supplementary Figures S4A–F). To rule out that overexpression of TrkB SHC- reduces neuronal survival due to impaired Akt signaling (Atwal et al., 2000), the percentage of caspase 3-positive cells in the population of EGFP-transduced neurons was determined. Quantification revealed no impact on the viability of hippocampal neurons by any TrkB mutant (Supplementary Figures S4G–H).

Based on our observation that TrkB is required for gephyrin clustering on cell bodies in vivo, we analyzed gephyrin clustering on somata after transduction of TrkB WT as well as TrkB mutants in hippocampal neurons in vitro (Figure 2). EGFP-positive transduced neurons expressing TrkB WT (Figure 2A), TrkB SHC- (Figure 2B), TrkB PLC- (Figure 2C) or TrkB KD (Figure 2D) were stained for gephyrin (Gphn) and VGat. In particular, we determined gephyrin cluster densities across the whole somatic compartment (total, Figure 2E) and of the subpopulation of synaptic gephyrin clusters defined by the apposition to presynaptic VGat punctae (syn, Figure 2F). Additionally, gephyrin cluster size was measured on whole neuronal somata (Figure 2G). Overview images are given in Supplementary Figures S5A–D. Analysis of total gephyrin cluster density revealed an increase after expression of the TrkB SHC- mutant, while TrkB PLC- or TrkB KD overexpression remained ineffective (Figure 2E). Focusing on synaptic gephyrin cluster density, TrkB PLC- and TrkB KD reduced synaptic gephyrin density while the TrkB SHC- mutant was ineffective (Figure 2F). Hence, TrkB activity and TrkB-dependent activation of PLCγ is required for synaptic gephyrin clustering. Examination of gephyrin cluster size revealed an increase after expression of TrkB SHC- and TrkB KD (Figure 2G). We conclude that TrkB-dependent Shc signaling is involved in the negative control of gephyrin cluster size. Gephyrin is required for the localization of GABA_A receptors to postsynaptic sites (Kneussel et al., 1999). Expression of glutamatergic marker Psd95, both total and synaptic remained unaffected after overexpression of TrkB WT and mutants, indicating selectivity for gephyrin clustering (Supplementary Figures S6A–D). In conclusion, the results imply a dual role of TrkB induced signaling to gephyrin regulation. Synaptic gephyrin clustering is positively regulated by TrkB-PLCγ signaling while gephyrin cluster size is negatively regulated by TrkB-Shc-dependent mechanisms.

Calcium signaling was previously shown to induce gephyrin clustering through calmodulin-dependent protein kinase II (CaMKII) (Petrini et al., 2014; Flores et al., 2015). Since Bdnf and TrkB-PLCγ signaling are known to activate CaMKII (Minichiello et al., 2002; Zeng et al., 2010), we asked whether the TrkB mutants alter the activation of CaMKII (Zeng et al., 2010). CaMKII autophosphorylation and activation was analyzed by quantitative immunofluorescence analysis with an antibody specific for phosphoThr286 (Figures 2H–K). Overexpression of TrkB PLC- and TrkB KD significantly reduced CaMKII phosphorylation indicating that TrkB activity is indeed required for the activation of CamKII (Figure 2L; for overview see Supplementary Figure S7A–D). Overexpression of the TrkB SHC- - did not impair CaMKII phosphorylation, significantly. Thus, loss of synaptic gephyrin clusters after TrkB PLC- and KD overexpression is accompanied by reduced CaMKII activity.

Impaired synaptic gephyrin and concomitant GABA_A receptor clustering may result in a change in neuronal activity of glutamatergic neurons. TrkB WT and TrkB mutants expressed under the control of the CaMKII promoter were transduced into hippocampal neurons. Principal neurons were identified through EGFP fluorescence (Figure 3A) and subsequently submitted to calcium imaging for measuring neuronal activity (Figure 3B). In accordance with reduced inhibitory synapse formation (Figure 2F), overexpression of TrkB PLC- significantly increased the frequency of calcium transients, while the amplitude was decreased (Figure 3C). However, no changes of calcium transients were observed after overexpression of the TrkB SHC- mutant. Eventually, opposing effects of SHC signaling reducing gephyrin cluster size (Figure 2G) and PLCγ signaling inducing synaptic gephyrin clustering (Figure 2F) mask each other after the combined inhibition of both pathways with the TrkB KD mutant.

Chemically induced inhibitory long-term potentiation (chem iLTP) utilizes a moderate calcium elevation protocol with the application of NMDA and CNQX. This method induces gephyrin clustering dependent on CaMKII activity, to simulate a homeostatic response (Petrini et al., 2014). To further elucidate a role for TrkB to chem iLTP-induced gephyrin clustering, we transduced expression vectors for TrkB WT and its mutants into primary hippocampal neurons and asked for an impact of TrkB on chem iLTP-induced gephyrin clustering. In the presence of TrkB WT, chem iLTP increased synaptic gephyrin clustering in our experimental system (Figures 4A,B). Overexpression of TrkB KD as well as of TrkB SHC- and TrkB PLC- reduced total and synaptic gephyrin clustering to base line levels, suggesting that TrkB activity as well as activation of both signaling pathways are a requirement for chem iLTP-induced gephyrin clustering (Figures 4C,D). The findings suggest Shc signaling to induce gephyrin clustering in chem iLTP while it reduces gephyrin clustering under unstimulated basal conditions (Figures 2E,G). In contrast, chem iLTP-induced increase in gephyrin cluster size was not affected by TrkB PLC- while this was completely abolished by TrkB SHC- and TrkB KD (Figure 4E). Therefore, chem iLTP-induced gephyrin cluster expansion is mainly triggered by TrkB-dependent Shc signaling. For further confirmation, we added selective antagonists for Mek1 (PD98059), CaMKII (KN-62), and TrkB (Cyclotraxin B) to TrkB WT overexpressing neurons to interfere with TrkB activity, Shc- and PLC-dependent signaling with an independent approach (Figure 4F). The results show that total gephyrin cluster size is reduced after application of each of the three compounds. Compared to TrkB PLC- overexpression (Figure 4E) a pronounced inhibition was observed by the CaMKII inhibitor KN-62. The results may be explained by additional mechanisms of chem iLTP-induced calcium increase, such as the calcium influx through NMDA receptors (Petrini et al., 2014).