VAMP8-dependent fusion of recycling endosomes with the plasma membrane facilitates T lymphocyte cytotoxicity.
Although multiple SNARE proteins have been implicated in cytotoxic granule exocytosis, the role of vesicular SNARE proteins, i.e., vesicle-associated membrane proteins (VAMPs), remains enigmatic.In primary human CTLs, however, VAMP8 colocalized with Rab11a-positive recycling endosomes.Our findings imply that secretory granule exocytosis pathways in other cell types may also be more complex than previously appreciated.
Affiliation: Cellular Neurophysiology, Center for Integrative Physiology and Molecular Medicine, Saarland University, 66421 Homburg, Germany Department of Medicine, Center For Infectious Medicine, 14186 Stockholm, Sweden.
- Cell Membrane/metabolism*
- R-SNARE Proteins/physiology*
- T-Lymphocytes, Cytotoxic/immunology*
- Cell Degranulation
- Cells, Cultured
- Cytotoxicity, Immunologic
- Immunological Synapses/metabolism
- Membrane Fusion
- Protein Transport
- Qa-SNARE Proteins/metabolism
- Signal Transduction
© Copyright Policy
License 1 - License 2
fig6: VAMP8 is required for fusion of recycling endosomes and cytotoxic granules at the immune synapse. (A and B) Bead-stimulated human CD8+ T cells were transfected with siRNA, as indicated. VAMP8 expression was determined by Western blot analysis. (A) Expression of VAMP8 relative to GAPDH in transfected CTLs, as indicated, from one representative donor. (B) Graphs represent means, normalized expression of VAMP8 in transfected CTLs, as indicated, from four individuals. Bars indicate SDs. (C–G) Bead-stimulated human CD8+ T cells were transfected with control siRNA or siVAMP8#2 siRNA, as indicated, and mCherry-Rab11a (C and D) or granzyme B (GzmB)–mCherry (E–G) encoding constructs and imaged by TIRF microscopy on anti-CD3– and anti-CD28–coated coverslips 16–20 h after transfection. (C) Mean dwell time of mCherry-Rab11a vesicles in the TIRF plane per cell (n = 15). (D) Mean fluorescence dispersion events for mCherry-Rab11a vesicles in the TIRF plane per cell (n = 15, unpaired t test, **, P > 0.01). (E) Selected live-cell TIRF microscopy images of granzyme B–mCherry in representative CTLs transfected with siRNA as indicated. Fusion event are indicated with arrowheads; vesicle 1 is indicated by closed arrowheads, and vesicle 2 is indicated by open arrowheads. Bars, 5 µm. (F) Mean dwell time of granzyme B–mCherry vesicles in the TIRF plane per cell (n = 8). Error bars indicate SDs. (G) Mean fluorescence dispersion events for granzyme B–mCherry vesicles in the TIRF plane per cell (n = 15, unpaired t test, **, P > 0.01). Error bars indicate SDs. (H and I) Bead-stimulated human CD8+ T cells were transfected with siRNA, as indicated, and analyzed by flow cytometry 18 h after transfection. CTLs were stimulated with anti-CD3 and anti-CD28 antibodies and then surface stained with the anti-CD107a antibody. (H) Representative histograms of CD107a staining on gated CTLs, as indicated, experiments were repeated with six individual donors, and data are quantified in I. CD8+ cells were isolated and stimulated with CD3/CD28 beads for 48 h and then transfected with control or VAMP8 siRNA. After 18 h, the cells were supplemented with anti-CD3 antibody for 3 h. After stimulation, the lymphocytes were surface stained for CD107 expression. Percentages indicate the increase in CD107 expression on the cell surface compared to nonstimulated conditions. (I) Graphs depict the mean frequency of CTLs, transfected and stimulated as indicated, expressing surface CD107a in six individual donors (ANOVA, **, P > 0.01). Error bars indicate SDs. siCTRL, control siRNA.
To determine whether VAMP8 regulates human CTL vesicle exocytosis, we attempted to knockdown the expression of VAMP8 by transfection of CTL with four VAMP8-specific siRNAs. Western blot and densitometric analyses showed that VAMP8 expression was reduced by all siRNAs (Fig. 6, A and B). siVAMP8#2 consistently provided the strongest knockdown of VAMP8 expression (with a knockdown efficiency of 70.8 ± 12.6). Staining of transfected cells with a VAMP8-specific antibody confirmed that the expression of VAMP8 was uniformly reduced in CTLs (Fig. S5, A and B). siVAMP8#2 was therefore used for all the following experiments.