STIM1-Directed Reorganization of Microtubules in Activated Mast Cells

Changes in the spatial distribution of microtubules in resting and activated BMMCs. Resting cells (A, B; −Ag), cells activated by FcεRI aggregation (C, D; +Ag), pervanadate (E, F; +Pv), or thapsigargin (G, H; +Tg) were fixed and stained for β-tubulin. The preparations were imaged by laser scanning confocal microscopy. The stacks of confocal sections were deconvoluted and subjected to three-dimensional reconstruction. Resulting three-dimensional images viewed from top of the cells (A, C, E, G) and from the plane perpendicular to the plane of cell adhesion (B, D, F, H). Each pair (A–B, C–D, E–F, and G–H) represents the same cells. Scale bars, 5 μm.

Effect of extracellular Ca²⁺ on generation of microtubule protrusions. A, BMMCL cells activated by FcεRI aggregation in the presence (a, b; +Ca²⁺) or absence (c, d; –Ca²⁺) of extracellular Ca²⁺ (1.8 mM) were fixed and stained for β-tubulin. Scale bars, 20 μm (c) and 10 μm (d). Comparable magnifications are in (a, c) and in (b, d). B, Statistical analysis of the frequency of microtubule protrusions in BMMCL cells. Cells activated by FcεRI aggregation (+Ag), pervanadate (+Pv), or thapsigargin (+Tg) in the presence (+Ca²⁺) or absence (−Ca²⁺) of extracellular Ca²⁺. Three independent experiments were performed, each involving 500 cells, and examined for the presence of microtubule protrusions. Values indicate means ± SD, n = 3; ***p < 0.001.

Activation of mast cells increases the number of growing microtubules in cell periphery as determined by TIRFM time-lapse imaging. A, Time-lapse imaging of resting (a, b) and thapsigargin-activated (c, d) BMMCL cells expressing EB1-GFP. Still images of EB1 (a, c) and tracks of EB1 comets over 20 s created by maximum intensity projection of the 20 consecutive frames (b, d). Scale bar, 5 μm. B, Histogram of microtubule growth rates in cell periphery of resting (−Tg) and thapsigargin-activated (+Tg) cells. A total of 15 different cells were tracked in five independent experiments. Values indicate mean ± SE, n = 15; **p < 0.01; ***p < 0.001.

Characterization of mast cells with reduced level of STIM1. A, Immunoblots of whole cell lysates from BMMCs or BMMCL cells probed with anti-STIM1 and anti-actin (loading control) Abs. Control cells infected with empty pLKO.1 vector (Con), noninfected wild-type cells (WT), cells selected after KD of STIM1 by shRNA1 (KD1), or shRNA2 (KD2). Numbers under the blots indicate relative amounts of STIM1 normalized to control cells (Con) and to the amount of actin in individual samples (Fold). B, Comparison of STIM1 expression levels in control and STIM1 KD BMMCs or BMMCL cells. Values indicate means ± SD from independent experiments (n = 6 for controls; n = 3 for KD1; n = 5 for KD2).

Decreased expression of STIM1 inhibits the generation of microtubule protrusions in activated cells. A, Control BMMCs, carrying empty pLKO.1 vector (a, c) or STIM1 KD2 cells (b, d) were activated by thapsigargin, fixed, and stained for β-tubulin. Scale bars, 20 μm (a, b) and 10 μm (c, d). B, Statistical analysis of the frequency of microtubule protrusions in control cells (carrying empty pLKO.1 vector) and STIM1 KD2 cells activated by FcεRI aggregation (+Ag), pervanadate (+Pv), or thapsigargin (+Tg). Three independent experiments were performed, each involving 500 BMMCs or BMMCL cells, and examined for the presence of microtubule protrusions. Values indicate means ± SD, n = 3; **p < 0.01; ***p < 0.001.

KD of STIM1 prevents changes in microtubule dynamics in activated cells as determined by TIRFM time-lapse imaging. STIM1 KD2 cells were nucleofected with EB1-GFP and used for time-lapse imaging. A, Resting (a, b) and thapsigargin-activated (c, d) STIM1 KD2 cells. Still images of EB1 (a, c) and tracks of EB1 comets over 20 s created by a maximum intensity projection of 20 consecutive frames (b, d). Scale bar, 5 μm. B, Histogram of microtubule growth rates in the cell periphery of resting (STIM1 KD) and thapsigargin-activated (STIM1 KD +Tg) cells. A total of nine different cells were tracked in three independent experiments. Values indicate means ± SE, n = 9; *p < 0.05.

Phenotype rescue of STIM1 KD2 BMMCL cells after introduction of human STIM1. A, Localization of mCherry-tagged human STIM1 during TIRFM time-lapse imaging of resting cells expressing EB1-GFP. Still images of mCherry-hSTIM1 (a) EB1-GFP (b) and superposition of images (c; mCherry, red; GFP, green). Arrows indicate the same positions. Scale bar, 10 μm. B, STIM1 KD2 cells were nucleofected with mCherry-hSTIM1 (a–c) or mCherry vector alone (d–f; control) and activated by thapsigargin. Microtubules in fixed cells stained with anti-α-tubulin Ab (a, d). Fluorescence of nucleofected mCherry vectors (b, e). Superposition of images (c, f; tubulin, green; mCherry, red). The preparations were imaged by fluorescence microscopy; a–c and d–f represent the same cells. Scale bar, 10 μm. C, Statistical analysis of the frequency of microtubule protrusions in thapsigargin-activated control cells (1), STIM1 KD2 cells (2), STIM1 KD2 cells nucleofected with pYFP-hSTIM1 (3) and STIM1 KD2 cells nucleofected with pYFP empty vector (4). Three independent experiments were performed, each involving 500 (1, 2) or 100 (3, 4) cells examined for the presence of microtubule protrusions. Values indicate means ± SD, n = 3; ***p < 0.001. D, Changes in intracellular Ca²⁺ mobilization. KD2 cells were nucleofected with pYFP-hSTIM1 (green line) or with pYFP empty vector (red line). Nontransfected cells (pink line), cells transfected with pLKO.1 (blue line) and STIM1 KD2 cells (black line) served as controls. The arrow indicates activation by 2 μM thapsigargin. The extent of activation is expressed as a ratio of Fura Red fluorescence intensity induced with 406- and 488-nm lasers. Representative curves are plotted against time. The line below the asterisks indicates the time interval of significant differences between STIM1 KD2 cell transfected with pYFP-hSTIM1 or with pYFP empty vector; **p < 0.01; n = 3.

Effect of microtubule depolymerization on Ca²⁺ uptake and degranulation. A, Effect of nocodazole on Ca²⁺ uptake. BMMCs were treated or not with nocodazole (10 μM) for 30 min and then exposed to thapsigargin (2 μM; Tg) or BSS-BSA alone (Control) for 15 min in the presence of extracellular ⁴⁵Ca²⁺ (1 mM) and nocodazole (10 μM). B, Effect of nocodazole on degranulation. BMMCs were treated with nocodazole and exposed to thapsigargin as in A, and the release of β-glucuronidase was determined. Data in A and B represent means ± SD, n = 6–8; *p < 0.05; **p < 0.01.