Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2018 Mar 5;217(3):1129-1142.
doi: 10.1083/jcb.201703206. Epub 2018 Jan 16.

Quantifying Exosome Secretion From Single Cells Reveals a Modulatory Role for GPCR Signaling

Affiliations
Free PMC article

Quantifying Exosome Secretion From Single Cells Reveals a Modulatory Role for GPCR Signaling

Frederik Johannes Verweij et al. J Cell Biol. .
Free PMC article

Erratum in

  • Correction: Quantifying exosome secretion from single cells reveals a modulatory role for GPCR signaling.
    Verweij FJ, Bebelman MP, Jimenez CR, Garcia-Vallejo JJ, Janssen H, Neefjes J, Knol JC, de Goeij-de Haas R, Piersma SR, Baglio SR, Verhage M, Middeldorp JM, Zomer A, van Rheenen J, Coppolino MG, Hurbain I, Raposo G, Smit MJ, Toonen RFG, van Niel G, Pegtel DM. Verweij FJ, et al. J Cell Biol. 2018 Mar 5;217(3):1157. doi: 10.1083/JCB.20170320601192018c. Epub 2018 Jan 23. J Cell Biol. 2018. PMID: 29362224 Free PMC article. No abstract available.

Abstract

Exosomes are small endosome-derived extracellular vesicles implicated in cell-cell communication and are secreted by living cells when multivesicular bodies (MVBs) fuse with the plasma membrane (PM). Current techniques to study exosome physiology are based on isolation procedures after secretion, precluding direct and dynamic insight into the mechanics of exosome biogenesis and the regulation of their release. In this study, we propose real-time visualization of MVB-PM fusion to overcome these limitations. We designed tetraspanin-based pH-sensitive optical reporters that detect MVB-PM fusion using live total internal reflection fluorescence and dynamic correlative light-electron microscopy. Quantitative analysis demonstrates that MVB-PM fusion frequency is reduced by depleting the target membrane SNAREs SNAP23 and syntaxin-4 but also can be induced in single cells by stimulation of the histamine H1 receptor (H1HR). Interestingly, activation of H1R1 in HeLa cells increases Ser110 phosphorylation of SNAP23, promoting MVB-PM fusion and the release of CD63-enriched exosomes. Using this single-cell resolution approach, we highlight the modulatory dynamics of MVB exocytosis that will help to increase our understanding of exosome physiology and identify druggable targets in exosome-associated pathologies.

Figures

Figure 1.
Figure 1.
CD63-pHluorin is sorted into acidic MVBs and released via exosomes. (a) Proposed model for the visualization of MVB–PM fusion: a pH-sensitive optical reporter (CD63-pHluorin) is quenched when facing the acidic lumen of the MVB. Upon fusion, low luminal pH is immediately neutralized, resulting in a sudden increase in fluorescent intensity. EC, extracellular. (b) Immunofluorescent colabeling of total CD63 (red) and CD63-pHluorin (green) in HeLa cells. PCC, Pearson’s correlation coefficient. (c) TIRF images of a CD63-pHluorin–expressing HeLa cell at normal and elevated intracellular pH (NH4Cl superfusion). On the right, a heat map revealing acidic vesicles close to the PM was obtained by subtracting the fluorescent intensity values of the normal pH from the high-pH condition. (d) EM images of an MVB close to the PM (left) and EVs aligning the PM (right) labeled with gold particles directed to GFP (10 nm) in CD63-pHluorin–expressing HeLa cells. (e) Imaging flow cytometry of the number of late endosomes per cell in a 2.5-µm optical section in CD63-pHluorin–expressing cells (left) or immunostaining against LAMP1 in nontransfected cells (right; n > 2,000 cells). (f) Volume distribution of endosomes based on analysis of whole-cell confocal scans (error bars represent SD; n = 3). The blue area accounts for 75% of the total number of endosomes and covers the 400–600-nm-diameter range. (g) Immunogold labeling on purified exosomes with gold particles (10 nm) coupled to anti-GFP antibody. (h) Western blotting analysis on untransfected (−) and CD63-pHluorin–transfected (+) cells and purified exosomes for total CD63 and GFP. (i) Example of a localized sudden increase in fluorescence at the PM before the event (1), during the event (2), and right before disappearance of the signal (3). (j) Left: total projection of fusion events (bright spots) over a time course of 3 min onto two cells (blue). Right: representative example of CD63-pHluorin–expressing HeLa cell. N, nucleus. Bars: (b, c, and j) 10 µm; (i) 2.5 µm. (k) Effect of incubation with GW4896 (5 µM; n ≥ 8 cells per condition) and nSMase-2 knockdown (n ≥ 22 cells per condition) on fusion activity in HeLa cells. *, P < 0.05; ***, P < 0.001 using Student’s two-tailed two-sample t test. Whiskers in the box plots represent 1.5 times the interquartile distance or the highest or lowest point, whichever is shorter. (l) Western blotting analysis on purified exosomes from GW4896- and control-treated HeLa cells for CD63 and CD81.
Figure 2.
Figure 2.
CD63-pHluorin fusion events are derived from MVBs. (a) Left three panels: live imaging of fusion events (indicated by white arrows) over a time course of 12 s onto one cell before the event (left), at the start of the event (middle), and right before fixation of the cell (3). Right: inset showing a magnification of the localized sudden increase in fluorescence at the PM (highlighted by a dashed line square) right before fixation. (b) Left: correlation of light microscopy signal of a fusion event observed by live imaging with EM pictures of the first section of the cell facing the coverslip (low magnification). Right: correlation of light microscopy signal with the first slice of the electron tomographic reconstruction of the first section of the cell facing the coverslip. The orange circle indicates the error range (167 nm) of the correlation performed by eC-CLEM. (c) 3D model of the electron tomographic reconstruction. The ER is depicted in light violet. Dense compartments are depicted in brown. The structure of interest is depicted in red and orange. (d) Bottom side view of the 3D model of the compartment of interest in its surroundings. The white arrow indicates the opening of the MVB where ILVs are released. (e) 3D model showing the MVB isolated from its environment. ILVs secreted through the opening of the MVB are depicted in white. (f) Top view of the secretory profile of the MVB that correlates with the fluorescence burst of the CD63-pHluorin fusion event.
Figure 3.
Figure 3.
MVB–PM fusion is distinct from other forms of vesicle-mediated exocytosis. (a) Schematic model showing the markers used in this study for the different types of cargo delivery of vesicles fusing with the PM. (b) Time-lapse imaging (heat maps) of a fusion event of the exosomal protein CD63-pHluorin. (c) Time-lapse images of soluble (NPY-pHluorin) and membrane protein (VAMP2-pHluorin) fusion events. (d) Fluorescent signal duration of NPY (mean = 0.85 s), VAMP2 (mean = 2.12 s), and CD63 (mean = 106.55 s) fusion events. n ≥ 13 events per reporter. (e) 3D heat maps of three consecutive CD63-pHluorin fusion event frames. (f) Western blot for exosomal markers (CD63 and Alix) on EVs purified from the supernatant (soluble) and EVs attached to the cell surface (PM attached) isolated after short trypsinization of the cells. (g) Direct comparison between signal duration of fusion events of CD81- and CD9-pHluorin relative to CD63-pHluorin. n ≥ 20 events per reporter. ***, P < 0.001; ****, P < 0.0001 using Student’s two-tailed two-sample t test.
Figure 4.
Figure 4.
GPCR activation triggers MVB–PM fusion in single cells in a calcium-independent manner. (a) Schematic model of imaging setup. (b) Fusion activity of HeLa cells stimulated with KCl (70 mM), caffeine (20 mM), or histamine (100 µM). n ≥ 8 cells per condition. (c) Total projection of fusion events over a 60-s time course onto cells before (top) and after (bottom) stimulation with histamine (100 µM). Pseudocolored as in Fig. 1 j. (d) Measurement of individual HeLa cells (n = 14) before and during stimulation with histamine (100 µM). (e) Mean fusion kinetics of CD63-pHluorin HeLa cells (n = 6) showing the distribution of fusion events over time (dark blue line; SD is in light blue) and the calcium levels (red) during histamine stimulation (gray-shaded block). (f) Heat maps revealing calcium responses (measured by Fluo-4) upon histamine stimulation obtained by subtracting the fluorescent intensity values before stimulation from those after 8-s stimulation. Cells were nontreated or incubated with a buffer with fast (BAPTA) or slow (EGTA) calcium-binding kinetics. Bars, 10 µm. (g) Quantification of fusion activity of histamine-stimulated HeLa cells in the presence of EGTA (top) or BAPTA (bottom) buffers. n ≥ 10 cells per condition. (h) Measurement of individual HUVEC cells (n = 30) before and after stimulation with histamine (100 µM). *, P < 0.05; **, P < 0.01 using Student’s two-tailed two-sample t test. All t tests were unpaired except for d and h. Whiskers in the box plots (b and g) represent 1.5 times the interquartile distance or the highest or lowest point, whichever is shorter.
Figure 5.
Figure 5.
The GPCR downstream effector SNAP23 regulates MVB–PM fusion. (a) Network of proteins of interest together with direct interactors with altered phosphorylation levels upon histamine (100 µM) stimulation as identified by phosphoproteomics in HeLa and HUVEC cells. Proteins of interest are depicted with a blue rim. FC, fold change. (b) Graph showing the signal intensity values of phosphorylated peptides from proteins of interest before and after stimulation with 100 µM histamine. Data represent means ± SD of two technical replicates per condition. *, P < 0.05 (P = 0.048) using Student’s two-tailed two-sample t test. (c) Western blotting analysis on SNAP23 protein expression in six different cell lines. (d) Confocal analysis of FL (GFP-SNAP23-FL) and truncated (GFP-SNAP23-CΔ9) GFP-SNAP23 (in gray)–transfected SiHa cells labeled for CD63 (red). (e) Total projection of fusion events in CD63-pHluorin SiHa cells cotransfected with SNAP23-FL or SNAP23-CΔ9 over 3 min. Pseudocolored as in Fig. 1 j. Bars, 10 µm. (f) Quantification of fusion events in CD63-pHluorin SiHa cells cotransfected with SNAP23-FL or SNAP23-CΔ9. n ≥ 10 cells per condition. (g) Confirmation of SNAP23 knockdown (KD) at the protein level in HeLa cells. (h) Effect of SNAP23 knockdown on MVB–PM fusion in HeLa cells. n ≥ 17 cells per condition. (i) Confirmation of SNAP23 and syntaxin-4 knockdown in HeLa cells at the mRNA level. Data represent means ± SD. (j) Effect of the knockdown of SNAP23 or syntaxin-4 on the fusion activity of HeLa cells. n ≥ 11 cells per condition. ctrl, nontransfected; siCTRL, control siRNA. *, P < 0.05; **, P < 0.01 using Student’s two-tailed two-sample t test. Whiskers in the box plots (f, h, and j) represent 1.5 times the interquartile distance or the highest or lowest point, whichever is shorter.
Figure 6.
Figure 6.
GPCR activation triggers MVB–PM fusion in HeLa cells via SNAP23-Ser110 phosphorylation. (a) Fusion activity of histamine-stimulated cells nontreated or treated with the Gαq inhibitor UBO-QIC (1 µM). n ≥ 24 cells per condition. (b) Basal fusion activity in cells treated with PKC inhibitors GÖ6976 (1 µM) or GÖ6983 (1 µM). n ≥ 11 cells per condition. (c) Fusion activity of histamine-stimulated cells nontreated or preincubated with GÖ6983 (1 µM). n ≥ 11 cells per condition. (d) Schematic representation of SNAP23 with SNARE motifs, a membrane-anchoring domain (M), and all phosphosites with the posphosite targeted by histamine stimulation (Ser110) in bold. (e) Fusion activity of histamine-stimulated cells transfected with WT SNAP23, phosphomutant SNAP23-S110A, or phosphomimic SNAP23-S110D. n ≥ 16 cells per condition. (f) Left: fusion activity of CD63-, CD81-, and CD9-pHluorin HeLa cells cotransfected with SNAP23 WT or SNAP23-S110A. n ≥ 16 cells per condition. Western blot on exosomes isolated from SNAP23-WT and SNAP23-S110A HeLa cells labeled for CD63, CD9, CD81, flotillin-1, and syntenin-1. (g) Schematic representation of the histamine-stimulated pathway leading to exosome release as identified by phosphoproteomics and specific inhibitors. Blue-rimmed proteins represent the putative pathway implicated by both experiments. The IP3–Ca2+ pathway is represented in gray as a direct link with MVB–PM fusion is missing. **, P < 0.01; ****, P < 0.0001 using Student’s two-tailed two-sample t test. All t tests were paired except for b and f. Whiskers in the box plots in b and f represent 1.5 times the interquartile distance or the highest or lowest point, whichever is shorter.

Similar articles

See all similar articles

Cited by 27 articles

See all "Cited by" articles

References

    1. Baglio S.R., van Eijndhoven M.A.J., Koppers-Lalic D., Berenguer J., Lougheed S.M., Gibbs S., Léveillé N., Rinkel R.N.P.M., Hopmans E.S., Swaminathan S., et al. 2016. Sensing of latent EBV infection through exosomal transfer of 5'pppRNA. Proc. Natl. Acad. Sci. USA. 113:E587–E596. 10.1073/pnas.1518130113 - DOI - PMC - PubMed
    1. Bolte S., and Cordelières F.P. 2006. A guided tour into subcellular colocalization analysis in light microscopy. J. Microsc. 224:213–232. 10.1111/j.1365-2818.2006.01706.x - DOI - PubMed
    1. Colombo M., Raposo G., and Théry C. 2014. Biogenesis, secretion, and intercellular interactions of exosomes and other extracellular vesicles. Annu. Rev. Cell Dev. Biol. 30:255–289. 10.1146/annurev-cellbio-101512-122326 - DOI - PubMed
    1. Contento R.L., Molon B., Boularan C., Pozzan T., Manes S., Marullo S., and Viola A. 2008. CXCR4-CCR5: a couple modulating T cell functions. Proc. Natl. Acad. Sci. USA. 105:10101–10106. 10.1073/pnas.0804286105 - DOI - PMC - PubMed
    1. Cox J., and Mann M. 2008. MaxQuant enables high peptide identification rates, individualized p.p.b.-range mass accuracies and proteome-wide protein quantification. Nat. Biotechnol. 26:1367–1372. 10.1038/nbt.1511 - DOI - PubMed

Publication types

MeSH terms

Feedback