Mechanism of evenness interrupted (Evi)-exosome release at synaptic boutons

Abstract

Wnt signaling plays critical roles during synaptic development and plasticity. However, the mechanisms by which Wnts are released and travel to target cells are unresolved. During synaptic development, the secretion of Drosophila Wnt1, Wingless, requires the function of Evenness Interrupted (Evi)/Wls, a Wingless-binding protein that is secreted along with Wingless at the neuromuscular junction. Given that Evi is a transmembrane protein, these studies suggested the presence of a novel vesicular mechanism of trans-synaptic communication, potentially in the form of exosomes. To establish the mechanisms for the release of Evi vesicles, we used a dsRNA assay in cultured cells to screen for genes that when down-regulated prevent the release of Evi vesicles. We identified two proteins, Rab11 and Syntaxin 1A (Syx1A), that were required for Evi vesicle release. To determine whether the same mechanisms were used in vivo at the neuromuscular junction, we altered the activity of Rab11 and Syx1A in motoneurons and determined the impact on Evi release. We found that Syx1A, Rab11, and its effector Myosin5 were required for proper Evi vesicle release. Furthermore, ultrastructural analysis of synaptic boutons demonstrated the presence of multivesicular bodies, organelles involved in the production and release of exosomes, and these multivesicular bodies contained Evi. We also used mass spectrometry, electron microscopy, and biochemical techniques to characterize the exosome fraction from cultured cells. Our studies revealed that secreted Evi vesicles show remarkable conservation with exosomes in other systems. In summary, our observations unravel some of the in vivo mechanisms required for Evi vesicle release.

FIGURE 1.

Exosome uptake assay and role of Rab11 in exosome release from Evi-GFP-S2 cells. A, mCherry-S2 recipient cell incubated with the cleared culture medium from Evi-GFP-S2 donor cells showing Evi-GFP puncta (arrows) in the cytoplasm. Image is shown after deconvolution. B, untransfected S2 recipient cell incubated with purified exosome fraction isolated from Evi-GFP-S2 donor cells. C, number of Evi-GFP puncta per cell inside mCherry-S2 recipient cells after incubation with increasing amounts of culture medium from Evi-GFP-S2 donor cells. Two independent experiments are shown. n = 100 for each point. D, Western blots showing that endogenous Rab11 protein levels are significantly decreased after Rab11-dsRNA treatment. E, quantification of Rab11 protein levels in Evi-GFP-S2 cell lysate after Rab11-dsRNA treatment. n = 3. F, Western blots of Evi-GFP-S2 cell lysate, showing that Evi-GFP levels are not significantly affected by Rab11-dsRNA treatment. Tub, tubulin. G, quantification of Evi-GFP levels in Evi-GFP-S2 cell lysate after Rab11-dsRNA treatment. n = 3. H, number of Evi-GFP puncta in untransfected recipient cells incubated with dsRNA-treated Evi-GFP-S2 donor cells normalized to control, showing that Rab11 dsRNA treatment drastically reduces the number of internalized Evi vesicles. n = 100 cells per condition, per experiment. Three independent experiments were performed for each dsRNA. I, Western blot of Evi-GFP-S2 or dSiaT-S2 cell culture medium showing that Evi-GFP, but not dSiaT levels, are significantly reduced by Rab11-dsRNA treatment. Quantification of Evi-GFP in the culture medium of Evi-GFP-S2 (J) and, dSiaT in the culture medium of dSiaT-S2 cells after treatment with Rab11-dsRNA (K). n = 3. Calibration bar is 8 μm in A and B.

FIGURE 2.

dsRNA screen of exosomal candidates and role of Syx1A in Evi vesicle release from Evi-GFP-S2 cells. A, number of exosomes internalized by untransfected S2 cells after treatment of Evi-GFP-S2 donor cells with the indicated dsRNA, showing that Syx1A down-regulation significantly decreases the number of internalized exosomes. Data are normalized to untreated control cells. n = 100 cells per experiment, per condition. Three independent experiments were performed for each dsRNA. B, Western blots of Evi-GFP and dSiaT-S2 cell lysates treated with Syx1A-dsRNA constructs showing that Syx1A protein levels are significantly reduced by this treatment. C, quantification of Syx1A levels in Evi-GFP-S2 cell lysates after dsRNA treatment. n = 3. D, Western blots of Evi-GFP and dSiaT-S2 cell culture medium showing that Syx1A-dsRNA treatment decreases the levels of secreted Evi-GFP but not dSiaT. E, quantification of Evi-GFP in the cleared culture medium of Evi-GFP-S2 cells after dsRNA treatment. n = 3. F, quantification of dSiaT levels in the culture medium of dSiaT-S2 cells after dsRNA treatment. n = 3. G, Western blots of Evi-GFP-S2 cell lysate, showing that Syx1A-dsRNA treatment does not affect Evi-GFP levels in these cells. Tub, tubulin. H, quantification of Evi-GFP levels in cell lysates of Evi-GFP-S2 cells after dsRNA treatment. n = 3.

FIGURE 3.

Rab11 is required for Evi-HA release at the NMJ and is localized at synaptic boutons. A–C, endogenous Evi localization at the NMJ of the following genotypes, in preparations labeled with anti-Evi (N-terminal epitope) and anti HRP. Insets are high magnification views of synaptic boutons. A, control (C380-Gal4/+) animal; B, larva expressing Rab11DN^N124I in motoneurons (C380-Gal4 > Rab11DN^N124I), showing a decrease in endogenous Evi levels; C, larva expressing Rab11-RNAi in motoneurons (C380-Gal4 > Rab11-RNAi), showing a similar decrease in endogenous Evi levels. D–F, localization of Evi-HA at the NMJ in larvae expressing Evi-HA in motoneurons in the following genotypes: D, control larva expressing Evi-HA in motoneurons (C380-Gal4 > Evi-HA); E, larva expressing Rab11DN^N124I and Evi-HA in motoneurons (C380-Gal4 > Evi-HA, Rab11DN^N124I), showing the localization of Evi-HA in large aggregates within the boutons and absent from the postsynaptic region; F, larva expressing Rab35DN^S22N and Evi-HA in motoneurons (C380-Gal4 > Evi-HA, Rab35DN^S22N) showing the normal presence of Evi-HA at the postsynaptic region. G and H, endogenous Wg levels at the NMJ of a control larva (C380-Gal4/+), showing the presence of Wg at the postsynaptic region (G), and larva expressing Rab11DN^N124I in motoneurons (C380-Gal4 > Rab11DN^N124I), showing a decrease in Wg levels at the postsynaptic region (H). J and K, deconvolved images showing the localization of endogenous Rab11 at synaptic boutons of a wild type larva in relationship to HRP (J) and Bruchpilot (Brp) and dPak (K). I, quantification of Evi-HA, endogenous Evi and Wg immunoreactivity levels, shown as ratios of postsynaptic to presynaptic intensities, normalized to controls. At least five larvae and 10 NMJs were analyzed for each genotype, and the following number of boutons were quantified for Evi-HA: 48 boutons for control C380-Gal4 > Evi-HA and 72 boutons for C380-Gal4 > Evi-HA, Rab11DN^N124I; 46 boutons for control C380-Gal4 > EviHA and 50 boutons for C380-Gal4 > EviHA, Rab35DN; endogenous Evi, 68 boutons for control (C380-Gal4/+) and 45 boutons for C380-Gal4 > Rab11DN^N124I; endogenous Evi, 75 boutons for C380-Gal4/+ and 56 boutons for C380-Gal4 > Rab11-RNAi; endogenous Wg, 47 boutons for C380-Gal4/+ and 38 boutons for C380-Gal4 > Rab11DN^N124I. Calibration bar is 5 μm for J–K, 15 μm for A–H, and 5 μm for insets on A–H. Images are confocal Z-stack projections of NMJs on muscles 6/7 at segment A3. NS, not statistically significant.

FIGURE 4.

Syntaxin 1A is involved in Evi vesicle release at the larval NMJ. A–C, endogenous Evi localization at the larval NMJ of a y w control, showing endogenous Evi localization in both the pre- and postsynaptic region (A) and a y w/w; syx1A⁴/syx1A^Δ229 animal, showing virtual absence of endogenous Evi (B). C, larva expressing Syx1A-RNAi in motoneurons (C380-Gal4 > Syx1A-RNAi), showing a decrease in postsynaptic Evi. D, quantification of endogenous Syx1A levels in synaptic boutons of larvae expressing Syx1A-RNAi in motoneurons (C380-Gal4 > Syx1A-RNAi). Four larvae (eight NMJs) were quantified for control C380-Gal4/+ (38 boutons) and C380-Gal4 > Syx1A-RNAi (38 boutons). E, quantification of endogenous Evi levels in the indicated genotypes, shown as ratios of postsynaptic to presynaptic intensities, normalized to controls. 75 boutons were quantified for control C380-Gal4/+ and 56 boutons for C380-Gal4 > Syx1A-RNAi. F–H, Evi-HA localization at the larval NMJ of a control larva expressing Evi-HA in motoneurons (C380-Gal4 > Evi-HA), showing transfer of Evi-HA to the postsynaptic region (F); larva expressing both Evi-HA and Syx1A-RNAi in motoneurons (C380-Gal4 > Evi-HA, Syx1A-RNAi), showing a decrease in postsynaptic Evi-HA levels (G); larva expressing overactive Syx1A(T254I) in motoneurons (C380-Gal4 > Evi-HA, Syx1A(T254I)), showing an increase in postsynaptic Evi-HA levels (H). I, Evi-HA immunoreactivity levels shown as ratios of postsynaptic to presynaptic intensities, normalized to controls. At least five larvae and 10 NMJs were analyzed for each genotype, and the following number of boutons were quantified: 55 boutons for control C380-Gal4 > Evi-HA; 83 boutons for C380-Gal4 > Evi-HA, Syx1A-RNAi; 61 boutons for control C380-Gal4 > Evi-HA; 76 for C380-Gal4 > Syx1A, Evi-HA; 66 for C380-Gal4 > Syx1A(T254I), Evi-HA. NS, not statistically significant. Calibration bar is 8 μm in A–G (left and middle panel), 3 μm in A–E, insets. Images are confocal Z-stack projections of NMJs on muscles 6/7 at segment A3.

FIGURE 5.

Myo5 is required for Evi secretion at the NMJ. A and B, endogenous Evi localization at the larval NMJ in a control (C380-Gal4/+) larva (A); larva expressing both Myo5DN in motoneurons (C380-Gal4 > Myo5DN) showing a marked reduction of synaptic Evi levels at the postsynaptic region (B). C, control (C380-Gal4 > Evi-HA) larva expressing Evi-HA in motoneurons; D, larva expressing both Evi-HA and Myo5DN in motoneurons (C380-Gal4 > Myo5DN, Evi-HA), showing a decrease in postsynaptic Evi-HA. E, Evi-HA or endogenous Evi immunoreactivity levels shown as ratios of postsynaptic to presynaptic intensities that were normalized to controls. At least five larvae and 10 NMJs were analyzed for each genotype, and the following number of boutons were quantified: 41 boutons for control C380-Gal4 > Evi-HA and 19 for C380-Gal4 > Evi-HA, Myo5DN; 35 boutons for control C380-Gal4 > Evi-HA and 46 for C380-Gal4 > Evi-HA, Myo5DN; 43 boutons for control C380-Gal4/+ and 54 boutons for C380-Gal4 > Myo5DN. Calibration bar is 8 μm for A–E (left and middle panels), 3 μm for insets. Images are confocal Z-stack projections of NMJs on muscles 6/7 at segment A3.

FIGURE 6.

Electron microscopy of multivesicular bodies and exosomes at larval NMJs. A–E, transmission electron micrographs of type I synaptic boutons (b) from a third instar larva, showing the presence of a multivesicular body (arrow) inside the bouton (A), shown at high magnification in C. Arrowheads point to small vesicular structures at the SSR. B, multivesicular body (arrow) in the presynaptic bouton appearing closely attached to the presynaptic membrane, shown at high magnification in D. E, a multivesicular body (arrow) appearing to fuse with the presynaptic membrane of a synaptic bouton (b). Arrowheads point to vesicular structures in the SSR that might have been released from the presynaptic terminal. F–I, immunoelectron micrographs of body wall muscles from larvae expressing Evi-GFP in motoneurons labeled with GFP antibodies showing a type I synaptic bouton (b) with gold particles representing Evi-GFP immunoreactivity at a presynaptic multivesicular body (arrow) (F), shown at high magnification in H, and at the postsynaptic SSR. G, Evi-GFP localization in the postsynaptic SSR; white arrowheads point to the synaptic region, recognized by the size of the synaptic cleft and its increased electron density. I, high magnification view of the Evi-GFP immunogold label in G, showing that in the crevices of SSR Evi-GFP appears to be present on a membrane encapsulated vesicle (black arrowhead), consistent with the view that Evi-GFP is released via exosomes. Calibration bars: 450 nm in A and B; 150 nm in C and D; 170 in E; 400 nm in F; 130 nm in H; 250 nm in G; and 85 nm in I.

FIGURE 7.

Electron microscopic analysis of multivesicular bodies and exosomes in S2 cells. A, immunoelectron micrograph showing an Evi-GFP immunoreactive multivesicular body (mvb) within an S2 cell. B, transmission electron micrographs of negatively stained exosome preparations from S2 cells showing the characteristic cup-shaped appearance of exosomes. C, immunoelectron micrograph of exosomes from Evi-GFP-S2 cells stained with an antibody against the N terminus of Evi, showing the presence of gold particles decorating the exosomes. D, Western blot of Evi-GFP-S2 cell lysate and the purified exosome fraction showing enrichment of Evi-GFP in the exosomal fraction. 30 ng of Evi-GFP cell lysate and 0.3 ng of Evi-GFP exosome fraction were loaded. E, Western blot of S2 cell lysate and purified exosomes from Evi-GFP-S2 or untransfected S2 cells showing Wg enrichment in the exosome fraction. tub, tubulin. F, Western blot of larval body walls containing the central nervous system (CNS), as well as of untransfected S2 and Evi-GFP-S2 cell lysates and their corresponding exosome fractions. Note the enrichment of Lbm in the exosome fraction. Equal amounts of protein were loaded for each sample. Calibration bar: 300 nm in A; 200 nm in B; and 130 nm in C.