Supporting Materials and Methods

Cell Cultures, Transfection, and Immunolabeling. Primary hippocampal cultures were prepared from brains of rats from embryonic days 18–21 and maintained in serum-based medium with B27 supplement (Life Technologies). PC12 cells were maintained in DMEM (GIBCO) with 10% horse serum and 5% FBS. PC12 cells were cultured in six-well plates and transfected by using Lipofectamine 2000 (Invitrogen) according to standard protocol (4 m g of DNA in 250 m l of Opti-MEM and 10 m l of Lipofectamine in 250 m l of Opti-MEM per well). To determine the two-plasmid cotransfection efficiency, we transfected PC12 cells with GFP and DsRed and found that in all transfected PC12 cells examined, both GFP and DsRed were expressed, indicating that most of the DsRed-expressing PC12 cells in our coculture system also expressed b -neurexin. Neurons were also transfected with Lipofectamine 2000 according to the manufacturer’s recommendation except that 0.8 m g of DNA in 12.5 m l of Opti-MEM and 0.5 m l of Lipofectamine in 12.5 m l of Opti-MEM were mixed and added to each cover slip in 24-well plates (adapted from the protocol used previously for neuronal cultures; see ref. 1). PC12 (or HEK) cells and neurons were initially cultured separately. PC12 (or HEK) cells were transfected at 80% confluency, and neurons from days 9–11 in vitro were transfected on the same day. One day after transfection, PC12 (or HEK) cells were resuspended and plated into transfected neuronal cultures. Confocal imaging was done 2 days after coculturing. Immunocytochemistry was performed on cultures fixed with 2% paraformaldehyde (15 min at room temperature) and washed with PBS containing 0.3% Triton X-100 before incubation with primary and secondary antibodies.

Image Acquisition and Quantification. For fluorescent image analysis, cells were chosen randomly from three or more independent batches of cultures with four or more cover slips per batch for each construct. Fluorescent images were acquired at room temperature with a Zeiss 510 META NLO Axioplan2 confocal microscope, by using a Zeiss Achroplan ´ 63/0.95, working distance 2.0 objective with sequential acquisition setting at 1,024 ´ 1,024-pixel resolutions. Laser power and photomultipliers were set such that no detectable bleed-through occurred between different channels. Digital images of the cells were captured with a C4742-95 charge-coupled device camera (Hamamatsu Photonics, Hamamatsu City, Japan) and LSM 5 imaging software (Zeiss). Eight to 10 sections were taken from top to bottom of the specimen, and brightest point projections were made. Images for the same experiments were taken by using identical settings for laser power, photomultiplier gain, and offset. These settings were chosen such that the pixel intensities for the brightest samples were just below saturation, with the exception that when contours of the cell (DsRed signal from PC12 cells) or contours of the neuronal processes (GFP signal from neurons in Fig. 5) has to be clearly determined, signals from certain areas (center of the PC12 cell body or soma of the neurons) were saturated to obtain clear signals from the periphery of the cell body or neuronal dendrites. For quantification of FM staining, postsynaptic density (PSD)-95, a -amino-3-hydroxyl-5-methyl-4-isoxazolepropionate receptor (AMPAR), and NMDA receptor (NMDAR) puncta, images collected by using identical confocal settings were thresholded by intensity (using image analysis software such as METAMORPH) to exclude the diffuse/intracellular pool, and puncta were quantified by counting the number of thresholded areas of sizes between 0.25 and 4 m m2. Results were averaged for at least 15 images for each condition, and the mean and standard error were calculated. Puncta distance was calculated by averaging the distance between the centers of each puncta and its nearest neighbor on the same dendrite. For quantification of PSD-95 immunostaining fluorescence, because the signal surrounding the b -neurexin-transfected HEK cells was very strong, and it was not always possible to discern individual puncta, we calculated the total number of pixels above the intensity threshold rather than the number of puncta. Image quantification was performed by experienced investigators who were "blind" to the experimental conditions.

FM 4-64 Labeling. FM 4-64 dye (Molecular Probes) staining was performed by using standard procedures. Dye loading was achieved by incubating cells in external solution containing 50 mM KCl and 10 m M FM 4-64 for 90 sec. After extensive washing, dye unloading was performed with external saline containing 90 mM KCl. Images of presynaptic releasing sites were obtained by subtracting the postunloading images from the preunloading ones.

Electrophysiology. Whole-cell patch-clamp recordings were made at room temperature with 3–7 MW patch pipettes filled with an internal solution containing 140 mM CsCl, 2 mM MgCl2, 5 mM EGTA, 10 mM Hepes, 0.3 mM Na3-GTP, and 4 mM Na2-ATP (pH 7.35). Cultures were continuously superfused with external solution (119 mM NaCl/26 mM NaHCO3/2.5 mM KCl/10 mM glucose/2.5 mM CaCl2/1.3 mM MgSO4/1 mM NaH2PO4). AMPA miniature excitatory postsynaptic currents were recorded in the presence of tetrodotoxin (1 m M), picrotoxin (100 m M), and 2-amino-5-phosphonovaleric acid (100 mM); NMDA spontaneous excitatory postsynaptic currents were recorded in the presence of 6-cyano-7-nitroquinoxaline-2,3-dione (10 mM) and picrotoxin (100 m M); and glyceraldehyde-3-phosphate dehydrogenase miniature inhibitory postsynaptic currents were recorded with 6-cyano-7-nitroquinoxaline-2,3-dione (10 mM) and 2-amino-5-phosphonovaleric acid (100 mM). Cells were held at –60 mV. For most experiments, neighboring transfected and untransfected cells were selected for recording in an attempt to minimize any difference that may occur across cover slips and plating, such as the degree of connectivity.

All results are presented as mean ± SEM.

1. Blanpied, T. A., Scott, D. B. & Ehlers, M. D. (2002) Neuron 36, 435–449.