VCAM-1 colocalizes with moesin and ezrin at the apical surface of activated HUVEC. Confluent HUVEC were activated with 20 ng/ml TNF-α for 20 h. Thereafter, cells were fixed, permeabilized, and stained with the anti-ezrin pAb 90/3 (a and i, green), anti-moesin pAb 95/2 (e, green), anti–VCAM-1 mAb P8B1 (b and f, red), or anti–VE-cadherin mAb Tea1/31 (j, red). Merged images are shown in c, g, and k, where colocalizations are observed (yellow). Images represent confocal laser scanning micrographs showing horizontal projections or the corresponding orthogonal section of the same field. Insets correspond to the amplified image of the zones pointed to by arrows. Colocalization histograms of green and red signals corresponding to these images are shown on the right (d, h, and l). The corresponding colocalization percentages are 65.7% (d), 67.6% (h), and 13.3% (l). Bar, 20 μm.

Distribution of endogenous VCAM-1, ICAM-1, and ezrin during lymphoblast TEM. (A) Expression of α4-integrin (thick line), and αL-integrin (thin line) on T lymphoblasts as determined by flow cytometry analysis. P3×63 (dotted line) was used as negative control. (B) Transendothelial migration assay of T lymphoblasts pretreated with the blocking anti-α4 mAb HP2/1, the blocking anti-αL mAb TS1/11, the mixture of anti-α4 plus αL mAb, or the anti–ICAM-3 mAb TP1/24 as negative control. Values correspond to the arithmetic mean ± SD of a representative experiment run by duplicate out of three independent ones. Statistically significant values, as defined by unpaired Student's t test, are indicated with * (P < 0.05) or ** (P < 0.015) compared with no mAb treatment. (C) T lymphoblasts were allowed to transmigrate across an activated HUVEC monolayer, and then cells were fixed, permeabilized, and stained with the anti–VCAM-1 mAb P8B1 (a–c, red), the anti-ICAM-1 mAb Hu5/3 (d–f, red), or the anti-ezrin pAb 90/3 (c and f, green). Representative confocal horizontal images showing an apical section of endothelium with adhered lymphoblasts on top (a and d) and a basal section with transmigrated lymphoblasts beneath the endothelium (b and e) are presented. DIC images are shown overlaid with VCAM-1 (a and b) or ICAM-1 staining (d and e). Arrows point to VCAM-1 or ICAM-1 clusters at the contact area. Representative orthogonal sections corresponding to the white line in panel a or panel d are shown in panels c and f, respectively. Green signal corresponds to ezrin staining both in lymphoblasts and endothelium. Arrowheads point to the sites of VCAM-1/ezrin or ICAM-1/ezrin clustering at the apical surface of endothelial cells. Bars: (a and b) 20 μm; (d and e) 8 μm.

Localization of VCAM-1 and ezrin at the contact area of activated HUVEC with leukocytes. (A) Expression of α4-integrin (thick line), and αL-integrin (thin line) on 4M7 cells as determined by flow cytometry analysis. P3×63 (dotted line) was used as negative control. (B) Adhesion to activated HUVEC of 4M7 cells pretreated with the blocking anti-α4 mAb HP2/1, the activating anti-b1 mAb TS2/16, or the blocking anti-αL mAb TS1/11. Values correspond to the arithmetic mean ± SD of a representative experiment run by triplicate out of three independent ones. Statistically significant values, as defined by unpaired Student's t test, are indicated with * (P < 0.005) or ** (P < 0.0001), compared with no mAb treatment. (C) 4M7 cells interacting with activated endothelial cells were fixed, permeabilized, and stained with the anti-ezrin pAb 90/3 (green) and the anti–VCAM-1 mAb P8B1 (red). Representative horizontal sections of confocal laser scanning images (b and c) are merged in d. The corresponding DIC image is shown in a. The corresponding three-dimensional reconstruction is presented in e–g. Bar, 5 μm. (D) 4M7 cells were allowed to adhere to activated HUVEC transfected with moesin- or VCAM-1–GFP. GFP staining was monitored using live time-lapse fluorescence confocal microscopy. Horizontal sections showing the staining of moesin- (a) or VCAM-1–GFP (b) after 60 min of leukocyte–endothelium interaction. Arrows point to the GFP proteins clustered in the anchoring structure. Bar, 20 μm.

Characterization of the endothelial docking structure formed during leukocyte adhesion. (A) 4M7 cells were allowed to adhere to activated HUVEC cells, then fixed, permeabilized and stained for VCAM-1 (a, green), F-actin (b–d, f, and h, red), and tubulin (e and g, green). Representative horizontal sections of confocal image-stacks are presented in a and b and e and f. The panel c shows the projection of all the horizontal sections corresponding to the image presented in a and b. Three-dimensional reconstructions of F-actin (d and h) and tubulin (g) stainings are also shown. Bars, 20 μm. (B) 4M7 cells adhered to activated HUVEC were fixed, permeabilized, and stained for talin (a and b, green), VCAM-1 (a and b, red), or vinculin (c and d, green). Representative horizontal sections of confocal images are presented in a and c. Three-dimensional reconstructions are shown in b and d. Bar, 5 μm. (C) 4M7 cells were allowed to adhere to activated HUVEC transfected with α-actinin, paxillin-, and VASP-GFP. Thereafter, cells were fixed and stained with the anti-VCAM-1 mAb P8B1 (a, d, and f, red). Green signal corresponds to GFP fusion proteins (b, c, e, and g). Representative horizontal sections of confocal image stacks are presented in all panels except for c, which shows the three-dimensional reconstruction of α-actinin–GFP signal. Bars, 5 μm.

Rho/p160 ROCK pathway regulates the generation and maintenance of the VCAM-1–mediated docking structure. (A) Activated HUVEC were pretreated with Y27632 (30 μM), Ly294002 (20 μM), or Gö6976 (1 μM) for 20 min before or 30 min after the addition of 4M7 cells. Total adhesion time was in both cases 60 min. Quantification of leukocyte adhesion and endothelial docking structure formation was carried out by staining with the mAb anti–VCAM-1 P8B1 and counting 300 adhered cells of each treatment. A representative experiment out of four independent ones is presented. (B) Kinetics of the endothelial anchoring structure dissolution after the addition of the p160 ROCK inhibitor Y-27632. The inhibitor Y-27632 (30 μM) was added after the formation of the docking structure in a 4M7 adhesion assay performed as above. Representative horizontal sections captured every 15 min are shown in a–d. DIC and fluorescence images are merged and presented in the lower side of each panel. Arrows indicate the clustering of GFP proteins. Bar, 10 μm. (C) Effect of Y-27632 on peripheral blood lymphocyte adhesion and TEM under flow conditions. Activated endothelium was pretreated or not with Y-27632 (30 μM) for 30 min. Thereafter, PBLs were allowed to adhere and transmigrate under flow conditions for 10 min. Quantification of rolling (R), adhesion (A), transmigration (T), and detachment (D) events during the last minute of perfusion was carried out. Values correspond to the arithmetic mean ± SD of four different fields belonging to a representative experiment. Statistically significant values, as defined by unpaired Student's t test, are indicated with *(P < 0.01) or **(P < 0.002) compared with no inhibitory treatment.

Association of VCAM-1 with moesin and ezrin. (A) Cytokine-activated HUVEC were lysed and immunoprecipitated with the anti–VCAM-1 mAb 4B9 or Gly-Sepharose. Immunoprecipitates were then resolved on a 10% SDS-PAGE, and sequentially immunoblotted with the anti-VCAM-1 mAb 4B9, the anti-moesin pAb 95/2, and the anti-ezrin pAb 90/3. Molecular weights (kD) are indicated on the right side. (B) GST or the GST fusion proteins GST–VC, GST–IC3 and GST–Y9 were bound to glutathione-Sepharose beads and incubated with ³⁵S-Met-N-moesin or ³⁵S-Met-N-ezrin (top and bottom, respectively). After incubation, beads were boiled in sample buffer and eluted proteins were analyzed by 10% SDS-PAGE, autoradiography, and fluorography. Lanes with isotope-labeled N-moesin and N-ezrin (*) indicate the molecular mass of these truncated proteins. Densitometric diagrams normalized to the loading controls of GST proteins are shown on the right.

Dynamic changes in the localization of VCAM-1, ICAM-1, and moesin during lymphoblast TEM. (A) Lymphoblasts were allowed to transmigrate across activated HUVEC transfected with moesin- or VCAM-1–GFP. Video sequences tracking the spatial and temporal distribution of moesin (a–f) and VCAM-1 (g–l) were obtained using live time-lapse fluorescence confocal microscopy. Each image represents a projection of several representative horizontal sections of a confocal image stack depicted from the video sequence at the specified times. DIC and fluorescence images are merged and presented at the lower side of each panel. Arrows point to the GFP proteins clustering during the lymphoblast–endothelium interaction. Arrowheads indicate the absence of VCAM-1–GFP from the contact area between the transfected endothelial cell and a migrated lymphoblast placed beneath the endothelial monolayer. Corresponding digital video sequences are available at http://www.jcb.org/cgi/content/full/jcb.200112126/DC1. Bars, 20 μm. (B) Lymphoblast transmigration across activated HUVEC transfected with ICAM-1–GFP was analyzed by live time-lapse fluorescence confocal microscopy. Two representative horizontal sections from the apical and the basal side of the endothelial cell belonging to the same confocal stack depicted from the video sequence are presented. ICAM-1–GFP signal is shown in panels a and d. DIC images and the overlaid images are presented in panels b and e, and c and f, respectively. The corresponding video sequence is available at http://www.jcb.org/cgi/content/full/jcb.200112126/DC1. Bar, 5 μm.

Formation of the endothelial docking structure for adhered lymphocytes under flow. (A) Transendothelial migration assay of peripheral blood lymphocytes under fluid shear conditions. After 10 min of perfusion, cells were fixed and stained for ICAM-1 (b and d, green), VCAM-1 (c, d, g, and h, red), and ezrin (f and h, green). The corresponding DIC images are shown in panels a and e. Merged images are shown in panels d and h. Bar, 3.5 μm. (B) Three-dimensional reconstruction of VCAM-1 staining during 4M7 cell (a), T lymphoblast (b), or PBL under flow (c) adhesion.

Localization of phosphorylated ERM proteins and phosphoinositides at the anchoring structure. (A) 4M7 cells adhered to activated HUVEC were fixed, permeabilized and stained with the mAb 297S. Representative horizontal section of a confocal micrograph (a), and a three-dimensional reconstruction of a series of horizontal sections (b) are shown. Bar, 5 μm. (B) HUVEC cells were transfected with PLCs-PH- and GRP1-PH-GFP and then, 4M7 cells were allowed to adhere. Thereafter, cells were fixed, permeabilized and stained with the anti-VCAM-1 mAb P8B1 (a and d). Green signal corresponds to GFP fusion proteins (b and e). Panels c and f show three-dimensional reconstructions of horizontal sections corresponding to the green signal. Bars, 5 μm.