Fig 2: Aging antibody-labeled synaptic vesicle proteins cease to participate in exocytosis ATo determine whether the antibody-tagged synaptic vesicle proteins that are present in synapses at different time intervals after labeling are exocytosis-competent, we labeled neurons with CypHer5E-conjugated antibodies for Synaptotagmin 1 or VGAT for 1 h exactly as described in Fig 1A. At different time intervals after labeling, individual cultures were removed from the incubator, mounted in a live-stimulation chamber, and subjected to a stimulation paradigm designed to release the entire recycling pool (20 Hz, 30 s; Wilhelm et al, 2010). Exocytosis is determined by monitoring the fluorescence of the pH-sensitive fluorophore CypHer5E, which is fluorescent at low pH within vesicles, but is quenched by exposure to the neutral extracellular pH. To determine the proportion of the labeled vesicles that is still able to exocytose, we measured the fluorescence signal corresponding to all labeled vesicles by fixing and permeabilizing the neurons and then exposing them to a pH of 5.5, obtained with TES-buffered solutions.B, CTypical images of fluorescence either before or after stimulation, for Synaptotagmin 1 (B) or VGAT (C). Numerous individual images were taken of different fields for every condition and were averaged for the statistics, as imaging the same field under all conditions proved unfeasible, due to the high bleaching tendency of CypHer5E. Scale bar: 50 µm. All images in (B) were taken using an Olympus epifluorescence microscope, and all images in (C) were taken using a Nikon Ti-E epifluorescence microscope.D, EThe following numbers of experiments were quantified: for Synaptotagmin 1, n = 4, 3, 4, 4, 2, and 2 independent experiments per respective time point, with at least 12 neurons sampled per experiment; and for VGAT, n = 4, 4, 2, 4, 4, and 3 independent experiments per respective time point, with at least eight neurons sampled per experiment. The VGAT signal follows the same approximate dynamic as Synaptotagmin 1, but is faster. VGAT is present in only ˜ 5–10% of the neurons in our cultures, and therefore, this difference may reflect a cell type-specific effect. All data represent the mean ± SEM. Source data are available online for this figure.

Fig 3: Clathrin-independent endocytic retrieval of select synaptic vesicle (SV) cargos by the clathrin adaptor adaptor protein complex 2 (AP-2) at physiological temperature.(A–D) Poststimulus retrieval of Syt1 and SV2A in the absence of AP-2 persists unaffected in response to 200 APs applied at 40 Hz. Average normalized traces of WT and AP-2µ KO derived neurons cotransfected with Syt1-SEP (A) or SV2A-SEP (C) and mRFP or rescued by reexpression of untagged AP-2µ subunit together with soluble mRFP (AP-2µ) to identify transfected neurons in response to 200 APs applied at 40 Hz. Quantification of the endocytic decay time constant (t) of neurons expressing Syt1-SEP (B) (tWT = 17.17 ± 1.35 s, tAP-2µ KO = 21.53 ± 3.43 s, tAP-2µ KO+AP-2µ = 21.51 ± 3.91 s) or SV2A-SEP (D) (tWT = 18.33 ± 0.99 s, tAP-2µ KO = 20.45 ± 1.56 s, tAP-2µ KO+AP-2µ = 18.04 ± 1.16 s). Data shown represent the mean ± standard error of the mean (SEM): Syt1 (nWT = 7 images, nAP-2µ KO = 7 images, nAP-2µ KO+AP-2µ = 4 images; p(WT vs AP-2µ KO) = 0.4994; p(AP-2µ KO vs AP-2µ KO+AP2-µ) > 0.9999); SV2A (nWT = 44 images, nAP-2µ KO = 23 images, nAP-2µ KO+AP-2µ = 37 images; p(WT vs AP-2µ KO) = 0.4665; p(AP-2µ KO vs AP-2µ KO+AP-2µ) = 0.3943). One-way analysis of variance (ANOVA) with Tukey’s post-test. (E–L) Loss of AP-2 significantly delay the endocytic retrieval of other major SV proteins. Average normalized traces of WT and AP-2µ KO neurons cotransfected with VGLUT1-SEP (E), VGAT-SEP (G), Syp-SEP (I), Syb2-SEP (K), and mRFP or AP-2µ to rescue AP-2µ expression stimulated with 200 APs at 40 Hz. Endocytic decay time constants (t) were calculated from WT, AP-2µ KO neurons, and AP-2µ KO neurons rescued by reexpression of AP-2µ expressing VGLUT1-SEP (F) (tWT = 13.52 ± 0.90 s, tAP-2µ KO = 30.33 ± 2.01 s, tAP-2µ KO+AP-2µ = 16.38 ± 1.00 s), VGAT-SEP (H) (tWT = 16.76 ± 0.87 s, tAP-2µ KO = 40.56 ± 9.21 s, tAP-2µ KO+AP-2µ = 21.77 ± 2.15 s), Syp-SEP (J) (tWT = 13.05 ± 0.74 s, tAP-2µ KO = 23.76 ± 1.72 s, tAP-2µ KO+AP-2µ = 14.89 ± 1.04 s), VGAT-SEP (D) (tWT = 18.33 ± 0.99 s, tAP-2µ KO = 20.45 ± 1.56 s, tAP-2µ KO+AP-2µ = 18.04 ± 1.16 s), and Syb2-SEP (L) (tWT = 20.27 ± 1.48 s, tAP-2µ KO = 46.34 ± 9.43 s, tAP-2µ KO+AP-2µ = 22.46 ± 1.21 s). Data shown represent the mean ± SEM: VGLUT1 (nWT = 37 images, nAP-2µ KO = 24 images, nAP-2µ KO+AP-2µ = 23 images; ****p(WT vs AP-2µ KO) < 0.0001; ****p(AP-2µ KO vs AP-2µ KO+AP-2µ) < 0.0001); VGAT (nWT = 34 images, nAP-2µ KO = 11 images, nAP-2µ KO+AP-2µ = 32 images; ****p(WT vs AP-2µ KO) < 0.0001; ***p(AP-2µ KO vs AP-2µ KO+AP-2µ) = 0.0008); Syp (nWT = 33 images, nAP-2µ KO = 29 images, nAP-2µ KO+AP-2µ = 37 images; ****p(WT vs AP-2µ KO) < 0.0001; ****p(AP-2µ KO vs AP-2µ KO+AP-2µ) < 0.0001); Syb2 (nWT = 15 images, nAP-2µ KO = 20 images, nAP-2µ KO+AP-2µ = 26 images; **p(WT vs AP-2µ KO) = 0.0083; **p(AP-2µ KO vs AP-2µ KO+AP-2µ) = 0.0052). One-way ANOVA with Tukey’s post-test. (M–P) Delayed poststimulus retrieval of major SV proteins in the absence of AP-2 is not caused by defects in re-acidification of endocytosed vesicles. Representative normalized traces of WT and AP-2µ KO neurons expressing VGLUT1-SEP (M) or SV2A-SEP (O) stimulated with 200 APs applied at 40 Hz and subjected to low pH imaging buffer before and after train stimulation. Fluorescence quenching by application of acidic buffer poststimulus (?F2) vs prestimulus (?F1) of VGLUT1-SEP (N) (nWT = 10 images, nAP-2µ KO = 6 images) or SV2A-SEP (P) (nWT = 8 images, nAP-2µ KO = 13 images) is taken as a measure to probe the SEP surface pool in WT and AP-2µ KO hippocampal neurons. Values for WT were set to 1. Data shown represent the mean ± SEM. VGLUT1: WT = 1.0 ± 0.1; AP-2µ KO = 1.8 ± 0.2. p = 0.0009. SV2: WT = 1.0 ± 0.1; AP-2µ KO = 1.1 ± 0.1. p = 0.4267. Two-sided unpaired t-test. Raw data can be found in Figure 3—source data 1. Figure 3—source data 1.Source data for Figure 3A-P.

Fig 4: Adaptor protein complex 2 (AP-2) depletion results in surface stranding of endogenous vesicular neurotransmitter transporters but not of Synaptotagmin 1 and SV2A.(A) Schematic diagram of the workflow for cell surface protein enrichment. (B–I) AP-2 participates in the surface retrieval of the endogenous synaptic vesicle (SV) proteins such as VGLUT1 and VGAT but not of SV2 and Syt1. Cell surface proteins from WT and AP-2µ KO cerebellar granule cells were biotinylated and affinity-purified using streptavidin beads. Total (Input) and biotinylated proteins (Surface) were analyzed by western blot using antibodies against Syt1 (B), SV2 (D), VGLUT1 (F), and VGAT (H). N-cadherin and GAPDH were used as control of cell surface membrane and cytosol fraction, respectively. The fold surface enrichment of select proteins (C, E, G, I) in the absence of AP-2 was quantified. Values for WT neurons were set to 1. Data represent the mean ± standard error of the mean (SEM). nSyt1 = 8, nSV2 = 5, nVGLUT1 = 8, nVGAT = 6 independent experiments. pSyt1 = 0.4456, pSV2 = 0.6736, **pVGLUT1 = 0.0049, *pVGAT = 0. 0279; two-sided one-sample t-test. Raw data can be found in Figure 6—source data 1, and Figure 6—source data 2. Figure 6—source data 1.Source data for Figure 6C, E, G, I. Figure 6—source data 2.Raw uncropped immunoblot images for Figure 6B, D, F, H.

Fig 5: Adaptor protein complex 2 (AP-2)-binding deficient mutants of vesicular neurotransmitter transporters phenocopy loss of AP-2.(A–C) Association of the cytoplasmic domain of VGLUT1 with the clathrin adaptor complex AP-2 is abolished upon mutational inactivation of the putative AP-2-binding dileucine motif, that is F510A/V511A (FV/AA). (A) Acidic cluster dileucine-like motifs identified in the C-terminal cytoplasmic tail of mouse VGLUT1 and in the N-terminal cytoplasmic tail of mouse VGAT. Numbers indicate amino acid numbers of the respective proteins. Green and magenta indicate two acidic amino acids and two hydrophobic amino acids conserved within the motifs, respectively. (B) Immunoblot analysis of material affinity-purified via GST-VGLUT1 C-terminus-WT, GST-VGLUT1 C-terminus-FV/AA or GST alone, and brain lysate using specific antibodies against AP-2, AP-1, and AP-3 shows that dileucine-like motif found in the C-terminus of VGLUT1 binds preferentially to AP-2. (C) Quantified data exhibit VGLUT1 C-terminus-FV/AA variant to significantly disrupt interaction with AP-2. Data represent the mean ± standard error of the mean (SEM) from n = 3 independent experiments. **p = 0.0027, two-sided one-sample t-test. (D–G) Mutant variants of VGLUT1 or VGAT defective in AP-2-binding display significantly slower endocytosis kinetics in response to stimulation in a similar manner to be observed in the absence of AP-2. (D) Average normalized traces of neurons from WT and AP-2µ KO mice transfected with either the WT or the mutant variant (FV/AA) of VGLUT1-SEP in response of a stimulus train of 200 APs applied at 40 Hz at physiological temperature. (E) Quantification of the endocytic decay time constant (t) of VGLUT1-SEP-expressing neurons (tWT-WT = 14.82 ± 1.98 s, tWT-FV/AA = 29.82 ± 1.82 s, tAP-2µ KO-WT = 26.15 ± 1.34 s, tAP-2µ KO-FV/AA = 34.80 ± 3.18 s). Data represent the mean ± SEM of nWT-WT = 16 images, nWT-FV/AA = 18 images, nAP-2µ KO-WT = 12 images, and nAP-2µ KO-FV/AA = 11 images. ****pWT-WT vs WT-FV/AA < 0.0001, **pWT-WT vs AP-2µ KO-WT = 0.0023, ****pWT-WT vs AP-2µ KO-FV/AA < 0.0001, pAP-2µ KO-WT vs AP-2µ KO-FV/AA = 0.0530, one-way analysis of variance (ANOVA) with Tukey’s post-test. (F) Average normalized traces of WT and AP-2µ KO neurons transfected with either the WT or the mutant variant (FA/AA) of VGAT-SEP in response of 200 APs applied at 40 Hz at physiological temperature. (G) Quantification of the endocytic decay time constant (t) of VGAT-SEP-expressing neurons (tWT-WT = 18.86 ± 1.79 s, tWT-FA/AA = 49.56 ± 5.95 s, tAP-2µ KO-WT = 47.43 ± 4.19 s, tAP-2µ KO-FA/AA = 57.60 ± 6.92 s). Data represent the mean ± SEM of nWT-WT = 30 images, nWT-FA/AA = 19 images, nAP-2µ KO-WT = 42 images, and nAP-2µ KO-FA/AA = 16 images. ***pWT-WT vs WT-FA/AA = 0.0001, ****pWT-WT vs AP-2µ KO-WT < 0.0001, ****pWT-WT vs AP-2µ KO-FA/AA < 0.0001, pAP-2µ KO-WT vs AP-2µ KO-FA/AA = 0.4561, one-way ANOVA with Tukey’s post-test. (H) Illustrated model proposing a clathrin-independent role for dedicated endocytic adaptors such as AP-2 which recognize select exocytosed synaptic vesicle (SV) proteins (e.g., VGLUT1 and VGAT) present on the neuronal surface to facilitate their clathrin-independent endocytic internalization while clathrin operates downstream facilitating the reformation of functional SVs by budding from internal endosome-like vacuoles (ELVs) in a process that also depends on AP-2 and other clathrin-associated endocytic proteins. Raw data can be found in Figure 8—source data 1 and Figure 8—source data 2. Figure 8—source data 1.Source data for Figure 8C-G. Figure 8—source data 2.Raw uncropped immunoblot images for Figure 8B.