TY - JOUR
AU - Luhmann, Heiko J
AB - Abstract Neuroligin-4 (Nlgn4) is a cell adhesion protein that regulates synapse organization and function. Mutations in human NLGN4 are among the causes of autism spectrum disorders. In mouse, Nlgn4 knockout (KO) perturbs GABAergic synaptic transmission and oscillatory activity in hippocampus, and causes social interaction deficits. The complex profile of cellular and circuit changes that are caused by Nlgn4-KO is still only partly understood. Using Nlgn4-KO mice, we found that Nlgn4-KO increases the power in the alpha frequency band of spontaneous network activity in the barrel cortex under urethane anesthesia in vivo. Nlgn4-KO did not affect single-whisker-induced local field potentials, but suppressed the late evoked multiunit activity in vivo. Although Nlgn4-KO did not affect evoked EPSCs in layer 4 (L4) spiny stellate cells in acute thalamocortical slices elicited by electrical stimulation of thalamocortical inputs, it caused a lower frequency of both miniature (m) IPSCs and mEPSCs, and a decrease in the number of readily releasable vesicles at GABAergic and glutamatergic connections, weakening both excitatory and inhibitory transmission. However, Nlgn4 deficit strongly suppresses glutamatergic activity, shifting the excitation–inhibition balance to inhibition. We conclude that Nlgn4-KO does not influence the incoming whisker-mediated sensory information to the barrel cortex, but modifies intracortical information processing. adhesion proteins, autistic spectrum disorders, short-termed plasticity, single-whisker stimulation Introduction Neuroligin-4 (Nlgn4) is a member of the neuroligin family of postsynaptic adhesion proteins (Bolliger et al. 2001; Jamain et al. 2003, 2008), which are expressed throughout the central nervous system (CNS) and play an important role in synaptic organization and function (Varoqueaux et al. 2006; Sudhof 2008). Human loss-of-function mutations of Neroligin-4X are among the most frequent monogenic causes of autism spectrum disorders (ASD) (Jamain et al. 2003; Laumonnier et al. 2004; Bemben et al. 2015). Genetic knockout (KO) of Nlgn4, a homologue of human Nlgn4X, in mice results in autism-like behavioral changes, including reduced social interaction and stereotypies, indicating Nlgn4-KO as a valid model of ASD (El-Kordi et al. 2013; Ju et al. 2014). Four neuroligin isoforms are expressed at specific subpopulations of synapses. Neuroligin-1 is localized at glutamatergic synapses (Song et al. 1999), Neuroligin-2 at GABAergic synapses (Varoqueaux et al. 2004), and Neuroligin-3 at both excitatory and inhibitory synapses (Budreck and Scheiffele 2007). The question whether Nlgn4 is localized in a specific subpopulation of synapses remains partly unresolved. In the retina, spinal cord, brainstem, thalamus, and globus pallidus, Nlgn4 is predominantly present at inhibitory synapses. In retina Nlgn4 is involved in the postsynaptic clustering of glycine receptors, consequently glycinergic transmission is weakened in the Nlgn4-KO retina (Hoon et al. 2011). Currently available antibodies demonstrated a weak staining in cerebral cortex, hippocampus and most other forebrain areas, but all antibodies failed to detect a large proportion of the forebrain Nlgn4 pool. However, loss of Nlgn4 causes hippocampal deficits at GABAergic synapses, suggesting a prominent function in maturation of inhibition (Hammer et al. 2015). Further, genetic ablation of Nlgn4 has been reported to attenuate both glutamatergic and GABAergic synaptic transmission in L2/3 of mouse primary somatosensory (S1) cortex (Delattre et al. 2013), indicating that Nlgn4 may contribute to information processing and function in the somatosensory barrel field (S1BF). However, the mechanisms underlying the observed weakening of synaptic transmission have remained elusive and the question as to whether Ngln4-KO affects neuronal activity in vivo has not yet been addressed. The present study aimed to investigate effects of genetic deletion of Nlgn4 on neuronal activity in the barrel cortex in vivo and at the cellular level in vitro. We report that in the barrel cortex in vivo Nlgn4 loss leads to a potentiation of spontaneous neuronal activity in the alpha frequency band, but does not affect sensory-evoked responses following single-whisker stimulation. In vitro experiments in acute thalamocortical brain slices revealed that Nlgn4-KO weakens, mostly presynaptically, both GABAergic and glutamatergic synapses, shifting, at the same time, the excitatory/inhibitory balance to inhibition. Further analyses of evoked multiunit activity (MUA) and current source density (CSD) analyses revealed a decrease in late evoked activity in Nlgn4-KO animals. Thus, Nlgn4 plays an essential role in intracortical information processing. Materials and Methods Nlgn4 knockout mice (Jamain et al. 2008) and wild type (WT) littermates were used on postnatal day (P) 15–30 or P30–60. All experiments were conducted in accordance with EU directive 86/609/EEC and were approved by the local ethical committee (Landesuntersuchungsanstalt RLP, Koblenz, Germany). All efforts were made to minimize the number of animals used and their suffering. Immunohistochemistry, Image Capture, and Analysis Immunohistochemistry was carried out on 8–12 weeks old WT and Nlgn4-KO mice. Animals were deeply anesthetized with isoflurane (DeltaSelect) and then rapidly decapitated. The whole brain was removed carefully from the skull. Brains were then quickly frozen in a −35°C isopentane bath (Carl-Roth). The brain was subsequently mounted on a specimen holder and embedded in Tissue-Tek using a cryostat (Leica), and coronal brains sections (16 μm) were cut. The sections were directly mounted on glass slides and dried for 30 min at room temperature. Slides were submerged in a −20°C methanol bath for 5 min for the fixation (Gasser et al. 2006). Following fixation and subsequent washing steps 3 × 5 min with PBS, the sections were blocked with 3% goat serum and 0.2% Triton X-100 (Roche Applied Science) for 1 h at room temperature. This blocking solution was also used as vehicle for the primary and secondary antibody incubations. The following dilutions were used for the primary antibodies (Gephyrin 1/1000 3b11 Mouse Synaptic System, GABAAR-γ2 subunit 1/3000 Guinea Pig kindly provided by Dr J.-M. Fritschy and PSD95 1/3000 K28/43 Mouse Neuromab) and for the secondary antibodies (Mouse, Guinea Pig Alexa Fluor 488 nm or 555 nm 1/1200 Molecular Probes). Incubation of the primary antibodies was carried out at 4°C overnight, followed by incubation with secondary antibodies for 2 h at room temperature, with 3 × 5 min washes using PBS between every step. Lastly, sections were mounted using a water based mounting medium (Polysciences Inc.). Single-plane confocal images were recorded using TCS-SP5 inverted confocal microscope (Leica Microsystems) and taken with a ×63 objective (NA 1.4). A digital zoom factor of 8 was used for the high magnification images that were necessary for the perisomatic synaptic quantification. For quantification experiment, images were processed using Image J (Gasser et al. 2006). Each image was first thresholded to generate a binary image, in which individual synapses were represented by single particles. The same threshold was applied to all images and genotypes within the same brain region. The binarized images were then subjected to the Watershed Segmentation algorithm to automatically separate overlapping particles. Finally, the command “Analyze Particles” was used to count the number of particles per image that were considered to be synapses. The perisomatic area was determined by selecting the cell surface then enlarging the selection with a predetermined value to include the perisomatic area and then finally excluding the nucleus. Voltage-Sensitive Dye Imaging Voltage-sensitive dye imaging (VSDI) was performed as described previously (Yang et al. 2013). VSD RH1691 (Optical Imaging, Rehovot, Israel) was dissolved at 1 mg/mL in a Ringer’s solution containing (in mM): 135 NaCl, 5.4 KCl, one MgCl2, 1.8 CaCl2 and 5 HEPES (pH was set to 7.2 with NaOH). A 3 × 3 mm2 area of craniotomy (1–4 mm posterior to bregma and 1–4 mm from the midline) was performed over the barrel cortex. The dura matter was removed. The VSD was applied to the surface of the barrel cortex and allowed to diffuse into the cortex for 1 h. Subsequently, unbound dye was carefully washed-out with the Ringer’s solution. The cortex was covered with a 1.5% low-melting agarose and a glass cover slip was placed on top to stabilize the tissue. A 25 mm focal length lens was used for the recording (Navitar video lens, Stemmer Imaging, Puchheim, Germany). A red LED was used for excitation (MRLED 625 nm, Thorlabs GmbH, Dachau, Germany), and MiCam Ultima L high-speed camera was used to record fluorescence signals (Scimedia, Costa Mesa, CA, USA). RH1691 was excited at 630 nm. The excitation and emission light was separated using a 650-nm dichroic mirror. Emitted light was filtered using a 660 nm long-pass filter. All measurements were performed using 100 × 100 pixels resolution. Acquisition rate was 500 Hz. Duration of acquisition was set to 2 s. Fluorescence signals were expressed as relative changes from pre-stimulus levels (ΔF/F0). Surgical Preparation, In Vivo Electrophysiological Recordings and Analysis Extracellular recordings were performed in the barrel cortex of P30–60 Nlgn4-KO and WT mice in vivo. Under initial intraperitoneal urethane anesthesia (1.6–2 g/kg, Sigma-Aldrich), the head of the mouse was fixed and the bone, but not the dura mater, above the barrel cortex was carefully removed. Animals were kept at a constant temperature of 37°C by placing them on a heating blanket. During recordings, additional urethane (10–20% of the initial dose) was given when the mice showed any sign of distress. A 2 × 2 mm2 area of craniotomy (1–3 mm posterior to bregma and 2–4 mm from the midline) was performed over the left barrel cortex. For this set of experiments the dura was kept intact. After craniotomy single whisker stimulation and VSDI were applied to functionally identify the cortical representation of the stimulated whisker, the barrel-related column (Fig. 1A). An 8-shank 128-channel electrode (200 μm horizontal shank distance and 75 μm vertical interelectrode distance, 1–2 MΩ, NeuroNexus Technologies, Ann Arbor, MI) was labeled with DiI (1,1′-dioctadecyl-3,3,3′,3′-tetramethyl indocarbocyanine, Molecular Probes, Eugene, OR, USA) and then inserted perpendicularly into the barrel cortex according to the VSDI-identified barrel positions (Fig. 1B) to record local field potentials (LFPs) and MUA. After histological identification of the barrels with cytochrome oxidase staining and evoked CSD pattern identification of cortical layers, we chose the channel with the maximal amplitude of evoked LFP in L4 as standard channel to compare the difference of single whisker-evoked responses between WT and Nlgn4-KO mice (Fig. 1C,D). Figure 1. View largeDownload slide Identification of barrel-related columns and neocortical layers by voltage-sensitive dye imaging (VSDI) and current-source density (CSD) analysis, respectively. (A) A representative VSDI response elicited by B3 whisker stimulation in a P47 WT mouse. The localization of the B3 whisker representation in the barrel cortex and 3 successive poststimulus VSD images are shown in the upper row. Red dot in the most left image indicates the center of the B3 whisker-evoked response. Lower trace shows a 120 ms long VSDI recording. The red dashed line indicates the stimulation onset. Red arrow heads mark time points at which the upper VSD images were obtained. (B) Schematic illustration of the eight-shank 128-channel electrode and photograph of cytochrome oxidase stained barrel map from a P46 Nlgn4-KO mouse. The eight red arrow heads point to the insertion sites of the eight shanks in the C row of the barrel cortex. (C) LFPs and corresponding CSD pattern elicited by C1 whisker stimulation. Data were obtained from the same mouse as in B. Note maximum of evoked response at shank 3 (S3). Cortical layers were functionally identified according to the CSD pattern at shank 3. The blue color represents current sinks, red color represents current sources. (D) LFPs and corresponding CSD pattern induced by C2 whisker stimulation. Data from the same mouse as in B. Note maximum of evoked response at shank 4 (S4). Cortical layers were functionally identified according to the CSD pattern at S4. Figure 1. View largeDownload slide Identification of barrel-related columns and neocortical layers by voltage-sensitive dye imaging (VSDI) and current-source density (CSD) analysis, respectively. (A) A representative VSDI response elicited by B3 whisker stimulation in a P47 WT mouse. The localization of the B3 whisker representation in the barrel cortex and 3 successive poststimulus VSD images are shown in the upper row. Red dot in the most left image indicates the center of the B3 whisker-evoked response. Lower trace shows a 120 ms long VSDI recording. The red dashed line indicates the stimulation onset. Red arrow heads mark time points at which the upper VSD images were obtained. (B) Schematic illustration of the eight-shank 128-channel electrode and photograph of cytochrome oxidase stained barrel map from a P46 Nlgn4-KO mouse. The eight red arrow heads point to the insertion sites of the eight shanks in the C row of the barrel cortex. (C) LFPs and corresponding CSD pattern elicited by C1 whisker stimulation. Data were obtained from the same mouse as in B. Note maximum of evoked response at shank 3 (S3). Cortical layers were functionally identified according to the CSD pattern at shank 3. The blue color represents current sinks, red color represents current sources. (D) LFPs and corresponding CSD pattern induced by C2 whisker stimulation. Data from the same mouse as in B. Note maximum of evoked response at shank 4 (S4). Cortical layers were functionally identified according to the CSD pattern at S4. Evoked CSD elicited by single-whisker stimulation was calculated from the averaged LFPs (50 or 100 trials). First, the uppermost and lowermost channels were duplicated in each shank. Then, the second derivative was computed as follows: D=1h2(ϕ(r+h)−2ϕ(r)+ϕ(r−h)), where ϕ(r) is the LFP at depth r, and h is the vertical interelectrode distance (75 μm). Finally, pseudocolor image plots were generated by linear interpolation along the depth axis (Fig. 1C,D). Data were analyzed offline using MATLAB software version 7.7 (MathWorks). In each experiment 30 min of continuous recording of spontaneous LFPs was used for further analysis applying the fast Fourier transform (FFT). Firstly, the 30 min LFP recording was separated into 360 segments, that is, each segment consisted of a 5 s LFP recording. Next, each 5 s raw LFP data were band-pass filtered 0.1–499 Hz and FFT analysis was applied (the frequency resolution was 0.2 Hz). Finally, mean FFT was calculated by averaging the FFT spectra in all 360 segments. MUA was detected in 800–5000 Hz filtered LFP signals by applying a threshold at 7.5 times the baseline standard deviation (SD). Histology At the end of each experiment, animals were deeply anesthetized with an intraperitoneal injection of ketamine (100 mg/kg) and xylazine (8 mg/kg) mixture, perfused with 0.1 M sodium phosphate buffer (PB), brains were carefully removed and fixed with 4% paraformaldehyde for 24 h. Subsequently the cortex was washed in phosphate buffer (PB) and incubated in 30% sucrose overnight. After washing with PB, the barrel cortex was tangentially sectioned in 200 μm slices, rinsed in PB, then incubated for 2–7 h at 39°C in a solution of 0.6 mg cytochrome C, 0.5 mg DAB, and 44 mg saccharose per mL, with 0.3% catalase (Sigma, Deissenhofen, Germany). Finally, the stainings were intensified with 0.5% copper(II)sulfate (Sigma) for 2–3 min, dried and coverslipped. Detection and Phase Analysis of Spindle Oscillations Spindle oscillations were detected based on the following procedures modified from (Andrillon et al. 2011). Briefly, raw LFP signals were band pass filtered (8–13 Hz) using a third order Butterworth filter. The instantaneous amplitudes and phases were computed via the Hilbert transformation. In order to reduce the influence of variability of mean and SD values from different recordings, we pooled the data from all recordings in 11 barrel-related columns in 5 Nlgn4-KO mice and 7 columns in 4 WT mice. In each experiment we recorded spontaneous activity for 30 min. After Hilbert transformation, a detection threshold was set as mean value of recordings ± 3SD. The peaks of Hilbert transformed signals exceeding this threshold were considered to be potential spindle oscillations. A start/end threshold was set at mean ± 1SD and the duration of each spindle oscillation was determined by measuring the difference between onset and end of the signal (see Supplementary Fig. S1A). Phases of spindle oscillations were computed as described elsewhere (Tsintsadze et al. 2015). First, LFP signals were band pass filtered at 8–13 Hz. Second, local phases were computed using Hilbert transformation of the band pass filtered LFP signals. Subsequently, each MUA event was assigned to the corresponding phase (see Supplementary Fig. S1B). The MUA-phase circular plots were made using CircStat toolbox (Berens 2009). Phase modulation of MUA by alpha-spindle oscillations was analysed using Rayleigh circular statistics. Watson–Williams multisample test was applied. Single Whisker Stimulation A single whisker was stimulated approximately 1 mm from the snout using a miniature solenoid actuator modified from (Krupa et al. 2001) that generated a 16 ms long deflection in the rostral-to-caudal direction (about 1 mm in amplitude). Paired-pulse stimuli were applied at 100, 250, and 500 ms interstimulus intervals (ISIs). At least 20 trials were performed at each ISI. The interval between trials was 10 s. Spike Sorting Spike detection and sorting was performed from the LFPs recorded from channels located in a principal barrel column from L2–4 similar as described previously (Reyes-Puerta, Kim, et al. 2015; Reyes-Puerta, Sun, et al. 2015). Continuously recorded data were high-pass filtered (0.8–5 kHz). Nonoverlapping groups of 3–4 adjacent channels were selected and defined as “virtual tetrodes”. Spike detection was performed in each recorded group of channels independently using amplitude thresholding (7.5SD). The detected spikes contained the sampled amplitude values from all channels in the group in the time range from −0.5 to +0.5 ms relative to the waveform negative peak. These spike waveforms were used to calculate feature vectors (negative peak amplitude and 2 first principal components derived from the waveforms). The (n × 3)-dimensional vectors (where n represents the number of channels within a group) were sorted using KlustaKwik software (Harris et al. 2000; Hazan et al. 2006). After automatic unit clustering, the units were manually inspected and isolated using custom-made Matlab code. Several criteria were used to ensure the isolation quality of the sorted units. First, the coefficient of variation (CV) of the spontaneous firing rate computed over the whole recording period using 100 s bins had to be <1.5. Second, the “isolation distance” had to be <10 (Schmitzer-Torbert et al. 2005). Third, <1% of the spikes should occur within 1 ms of each other. For each isolated unit, all spikes were aligned by their troughs and averaged. Cells were subsequently classified as putative inhibitory interneurons (INH) or putative excitatory neurons (EXC) using mean spike waveform duration defining the latter as trough-to-peak latency as previously reported (see Supplementary Fig. S1C and Bartho et al. 2004; Mitchell et al. 2007). Slice Preparation Nlgn4 knockout and WT P15–30 mice were deeply anesthetized with isoflurane. After decapitation, the brains were quickly removed and immersed in ice-cold artificial cerebrospinal fluid (ACSF, composition see below). Coronal slices (400 μm thickness) were cut on a vibratome (Campden Instruments Ltd., UK or HR2, Sigmann Elektronik, Hüffenhardt, Germany). For recordings of evoked excitatory postsynaptic currents (eEPSCs) thalamocortical paracoronal slices (400 μm) were prepared at an angle of 55° relative to the middle on a ramp at an angle of 13° as described before (Agmon and Connors 1991; Schubert et al. 2003; Staiger et al. 2004). Slices were stored in an incubation chamber filled with oxygenated ACSF at room temperature before they were transferred to the recording chamber. Data Acquisition and Analysis Whole-cell patch-clamp recordings were performed at about 31°C in a submerged-type recording chamber attached to the fixed stage of a microscope (BX51 WI, Olympus or Axioscope FS, Zeiss). Spiny stellate cells were identified by their location, morphological appearance by infrared differential interference contrast image and spike firing properties. Biocytin (0.5%, Sigma, Deisenhofen, Germany) was added in the patch pipette solution for later morphological cell reconstruction (see Supplementary Fig. S2). Patch-pipettes (3–5 MΩ) were pulled from borosilicate glass capillaries (Science Products, Hofheim, Germany) on a vertical puller (PP-830, Narishige). GABAergic and glutamatergic postsynaptic currents were recorded using either KCl- or K-gluconate-based pipette solution, respectively. The solutions contained (in mM) 130 KCl or K-gluconate, 1 CaCl2, 2 MgCl2, 10 EGTA, 10 HEPES, 2 Na2-ATP, 0.5 Na-GTP (pH adjusted to 7.4 with KOH and osmolarity was set to 306 mOsm with sucrose). Signals were low-pass filtered at 3 kHz, recorded with a discontinuous voltage-clamp/current-clamp amplifier (SEC05L, NPI, Tamm, Germany), and stored using an ITC-1600 AD/DA board (HEKA). Data were recorded and then analyzed offline using TIDA 5.24 software (HEKA Elektronik, Lambrecht, Germany). Miniature postsynaptic currents (mPSCs) were processed using PeakCount V3.2 software (C. Henneberger, Institute of Neurophysiology, Charité, Berlin). The program applies a derivative threshold detection of individual mPSC. Each automatically detected event was visually inspected. Rise times (10–90%) and decay time constants (a single exponential fit) of individual mEPSCs or mIPSCs were calculated. Electrical Stimulation To prevent activation of voltage-gated sodium channels N-(2,6-dimethylphenylcarbamoylmethyl)-triethylammonium bromide (QX-314, 2 mM) was added to the intracellular solution. An isolated stimulation unit (A365, WPI) was used to generate rectangular electrical pulses. Pulse duration was set to 200 μs. Evoked inhibitory postsynaptic currents (eIPSCs) were elicited by focal electrical stimulation through a glass pipette filled with ACSF (about 4 MΩ) and recorded in the presence of the NMDA receptor blocker APV (40 μM), the AMPA receptor blocker CNQX (10 μM), and the GABAB-receptor blocker CGP55845 (1 μM). The stimulation pipette was positioned in the vicinity of the recorded cell (~100 μm) within L4 of the barrel cortex. Pulse intensity was adjusted to evoke a stable response with the maximal amplitude +20% of intensity (supramaximal stimulation). Typical pulse intensity was between 50 and 100 μA. Paired-pulse stimulation was performed at ISIs of 50, 100, and 200 ms. Pairs of pulses were applied at 0.1 Hz. Paired-pulse ratio (PPR) was calculated as the ratio of the averaged amplitude of the second response to the mean amplitude of the first one. Evoked excitatory PSCs (eEPSC) were elicited at a frequency of 0.1 Hz using a bipolar tungsten electrode placed in the ventral posteromedial nucleus (VPM) of the thalamus as described (Unichenko et al. 2016). Pulse intensity was adjusted to activate a unitary synaptic input (minimal stimulation). Stimulation was accepted as minimal if eEPSCs occurred at a short and constant latency (<3.5 ms), demonstrated paired-pulse depression and showed no decrease in latency of the second eEPSC induced 50 ms after the first (Beierlein et al. 2003). Typical pulse intensity was between 10 and 30 μA. Readily Releasable Pool Estimation Estimation of readily releasable pool of GABAergic and glutamatergic synapses was performed by focal application of ACSF supplemented with 500 mM sucrose (hyperosmotic solution) via a glass pipette (1 μm tip diameter) placed at about 20 μm distance from the soma of L4 spiny stellate neuron (Rosenmund and Stevens 1996). Duration of application was set to 2.5–4 s. A LHDA0533115H pressure application system (Lee, Westbrook, CT, USA) was used. The amount of released vesicles was estimated as the total charge transfer normalized by the area of averaged miniature event. Solutions and Drugs Standard ACSF contained (in mM) 126 NaCl, 26 NaHCO3, 1.25 NaH2PO5, 1 MgCl2, 2 CaCl2, 2.5 KCl, 10 glucose (pH 7.4, osmolarity 306 mOsm). All solutions were equilibrated with 95% O2/5% CO2 at least 1 h before use. Miniature EPSCs (mEPSCs) were recorded in the presence of the voltage-gated Na+-channel blocker tetrodotoxin (TTX, 1 μM), the GABAA-receptor (GABAAR) blocker gabazine (10 μM), and the GABAB-receptor (GABABR) blocker CGP55845 (1 μM). Miniature inhibitory postsynaptic currents (mIPSCs) were recorded in the presence of TTX (1 μM), APV (40 μM), CNQX (10 μM) and CGP55845 (1 μM). 6-Imino-3-(4-methoxyphenyl)-1(6H)-pyridazinebutanoic acid hydrobromide (gabazine), (2S)-3-[[(1S)-1-(3,4-dichlorophenyl)ethyl]amino-2-hydroxypropyl](phenylmethyl)phosphinic acid (CGP55845), DL-2-Amino-5-phosphonopentanoic acid (APV), 6-Cyano-7-nitroquinoxaline-2,3-dione (CNQX) and tetrodotoxin (TTX) were obtained from Biotrend (Cologne, Germany). All other chemicals were obtained from Sigma-Aldrich (Munich, Germany). All substances were added to the solutions shortly before the experiment. CNQX and CGP55845 were dissolved in dimethylsulfoxide (DMSO). The DMSO concentration of the final solution never exceeded 0.2%. Data Evaluation and Statistics GraphPad Prism (GraphPad Software, Inc.) was used for statistical analysis. Whiskers on box plot diagrams represent 10–90 percentile, dots represent outliers. Error levels on bar diagrams represent the standard error of mean (SEM). Differences between groups were tested using unpaired Student’s t-test for values demonstrated normal distribution and Mann–Whitney-test for values demonstrated non-normal distribution. Distributions of experimental values were tested for normality using Kolmogorov–Smirnov test. *P < 0.05, **P < 0.01, ***P < 0.001; ns, not significant. Results Nlgn4-KO Influences Spontaneous Neuronal Activity in the Barrel Cortex In Vivo Recent studies performed in cortical and hippocampal slices demonstrated deficits in synaptic and neuronal activity in Nlgn4-KO mice (Hoon et al. 2011; Delattre et al. 2013; Hammer et al. 2015). We tested whether genetic Nlgn4 ablation alters neocortical network activity in vivo. Spontaneous LFPs in the barrel cortex of anesthetized Nlgn4-KO and WT mice were recorded using an 8-shank 128 electrode array (Figs 1 and 2A). Spectral analysis of spontaneous activity in L4 of the barrel cortex revealed a significantly higher power in the alpha frequency band in Nlgn4-KO mice as compared with WT controls (Nlgn-4 KO: 12.6 ± 1.1 × 106 μV2, n = 11 barrels, 5 mice and WT: 8.5 ± 0.8 × 106, n = 7 barrels, 4 mice; P = 0.024, Mann–Whitney test, Fig. 2B). In addition, an increase in the occurrence (Nlgn4-KO: 7.6 ± 1.1 min−1, n = 11 barrels, 5 mice and WT: 3.8 ± 0.8 min−1, n = 7 barrels, 4 mice; P = 0.023, Mann–Whitney test, left panel) and prolongation of spindle oscillations (Nlgn-4 KO: 0.72 ± 0.03 s, n = 11 barrels, 5 mice and WT: 0.56 ± 0.06 s, n = 7 barrels, 4 mice; P = 0.037, Mann–Whitney test, right panel) was observed in Nlgn4-KO mice (Fig. 2C). Circular statistics revealed a significant level of MUA modulation in the alpha-range activity, both in Nlgn4-KO and WT mice (Rayleigh test, P < 0.001, Fig. 2D). Although Nlgn4-KO animals demonstrated an increase in the alpha spectrum band activity, only a small change in the mean direction of the resultant vector was observed (Nlgn4-KO: 151.8° ± 0.2°; WT: 153.1° ± 0.3°, P = 0.006, Watson–Williams multisample test). The mean resultant vector length was not significantly different between Nlgn4-KO mice and WT littermates, while the ratio of MUA within alpha-spindle was substantially higher in Nlgn4-KOs as compared with WT mice (Fig. 2D,E). In addition the laminar CSD pattern and amplitudes of alpha-spindles were similar in Nlgn4-KO and WT mice (see Supplementary Fig. 3), suggesting that the amount but not the general properties of alpha-range activity is modified by Nlgn4 loss. These data are in agreement with an increase in power of the alpha band in EEG recordings obtained from patients with ASD (Cornew et al. 2012), further supporting the notion that the Nlgn4-KO mice represent an adequate model of human ASD. We next investigated whether genetic Nlgn4 deletion influences evoked sensory responses elicited by single-whisker stimulation. In contrast to spontaneous network activity, no significant differences were observed in the properties of evoked LFPs recorded in Nlgn4-KO animals and WT littermates (Fig. 3A). Also PPR at 3 different interstimulus intervals did not significantly differ between the 2 genotypes (Fig. 3B). To assess the question whether excitation transfer and activity propagation is altered in the Nlgn4-KO barrel cortex we recorded VSDI transients (ΔF/F0) evoked by single-whisker stimulation (Fig. 4A1–2). Similar to the evoked LFP responses neither amplitudes of VSDI transients nor the area of the response spread were significantly different between WT (n = 6 C2 barrels) and Nlgn4 deficient animals (n = 6 C2 barrels, P > 0.05, Mann–Whitney test) (Fig. 4B1–2, Supplementary Fig. 4). These data indicate that Nlgn4 loss potentiates spontaneous neuronal activity in the somatosensory cortex, but does not affect whisker-mediated sensory-evoked responses. Figure 2. View largeDownload slide Nlgn4-KO influences spontaneous neuronal activity in the barrel cortex in vivo. (A) Continuous 60 s recordings of spontaneous activity in L4 of WT (top 2 traces) and Nlgn4-KO (bottom 2 traces) mice. (B) Averaged FFT spectra (left panel) in cortical L4 of WT (4 mice, 7 barrels, black trace) and Nlgn4-KO mice (5 mice, 11 barrels, grey trace). Right panel shows summation of averaged FFT power in different frequency bands. Note significantly higher power in the alpha frequency band in Nlgn4-KO mice compared with WT littermates. (C) Significantly higher occurrence rate and duration of alpha-spindle oscillations in Nlgn4-KO compared with WT mice. (D) Unit’s phase histogram within alpha-spindle activity in WT (n = 4052 spikes, 7 barrels, 4 mice, left) and Nlgn4-KO animals (n = 9264 spikes, 11 barrels, 5 mice, right). (E) Similar level of MUA modulation was observed in both genotypes. Mean direction (left bar diagram) and mean resultant vector length (middle bar diagram) were not significantly different between the 2 genotypes, however, significantly higher number of MUA spikes in alpha spindle normalized to the total MUA spike number was detected in Nlgn4-KO mice (right bar diagram, n = 7 barrels, 4 WT mice and n = 11 barrels, 5 Nlgn4-KO mice). Asterisks mark significant differences, *P < 0.05, **P < 0.01; and ns, not significant; Mann–Whitney test. Watson–Williams multisample test for circular data. Figure 2. View largeDownload slide Nlgn4-KO influences spontaneous neuronal activity in the barrel cortex in vivo. (A) Continuous 60 s recordings of spontaneous activity in L4 of WT (top 2 traces) and Nlgn4-KO (bottom 2 traces) mice. (B) Averaged FFT spectra (left panel) in cortical L4 of WT (4 mice, 7 barrels, black trace) and Nlgn4-KO mice (5 mice, 11 barrels, grey trace). Right panel shows summation of averaged FFT power in different frequency bands. Note significantly higher power in the alpha frequency band in Nlgn4-KO mice compared with WT littermates. (C) Significantly higher occurrence rate and duration of alpha-spindle oscillations in Nlgn4-KO compared with WT mice. (D) Unit’s phase histogram within alpha-spindle activity in WT (n = 4052 spikes, 7 barrels, 4 mice, left) and Nlgn4-KO animals (n = 9264 spikes, 11 barrels, 5 mice, right). (E) Similar level of MUA modulation was observed in both genotypes. Mean direction (left bar diagram) and mean resultant vector length (middle bar diagram) were not significantly different between the 2 genotypes, however, significantly higher number of MUA spikes in alpha spindle normalized to the total MUA spike number was detected in Nlgn4-KO mice (right bar diagram, n = 7 barrels, 4 WT mice and n = 11 barrels, 5 Nlgn4-KO mice). Asterisks mark significant differences, *P < 0.05, **P < 0.01; and ns, not significant; Mann–Whitney test. Watson–Williams multisample test for circular data. Figure 3. View largeDownload slide Nlgn4-KO does not influence single whisker-mediated evoked responses in the barrel cortex in vivo. (A) Grand average of responses from WT (black trace, n = 700 events, 7 barrels, 4 mice) and Nlgn4-KO mice (grey trace, n = 1100 events, 11 barrels, 5 mice) evoked by single-whisker stimulation. No significant differences in the peak amplitude and slope of evoked response between Nlgn4-KO and WT mice. (B) Grand average of evoked paired-pulse responses (at 100, 250, and 500 ms ISIs) in Nlgn4-KO mice (grey trace) and WT (black trace). Plot illustrates paired-pulse ratios at corresponding ISIs in Nlgn4-KO and WT mice; ns, not significant, Mann–Whitney test. Figure 3. View largeDownload slide Nlgn4-KO does not influence single whisker-mediated evoked responses in the barrel cortex in vivo. (A) Grand average of responses from WT (black trace, n = 700 events, 7 barrels, 4 mice) and Nlgn4-KO mice (grey trace, n = 1100 events, 11 barrels, 5 mice) evoked by single-whisker stimulation. No significant differences in the peak amplitude and slope of evoked response between Nlgn4-KO and WT mice. (B) Grand average of evoked paired-pulse responses (at 100, 250, and 500 ms ISIs) in Nlgn4-KO mice (grey trace) and WT (black trace). Plot illustrates paired-pulse ratios at corresponding ISIs in Nlgn4-KO and WT mice; ns, not significant, Mann–Whitney test. Figure 4. View largeDownload slide Nlgn4-KO does not change VSD responses induced by single-whisker stimulation. (A1) VSDI pattern evoked by deflection of C2 whisker at different time points after the stimulus onset (P34 Nlgn4-KO mouse). (A2) Representative VSD transient evoked by deflection of C2 whisker, dashed line represents 50% of peak VSD amplitude threshold which was used for calculation of evoked VSD response area. (A3) Solid contour represents area of the evoked VSD response at half-maximal amplitude at the time of peak amplitude. (B1) Average peak amplitude and mean area (B2) of VSD responses induced by single-whisker stimulation in WT (open bars) and Nlgn4 deficient animals (filled bars, n = 6 mice of each genotype). ns, not significant, Mann–Whitney test. Figure 4. View largeDownload slide Nlgn4-KO does not change VSD responses induced by single-whisker stimulation. (A1) VSDI pattern evoked by deflection of C2 whisker at different time points after the stimulus onset (P34 Nlgn4-KO mouse). (A2) Representative VSD transient evoked by deflection of C2 whisker, dashed line represents 50% of peak VSD amplitude threshold which was used for calculation of evoked VSD response area. (A3) Solid contour represents area of the evoked VSD response at half-maximal amplitude at the time of peak amplitude. (B1) Average peak amplitude and mean area (B2) of VSD responses induced by single-whisker stimulation in WT (open bars) and Nlgn4 deficient animals (filled bars, n = 6 mice of each genotype). ns, not significant, Mann–Whitney test. Disturbed Intracortical Information Processing in Nlgn4-KO Mice Although no major changes in the whisker-mediated evoked activity were detected by VSDI, this method records the activity only in the superficial neocortical layers. Therefore, we next explored whether Nlgn4 loss influences information flow between barrel-related columns and within one column in the barrel cortex in vivo by analyzing MUA. The late component of evoked MUA occurring between 50 and 100 ms was significantly smaller in Nlgn4-KO animals compared with WT within L2–4 of the principal column and in L4 of the neighboring column (Fig. 5A). These data were supported by the calculation of the late amplitudes of the current sinks in the CSD analyses (Fig. 5B). Furthermore, to assess the contribution of excitatory and inhibitory neurons to the observed abnormal neuronal activity, we classified neurons as putative excitatory and inhibitory according to their mean spike waveform duration (Fig. 5C, Supplemental Fig. 1C). Both, excitatory and inhibitory neurons showed a significant decrease in the late MUA component of Nlgn4-KO compared with WT animals, indicating a major impact of Nlgn4 loss on both glutamatergic and GABAergic synapses in the somatosensory cortex. Figure 5. View largeDownload slide Genetic ablation of Nlgn4 affects late whisker-mediated evoked multiunit activity (MUA) and current sink responses in layer 2–5 of the barrel cortex in vivo. (A) Averaged traces representing evoked MUA recordings in Nlgn4-KO (grey, n = 11 barrels, 5 mice) and WT (black, n = 7 barrels, 4 mice) mice. (B) Mean evoked current sink responses from Nlgn4-KO (grey, n = 11 barrels, 5 mice) and WT (black n = 7 barrels, 4 mice) animals. (C) Isolation of putative excitatory (EXC) and inhibitory (INH) neurons based on mean spike waveform duration. (D) Averaged traces representing evoked MUA recorded from putative excitatory and inhibitory L2–5 neurons in Nlgn4-KO (grey) and WT (black) mice. Asterisks mark significant differences, *P < 0.05, **P < 0.01. Figure 5. View largeDownload slide Genetic ablation of Nlgn4 affects late whisker-mediated evoked multiunit activity (MUA) and current sink responses in layer 2–5 of the barrel cortex in vivo. (A) Averaged traces representing evoked MUA recordings in Nlgn4-KO (grey, n = 11 barrels, 5 mice) and WT (black, n = 7 barrels, 4 mice) mice. (B) Mean evoked current sink responses from Nlgn4-KO (grey, n = 11 barrels, 5 mice) and WT (black n = 7 barrels, 4 mice) animals. (C) Isolation of putative excitatory (EXC) and inhibitory (INH) neurons based on mean spike waveform duration. (D) Averaged traces representing evoked MUA recorded from putative excitatory and inhibitory L2–5 neurons in Nlgn4-KO (grey) and WT (black) mice. Asterisks mark significant differences, *P < 0.05, **P < 0.01. Impaired GABAergic Transmission in the Nlgn4-Deficient Barrel Cortex To assess the mechanisms underlying the changes of spontaneous activity observed in vivo, we performed whole-cell patch clamp recordings in acute brain slices. Since Nlgn4 loss was reported to affect inhibitory neurotransmission (Hoon et al. 2011; Hammer et al. 2015), we first recorded mIPSCs in L4 spiny stellate neurons. No difference in the mean mIPSC amplitudes was observed (Nlgn4-KO: 34 ± 4 pA, n = 22 and WT: 40 ± 5 pA, n = 15; P = 0.35, Mann–Whitney test, Fig. 6A,B), but mean mIPSC frequency was significantly lower in Nlgn4-KO compared with WT littermates (Nlgn4-KO: 1.2 ± 0.2 Hz, n = 22 and WT: 2.7 ± 0.5 Hz, n = 15; P = 0.0009, Mann–Whitney test, Fig. 6A,B). In addition, Nlgn4 ablation led to larger mIPSC rise time (Nlgn4-KO: 1.63 ± 0.14 ms, n = 22 and WT: 1.04 ± 0.12 ms, n = 15; P = 0.0008, Mann–Whitney test) and decay time (Nlgn4-KO: 14.4 ± 0.9 ms, n = 22 and WT: 10.4 ± 0.9 ms, n = 15; P = 0.002, Mann–Whitney test) constants (Fig. 6B). Figure 6. View largeDownload slide Decreased release probability at GABAergic synapses in Nlgn4 KO in vitro. (A) Representative mIPSCs recorded in WT (top trace) and Nlgn4-KO (bottom trace) mice and cumulative distributions of mIPSC amplitudes. (B) Significantly lower mean frequency of mIPSCs and significant prolongation of rise time and decay time constants in Nlgn4-KO mice compared with WT littermates (Mann–Whitney test). (C) Averaged eIPSCs (20 trials in both cases) elicited by paired-pulse electrical stimulation (50 ms ISI) of L4 spiny stellate neuron in WT (upper trace) and KO (lower trace) mice. Significantly smaller mean eIPSC amplitude in KO versus WT mice. Significantly larger PPR in Nlgn4-KO animals was observed at 50 ms ISI, but not at 100 and 200 ms ISIs (unpaired Student’s t-test). Numbers in (B,C) indicate number of cells. Asterisks mark significant differences, *P < 0.05, **P < 0.01, ***P < 0.001. Figure 6. View largeDownload slide Decreased release probability at GABAergic synapses in Nlgn4 KO in vitro. (A) Representative mIPSCs recorded in WT (top trace) and Nlgn4-KO (bottom trace) mice and cumulative distributions of mIPSC amplitudes. (B) Significantly lower mean frequency of mIPSCs and significant prolongation of rise time and decay time constants in Nlgn4-KO mice compared with WT littermates (Mann–Whitney test). (C) Averaged eIPSCs (20 trials in both cases) elicited by paired-pulse electrical stimulation (50 ms ISI) of L4 spiny stellate neuron in WT (upper trace) and KO (lower trace) mice. Significantly smaller mean eIPSC amplitude in KO versus WT mice. Significantly larger PPR in Nlgn4-KO animals was observed at 50 ms ISI, but not at 100 and 200 ms ISIs (unpaired Student’s t-test). Numbers in (B,C) indicate number of cells. Asterisks mark significant differences, *P < 0.05, **P < 0.01, ***P < 0.001. Decreased mIPSC frequency could be a consequence of decreased presynaptic neurotransmitter release probability in GABAergic synapses. Therefore we tested whether Nlgn4-KO affects eIPSCs. Indeed, a reduction of eIPSC amplitudes was observed in Nlgn4-KO animals (Nlgn4-KO: 86 ± 7 pA, n = 20 and WT: 133 ± 25 pA, n = 9; P = 0.02, unpaired Student’s t-test, Fig. 6C). In order to further investigate the possibility that Nlgn4 loss results in a decrease of release probability, we applied paired-pulse stimulation at 3 different ISIs, namely 50, 100, and 200 ms (Fig. 6C). The PPR at an ISI of 50 ms was significantly higher in Nlgn4-KO mice (Nlgn4-KO: 1.17 ± 0.14, n = 20 and WT: 0.76 ± 0.14, n = 9; P = 0.04, unpaired Student’s t-test, Fig. 6C). Thus both a decrease in mIPSC frequency and an increase in PPR demonstrate that presynaptic GABA release probability is reduced in Nlgn4-KO mice. NCAM, another neural cell adhesion molecule, has been reported to affect the number of presynaptic ready-for-release vesicles, the so-called readily releasable pool of vesicles (RRP) (Chan et al. 2005). Therefore we tested whether Nlgn4-KO influences the RRP in GABAergic synapses. Local pressure application of hyperosmotic (+500 mM sucrose) ACSF stimulated synaptic release of GABA, which was detected on the postsynaptic cell as a barrage of events. The number of released vesicles was significantly reduced in Nlgn4-KO animals (Nlgn4-KO: 1661 ± 285, n = 11 and WT: 4531 ± 457, n = 6; P = 0.0015, unpaired Student’s t-test, Fig. 7A). These data further support the notion that the observed decrease in mIPSC frequency in Nlgn4 KOs results from both, a reduction in RRP size and a decrease in release probability. Figure 7. View largeDownload slide Reduced RRP size at GABAergic synapses in Nlgn4 KO in vitro. (A) Representative responses to application of sucrose (500 mM) in WT (left trace) and Nlgn4-KO (right trace) recorded in the presence of APV, CNQX and TTX. Significantly smaller number of vesicles in RRP of Nlgn4-KO animals compared with WT littermates. (B) Confocal images showing gephyrin immunoreactivity in the barrel cortex of WT (left panel) and KO (right panel) mice. No significant changes were observed in the density of gephyrin puncta in the barrel cortex of Nlgn4-KO mice. Numbers in (A) indicate number of cells, in (B) number of pairs. Asterisks mark significant differences, **P < 0.01 and ns, not significant; unpaired Student’s t-test. Figure 7. View largeDownload slide Reduced RRP size at GABAergic synapses in Nlgn4 KO in vitro. (A) Representative responses to application of sucrose (500 mM) in WT (left trace) and Nlgn4-KO (right trace) recorded in the presence of APV, CNQX and TTX. Significantly smaller number of vesicles in RRP of Nlgn4-KO animals compared with WT littermates. (B) Confocal images showing gephyrin immunoreactivity in the barrel cortex of WT (left panel) and KO (right panel) mice. No significant changes were observed in the density of gephyrin puncta in the barrel cortex of Nlgn4-KO mice. Numbers in (A) indicate number of cells, in (B) number of pairs. Asterisks mark significant differences, **P < 0.01 and ns, not significant; unpaired Student’s t-test. The observed decrease in RRP size may reflect not only a decrease in the number of readily releasable vesicles but also a decrease in the number of GABAergic synapses. To investigate whether Nlgn4-KO causes changes of synaptic protein levels in L4 neurons we performed immunostaining for gephyrin, a scaffold protein of inhibitory postsynapses (Fig. 7B). The puncta number corresponding to gephyrin per 10 μm cell perimeter in the perisomatic region was not changed in KO animals as compared with WT littermates (Fig. 7B). Thus genetic ablation of Nlgn4 results in a robust presynaptic weakening of GABAergic transmission, but does not affect the number of GABAergic synapses in L4, at least not in the perisomatic region. Similar to previous studies (Delattre et al. 2013; Hammer et al. 2015), our data demonstrate that Nlgn4 acts at GABAergic synapses. Impaired Glutamatergic Transmission in Nlgn4-Deficient Barrel Cortex Changes in excitation/inhibition balance were previously reported in multiple mouse models of ASD (Tabuchi et al. 2007; Bateup et al. 2013; Vogt et al. 2015). Therefore, we recorded miniature EPSCs (mEPSCs) to explore whether Nlgn4-KO also affects glutamatergic transmission in L4. Mean mEPSC amplitudes did not differ in Nlgn4-KO as compared with WT animals (Nlgn4-KO: 8.4 ± 0.7 pA, n = 15 and WT: 9.7 ± 1.3 pA; n = 13, P = 0.003, Mann–Whitney test, Fig. 8A), while analysis of averaged mEPSC frequencies revealed a significant decrease in Nlgn4-KO animals (Nlgn4-KO: 0.6 ± 0.3 Hz, n = 15 and WT: 2.0 ± 0.6 Hz, n = 13; P = 0.003, Mann–Whitney test, Fig. 8A,B). In contrast to mIPSCs, neither the rise time constant (Nlgn4-KO: 1.7 ± 0.2 ms, n = 15 and WT: 1.5 ± 0.2 ms, n = 13; P = 0.58, Mann–Whitney test) nor the decay time constant (Nlgn4-KO: 9 ± 1 ms, n = 15 and WT: 7 ± 1 ms, n = 13; P = 0.18, Mann–Whitney test) of mEPSCs differed significantly between Nlgn4-KO and WT littermates (Fig. 8B). Figure 8. View largeDownload slide Nlgn4 KO affects mEPSCs in layer 4 spiny stellate neurons without affecting thalamocortical excitatory transmission in vitro. (A) Representative mEPSCs recorded in WT (top trace) and Nlgn4-KO (bottom trace) mice. Cumulative distributions of mEPSC amplitudes. (B) Significantly lower mean frequency of mEPSCs in Nlgn4-KO mice compared with WT littermates, while rise time and decay time constants did not vary between Nlgn4-KO and WT groups (Mann–Whitney test). (C) Representative mean eEPSCs elicited by paired-pulse electrical stimulation (50 ms ISI) in the VPM (20 trials in both cases) recorded in L4 spiny stellate neurons in WT (upper) and Nlgn4-KO (lower trace) mice. Neither the mean eEPSC amplitude, nor failure rates, nor PPR differed in Nlgn4-KO mice as compared with WT littermates (unpaired Student’s t-test). Numbers in (B,C) indicate number of cells. Asterisks mark significant differences, **P < 0.01 and ns, not significant. Figure 8. View largeDownload slide Nlgn4 KO affects mEPSCs in layer 4 spiny stellate neurons without affecting thalamocortical excitatory transmission in vitro. (A) Representative mEPSCs recorded in WT (top trace) and Nlgn4-KO (bottom trace) mice. Cumulative distributions of mEPSC amplitudes. (B) Significantly lower mean frequency of mEPSCs in Nlgn4-KO mice compared with WT littermates, while rise time and decay time constants did not vary between Nlgn4-KO and WT groups (Mann–Whitney test). (C) Representative mean eEPSCs elicited by paired-pulse electrical stimulation (50 ms ISI) in the VPM (20 trials in both cases) recorded in L4 spiny stellate neurons in WT (upper) and Nlgn4-KO (lower trace) mice. Neither the mean eEPSC amplitude, nor failure rates, nor PPR differed in Nlgn4-KO mice as compared with WT littermates (unpaired Student’s t-test). Numbers in (B,C) indicate number of cells. Asterisks mark significant differences, **P < 0.01 and ns, not significant. Layer 4 spiny stellate neurons receive direct innervation from the ventrobasal thalamus (Agmon and Connors 1991; Schubert et al. 2003; Staiger et al. 2004). To address the question as to whether Nlgn4 deletion affects thalamocortical excitatory inputs, we recorded eEPSC. Paired-pulse simulation at an ISI of 50 ms was performed in the VPM of the thalamus using bipolar tungsten electrode (Fig. 8C). Surprisingly, even though a decrease in mEPSC frequency was observed in Nlgn4-KO mice, no significant change in the mean amplitudes of thalamocortical eEPSCs in Nlgn4-KO animals was detected (Nlgn4-KO: 13 ± 3 pA, n = 9 and WT: 13 ± 4 pA, n = 8; P = 0.85, unpaired Student’s t-test). Similarly, no change was observed in the failure rate (part of trials in which the first stimulus did not elicit an eEPSC) of eEPSCs (Nlgn4-KO: 0.24 ± 0.05, n = 9 and WT: 0.21 ± 0.06, n = 8; P = 0.75, unpaired Student’s t-test) and in PPR (Nlgn4-KO: 0.77 ± 0.13, n = 9 and WT: 0.61 ± 0.1, n = 8; P = 0.38, unpaired Student’s t-test, Fig. 8C). Next, we estimated RRP at glutamatergic synapses applying hypertonic (+500 mM sucrose) ACSF. A significant decrease in the number of released vesicles was observed in Nlgn4-KO animals as compared with WT littermates (Nlgn4-KO: 1122 ± 127, n = 11 and WT: 2686 ± 323, n = 9; P = 0.006, unpaired Student’s t-test, Fig. 9A). Thus, the decrease in mEPSC frequency may be explained by a reduction in RRP size, indicating that excitatory transmission in L4 of the barrel cortex is reduced mostly presynaptically. Figure 9. View largeDownload slide Nlgn4 KO reduces RRP size at glutamatergic synapses in layer 4 spiny stellate neurons. (A) Typical responses elicited by application of the hypertonic solution in WT (left trace) and Nlgn4-KO (right trace). Number of vesicles in RRP in Nlgn4-KO mice was significantly reduced as compared with WT littermates (unpaired Student’s t-test). (B) Confocal images showing PSD-95 immunoreactivity in the barrel cortex of WT (left panel) and Nlgn4-KO (right panel) mice. No significant changes were observed in the density of PSD-95 puncta in the barrel cortex of Nlgn4-KO mice. Numbers in (A) indicate number of cells, in (B) number of pairs. Asterisks mark significant differences, ***P < 0.001 and ns, not significant, unpaired Student’s t-test. Figure 9. View largeDownload slide Nlgn4 KO reduces RRP size at glutamatergic synapses in layer 4 spiny stellate neurons. (A) Typical responses elicited by application of the hypertonic solution in WT (left trace) and Nlgn4-KO (right trace). Number of vesicles in RRP in Nlgn4-KO mice was significantly reduced as compared with WT littermates (unpaired Student’s t-test). (B) Confocal images showing PSD-95 immunoreactivity in the barrel cortex of WT (left panel) and Nlgn4-KO (right panel) mice. No significant changes were observed in the density of PSD-95 puncta in the barrel cortex of Nlgn4-KO mice. Numbers in (A) indicate number of cells, in (B) number of pairs. Asterisks mark significant differences, ***P < 0.001 and ns, not significant, unpaired Student’s t-test. We also tested if Nlgn4-loss influences the number of excitatory synapses. We performed immunostaining for PSD-95, a marker of postsynaptic density of excitatory synapses (Fig. 9B). The puncta number corresponding to PSD-95 per 10 μm cell perimeter in the perisomatic region was not different in Nlgn4-KO animals as compared with WT, indicating that Nlgn4 loss does not cause a loss of excitatory synapses in L4 of barrel cortex. We conclude that although Nlgn4-KO leads to RRP reduction in general, this change probably does not take place at thalamocortical synapses, preserving the strength of whisker-mediated sensory inputs. Discussion Using multichannel LFP recordings in vivo and whole-cell patch clamp recordings in vitro, we investigated the consequences of autism associated Nlgn4 loss in the mouse barrel cortex. We report that genetic deletion of Nlgn4 results in (1) a potentiation in the alpha (8–13 Hz) band of the power spectrum of spontaneous network activity in vivo, (2) a decrease in mIPSC and mEPSC frequencies in L4 spiny stellate neurons, (3) a reduction in the number of readily releasable vesicles in both GABAergic and glutamatergic synapses, although the number of perisomatic synapses remains unchanged, (4) normal early evoked sensory responses in vivo and thalamocortical eEPSCs in vitro, and (5) suppressed late component of evoked MUA and CSD. Our data indicate that Nlgn4 loss predominantly impairs intracortical synaptic transmission, and hence information processing in the cortex, but does not affect incoming sensory information. GABAergic Synaptic Transmission in Nlgn4-KO Animals Emerging evidence indicates that deficiency of inhibitory neurotransmission is crucially involved in functional alterations causing ASD (Chao et al. 2010; Han et al. 2012; Tyzio et al. 2014), for review see (Coghlan et al. 2012). Therefore, we studied the function of the inhibitory system at the single cell level in L4 of the barrel cortex. A recent study reported a ~50% decrease of eIPSC amplitude in Nlgn4-KO mice (Delattre et al. 2013). In the present study, the mean eIPSC amplitude was reduced in Nlgn4-KO to ~65% of WT value, which is in line with the previous work. A reduction of evoked responses may arise from various alterations, including changes in quantal amplitude, number of functional synapses or presynaptic release probability. Our immunostaining data revealed no alterations in the number of GABAergic synapses, at least not in the perisomatic region of spiny stellate cells (Fig. 7B). Furthermore, the mIPSC amplitude, that is, an IPSC elicited by release of a single GABA containing vesicle, was not influenced by the loss of Nlgn4. On the other hand, the mean mIPSC frequency was strongly reduced in Nlgn4-KO animals compared WT (to ~45%, Fig. 6B). This alteration most probably reflects a reduction of the presynaptic RRP, a decrease in GABA release probability, or both. Using a hypertonic solution to completely empty the RRP (Rosenmund and Stevens 1996), we report that the RRP size in Nlgn4-KO mice amounts to only ~35% of that in WT animals (Fig. 7A). Thus the observed reduction in mIPSC frequency can be attributed to the RRP decrease in Nlgn4-KO mice as compared with WT. Paired-pulse stimulation is generally applied to estimate release probability. In agreement with a previous study (Delattre et al. 2013), PPR was higher in Nlgn4-KO mice, indicating a decrease of presynaptic GABA release probability. PPR could be also influenced by the postsynaptic site, for instance via saturation and/or desensitization of postsynaptic receptors. However, mIPSC kinetics are significantly slower in Nlgn4-KO mice (Fig. 6B). Therefore, if postsynaptic receptors influenced PPR, one would rather expect a decrease in PPR due to receptor unavailability. Thus, the observed increase in PPR indicates that GABAergic release probability is reduced in Nlgn4-KO animals. We conclude that Nlgn4 loss causes a strong weakening of GABAergic inhibition in the somatosensory cortex. Most of the alterations caused by Nlgn4-KO originate presynaptically, but the prolongation of rise and decay time constants of mIPSCs indicate possible additional modifications of the postsynapse. Interestingly, Nlgn4 deficiency in CA3 region of hippocampus results mainly in postsynaptic changes (Hammer et al. 2015), while in the present study Nlgn4 loss affected predominantly presynaptic function. Although neuroligins are postsynaptic cell membrane molecules their interaction with presynaptic neurexin molecules is potentially capable to modulate presynaptic function (Sudhof 2008). The trans-synaptic signaling pathway was previously described for PSD-95–Nlgn1–β-Neurexin complex (Futai et al. 2007). Deletion of α-neurexins results in a severe impairment of presynaptic Ca2+ channels function and, hence, in impaired neurotransmitter release (Missler et al. 2003). Moreover interaction of neurexins with synaptotagmin, a vesicular Ca2+ binding protein (Hata et al. 1993), may modulate neurotransmitter release via targeting and docking of synaptic vesicles. Therefore, decrease in the RRP size and/or decrease in the release probability in both GABA- and glutamatergic system observed in the present study may result from trans-synaptic retrograde signaling alterations induced by Nlgn4 loss. Glutamatergic Synaptic Transmission in Nlgn4-KO Similar to the Nlgn4-KO-mediated effects on the GABAergic system, Nlgn4 ablation was reported to reduce eEPSCs. To investigate this issue in more detail, we inspected several parameters of the glutamatergic system. Loss of Nlgn4 did not affect the number of perisomatic synapses or the mean mEPSC amplitude. In contrast to our observations on the GABAergic system, Nlgn4-KO failed to change the kinetics of mEPSCs, indicating that Nlgn4-KO affects glutamatergic transmission mostly presynaptically. Both, mEPSC frequency and RRP size were reduced in Nlgn4-deficient animals compared with WT mice, to ~40% and ~30%, respectively. These results are in line with the reported decrease of eEPSC amplitude in Nlgn4-KO mice (Delattre et al. 2013). However, using the advantage of the thalamocortical in vitro preparation, we selectively stimulated thalamocortical projections. Neither the mean eEPSC amplitude nor the failure rate or the PPR were affected by Nlgn4-KO. Our data point to a preferential impairment of intracortical information processing rather than afferent sensory inputs by Nlgn4 loss, suggesting that Nlg4 may be expressed not at all glutamatergic synapses. Neuronal Activity in Nlgn4-Deficient Somatosensory Cortex In Vivo In agreement with our in vitro data, loss of Nlgn4 failed to affect either amplitude or PPR of single-whisker stimulation-induced LFPs as well as amplitudes of evoked VSDI transients in the barrel cortex in vivo (Figs 3 and 4). Thus, both in vitro and in vivo results indicate that neither the strength nor the short-term plasticity of whisker-mediated sensory inputs to the barrel cortex is affected by Nlgn4 loss. Interestingly, recent EEG data suggest that sensory inputs are not disturbed in ASD patients (Butler et al. 2017). In contrast, intracortical spontaneous activity in vivo revealed an increase in the alpha frequency band in Nlgn4-deficient animals (Fig. 2). In addition, a significantly higher occurrence rate and longer duration of alpha-spindle oscillations were observed in Nlgn4-KO mice. On the other hand Nlgn4-KO animals showed a similar profile of CSD and MUA within alpha-spindle oscillations. Hence, the deficits at glutamatergic and GABAergic synapses in Nlgn4-KO mice apparently enhance an amount but not basic characteristics of alpha-rhythm. Short-lasting alpha-spindles (200–400 ms) in EEG recordings are generally interpreted as a correlate of an inhibited network (Pfurtscheller 2003; Jokisch and Jensen 2007; Sonnleitner et al. 2012). Indeed, in Nlg4-KO animals delayed MUA (50–100 ms after stimulus) is not only more strongly suppressed in the entire principal column, but also in the neighboring columns, suggesting that information processing is impaired in the Nlgn4-KO barrel cortex. The lack of differences in the amplitude and shape of initial evoked component in vivo between Nlgn4-KO and WT animals is in line with the lack of alterations in thalamocortical eEPSCs. However, only about 15% of glutamatergic synapses in the L4 barrel cortex are formed by thalamic axons, while the other excitatory synapses represent intracortical connections (Benshalom and White 1986; Altmann et al. 1987; Feldmeyer et al. 1999). Thus, the deficits in synaptic transmission (Figs 8 and 9A) can be sufficient to reduce the late component evoked by single-whisker stimulation, which most likely results from intracortical processing. The late response induced by single-whisker stimulation under urethane anesthesia may represent UP-state of K-complex (Amzica and Steriade 1998, 2002). As UP-states are dependent on excitatory connections within L4 (Chauvette et al. 2010), our data indicate that intrabarrel excitatory reciprocal connections are weakened in the Nlgn4 deficient animals. Although Nlgn4-KO reduces MUA in both, excitatory and inhibitory neurons, it is tempting to speculate that excitatory activity is more strongly reduced (Fig. 5C), leading to a shift in excitation/inhibition (E/I) balance to inhibition. Excitation/Inhibition Balance in the Nlgn4-KO Mouse Model of ASD A reduction of the E/I ratio, defined as the ratio of mean amplitude of eEPSCs to mean amplitude of eIPSCs, has been previously reported in Nlgn4-deficient somatosensory cortex (Delattre et al. 2013). A stronger weakening of glutamatergic transmission and/or weaker reduction of GABAergic transmission may underlie this observation. Our data show that genetic Nlgn4 ablation leads to a reduction in mIPSC frequency to ~45% (Fig. 6), whereas mEPSC frequency was decreased to ~30% (Fig. 8), indicating a stronger suppression of the glutamatergic system in Nlgn4-deficient animals. Consequently the E/I ratio, defined as the ratio of mEPSC frequency to mIPSC frequency, was reduced from ~0.75 in WT animals to ~0.5 in Nlgn4-KO mice. A disturbed E/I balance is generally believed to contribute to many neurological and neuropsychiatric disorders. Interestingly, ASD is accompanied with epilepsy, a disorder occurring as a result of over-excitation, in up to 30% of the patients (Bolton et al. 2001; Amiet et al. 2008; Lai et al. 2014), suggesting that in ASD the E/I balance is shifted to excitation. Several studies confirmed an increased E/I ratio in animal models of ASD (Rubenstein and Merzenich 2003; Gogolla et al. 2009; Tang et al. 2014), but both in Delattre et al. (Delattre et al. 2013) and in this work a decrease of the E/I ratio has been observed. We suggest that the Nlgn4-KO-mediated decrease in E/I ratio can synchronize neuronal activity in the somatosensory cortex in a relatively “relaxed” alpha-rhythmus, causing an attention deficit. Indeed, alpha activity in EEG recordings is suggested to reflect an inhibition during information processing in the cerebral cortex (Ray and Cole 1985; Pfurtscheller 2001, 2003). As a consequence, incoming sensory information, despite its insensitivity to Nlgn4 ablation, cannot be correctly processed within intracortical networks and resulting in restricted interests or behavioral deficits typical for ASD. Supplementary Material Supplementary data is available at Cerebral Cortex online. Funding German Research Foundation (DFG) grants to H.J.L. and N.B. (CNMPB); the Max Planck Society to N.B.; the European Commission (EU-AIMS FP7-115300 to N.B. and Marie Curie IRG, D.K.-B.) and the Alexander von Humboldt Stiftung (to D.K.-B.). Notes We thank Dr Vicente Reyes-Puerta for valuable suggestions regarding this manuscript. The technical assistance of Mrs Beate Krumm is highly appreciated. Conflict of Interest: The authors declare no conflict of interest. References Agmon A , Connors BW . 1991 . Thalamocortical responses of mouse somatosensory (barrel) cortex invitro . Neuroscience . 41 : 365 – 379 . Google Scholar CrossRef Search ADS PubMed Altmann L , Luhmann HJ , Greuel JM , Singer W . 1987 . Functional and neuronal binocularity in kittens raised with rapidly alternating monocular occlusion . J Neurophysiol . 58 : 965 – 980 . Google Scholar CrossRef Search ADS PubMed Amiet C , Gourfinkel-An I , Bouzamondo A , Tordjman S , Baulac M , Lechat P , Mottron L , Cohen D . 2008 . Epilepsy in autism is associated with intellectual disability and gender: evidence from a meta-analysis . Biol Psychiatry . 64 : 577 – 582 . Google Scholar CrossRef Search ADS PubMed Amzica F , Steriade M . 1998 . Cellular substrates and laminar profile of sleep K-complex . Neuroscience . 82 : 671 – 686 . Google Scholar CrossRef Search ADS PubMed Amzica F , Steriade M . 2002 . The functional significance of K-complexes . Sleep Med Rev . 6 : 139 – 149 . Google Scholar CrossRef Search ADS PubMed Andrillon T , Nir Y , Staba RJ , Ferrarelli F , Cirelli C , Tononi G , Fried I . 2011 . Sleep spindles in humans: insights from intracranial EEG and unit recordings . J Neurosci . 31 : 17821 – 17834 . Google Scholar CrossRef Search ADS PubMed Bartho P , Hirase H , Monconduit L , Zugaro M , Harris KD , Buzsaki G . 2004 . Characterization of neocortical principal cells and interneurons by network interactions and extracellular features . J Neurophysiol . 92 : 600 – 608 . Google Scholar CrossRef Search ADS PubMed Bateup HS , Johnson CA , Denefrio CL , Saulnier JL , Kornacker K , Sabatini BL . 2013 . Excitatory/inhibitory synaptic imbalance leads to hippocampal hyperexcitability in mouse models of tuberous sclerosis . Neuron . 78 : 510 – 522 . Google Scholar CrossRef Search ADS PubMed Beierlein M , Gibson JR , Connors BW . 2003 . Two dynamically distinct inhibitory networks in layer 4 of the neocortex . J Neurophysiol . 90 : 2987 – 3000 . Google Scholar CrossRef Search ADS PubMed Bemben MA , Nguyen QA , Wang TG , Li Y , Nicoll RA , Roche KW . 2015 . Autism-associated mutation inhibits protein kinase C-mediated neuroligin-4X enhancement of excitatory synapses . Proc Natl Acad Sci USA . 112 : 2551 – 2556 . Google Scholar CrossRef Search ADS PubMed Benshalom G , White EL . 1986 . Quantification of thalamocortical synapses with spiny stellate neurons in layer IV of mouse somatosensory cortex . J Comp Neurol . 253 : 303 – 314 . Google Scholar CrossRef Search ADS PubMed Berens P . 2009 . CircStat: a MATLAB Toolbox for circular statistics . J Stat Softw . 31 : 1 – 21 . Google Scholar CrossRef Search ADS Bolliger MF , Frei K , Winterhalter KH , Gloor SM . 2001 . Identification of a novel neuroligin in humans which binds to PSD-95 and has a widespread expression . Biochem J . 356 : 581 – 588 . Google Scholar CrossRef Search ADS PubMed Bolton PF , Dennis NR , Browne CE , Thomas NS , Veltman MWM , Thompson RJ , Jacobs P . 2001 . The phenotypic manifestations of interstitial duplications of proximal 15q with special reference to the autistic spectrum disorders . Am J Med Genet . 105 : 675 – 685 . Google Scholar CrossRef Search ADS PubMed Budreck EC , Scheiffele P . 2007 . Neuroligin-3 is a neuronal adhesion protein at GABAergic and glutamatergic synapses . Eur J Neurosci . 26 : 1738 – 1748 . Google Scholar CrossRef Search ADS PubMed Butler JS , Molholm S , Andrade GN , Foxe JJ . 2017 . An examination of the neural unreliability thesis of autism . Cereb Cortex . 27 : 185 – 200 . Google Scholar CrossRef Search ADS PubMed Chan SA , Polo-Parada L , Landmesser LT , Smith C . 2005 . Adrenal chromaffin cells exhibit impaired granule trafficking in NCAM knockout mice . J Neurophysiol . 94 : 1037 – 1047 . Google Scholar CrossRef Search ADS PubMed Chao HT , Chen HM , Samaco RC , Xue MS , Chahrour M , Yoo J , Neul JL , Gong SC , Lu HC , Heintz N , et al. . 2010 . Dysfunction in GABA signalling mediates autism-like stereotypies and Rett syndrome phenotypes . Nature . 468 : 263 – 269 . Google Scholar CrossRef Search ADS PubMed Chauvette S , Volgushev M , Timofeev I . 2010 . Origin of active states in local neocortical networks during slow sleep oscillation . Cereb Cortex . 20 : 2660 – 2674 . Google Scholar CrossRef Search ADS PubMed Coghlan S , Horder J , Inkster B , Mendez MA , Murphy DG , Nutt DJ . 2012 . GABA system dysfunction in autism and related disorders: from synapse to symptoms . Neurosci Biobehav Rev . 36 : 2044 – 2055 . Google Scholar CrossRef Search ADS PubMed Cornew L , Roberts TPL , Blaskey L , Edgar JC . 2012 . Resting-state oscillatory activity in autism spectrum disorders . J Autism Dev Disord . 42 : 1884 – 1894 . Google Scholar CrossRef Search ADS PubMed Delattre V , La Mendola D , Meystre J , Markram H , Markram K . 2013 . Nlgn4 knockout induces network hypo-excitability in juvenile mouse somatosensory cortex in vitro . Sci Rep . 3 : 2897 . Google Scholar CrossRef Search ADS PubMed El-Kordi A , Winkler D , Hammerschmidt K , Kastner A , Krueger D , Ronnenberg A , Ritter C , Jatho J , Radyushkin K , Bourgeron T , et al. . 2013 . Development of an autism severity score for mice using Nlgn4 null mutants as a construct-valid model of heritable monogenic autism . Behav Brain Res . 251 : 41 – 49 . Google Scholar CrossRef Search ADS PubMed Feldmeyer D , Egger V , Lubke J , Sakmann B . 1999 . Reliable synaptic connections between pairs of excitatory layer 4 neurones within a single ‘barrel’ of developing rat somatosensory cortex . J Physiol . 521 ( Pt 1 ): 169 – 190 . Google Scholar CrossRef Search ADS PubMed Futai K , Kim MJ , Hashikawa T , Scheiffele P , Sheng M , Hayashi Y . 2007 . Retrograde modulation of presynaptic release probability through signaling mediated by PSD-95-neuroligin . Nat Neurosci . 10 : 186 – 195 . Google Scholar CrossRef Search ADS PubMed Gasser EMS , Straub CJ , Panzanelli P , Weinmann O , Sassoe-Pognetto M , Fritschy JM . 2006 . Immunofluorescence in brain sections: simultaneous detection of presynaptic and postsynaptic proteins in identified neurons . Nat Protoc . 1 : 1887 – 1897 . Google Scholar CrossRef Search ADS PubMed Gogolla N , LeBlanc JJ , Quast KB , Sudhof TC , Fagiolini M , Hensch TK . 2009 . Common circuit defect of excitatory-inhibitory balance in mouse models of autism . J Neurodev Disord . 1 : 172 – 181 . Google Scholar CrossRef Search ADS PubMed Hammer M , Krueger-Burg D , Tuffy LP , Cooper BH , Taschenberger H , Goswami SP , Ehrenreich H , Jonas P , Varoqueaux F , Rhee JS , et al. . 2015 . Perturbed hippocampal synaptic inhibition and gamma-oscillations in a neuroligin-4 knockout mouse model of autism . Cell Rep . 13 : 516 – 523 . Google Scholar CrossRef Search ADS PubMed Han S , Tai C , Westenbroek RE , Yu FH , Cheah CS , Potter GB , Rubenstein JL , Scheuer T , de la Iglesia HO , Catterall WA . 2012 . Autistic-like behaviour in Scn1a(+/−) mice and rescue by enhanced GABA-mediated neurotransmission . Nature . 489 : 385 – 390 . Google Scholar CrossRef Search ADS PubMed Harris KD , Henze DA , Csicsvari J , Hirase H , Buzsaki G . 2000 . Accuracy of tetrode spike separation as determined by simultaneous intracellular and extracellular measurements . J Neurophysiol . 84 : 401 – 414 . Google Scholar CrossRef Search ADS PubMed Hata Y , Davletov B , Petrenko AG , Jahn R , Sudhof TC . 1993 . Interaction of synaptotagmin with the cytoplasmic domains of neurexins . Neuron . 10 : 307 – 315 . Google Scholar CrossRef Search ADS PubMed Hazan L , Zugaro M , Buzsaki G . 2006 . Klusters, NeuroScope, NDManager: a free software suite for neurophysiological data processing and visualization . J Neurosci Methods . 155 : 207 – 216 . Google Scholar CrossRef Search ADS PubMed Hoon M , Soykan T , Falkenburger B , Hammer M , Patrizi A , Schmidt KF , Sassoe-Pognetto M , Lowel S , Moser T , Taschenberger H , et al. . 2011 . Neuroligin-4 is localized to glycinergic postsynapses and regulates inhibition in the retina . Proc Natl Acad Sci USA . 108 : 3053 – 3058 . Google Scholar CrossRef Search ADS PubMed Jamain S , Quach H , Betancur C , Rastam M , Colineaux C , Gillberg IC , Soderstrom H , Giros B , Leboyer M , Gillberg C , et al. . 2003 . Mutations of the X-linked genes encoding neuroligins NLGN3 and NLGN4 are associated with autism . Nat Genet . 34 : 27 – 29 . Google Scholar CrossRef Search ADS PubMed Jamain S , Radyushkin K , Hammerschmidt K , Granon S , Boretius S , Varoqueaux F , Ramanantsoa N , Gallego J , Ronnenberg A , Winter D , et al. . 2008 . Reduced social interaction and ultrasonic communication in a mouse model of monogenic heritable autism . Proc Natl Acad Sci USA . 105 : 1710 – 1715 . Google Scholar CrossRef Search ADS PubMed Jokisch D , Jensen O . 2007 . Modulation of gamma and alpha activity during a working memory task engaging the dorsal or ventral stream . J Neurosci . 27 : 3244 – 3251 . Google Scholar CrossRef Search ADS PubMed Ju A , Hammerschmidt K , Tantra M , Krueger D , Brose N , Ehrenreich H . 2014 . Juvenile manifestation of ultrasound communication deficits in the neuroligin-4 null mutant mouse model of autism . Behav Brain Res . 270 : 159 – 164 . Google Scholar CrossRef Search ADS PubMed Krupa DJ , Brisben AJ , Nicolelis MAL . 2001 . A multi-channel whisker stimulator for producing spatiotemporally complex tactile stimuli . J Neurosci Methods . 104 : 199 – 208 . Google Scholar CrossRef Search ADS PubMed Lai MC , Lombardo MV , Baron-Cohen S . 2014 . Autism . Lancet . 383 : 896 – 910 . Google Scholar CrossRef Search ADS PubMed Laumonnier F , Bonnet-Brilhault F , Gomot M , Blanc R , David A , Moizard MP , Raynaud M , Ronce N , Lemonnier E , Calvas P , et al. . 2004 . X-linked mental retardation and autism are associated with a mutation in the NLGN4 gene, a member of the neuroligin family . Am J Hum Genet . 74 : 552 – 557 . Google Scholar CrossRef Search ADS PubMed Missler M , Zhang W , Rohlmann A , Kattenstroth G , Hammer RE , Gottmann K , Sudhof TC . 2003 . Alpha-neurexins couple Ca2+ channels to synaptic vesicle exocytosis . Nature . 423 : 939 – 948 . Google Scholar CrossRef Search ADS PubMed Mitchell JF , Sundberg KA , Reynolds JH . 2007 . Differential attention-dependent response modulation across cell classes in macaque visual area V4 . Neuron . 55 : 131 – 141 . Google Scholar CrossRef Search ADS PubMed Pfurtscheller G . 2001 . Functional brain imaging based on ERD/ERS . Vision Res . 41 : 1257 – 1260 . Google Scholar CrossRef Search ADS PubMed Pfurtscheller G . 2003 . Induced oscillations in the alpha band: functional meaning . Epilepsia . 44 : 2 – 8 . Google Scholar CrossRef Search ADS PubMed Ray WJ , Cole HW . 1985 . Eeg alpha-activity reflects attentional demands, and beta-activity reflects emotional and cognitive-processes . Science . 228 : 750 – 752 . Google Scholar CrossRef Search ADS PubMed Reyes-Puerta V , Kim S , Sun JJ , Imbrosci B , Kilb W , Luhmann HJ . 2015 . High stimulus-related information in barrel cortex inhibitory interneurons . PLoS Comput Biol . 11 : e1004121 . Google Scholar CrossRef Search ADS PubMed Reyes-Puerta V , Sun JJ , Kim S , Kilb W , Luhmann HJ . 2015 . Laminar and columnar structure of sensory-evoked multineuronal spike sequences in adult rat barrel cortex in vivo . Cereb Cortex . 25 : 2001 – 2021 . Google Scholar CrossRef Search ADS PubMed Rosenmund C , Stevens CF . 1996 . Definition of the readily releasable pool of vesicles at hippocampal synapses . Neuron . 16 : 1197 – 1207 . Google Scholar CrossRef Search ADS PubMed Rubenstein JLR , Merzenich MM . 2003 . Model of autism: increased ratio of excitation/inhibition in key neural systems . Genes Brain Behav . 2 : 255 – 267 . Google Scholar CrossRef Search ADS PubMed Schmitzer-Torbert N , Jackson J , Henze D , Harris K , Redish AD . 2005 . Quantitative measures of cluster quality for use in extracellular recordings . Neuroscience . 131 : 1 – 11 . Google Scholar CrossRef Search ADS PubMed Schubert D , Kotter R , Zilles K , Luhmann HJ , Staiger JF . 2003 . Cell type-specific circuits of cortical layer IV spiny neurons . J Neurosci . 23 : 2961 – 2970 . Google Scholar CrossRef Search ADS PubMed Song JY , Ichtchenko K , Sudhof TC , Brose N . 1999 . Neuroligin 1 is a postsynaptic cell-adhesion molecule of excitatory synapses . Proc Natl Acad Sci USA . 96 : 1100 – 1105 . Google Scholar CrossRef Search ADS PubMed Sonnleitner A , Simon M , Kincses WE , Buchner A , Schrauf M . 2012 . Alpha spindles as neurophysiological correlates indicating attentional shift in a simulated driving task . Int J Psychophysiol . 83 : 110 – 118 . Google Scholar CrossRef Search ADS PubMed Staiger JF , Flagmeyer I , Schubert D , Zilles K , Kotter R , Luhmann HJ . 2004 . Functional diversity of layer IV spiny neurons in rat somatosensory cortex: quantitative morphology of electrophysiologically characterized and biocytin labeled cells . Cereb Cortex . 14 : 690 – 701 . Google Scholar CrossRef Search ADS PubMed Sudhof TC . 2008 . Neuroligins and neurexins link synaptic function to cognitive disease . Nature . 455 : 903 – 911 . Google Scholar CrossRef Search ADS PubMed Tabuchi K , Blundell J , Etherton MR , Hammer RE , Liu X , Powell CM , Sudhof TC . 2007 . A neuroligin-3 mutation implicated in autism increases inhibitory synaptic transmission in mice . Science . 318 : 71 – 76 . Google Scholar CrossRef Search ADS PubMed Tang GM , Gudsnuk K , Kuo SH , Cotrina ML , Rosoklija G , Sosunov A , Sonders MS , Kanter E , Castagna C , Yamamoto A , et al. . 2014 . Loss of mTOR-dependent macroautophagy causes autistic-like synaptic pruning deficits . Neuron . 83 : 1131 – 1143 . Google Scholar CrossRef Search ADS PubMed Tsintsadze V , Minlebaev M , Suchkov D , Cunningham MO , Khazipov R . 2015 . Ontogeny of kainate-induced gamma oscillations in the rat CA3 hippocampus in vitro . Front Cell Neurosci . 9 : 195 . Google Scholar CrossRef Search ADS PubMed Tyzio R , Nardou R , Ferrari DC , Tsintsadze T , Shahrokhi A , Eftekhari S , Khalilov I , Tsintsadze V , Brouchoud C , Chazal G , et al. . 2014 . Oxytocin-mediated GABA inhibition during delivery attenuates autism pathogenesis in rodent offspring . Science . 343 : 675 – 679 . Google Scholar CrossRef Search ADS PubMed Unichenko P , Kirischuk S , Yang JW , Baumgart J , Roskoden T , Schneider P , Sommer A , Horta G , Radyushkin K , Nitsch R , et al. . 2016 . Plasticity-related gene 1 affects mouse barrel cortex function via strengthening of glutamatergic thalamocortical transmission . Cereb Cortex . 26 : 3260 – 3272 . Google Scholar CrossRef Search ADS PubMed Varoqueaux F , Aramuni G , Rawson RL , Mohrmann R , Missler M , Gottmann K , Zhang WQ , Sudhof TC , Brose N . 2006 . Neuroligins determine synapse maturation and function . Neuron . 51 : 741 – 754 . Google Scholar CrossRef Search ADS PubMed Varoqueaux F , Jamain S , Brose N . 2004 . Neuroligin 2 is exclusively localized to inhibitory synapses . Eur J Cell Biol . 83 : 449 – 456 . Google Scholar CrossRef Search ADS PubMed Vogt D , Cho KK , Lee AT , Sohal VS , Rubenstein JL . 2015 . The parvalbumin/somatostatin ratio is increased in Pten mutant mice and by human PTEN ASD alleles . Cell Rep . 11 : 944 – 956 . Google Scholar CrossRef Search ADS PubMed Yang JW , An S , Sun JJ , Reyes-Puerta V , Kindler J , Berger T , Kilb W , Luhmann HJ . 2013 . Thalamic network oscillations synchronize ontogenetic columns in the newborn rat barrel cortex . Cereb Cortex . 23 : 1299 – 1316 . Google Scholar CrossRef Search ADS PubMed Author notes Petr Unichenko, Jenq-Wei Yang and Sergei Kirischuk are contributed equally to this work © The Author 2017. Published by Oxford University Press. All rights reserved. For Permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)
TI - Autism Related Neuroligin-4 Knockout Impairs Intracortical Processing but not Sensory Inputs in Mouse Barrel Cortex
JF - Cerebral Cortex
DO - 10.1093/cercor/bhx165
DA - 2017-07-05
UR - https://www.deepdyve.com/lp/oxford-university-press/autism-related-neuroligin-4-knockout-impairs-intracortical-processing-fLUN0xRMNh
SP - 1
EP - 2886
VL - Advance Article
IS - 8
DP - DeepDyve
ER -