Generation of mice with a disruption of the Shank3 gene
Animal procedures were approved by the Mount Sinai School of Medicine and the National Institute of Mental Health Animal Care and Use committees. We made use of gene targeting in Bruce4 C57BL/6 embryonic stem (ES) cells  to generate a mouse line that has loxP sites inserted before exon 4 and after exon 9 (encoding the ankyrin repeats), with the selection cassette (flanked by FRT sites) excised by FLP recombinase. This floxed strategy was chosen to allow us the option of doing conditional (region-specific) knockouts if needed. C57BL/6 was used as the chosen embryonic stem cell line because of the more robust social and cognitive abilities in this line as compared to many of the 129-derived ES lines. For all studies reported here, the floxed allele was first excised by crossing with a CMV-Cre transgenic line (again on a C57BL/6 background) that has ubiquitous Cre expression and a line maintained with a deletion of exons 4 through 9. This Shank3-deficient line was carried forward by crossing heterozygotes with the C57BL/6 strain to maintain a pure C57 background suitable for electrophysiology and behavioral analyses.
Quantitative polymerase chain reaction
RNA was extracted from brain cortex using the RNeasy Mini Kit (Qiagen, Valencia, CA, USA) according to the manufacturer's instructions. cDNA was synthesized with the High Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Carlsbad, CA, USA). The universal probe library (UPL) system (Roche, Indianapolis, IN, USA) was used to perform quantitative polymerase chain reaction (qPCR). Primers located in exons 6 and 7 of Shank3 (NM_021423) were designed using ProbeFinder Software (Roche). Three reference genes (Actb, Gapd and Rpl13a) were used for normalization, and relative expression levels were calculated using qBase software , now available from Biogazelle (Ghent, Belgium). Primer sequences and UPL probe numbers were Shank3, forward tggttggcaagagatccat, reverse ttggccccatagaacaaaag, #1; Actb, forward ggatgcagaaggagattactgc, reverse ccaccgatccacacagagta, #63; Gapd, Fw gccaaaagggtcatcatctc, reverse cacacccatcacaaacatgg, #29; Rpl 13a, forward tccctgctgctctcaagg, reverse gccccaggtaagcaaactt, #41. Unpaired t-tests were used for group comparisons.
PSD fractions were prepared as follows. Hemibrains of wild-type, heterozygous and homozygous Shank3 mice were homogenized in 2-[4-(2-hydroxyethyl)piperazin-1-yl]ethanesulfonic acid (HEPES)-A containing 4 mM HEPES, pH 7.4, 0.32 M sucrose, Protease Inhibitor Cocktail and PhoSTOP Phosphatase Inhibitor Cocktail (both from Roche). Nuclear fractions were precipitated by centrifuging twice at 700 g for 15 min, and the resulting supernatants were further centrifuged at 21,000 g for 15 min. The precipitates were resuspended in HEPES-B containing 4 mM HEPES, pH 7.4, Protease Inhibitor Cocktail and PhoSTOP Phosphatase Inhibitor Cocktail, homogenized and rotated at 4°C for 1 hour. The lysates were centrifuged at 32,000 g for 20 min and washed twice with HEPES-C containing 50 mM HEPES, pH 7.4, 0.5% Triton X-100, Protease Inhibitor Cocktail and PhoSTOP Phosphatase Inhibitor Cocktail. Finally, postsynaptic density fractions were resuspended in HEPES-C containing 1.8% sodium dodecyl sulfate (SDS) and 2.5 M urea. Fifty-two μg of each PSD fraction were loaded to 4-12% SDS-polyacrylamide gel electrophoresis (PAGE gel (Invitrogen, Carlsbad, CA, USA), transferred to polyvinylidene fluoride membrane and immunoblotted with either the N69/46 anti-Shank3 antibody directed against an epitope downstream of the PDZ domain (UC Davis/NIH NeuroMab Facility, Davis, CA) or the anti-ProSAP2 anti-Shank3 antibody directed against the last 100 amino acids of Shank3 (Millipore, Billerica, MA, USA). For βIII tubulin, the membrane was stripped and immunoblotted with an anti-βIII tubulin antibody (Abcam, Cambridge, MA, USA).
Hippocampal slice electrophysiology
Whole cell recording, two-photon time-lapse imaging and analysis
Methods of recording, imaging and analysis were carried out according to our previously published protocols [34, 35]. All experiments were conducted on CA1 pyramidal cells at 32°C in acute slices taken from Shank3 heterozygous mice and wild-type littermates. Spines were visualized using calcein contained in the patch pipette, making use of a two-photon laser scanning system modified from Olympus Fluoview FV 300 driven by a Chameleon two-photon laser (Coherent, Santa Clara, CA, USA). Baseline synaptic responses were evoked using a glass pipette positioned ~20 μm away from the imaged spines and recorded at the soma. Long-term potentiation (LTP) was induced with a θ-burst pairing (TBP) protocol in which two trains of θ-burst stimuli (each train, separated by 20 s, consisted of five bursts at 5 Hz, and each burst contained five pulses at 100 Hz) were paired with brief, small postsynaptic depolarization. Volume analysis of individual spines was performed as detailed previously . Briefly, the integrated fluorescence intensity inside a spine head was measured for individual spines at different time points and normalized to the fluorescence intensity of the dendrites from the same image stack to correct for potential changes in excitation . Spine volume in the θ-burst stimulation (TBS) experiments was also calculated using the Rayburst algorithm in NeuronStudio software (available from the Computational Neurobiology and Imaging Center, Mount Sinai School of Medicine, New York, NY, USA) following deconvolution of the data [37–39], and we obtained similar results using either approach.
Hippocampal slices (350 μm thick) were prepared from 4- to 6-week-old heterozygous mice and their wild-type littermate controls. Slices were perfused with Ringer's solution containing (in mM): NaCl, 125.0; KCl, 2.5; MgSO4, 1.3; NaH2PO4, 1.0; NaHCO3, 26.2; CaCl2, 2.5; and glucose, 11.0. The Ringer's solution was bubbled with 95% O2 and 5% CO2 at 32°C during extracellular recordings (electrode solution: 3 M NaCl). Slices were maintained for 1 hour prior to establishment of a baseline of field excitatory postsynaptic potentials (fEPSPs) recorded from stratum radiatum in area CA1, evoked by stimulation of the Schaffer collateral-commissural afferents (100-μs pulses every 30 s) with bipolar tungsten electrodes placed into area CA3 . Test stimulus intensity was adjusted to obtain fEPSPs with amplitudes that were one-half of the maximal response. The EPSP initial slope (mV/ms) was determined from the average waveform of four consecutive responses. Input-output (I/O) curves were generated by plotting the fEPSP slope versus fiber volley amplitude in low-Mg2+ (0.1 mM) solution. AMPA receptor-mediated and NMDA receptor-mediated I/O relationships were measured in the presence of 2-amino-5-phosphonopentanoic acid (APV; 50 μM) and 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX; 100 μM), respectively (Sigma, St. Louis, MO, USA).
Paired-pulse responses were measured with an interstimulus interval (ISI) of 50 ms and are expressed as the ratio of the average responses to the second stimulation pulse (FP2) to the first stimulation pulse (FP1). LTP was induced by either a high-frequency stimulus (four trains of 100-Hz, 1-s stimulations separated by 5 min) or TBS (15 bursts of four pulses at 100 Hz separated by 200 ms). To induce long-term depression (LTD), Schaffer collaterals were stimulated by low-frequency stimulation (LFS; 900 pulses at 1 Hz, 15 min) or by a paired-pulse low-frequency stimulation (PP-LFS; 1 Hz for 20 min, 50-ms interstimulus interval ) to induce mGlu receptor-dependent LTD. Data were expressed as means ± SD, and statistical analyses were performed using analysis of variance (ANOVA) or Student's t-test, with significance set at an α level of 0.05.
Measurement of GluR1-immunoreactive puncta
Three-month-old animals were anesthetized with 250 μl of 15% chloral hydrate and perfused transcardially with 1% paraformaldehyde for 1 min followed by 4% paraformaldehyde for a total of 13 min. The brains were then removed, hemisected, cut in 50-μm-thick sections using a Leica VT1000S Vibratome (Vibratome, Bannockburn, IL, USA) and subsequently stored in phosphate-buffered saline (PBS) until use. Sections were incubated at 37°C for 5 min, followed by incubation in acidified pepsin (1 ml in a 0.2 N HCl solution) for 6.5 min. The tissue was then washed at room temperature in PBS-B (3 × 20 min) and incubated in a 0.3% Triton X-100, 0.5% bovine serum albumin (BSA), 5% normal goat serum for 1 h on an orbital shaker. The blocking step was followed by overnight incubation in the primary antibody (rabbit polyclonal antiglutamate receptor 1 AB1504; Millipore, Billerica, MA, USA), which was made in blocking solution at the appropriate dilution (1 μg/ml). The tissue sections were then washed in PBS-B (5 × 5 min) and incubated in secondary antibody (goat antirabbit Alexa Fluor 488, Invitrogen) in a 2% BSA and 0.3% Triton PBS-B solution at the appropriate dilution (1:400) for 1 h at room temperature on an orbital shaker. Finally, the tissue was washed in PBS-B (3 × 5 min), stained with 4",6"-diamino-2-phenylindole-2HCl (DAPI) and mounted on charged Aqua ColorFrost slides using Vectashield mounting medium (Vector Laboratories, Burlingame, CA, USA).
To quantify puncta, we used a systematic random sampling approach whereby a 1:6 series of sections were stained, the stratum radiatum of the CA1 was contoured using SteroInvestigator (MBF Bioscience, Williston, VT, USA) and the sampling sites were determined using an optical fractionator with the size of the grid set at 18 μm2 (the dimension of the confocal image stacks to be later sampled with a ×100 lens objective), at a digital zoom of 5 on a Zeiss LSM510 META confocal microscope (Zeiss, Oberkochen, Germany). The trace of the contoured area with an optical fractionator sampling grid placed on it was used as a guide to obtain confocal image stacks of the above-mentioned dimensions that were 100 μm apart from each other (i.e., the size of the probe).
Confocal imaging and puncta quantification
The fluorescent puncta were visualized under a ×100 oil immersion objective (1.4 numerical aperture) in a series of Z-stacks using an Argon/2 laser (488 nm wavelength) at 50% output (tube current of 6.4 A and maximum power of 30 mW), with a collection band pass spectrum of 505-550 nm (with the following laser and microscope settings: image frame size of 512 × 512, 1 Airy unit, refractive index correction of 0.9144 and Z stack interval of 0.1 μm (x,y pixel size = 0.05 μm)) and their intensity, number and size quantified. These settings were optimized during pilot studies and held constant throughout the study.
The resultant stacks were then deconvolved using AutoDeblur 1.4.1 (Media Cybernetics, Bethesda, MD, USA), using an adaptive point-spread function (PSF) deconvolution method with a theoretical PSF, and then analyzed with custom Vamp2D software  that reliably calculates the size of individual puncta on the basis of three-dimensional estimates and circumvents the problem of object superimposition found with more traditional methods that collapse the stacks into two-dimensional projections. The relative density of puncta was then calculated per cubic micrometer, and the differences between groups were assessed using the nonparametric Mann-Whitney U test.
Shank3 wild-type and heterozygote breeding pairs were imported from Mount Sinai School of Medicine to the National Institute of Mental Health. Mice were maintained by breeding C57BL/6 wild-type mice with Shank3 heterozygotes and housed in a conventional temperature- and humidity-controlled vivarium. Littermates were housed by sex in mixed genotype groups of two to four per cage on a 12:12-h circadian cycle with lights on at 0600. Behavioral experiments were conducted between 1000 and 1600 in dedicated testing rooms.
Developmental milestones were tested across postnatal days 2-14, including measures of body weight, body length, tail length, pinnea detachment, eye opening, incisor eruption, fur development, righting reflex, negative geotaxis, cliff avoidance, grasping reflex, auditory startle, bar holding, level screen and vertical screen as previously described [43, 44]. In addition, the mice were evaluated in a standard, automated three-chambered social approach task as previously described .
Male-female social interactions were evaluated in a 5-min test session as previously described [43, 45], with the exception that subject males were group-housed and individually tested in clean cages with clean litter. Each of the 12 wild-type and 14 heterozygous male subject mice, ages 2.5-4 months, was paired with a different unfamiliar estrus C57BL/6J female. A digital closed-circuit television camera (Panasonic, Secaucus, NJ, USA) was positioned horizontally 30 cm from the cage. An ultrasonic microphone (Avisoft UltraSoundGate condenser microphone capsule CM15; Avisoft Bioacoustics, Berlin, Germany) was mounted 20 cm above the cage. Sampling frequency for the microphone was 250 kHz, and the resolution was 16 bits. The entire apparatus was contained in a sound-attenuating environmental chamber (ENV-018V; Med Associates, St. Albans, VT, USA) illuminated by a single 25-Watt red light. Videos from the male subjects were subsequently scored by an investigator uninformed of the subject's genotype on measures of nose-to-nose sniffing, nose-to-anogenital sniffing and sniffing of other body regions, using Noldus Observer software (Noldus Information Technology, Leesburg, VA, USA) as previously described. Ultrasonic vocalizations were played back and spectrograms were displayed using Avisoft software [43, 45]. Ultrasonic vocalizations were identified manually by two highly trained investigators blinded to genotype information, and summary statistics were calculated using the Avisoft package. Interrater reliability was 95%. Data were analyzed using an unpaired Student's t-test.
Olfactory habituation/dishabituation testing was conducted in male and female Shank3 wild-type and heterozygous mice ages 2.5-4 months using methods previously described [44, 46, 47]. Nonsocial and social odors were presented on a series of cotton swabs inserted into the home cage sequentially, each for 2 min, in the following order: water, water, water (distilled water); almond, almond, almond (1:100 dilution almond extract); banana, banana, banana (1:100 dilution artificial banana flavoring); social 1, social 1, social 1 (swiped from the bottom of a cage housing unfamiliar sex-matched B6 mice); and social 2, social 2, social 2 (swiped from the bottom of a second cage housing a different group of unfamiliar sex-matched 129/SvImJ mice). One-way repeated measures ANOVA was performed within each genotype for each set of habituation events and each dishabituation event, followed by a Tukey post hoc test.