Condensates of synaptic vesicles and synapsin-1 mediate actin sequestering and polymerization

Abstract

Neuronal communication relies on precisely maintained synaptic vesicle (SV) clusters, which assemble via liquid-liquid phase separation. This process requires synapsins, the major synaptic phosphoproteins, which are known to bind actin. Reorganization of SVs, synapsins, and actin is a hallmark of synaptic activity, but the molecular details of the interactions between these components remain unclear. Here, we combine in vitro reconstitution with expansion microscopy, super-resolution imaging, and cryo-electron tomography to dissect the roles of SV-synapsin-1 condensates in the organization of the presynaptic actin cytoskeleton. Our results indicate that condensation of synapsin-1 initiates actin polymerization. This process enables SV-synapsin-actin assemblies to facilitate the mesoscale organization of SV clusters along axons, which is similar to the native presynaptic organization observed at both lamprey and mammalian synapses. Understanding the relationship between the actin network and synapsin-synaptic vesicle condensates can help elucidate how coordinated neurotransmission along the axon enables circuit function and behavior.

Keywords: Actin; Liquid-Liquid Phase Separation; Presynapses; Synapsin; Synaptic Vesicles.

Figure 1. Synapsin-1 condensates are sites for actin polymerization.

(A) Representative confocal images from the in vitro reconstitution of EGFP-Syn1-FL (4 µM, 3% (w/v) PEG 8000) with ATTO647-labeled G-actin monomers (4 µM) at t = 0 (left) and 35 min (right). Buffer used in these experiments, hereafter referred to as ‘reaction buffer’, contained: 25 mM Tris-HCl (pH 7.4), 150 mM NaCl, 0.5 mM TCEP. Note the condensates growing over time and forming radial “asters.” Scale bar, 5 µm. (B) Confocal images of the EGFP-Syn1-IDR (4 µM, 3% (w/v) PEG 8000) and ATTO647-labeled G-actin reconstitution (4 µM) in reaction buffer at t = 0 (left) and 35 min (right). Scale bar, 5 µm. (C) Plot for the quantification of normalized enrichment of actin within EGFP-Syn1-FL phases. Gompertz growth curve: Y = YM*(Y0/YM)^(exp(-K*X)), YM = 0.9611,Y0 = 0.01032, K = 0.2958, 1/K = 3.381, R² = 0.9376. Data shown here is quantified from three independent experiments and 30 analyzed condensates (N = 3, n = 30). Each dot represents the mean and error bars, SD. (D) Plot showing the number of Sholl intersections for polymerized actin in the presence of EGFP-Syn1-FL, EGFP-Syn1-IDR, and EGFP-Syn1-Dom. C condensates as a function of distance (in µm). The number of Sholl intersections was counted in 0.5 µm radius increments from the center of the condensates. Data from three independent replicates (For Syn1-FL: N = 3; n = 21; for Syn1-IDR: N = 3; n = 10; for Syn1-Dom.C N = 3; n no condensates formed). Each dot represents the mean intersection count for each condition. Error bars represent SEM. (E) Plot depicting the quantification of pyrene-actin fluorescence intensity as a function of time. Pyrene-actin (10 µM) polymerization was assessed in the presence of EGFP-Syn1-FL or EGFP-Syn1-IDR (10 µM) in 1X actin polymerization buffer (10 mM Tris-HCl, pH = 7.5, 2 mM MgCl₂, 50 mM KCl, and 0.5 mM ATP). The data shown here is quantified from three independent experiments (N = 3). Each dot represents the mean and error bars, SD. Source data are available online for this figure.

Figure 2. Actin torus formation around Syn1-FL condensates precedes aster assembly.

(A) Scheme and corresponding representative confocal images from the in vitro Syn1-actin reconstitution assay, showing actin enrichment over a period of 20 min within EGFP-Syn1-FL condensates. Images were acquired at excitation wavelengths 488 and 647 nm for EGFP-synapsin-1-FL and ATTO647 G-actin, respectively. Scale bar, 5 µm. (B) Scheme and corresponding representative confocal images depicting stages of “actin-aster” assembly from EGFP-Syn1-FL condensates. Scale bar, 5 µm. Here, the same datasets are shown with two different LUTs to highlight “torus” and “aster” structures. Right: line profiles depict the enrichment of fluorescently labeled actin within the condensate. Data obtained from three independent reconstitutions, 30 condensates analyzed for each stage. Error bars represent SEM. (C) Scheme of “actin-asters” connecting and forming “web-like” assembly and corresponding confocal image of polymerized “actin-asters” forming “web-like” structures, but with different LUTs to highlight “torus” and “aster” structures. Scale bar, 10 µm. (D) Representative images from the EGFP-Syn1-FL and actin reconstitution acquired using TIRF microscopy show “actin-toruses” and “actin-fibers” as indicated with orange and blue arrows, respectively. Scale bar, 5 µm. Source data are available online for this figure.

Figure 3. The morphology of synapsin-1:actin assemblies as a function of synapsin 1 phosphorylation.

(A) 3D synapsin-1 protein structure prediction obtained from the AlphaFold protein structure database, entry AF-O88935-F1-model_v4. Arrows point to the serine residues that are targeted by PKA (S9) and CaMKII (S568 and S605) in mouse synapsin 1. For the reconstituted protein variants used in this study, Syn1-PKA has serine 9 mutated to glutamate, Syn1-CaMKII has both serines 568 and 605 mutated, and EGFP-Syn1-PKA/CaMKII has all three serines mutated. (B) Representative images for actin polymerization reactions in the presence of purified EGFP-Syn1, EGFP-Syn1-PKA, EGFP-Syn1-CaMKII, and EGFP-Syn1-PKA/CaMKII, each with 3% PEG 8000 in reaction buffer. All images were taken after 40 min of incubation. Images were acquired using a spinning-disk confocal microscope, employing the 647 nm wavelength for actin (represented in green). Scale bars, 10 µm. (C) Graph showing the number of Sholl intersections for polymerized actin in the presence of EGFP-Syn1, EGFP-Syn1-PKA, EGFP-Syn1-CaMKII, and EGFP-Syn1-PKA/CaMKII condensates as a function of distance (in µm). The number of Sholl intersections was counted in 0.5 µm radius increments from the center of the condensates, with at least 100 synapsin:actin assemblies analyzed for each condition. Each dot represents the mean intersection count for each condition. Data from three independent reconstitutions were analyzed. Error bars represent the SEM. For statistical analysis, an R code for linear mixed effects model was applied as in (Jackson et al, 2024); The original code is available at https://zenodo.org/records/1158612 (Wilson et al, 2017). ***p < 0.001, **p < 0.01, *p < 0.05. Source data are available online for this figure.

Figure 4. Actin forms toruses at the interface of SV condensates in vitro and in living synapses.

(A) Left: Representative images of the reconstituted EGFP-Syn1-FL (4 µM, 3% (w/v) PEG 8000) and SVs (3 nM, labeled with FM4-64, 1.65 µM) condensates after adding ATTO647-labeled G-actin monomers (4 µM) at t = 0 (top) and 35 min (bottom). Images were acquired using a spinning-disk confocal microscope. Right: Magnified images of polymerized actin-asters formed at 35 min. Scale bar, 5 µm. (B) Scheme of SV recruitment along the actin fibers. (C) Cartoon illustrating the lamprey spinal cord with giant reticulospinal (RS) axons and microinjection strategy. V, ventral; D, dorsal. (D) Laser-scanning confocal microscopy (LSM) image of a live lamprey reticulospinal axon co-injected with FM4-64 and Alexa Fluor™ 488-phalloidin for characterizing SVs and actin, respectively. Excitation wavelengths were 488 nm for phalloidin-actin and 560 nm for SVs labeled with FM4-64. The Inset shows the clear assembly of actin as a torus around SVs. Scale bars, 5 µm (overview) and 1 µm (insets). (E) LSM images of immunolabeled lamprey synapses. Synapses were stained for endogenous synapsin 1, endogenous SV2 (an SV marker), and actin (Alexa Fluor™ 488-phalloidin). Scale bar, 1 µm. (F) High-resolution confocal microscopy maximum z-projection image showing individual synapses connected together by phalloidin-labeled “aster-like” structures (white arrows). The spinal cord was bathed in Ringer solution containing 0.03% DMSO, followed by injection of Alexa Fluor™ 488-phalloidin. The same image is depicted twice using the indicated minimum and maximum values (16-bit scale). Scale bar, 2 µm. Source data are available online for this figure.

Figure 5. The effects of the inhibition of actin polymerization on presynaptic actin assemblies in vitro and in living synapses.

(A) Schematic illustrating the experimental outline for the reconstitution of G-actin monomers pretreated with Latrunculin A (Lat A) with the subsequent addition of synapsin-1 and 3% PEG 8000. (B) Representative maximum intensity projection from confocal microscopy images from the reconstitution of 4 µM Lat A-treated G-actin monomers with synapsin-1 condensates in reaction buffer at t = 0, 2, 15, and 30 min. Images were acquired at 488 and 647 nm wavelengths for EGFP-synapsin 1 and ATTO647 actin, respectively. Scale bar, 3 µm. (C) Schematic illustrating the experimental outline for the addition of 4 µM Lat A on top of preformed “actin-asters” formed from synapsin-1 condensates. (D) Representative maximum intensity projection from confocal microscopy images from the treatment of reconstituted “actin-asters” with 4 µM Lat A. Images were acquired both before and after (t = 0, 15, and 30 min) addition of 4 µM Lat A at 488 and 647 nm wavelengths for EGFP-synapsin-1 and ATTO647 actin, respectively. Scale bar, 3 µm. (E) Schematic illustrating the experimental outline for the Lat A-based experiments in lamprey reticulospinal (RS) axons ex vivo. (F) Treatment of lamprey RS synapses with Lat A before the injection of fluorescently conjugated phalloidin (Alexa Fluor 488-Phalloidin) leads to the disappearance of the presynaptic actin assemblies. Top: Representative maximum intensity z-projection image of presynaptic actin assemblies inside lamprey RS axons pretreated with 0.03% DMSO (Control), followed by microinjection with Alexa Fluor 488-Phalloidin. Bottom: Representative maximum intensity z-projection image showing the absence of pre-synaptic actin assemblies inside lamprey RS axons pretreated with 2.5 µM Lat A in 0.03% DMSO, followed by microinjection with Alexa Fluor 488-Phalloidin. Magnified regions (right) are illustrated in the overview section (left). White arrows point to aster formations. Scale bars = 10 μm (left) and 1 µm (right). (G) Treatment of lamprey RS synapses with Lat A after the injection of Alexa Fluor 488-Phalloidin does not affect pre-synaptic actin assemblies. Top: Representative maximum intensity z-projection image of pre-synaptic actin assemblies inside lamprey RS axons microinjected with Alexa Fluor 488-Phalloidin and post-treated with 0.03% DMSO (Control). Bottom: Representative maximum intensity z-projection image of pre-synaptic actin assemblies inside lamprey RS axons microinjected with Alexa Fluor 488-Phalloidin and post-treated with 2.5 µM Lat A in 0.03% DMSO. Magnified regions (right) are illustrated in the overview sections (left). White arrows point to aster formations. Scale bars = 10 μm (left) and 1 µm (right). (H) Bar chart showing the quantification of the density (number of objects per surface area of axon) of phalloidin-positive objects inside axons pretreated with Lat A (2.5 µM in 0.03% DMSO) or the corresponding control (0.03% DMSO). Each dot represents a z-stack image series through an axon. Error bars represent the standard error of the mean (SEM). Data were from n = 14 images; n = 3 axons from N = 3 animals. For statistical analysis, an unpaired t-test was performed in GraphPad PRISM.  The p value =  3.48 × 10⁻⁶; ****p < 0.0001 (Pre-treated control-Lat-A). (I) Bar chart showing the quantification of the density of phalloidin-positive objects inside axons post-treated with Latrunculin A (2.5 µM in 0.03% DMSO) or the corresponding control (0.03% DMSO). Each dot represents a z-stack image series through an axon. Error bars represent the standard error of the mean (SEM). Data were from n = 16–18 images; n = 3–4 axons from N = 3–4 animals. For statistical analysis, an unpaired t-test was performed in GraphPad PRISM. The p value = 0.53 (Post-treated Control-Lat-A). “ns” indicates not significant. Source data are available online for this figure.

Figure 6. Actin aster are stable structures connecting neighboring condensates.

(A) Representative confocal image from the reconstituted EGFP-synapsin-1:actin:GUV assemblies. Projection of the maximal intensity, 30 min post-incubation. Scale bar, 10 µm. (B) Magnified region from (A), indicating the long actin fibers connecting the adjacent condensates. (C) Top: Skeletonized actin fibrils from (A). Bottom: A single continuous network of synapsin condensates interconnected with a continuous actin network is highlighted in green. Scale bar, 20 µm. (D) Heatmap of co-dependencies (Pearson’s correlation) for synapsin1:actin assemblies’ node properties: (1) volume, (2) surface area, (3) diameter, (4) main-, (5) second-, (6) third- axis length, (7) mean-, (8) maximum, (9) minimum- intensity, (10) median Euclidean distance from other nodes). Data obtained from three independent reconstitutions. (E) Principal component (PC) analysis of synapsin-1:actin assemblies: scatter plot of PC1 vs. PC2 scores (three clusters identified by k-means clustering highlighted in different colors). Data obtained from three independent reconstitutions. (F) Representative confocal image from the reconstituted synapsin-1:actin assemblies spreading over the surface of a GUVs. Projection of the maximal intensity, 30 min post-incubation. Scale bar, 3 µm. (G) Schematic representation of the synapsin:actin:GUV association. (H) Representative confocal images of synapsin:actin assemblies before (top) and after (bottom) the addition of 1,6-hexanediol (10% w/v). The treatment of the synapsin:actin assemblies with 1,6-hexanediol partially disperses synapsin condensates without affecting the preformed actin-asters. Asterisks indicate synapsin condensates without actin, which do disperse upon the addition of 1,6-hexanediol, as expected. Scale bars: 10 µm (overview), 1 µm (magnified). (I) Quantification of the synapsin 1 and actin partitioning before and after treatment with 1,6-hexanediol. Data from three independent reconstitutions, >1500 condensates analyzed for each condition. The box stretches from the 25th to 75th percentile, the dots represent individual data points, the central line shows the median, and the whiskers represent the min and max values. The p values are 1.25 × 10⁻⁵⁰ for Syn1-FL before vs. after hexanediol and 2.74 × 10⁻⁵ for actin before vs. after hexanediol. ****p < 0.0001; Mann–Whitney U-test (two-tailed, non-parametric). Source data are available online for this figure.

Figure 7. Visualizing actin at pre- and post-synapses in murine neurons.

(A) Cryo-electron tomogram of a wild-type synapse. Scale bar, 100 nm. (B) 3D rendering of structures identified in (A). Annotation: yellow, synaptic vesicles (SVs); green, pre-synaptic plasma membrane; blue, post-synaptic plasma membrane; red, actin filament; colored arrowheads point to actin filaments at the presynapse. Scale bar, 100 nm. (C, D) Exemplary region highlighting actin filaments associated with the SV cluster from (A). Note the traced actin filaments (red in C) and original density (arrowheads in D) juxtaposed to SVs. Scale bar, 50 nm. (E) Exemplary region highlighting actin filaments in the periactive zone associated with the plasma membrane. Scale bar, 50 nm. (F) Expansion microscopy images of wild-type hippocampal neurons in culture (14 DIV) immunostained for VGLUT1 (an integral SV protein), PSD95 (post-synaptic protein), and actin. The inset (white dotted line) focuses on resolved pre- and post-synaptic regions. Note the presence of actin in both compartments (yellow line outlines the presynapse). Expansion factor 4.3; scale bar, 1 µm (G). The same as in (F) but for neurons from synapsin triple knockout animals (SynTKO). Expansion factor 4.3; scale bar, 1 µm. (H) Quantification of the actin signal present in VGLUT1-positive regions in synapses from WT and SynTKO neurons. Data from three independent neuronal preparations, 731/1148 synapses analyzed for WT/TKO, respectively. Bars represent mean ± SD; Mann–Whitney U-test (two-tailed, non-parametric); p value = 2.73 × 10⁻⁵ (WT-TKO). ****p < 0.0001. (I) Quantification of the actin signal present in PSD95-positive regions in synapses from WT and SynTKO neurons. Data from three independent neuronal preparations, 1036/1215 synapses analyzed for WT/TKO, respectively. Bars represent mean ± SD; Mann–Whitney U-test (two-tailed, non-parametric); p value = 0.138 (WT-TKO); n.s., non-significant. Source data are available online for this figure.

Figure 8. Synapsin-SV condensates are necessary to sequester actin in living synapses.

(A, B) Two-color super-resolution STED images of hippocampal neurons (14 DIV) from (A) wild-type and (B) synapsin triple knockout neurons stained with StarRed-phalloidin and anti-synaptophysin (Star580). Scale bars, 1 µm. (C) Magnified region from (A) showing a cohort of synaptic vesicles (anti-synaptophysin) colocalizing with actin (StarRed-phalloidin). Scale bar, 500 nm. (D) Magnified region from (B) indicating a more dispersed signal of SVs (anti-synaptophysin) lacking the colocalization with actin (StarRed-phalloidin). Scale bar, 500 nm. (E, F). Representative images of synapsin triple knockout neurons rescued with (E) full-length EGFP-synapsin-1 or (F) EGFP-synapsin-1 IDR and stained with StarRed-phalloidin and anti-synaptophysin (Star580). Scale bars, 1 µm. (G, H). Magnified regions from (E, F), respectively, showing colocalization of SVs with actin. Scale bars, 500 nm. (I) Autocorrelation of the actin signal along the axons (StarRed-phalloidin; see lines in the images in A, B) indicates that actin rings remain unaltered in the absence of synapsins. (J) Actin enrichment in synaptic vesicle cohorts is defined as a normalized intensity signal of actin channel within regions positive for synaptophysin (wild-type and synapsin triple knockout) or both synaptophysin and EGFP (in rescue experiments). Data from at least three independent neuronal preparations (for WT: N = 3; n = 36; for SynTKO: N = 3; n = 33; for SynTKO+FL: N = 4; n = 16; for SynTKO+IDR: N = 3; n = 34). Data points present median maximum actin intensity per image; the box represents mean ± 1.0 SE and whiskers represent mean ± 1.0 SD. Mann–Whitney U-test (two-tailed, non-parametric); WT-SynTKO (p = 9.82 × 10⁻⁷); SynTKO-SynTKO+FL (p = 2.21 × 10⁻⁶); SynTKO-SynTKO+IDR (p = 3.92 × 10⁻⁶) are all p < 0.0001 (****). Source data are available online for this figure.

Figure EV1. Recombinant EGFP-synapsin-1 and domains used for in vitro reconstitutions.

(A) Schematic cartoon representing the domain organization of synapsin-1 (Syn1). Syn1 full length (FL), Domain C (Dom. C), and intrinsically disordered region (IDR) are depicted. Actin-binding regions are depicted with magenta (a.a. 222–370 and 371–705). (B) SDS-PAGE gel of the purified proteins employed for in vitro reconstitutions in this study: EGFP-Syn1-FL (102.235 kDa), EGFP-Syn1-IDR (57.621 kDa), and EGFP-Syn1-Dom. C (63.116 kDa). (C) Schematic illustration of the Syn1-actin reconstitution assay flow. Actin polymerization from Syn1 phases was examined by first preforming 6 µM EGFP-Syn1-FL condensates with 3% (w/v) PEG 8000 on a glass-bottom dish. After incubating for 5 min, when EGFP-Syn1-FL condensates became 3–4 µm in size, ATTO647-labeled G-actin monomers were added from the top into these preformed EGFP-Syn1-FL condensates such that the final concentration of actin and EGFP-Syn1-FL in the final reaction mix was 4 µM for both components. Source data are available online for this figure.

Figure EV2. Reconstitution of actin with synapsin-1 in two distinct reaction orders.

(A) Top: Schematic illustration showing the order-1 of the reconstitution assay. Actin polymerization from synapsin-1 phases was examined by first adding 4 µM synapsin-1 to a glass-bottom dish, followed by the addition of 4 µM ATTO647 G-actin monomers 2 min later. Subsequently, 3% (w/v) PEG 8000 was added on top of the reaction mix after 10 min, and actin polymerization was followed for 30 min. Bottom: Representative SD confocal microscopy images from the reconstitution of actin with synapsin 1 liquid phases in reaction buffer at t = 0, 2, and 30 min. Zoomed-in regions towards the right side show actin-asters. Images were acquired at 488 and 647 nm wavelengths for EGFP-Syn1-FL and actin, respectively. Scale bar, 5 µm. (B) Top: Schematic illustration showing the order-2 of the reconstitution assay. Actin polymerization from synapsin-1 phases was assessed by first adding 4 µM ATTO647 G-actin monomers to a glass-bottom dish. Two minutes later, 4 µM synapsin 1 was added on top of the reaction mix. Subsequently, 3% (w/v) PEG 8000 was added on top of the reaction mix after 2 min, and actin polymerization was followed for 30 min. Bottom: Representative confocal microscopy images from the reconstitution of actin with synapsin-1 liquid phases in reaction buffer at t = 0, 2, and 30 min. Zoomed-in regions towards the right side show actin-asters. Images were acquired at 488 and 647 nm wavelengths for EGFP-Syn1-FL and actin, respectively. Scale bar, 5 µm. Source data are available online for this figure.

Figure EV3. In vitro reconstitutions and turbidity measurements of Syn1 variants with actin.

(A) Representative confocal images from the in vitro reconstitution of EGFP-Syn1-Dom. C (4 µM, 3% (w/v) PEG 8000) with ATTO647-labeled G-actin monomers (4 µM) at t = 0 (left) and t = 35 min (right). Image acquisition for EGFP-Syn1-Dom. C and ATTO647 G-actin was carried out at excitation wavelengths 488 and 647 nm, respectively. Scale bar, 5 µm. (B) Representative confocal images from the in vitro reconstitution of EGFP-G3BP2 (4 µM, 3% PEG 8000) with ATTO647-labeled G-actin monomers (4 µM) at t = 0 (left) and t = 35 min (right). Images were acquired at excitation wavelengths 488 and 647 nm for EGFP-G3BP2 and ATTO647 G-actin, respectively. Scale bar, 5 µm. (C) Quantification of the turbidity assay. Plot comparing the turbidity measurements for EGFP-Syn1-FL with actin, EGFP-Syn1-FL alone, and actin alone in the presence of 3% (w/v) PEG 8000. Actin polymerization was assessed as an increase in optical density after a 48 h incubation period. Actin used for the assay was Mg²⁺-exchanged and supplemented with 0.5 mM ATP. Data shown here is quantified from four independent experiments (N = 4). The p values are: 8.53 × 10⁻⁴ for Syn1-FL+Actin Vs Syn1-FL and 1.13 × 10⁻⁴ for Syn1-FL+Actin Vs Syn1-FL Vs Actin; ****p < 0.0001; one-way ANOVA test. (D) Top: representative brightfield images of reaction mixes after turbidity assay from (A). Bottom: epifluorescence images of the same regions at 488 nm excitation wavelength. Scale bar, 50 µm. Source data are available online for this figure.

Figure EV4. Recombinant synapsin 1 phosphomimetics used for in vitro reconstitutions.

(A) Schematic cartoon depicting the purification steps followed to obtain synapsin 1 (Syn1) from transfected Expi293^TM cells expressing His-SUMO-EGFP-Syn1. This same pipeline is used for all Syn1 variants. (B) Exemplary SDS-PAGE gel for His-SUMO-EGFP-Syn1-CaMKII (S568/605E) construct, showing the fractions from each purification step from 1 to 8 in (A). (C) SDS-PAGE gel of the final purified EGFP-Syn1 versions employed for in vitro reconstitutions in this study: EGFP-Syn1 WT, EGFP-Syn1-PKA (S9E), EGFP-Syn1-CaMKII (S568/605E), and EGFP-Syn1-PKA+CaMKII (S9/568/605E). All proteins weigh 102.235 kDa. (D) Representative images of EGFP-Syn1 WT, EGFP-Syn1-PKA (S9E), EGFP-Syn1-CaMKII (S568/605E), and EGFP-Syn1-PKA+CaMKII (S9/568/605E) when reconstituted in SEC buffer and in the presence of 3% PEG 8000 at t = 10 and t = 25 min. Images were acquired using a spinning-disk confocal microscope, employing the 488 nm wavelength for EGFP-Syn1. Scale bar, 5 µm. (E) Cumulative frequency indicating the size distribution of at t = 0 (full line) and t = 45 min (dotted line); color code as in (D). Data from three independent reconstitutions for each condition. Source data are available online for this figure.

Figure EV5. Reconstitution of actin with synapsin-1 and native SVs.

(A) Representative images for actin reconstitution in the presence of natively purified 3 nM SVs labeled with FM4-64 (1.65 µM final concentration), 3% PEG 8000 in reaction buffer at t = 0 and t = 45 min. Images were acquired using a spinning-disk confocal microscope at 561 and 647 nm for SVs and actin, respectively. Scale bar, 5 µm. (B) Quality control of synaptic vesicles. Left: Coomassie-stained 12% SDS-PAGE of fractions, brain homogenate (H) and final synaptic vesicle fraction (SV), from the synaptic vesicle isolation procedure from native rat brain (5 µg per lane). Right: Western blot of brain homogenate and final SV fraction detecting classical protein markers of SV enrichment and absence of proteins regarded as contaminants. (C) Diffusion coefficients of actin bound to SVs immobilized on a functional surface as described in Perego et al, . Data from three independent replicates (N = 3). The box stretches from the 25th to the 75th percentile, the dots represent individual data points, the central line shows the median, and the whiskers represent the min and max values. Green boxes, diffusion of actin alone; magenta boxes, diffusion of actin in the presence of EGFP-synapsin 1. (D) Reconstitution of actin with synapsin-1 IDR-SV phases and actin-asters. Representative confocal images of the reconstituted EGFP-Syn1-IDR (4 µM, 3% PEG 8000) and SVs (3 nM, labeled with FM4-64, 1.65 µM) condensates after adding ATTO647-labeled G-actin monomers (4 µM) at t = 0 (top) and 90 min (bottom). Excitation at 488 nm for EGFP-synapsin 1-IDR, 560 nm for SVs labeled with FM4-64 and 647 nm for ATTO647 G-actin. Scale bar, 5 µm. Source data are available online for this figure.

Figure EV6. 1,6-Hexanediol disperses EGFP-synapsin 1:actin condensates lacking an apparent actin polymerization.

(A) Representative confocal images from the in vitro reconstituted EGFP-synapsin-1:actin:GUV assemblies before and after 1,6-hexanediol treatment. The dashed line represents the line used for plotting the intensity profile. Scale bar, 5 µm. (B) Fluorescence intensity profiles of EGFP-synapsin-1 and actin along the dashed line from (A). Solid lines, fluorescence intensity before, and dashed lines, after treatment with 1,6-hexanediol. (C) Quantification of synapsin-1 and actin partitioning in EGFP-synapsin-1:actin:GUV assemblies before and after 1,6-hexanediol treatment. Data from three independent reconstitutions, 30 condensates analyzed for each condition. The p values are 6.76 × 10⁻¹⁷ for Syn1-FL before vs. after treatment and 6.31 × 10⁻¹⁵ for actin before vs after treatment. ****p < 0.0001; Mann–Whitney U-test (two-tailed, non-parametric). Source data are available online for this figure.

Figure EV7. Synaptic vesicle condensates are necessary for concentrating actin at the presynaptic boutons.

(A) Primary hippocampal neurons in culture (DIV 14) stained for actin (phalloidin-AberriorStarRed); images as in Fig. 8, but here shown with the same intensity scaling. Scale bars, 1 µm. (B) Actin enrichment in synaptic vesicle cohorts is defined as a normalized intensity signal of actin channel within regions positive for synaptophysin (wild-type and synapsin triple knockout) or both synaptophysin and EGFP (in rescue experiments). Data from at least three independent neuronal preparations (for WT: N = 3; n = 3; for SynTKO: N = 3; n = 3; for SynTKO+FL: N = 4; n = 4; for SynTKO+IDR: N = 3; n = 3); data points represent median maximum actin intensity per neuronal preparation; the box represents mean ± 1.0 SE, whiskers represent mean ±  1.0 SD, and the central line represent median two sample t-test (equal variance not assumed). WT-SynTKO p = 0.0049, *; SynTKO-SynTKO+FL p = 0.0021, **; SynTKO-SynTKO+IDR p = 0.030, *. (C) Actin enrichment in synaptic vesicle cohorts is defined as a normalized intensity signal of actin channel within regions positive for synaptophysin (wild-type and synapsin triple knockout) or both synaptophysin and EGFP (in rescue experiments). Data from three independent neuronal preparations (for WT: N = 3; n = 1592; for SynTKO: N = 3; n = 7096; for SynTKO+FL: N = 4; n = 209; for SynTKO+IDR: N = 3; n = 460); data points represent maximum actin intensity per region positive for synaptophysin. The box represents mean ± 1.0 SE, and whiskers represent mean ± 1.0 SD, and the central line represents median. Mann–Whitney test; WT-SynTKO, p = 3.06 x 10^(-251); SynTKO-SynTKO+FL, p = 2.12 × 10⁻⁷⁹; SynTKO-SynTKO+IDR, p = 1.96 × 10⁻¹³⁵; ****p < 0.0001. (D) Area of synaptophysin-positive regions in different conditions. Note in SynTKO neurons the presence of regions of varying size, particularly the smaller ones, presumably a consequence of dispersed SVs into smaller cohorts and/or individual vesicles. The red square represents mean values, and the central line shows the median. Data from at least three independent replicates (for WT: N = 3; n = 1592; for SynTKO: N = 3; n = 7096; for SynTKO+FL: N = 4; n = 209; for SynTKO+IDR: N = 3; n = 460). Source data are available online for this figure.