SV immunoisolation

All buffers and equipment were cooled to 0–1°C before beginning experiments, and all operations from homogenization until bead elution were conducted in a cold room. One to two C57B6/J mice, postnatal day (P) 14-20, were killed, and the brains, including cerebellum and brainstem, were rapidly removed. Each brain was homogenized in 4.2 ml of homogenization buffer [125 mm KCl, 20 mm potassium phosphate buffer, 5 mm EGTA, and protease inhibitors (cOmplete Mini EDTA-free, 1 tablet/10 ml), pH 7.3, at 0°C], using 10 strokes in a Teflon-glass Dounce homogenizer with rotation at 850–900 rpm using a digital overhead mixer (IKA). The homogenate was then centrifuged (20 min, 35,000 × g, 1°C). During centrifugation, 3 mg of Ab-Dynabeads (∼100 μl slurry) were transferred to a fresh 2 ml microcentrifuge tube, washed with 1 × 1 ml wash buffer (150 mm KCl, 10 mm potassium phosphate buffer, pH 7.2, at 0°C), and resuspended with 100 μl homogenization buffer. Following centrifugation, the supernatants (∼2.5 mg/ml protein) were pooled, and 1.9 ml was added to each tube containing Ab-Dynabeads. The tubes were incubated with rotation for 25 min, with the temperature maintained at 0°C by placing the tubes inside 50 ml conical tubes packed with ice. The beads were then collected using a magnetic stand, the supernatant was discarded, and the beads were washed four times by gently resuspending and triturating in 1 ml ice-cold wash buffer. In all cases, the final wash was used to transfer the beads to fresh 1.7 ml microcentrifuge tubes for elution. For Figure 1, the beads were split into two equal portions (1.5 mg each). For protein analysis, one portion was eluted by adding 30 μl 2% SDS and 25 mm Tris, pH 8.0, and heating to 50°C for 5 min. For polar amine analysis, the other portion was eluted by adding 30 μl of 50:50 MeOH:borate buffer and incubating on ice for 5 min. In Figure 1, immunoprecipitations using all four Ab-Dynabead conjugates were conducted in parallel for each experiment. For native elution experiments, the rho1D4 beads (3 mg; see Fig. 3) bearing SVs were washed as above and incubated first with 50 µl 200 μm 1D4 peptide (Cube Biotech) for 30 min on ice. This eluate was transferred to a fresh tube, and the beads were then eluted in 50 µl 2% SDS with heating to 50°C for 5 min. The elution buffers included 100 mm sodium borate, pH 8.5 (see Fig. 3) or 140 mM NaCl and 25 mM HEPES-NaOH, pH 7.4 (see Fig. 4). SVs were readily eluted with the 1D4 peptide in all buffers tested, for example, containing the following (in mm): 135 NaCl; 25 HEPES-NaOH, pH 7.4; 5 EGTA; 100 KCl; 25 HEPES-NaOH, pH 7.4; and 200 ammonium acetate (see Fig. 4). In later experiments (data not shown), higher concentrations of peptide (∼1 mm) appeared more effective for eluting SVs, and so we encourage investigators using this method to titrate this peptide elution step during initial studies.

Immunopurification of synaptic vesicles. A, Scheme for vesicle immunoprecipitation and analysis. Each mouse brain provided sufficient material for analysis of protein and neurotransmitter using two different mAbs. B, Staining of proteins separated by SDS-PAGE demonstrates broad similarity among anti-SV2, anti-syt1, and rho1D4 immunoprecipitates, with minimal protein binding by control beads bearing pooled bovine IgG. Note the dominant band at 38 kDa, corresponding to synaptophysin. C, Immunoblot analysis of precipitated material. Each antibody yields strong enrichment of SV proteins, but only the expected weak enrichment of the plasma membrane t-SNARE syntaxin-1 and no detectable contamination from the mitochondrial protein VDAC. D, E, For each experiment, the area of the HPLC fluorescence peak corresponding to NBD-derivatized glutamate (D) or GABA (E) was plotted against the normalized intensity of the synaptophysin band on immunoblot (n = 4 biological replicates using two separately prepared batches of Ab-Dynabeads for each mAb; B–E). Example raw chromatograms used for GABA and glutamate measurements are shown in Extended Data Figure 1-1. “LU*s”, luminance units * seconds.

rho1D4-IP enables peptide elution of native SVs. A, Purification scheme for peptide elution. The amino acid sequence of the 1D4 peptide sequence is shown. B, SDS-PAGE with fluorescent stain of protein eluted from rho1D4 beads using the 1D4 peptide followed by 2% SDS. SV proteins were readily eluted from the beads with the 1D4 peptide. C, Dynamic light scattering measurements (n = 3 biological replicates) of eluted material indicates a single population of particles ∼40–50 nm in diameter. This population represented >99% of particles detected in each sample. D, Negative-stain TEM of 1D4-eluted material demonstrates vesicles of the appropriate size, decorated with expected structures, with minimal contamination by nonvesicular structures. E, Immunoblot of SV proteins in input fraction along with peptide and SDS eluates from rho1D4 beads, with approximate amount of total protein loaded per lane. F, Immunoblot of peptide and SDS eluates probed with anti-mouse secondary mAb demonstrating the relative absence of eluted mAb in peptide eluates.

Cryo-electron tomography of synaptic vesicles obtained by rho1D4-IP. A, Varied morphology of SVs as seen by cryo-electron microscopy and their corresponding three-dimensional reconstructions. Scale bar, 25 nm. B, Frequency distribution of SV diameters measured along the longest axis (n = 421). C, Frequency distribution of SV ellipticity defined as the ratio between the longest and orthogonal diameters of each vesicle.