Table 1.
qPCR primers for Kinesin-1 cargo adaptor proteins.
Fig 1.
Live-cell imaging of cytoskeletal dynamics and granule movement during stimulation of mast cells.
RBL-2H3 cells were transfected with Lifeact-GFP (A and B) or EB3-GFP (C and D) to label nascent F-actin and microtubules respectively, and granules were stained with LysoTracker Red. A) Representative confocal images from se S1 Video (https://doi.org/10.6084/m9.figshare.19349555.v1) of F-actin remodeling and granule movement during a time course after antigen-stimulation. B) Magnified images from the 5 min time point at 10 s intervals showing the movement of a granule into a pre-existing actin protrusion (arrow), and newly forming lamellipodia devoid of granules (asterisk). C) Representative confocal images from S2 Video (https://doi.org/10.6084/m9.figshare.19349552.v1) of nascent microtubule formation and granule movement during a time course after antigen-stimulation. D) Magnified images from the 5 min time point at 10 s intervals showing the movement of granules (arrows) is coordinated with the growth of microtubules (arrowheads). Scale bar, 10 μm.
Fig 2.
Microtubule drugs inhibit mast cell granule exocytosis and affect granule motility.
A) Exocytosis assay of RBL-2H3 cells that were pre-treated with nocodazole, colchicine and paclitaxel for 30 min prior to antigen-stimulation. Exocytosis was assayed as the percent β-hexosaminidase released of total, normalized to vehicle (DMSO) controls (n = 3). B–D) Representative confocal images from live-cell imaging of microtubule dynamics and granule movement of mast cells treated with 1 μM nocodazole (B, S3 Video via https://doi.org/10.6084/m9.figshare.19349558.v1), 10 μM colchicine (C, S4 Video via https://doi.org/10.6084/m9.figshare.19349564.v1), or 10 μM paclitaxel (D, S5 Video via https://doi.org/10.6084/m9.figshare.19349567.v1). RBL-2H3 cells were transfected with EB3-GFP to label nascent microtubules and incubated with LysoTracker red to label granules. Cells were imaged for 1 min, then antigen-stimulated and concurrently drugs were added, followed by 15 min of imaging. Scale bar, 10 μm; * indicates untransfected cells. E) Granules at the cell periphery were analyzed by particle tracking software. A minimum of 15 granules from 5 cells were tracked. Shown is the mean granule speed +/- standard error. The effect of drugs was compared to the vehicle (0.5% DMSO) control by Student’s t-test (NS, not significant; **, p = 0.0074; n = 5).
Fig 3.
Kinesore, a small-molecule activator of microtubule motors, inhibits mast cell granule exocytosis.
A) Exocytosis assay of RBL-2H3 cells and BMMCs that were pre-treated with kinesore for 30 min prior to antigen-stimulation. Exocytosis was assayed as the percent β-hexosaminidase released of total, normalized to vehicle (DMSO) controls. 100 μM kinesore showed statistically significant inhibition of both RBL-2H3 cell and BMMC exocytosis via Student’s t-test (**, p = 0.0104 and 0.0088 for RBL-2H3 and BMMC respectively; n = 3). B) Representative confocal images from live-cell imaging of microtubule dynamics and granule movement of mast cells treated with 100 μM kinesore (see S6 Video via https://doi.org/10.6084/m9.figshare.19349573.v1). RBL-2H3 cells were transfected with EB3-tdTomato to label nascent microtubules and incubated with LysoTracker green to label granules. Cells were imaged for 1 min, then antigen-stimulated and concurrently 100 μM kinesore was added, followed by 15 min of imaging showing the movement of granules to the perinuclear region while microtubules project to the cell periphery. C) Magnified images from the 5 min– 15 min time points showing the formation of EB3-tdTomato puncta are not affected by kinesore. D) Representative confocal images from live-cell imaging of actin dynamics and granule movement of mast cells treated with 100 μM kinesore (see S7 Video via https://doi.org/10.6084/m9.figshare.19349579.v1). RBL-2H3 cells were transfected with Lifeact-Ruby to label nascent F-actin and incubated with LysoTracker green to label granules. Cells were imaged for 1 min, then antigen-stimulated and concurrently 100 μM kinesore was added followed by 15 min of imaging. Scale bar, 10 μm; * indicates untransfected cells.
Fig 4.
Microtubule drugs and the microtubule motor drugs, kinesore, affects cell morphology and granule distribution but not F-actin remodeling.
RBL-2H3 cells were pretreated for 30 min with vehicle (A, DMSO), 1 μM nocodazole (B), 10 μM colchicine (C), 10 μM paclitaxel (D), or 100 μM kinesore (E). Cells were left unstimulated or antigen-stimulated for 30 min, fixed and stained for nuclei (blue), F-actin (green), or CD63+ granules (red). Scale bar, 10 μm. Profile plots (right panels) show the levels of CD63+ granules distributed in the indicated cross-sections. Drug pretreatment did not affect the formation of F-actin rich lamellipodia after stimulation, however CD63+ granules accumulated in the perinuclear region of drug treated cells (B’-E’, red), while control cells (A’, red) showed CD63+ granules were distributed across the entire cross-section.
Table 2.
Quantification of perinuclear-retained CD63 signal in antigen-stimulated RBL-2H3 cells.
Fig 5.
Knock-down of the kinesin-1 heavy chain, Kif5b, inhibits granule translocation to the plasma membrane and exocytosis.
A) Comparison of Kif5b levels in different RBL-2H3 knock-down strains. Kif5b knock-down strains were generated with three different specific shRNAs and a non-specific scrambled shRNA as a control. Immunoblot analysis shows reduction in Kif5b in all three strains treated with specific shRNAs, with strain 479 showing the greatest reduction. Tubulin levels remained similar. B) Exocytosis assay of RBL-2H3 strains that were unstimulated (black bars), or antigen-stimulated for 30 min (hatched bars). The Kif5b knock-down strain showed a significant reduction in exocytosis compared to the scrambled control and wild-type strains. Exocytosis was assayed as the percent β-hexosaminidase released of total, normalized to wild-type stimulated samples. Statistical analysis was by two-tailed unpaired Student’s t-test (NS, not significant; ***, p = 0.0043; **, p = 0.029; n = 4). C and D) Immunofluorescence microscopy of RBL-2H3 strains showing the intracellular distribution of CD63+ granules. Control cells (C) or Kif5b knock-down cells (D) were left unstimulated or antigen-stimulated for 30 min, fixed and stained for nuclei (blue), microtubules (green), or CD63+ granules (red). Scale bar, 10 μm. Profile plots (right panels) show the levels of CD63+ granules distributed in the indicated cross-sections. Kif5b knock-down did not affect microtubule projections after stimulation, however CD63+ granules accumulated in the perinuclear region (D’, red), while in control cells CD63+ granules were distributed across the entire cross-section (C’, red).
Fig 6.
Kinesore treatment of mast cells affects microtubule structures independent of its effect on granule distribution.
A) RBL-2H3 cells were pretreated for 30 min with vehicle (DMSO), or 100 μM kinesore. Cells were left unstimulated, or antigen-stimulated for 30 min, fixed and stained for nuclei (blue), microtubules (green), or CD63+ granules (red). Granule distribution to the cell periphery is disrupted by kinesore treatment. Microtubule structures that normally project linearly to the cell periphery were disrupted by kinesore treatment. B) RBL-2H3 cells were pre-treated for 0 min, 15 min, 30 min or 60 min with kinesore then fixed (unstimulated), or antigen-stimulated for 30 min then fixed and stained for nuclei (blue) and microtubules (green). Microtubule linear structures were disrupted by kinesore after 30 min. Scale bar, 10 μm.
Fig 7.
Kinesore inhibits granule association of the microtubule motor kinesin-1.
A) Quantitative PCR analysis of mRNA isolated from RBL-2H3 cells. Gene expression level of cargo adaptors known to associate with secretory granules and lysosomes. B) Immunoblot of lysates from RBL-2H3, PC-12, HeLa and HEK293T cells (107 cells/ml lysate). Lysates were probed for Kif5b, and two cargo adaptors, Slp3 and SKIP. SKIP was not detected in RBL-2H3 cells. C) Granule-enriched fractions were prepared by differential centrifugation from unstimulated cells (0’), or from cells that were antigen-stimulated for 15 min and 30 min in the presence of kinesore or vehicle (0.5% DMSO). Fractions were probed by immunoblot for association of the microtubule motor subunit Kif5b, the cargo adaptor Slp3, the granule enzyme rat mast cell protease II (RMCP II) and the mitochondrial marker ATPIF1 (upper panels). Tubulin was not associated with the granule enriched fraction and instead was found in supernatant (lower panels). D) Levels of protein associated with granule-enriched fractions were examined by band densitometry of immunoblots. Values were normalized to unstimulated cells for each experiment (n = 3).