Figure 6

Rab5 is essential for IgE/Ag-triggered compound exocytosis. (a) A scheme illustrating the distinct modes of exocytosis and how they are measured by the de-quenching of endocytosed FITC-dextran. Cells are incubated with FITC-dextran, which is incorporated into the SGs. However, because of the acidic pH of the SGs, the fluorescence of FITC is quenched and the granule has no fluorescence (shown here as a black color) (A–C, I). When IgE sensitized cells are triggered to degranulate by specific antigen (Ag), the FITC fluorescence signal changes as a function of the mode of exocytosis. During full exocytosis, formation of a fusion pore between the single SG and the plasma membrane allows efflux of protons leading to the alkalization of the SG lumen. As a result, the fluorescence of FITC de-quenches and the SG becomes fluorescent (shown here as a green color) (A, II). This results in a short (less than 5 sec) burst of fluorescence that rapidly dissipates as the granule membrane fully merges with the plasma membrane and the contained FITC-dextran is released and diluted in the external medium (A, III). During kiss-and-run exocytosis, again, formation of the fusion pore leads to alkalization of the SG and de-quenching of the intra-granular FITC (shown here as an area of green color) (B, II). However, since the SG rapidly detaches and re-acidifies, FITC fluorescence quenches and the event is reflected in a small and short-lived fluorescence change (B, III- IV). Compound exocytosis starts, like the full exocytic event, with a fluorescent burst that corresponds to the alkalization of the single SG that fused with the plasma membrane (C, II). However, rather than emptying quickly and losing fluorescence upon release of the FITC probe, the SG, that is still connected to the membrane via a stable fusion pore, fuses with another SG, whose quenched fluorescence then de-quenches as it gains access to the external milieu and alkalinizes (C, III-IV). This results in a larger and significantly longer-lived burst of fluorescence (C, IV) than is seen in the other forms of degranulation, before fluorescence is lost due to the release of the FITC probe from the fused SGs (C, V). Therefore, while both full exocytosis and kiss-and-run exocytosis are associated with short, small bursts of fluorescence, compound exocytosis is reflected by the appearance of bursts of large and sustained fluorescent structures. Occasionally, the fusion event itself can be detected by the gradual de-quenching of a SG that is adjacent to and has fused with a fluorescent, hence exocytosing, SG (C, III). (b) RBL cells were co-transfected with 15 μg NPY-mRFP and either 30 μg of pSilencer or 15 μg of shRab5A, and 15 μg of shRab5B/C, as indicated. Cells were sensitized with 1 μg/ml of IgE after loading with FITC-dextran (1 mg/ml) for 48 h. Cells were triggered by 50 ng/ml of DNP-HSA (Ag) and visualized by time-lapse fluorescence microscopy as described in the Materials and Methods. The images corresponding to the indicated time periods after triggering with Ag are from Supplementary Videos 1 (b, pSilencer) and 5 (b, shRab5A/B/C) on line. The white arrow and white circle in (b, pSilencer) indicate transient bursts of large yellow fluorescent structures, indicative of exteriorization of fused SGs. The white arrow in (b, shRab5A/B/C) points to tiny fluorescent structures that persisted in Rab5-knockdown cells, likely to be indicative of single granules being exteriorized without compound fusion. Of note, the intensity in this image was increased to allow detection of the signal. (c–h) RBL cells were co-transfected with 10 μg of mStr-CA Rab5A and 20 μg of either empty vector (CA Rab5A) or HA-wt-SNAP23 (CA Rab5A/SNAP23). Cells were sensitized with 1 μg/ml of IgE after loading with FITC-dextran (1 mg/ml) for 48 h. Cells were triggered by 50 ng/ml of DNP-HSA (Ag) and visualized by time-lapse fluorescence microscopy as described in the Materials and Methods. (c) The average number of compound exocytosis events/cell was determined based on data derived from 4 separate videos for each transfection and a total of 15 cells for CA Rab5A and 13 cells for CA Rab5A/SNAP23. *P = 0.026 (unpaired two-tailed Student’s t-test). (d) The percentage of cells that failed to develop any compound exocytosis events, or developed 1–3 events/cell or developed 4 or more compound exocytosis events/cell was calculated. (e) Images captured at the designated time points after triggering with Ag (presented in Supplementary Video 4), demonstrate the de-quenching of FITC-dextran fluorescence (shown in green) during the fusion between a quenched SG (left, shown in black) and an adjacent SG (right, shown in green) that already fused with the plasma membrane and is therefore de-quenched and visible. The insets are the enlargements of the boxed areas. The white arrows point to the position of the SG that is in the process of de-quenching as it merges with the exocytosing, already de-quenched SG. (f) Quantification of the fluorescence changes during the fusion process. The right SG displays a constant level of fluorescence for approximately four minutes (13–17 min), implying that it is connected to the plasma membrane and therefore de-quenched and is in the process of releasing cargo until cargo discharge is complete at time point 17:6 min. The left SG is initially quenched but displays a 6–10 fold increase in FITC fluorescence after merging with the exocytosing SG. FITC fluorescence then drops upon the release of the cargo. (g) A schematic presentation of the process of homotypic SG fusion that is shown in (e). (h) Presentation of both the red and green channels of the designated time points derived from Video 4, showing the position of mSTR-CA Rab5A (red) around the fusing SGs that contain FITC-dextran (green, when de-quenched).