Targeted recycling of PECAM from endothelial cell surface-connected compartments during diapedesis
Table 1 explanatory text
We sought a complementary approach to that used in Figure 4 to study the redistribution of endothelial cell PECAM during TEM. Synchronized TEM assays were run in the absence of antibody or label of any kind and fixed at early time points (10-15 minutes) in order to catch monocytes in the act of transmigration. Permeabilized monolayers were then stained for PECAM expression. It is important to bear in mind that in this experiment a thick, reticular junction represents one in which membrane has not discharged into the junctional space. It is visualized by staining total PECAM on both the surface and in the intracellular membranes in fixed, permeabilized cells. In contrast, in text Figure 4a, an intensely staining junction was observed when membrane recycled because recycled PECAM was being selectively visualized by extracellular antibody applied to intact cells.
In HUVEC monolayers not exposed to monocytes, >80% of the cell borders displayed thick, reticular junctional staining, representing the stores of internalized PECAM-bearing membrane, as exemplified by text Figure 1b. When we examined HUVEC monolayers exposed to monocytes, >80% of the cell borders that were not being actively traversed by monocytes had a similar thick “reticular” staining pattern. This suggests that along these borders, PECAM recycling was proceeding constitutively, as in resting cells. In contrast, where monocytes were actively crossing, <19% of the junctions were “reticular”; the majority had a thin, but intense band of PECAM staining surrounding the transmigrating monocyte (Table 1, left column). This suggests that in these zones of diapedesis, the stores of PECAM in the surface-connected compartments had been (temporarily) depleted. Portions of the same cell border at a distance from the monocyte still bore reticular staining.
These results are most consistent with the hypothesis that near the zone of diapedesis PECAM is recycled directly to the endothelial cell-monocyte interface rather than recycling randomly to the junction followed by translocation in the plane of the plasmalemma to surround the monocyte. The ratio of monocytes to endothelial cells in this experiment was 1:1, and > 80% of the junctions were originally “reticular”, so monocytes did not choose to transmigrate thin junctions. If they had, the overall ratio of thick to thin junctions would not have changed in those cultures.
In the same experiments when TEM was blocked by anti-PECAM mAb hec71 we found that monocytes were arrested primarily over reticular type junctions, as expected if the network of sequestered membrane had not been exteriorized en mass, but was still mostly within the cell (Table 1, right column). The few monocytes that escaped the block and did transmigrate under these conditions were associated with thin junctions, as expected if the network of perijunctional membrane had returned to the cell border.
(Table 1 and its legend follow on the next page.)
Localized Loss of Reticular PECAM Staining Pattern Around Transmigrating Monocytes Percent “Reticular” Junctions
Transmigration Blocked by anti-PECAM mAb
Borders without monocytes
Borders with monocytes
(Table 1 Legend)
Synchronized TEM assays were fixed, permeabilized, and stained for PECAM expression. The staining pattern of the junctions was classified as either thin (similar to Fig. 1a) or reticular (similar to Fig. 1b). “Borders with monocytes” means that only those borders at which a monocyte was present were scored, while in “Borders without monocytes” borders of endothelial cells that had no monocytes either bound or transmigrating were scored. In resting endothelial monolayers to which no monocytes had been added 83.5 3.5% of the borders were reticular. Data are the mean SEM of three independent experiments in which a total of 3328 individual cell borders were scored. * p < 0.0001 vs. all others.
Human transferrin (Sigma) was iron-loaded and purified as described previously2. Transferrin was conjugated to Alexa Fluor 546 according to the manufacturer’s instructions (Molecular Probes, Eugene, OR). Rabbit polyclonal antibody against caveolin-1 was obtained from Transduction laboratories. Fluorescent goat anti-rabbit and goat anti-mouse Abs were purchased from Molecular Probes. The murine monoclonal antibody against decay accelerating factor (DAF) was a gift from Dr Stephen Tomlinson (Medical University of South Carolina).
Immunoelectron Microscopy Horseradish peroxidase (HRP) was conjugated to mAb P1.1 by published procedures3 and conjugated antibody was purified by FPLC. Fractions containing a molar ratio of 1-2 HRP molecules per IgG were used for the experiments. HUVEC monolayers grown on cover slip dishes4 were incubated for 1 hr. on ice or at 37C with HRP-P1.1 at 20 g ml-1 or the equivalent concentration of free HRP. The monolayers were washed in PBS and fixed in 4% glutaraldehyde (EM sciences, Ft. Washington, PA) in 0.1M sodium cacodylate buffer for 20 min on ice. Labeled antibody was detected with diaminobenzidine (DAB)-H2O2 by the method of Graham and Karnovsky5. The cells were postfixed in glutaraldehyde, embedded in plastic, and cross-sections and en face sections were cut and examined on a JEOL JSM 100CX II electron microscope.
Stereologic analysis Electron micrographs of representative randomly selected en face sections from specimens incubated at 37 and 4 were printed and overlaid with a lined grid. Quantitation was performed as described by standard stereologic techniques6. Briefly, the number of times the random lines crossed DAB-labeled internal membrane compartments or junctional membrane were recorded. The ratio is proportional to their relative surface area.
PECAM-1 and Transferrin co-localization:
HUVEC monolayers were incubated with 10 g ml-1 Alexa Fluor 546-transferrin for 1h at 37C. The cells were rinsed with cold divalent cation-containing PBS, fixed in 2% freshly prepared paraformaldehyde and permeabilized in 0.1% Triton X100 for 30 min on ice. PECAM-1 was visualized by indirect immunofluorescence using the murine P1.1 antibody specific for PECAM-1 and Alexa Fluor 488 goat anti-mouse antibody.
PECAM-1 and caveolin-1 co-localization:
Cells were incubated with Fab fragments of P1.1 antibody in conditioned medium for 1h at 37C. The cells were washed, fixed in 2% freshly diluted paraformaldehyde, permeabilized with 0.2% Triton X-100 for 10 min at room temperature and labeled with rabbit IgG against caveolin-1 for 1 hr on ice. Following extensive washing, the primary antibodies were detected with Alexa Fluor 488-labeled goat anti-mouse antibody, and Alexa Fluor 546-labeled goat anti-rabbit antibody, respectively, for an additional hour on ice.
HUVEC monolayers were washed twice with ice cold PBS and lysed in 2ml MES-buffered saline (MBS; 25mM Mes, pH 6.5, 0.15 M NaCl) containing 1% Triton X-100, 1mM PMSF and 1g ml-1 leupeptin7. Dishes were scrapped and cells were homogenized with 10 strokes of a loose-fitting Dounce homogenizer. The homogenate was mixed with an equal volume of 80% sucrose prepared in MBS, placed at the bottom of an ultracentrifuge tube and overlaid successively with 1ml of 35% sucrose, 1ml of 25% sucrose and 1ml of 15% sucrose in MBS lacking Triton X-100. The gradients were centrifuged at 39,000 rpm for 17h in a SW41 rotor (Beckman ultracentrifuge). 0.5 ml fractions were collected from top to bottom, diluted 3 fold in MBS containing Triton X-100and spun for 15 min. The pellets were solubilized in Tris saline buffer (150 mM NaCl, 10 mM Tris pH 8) containing 1mM PMSF and 1% Triton X-100. Samples containing equivalent amounts of proteins from each fraction were analyzed by Western blotting for PECAM-1 and Caveolin-1.
HUVEC monolayers were washed with ice-cold cytoskeleton stabilization buffer (CSB: 138 mM KCl, 3 mM MgCl2, 2 mM EGTA, 0.32 M sucrose, 10 mM MES, pH 6.1) and extracted on ice for 30min in CSB containing 0.5% Triton X-100 and protease inhibitors (10g ml-1 aprotinin, 0.1 mM PMSF). After washing in cold divalent cation-containing PBS, they were fixed in cold 6.6% paraformaldehyde freshly diluted in PBS for 10min8. The cells were then washed and labeled with mouse mAb against PECAM-1, ICAM-1 or DAF followed by Rhodamine-labeled rabbit anti-mouse IgG antibody. Control cells were washed with warm medium M199, fixed with 6.6% paraformaldehyde and exposed to the same antibodies as the triton extracted cells.
Fig 1S. PECAM-1 does not enter the endocytic recycling pathway. HUVEC monolayers were incubated with Alexa 546-transferrin for 1h at 37C to reach the steady state, then fixed, permeabilized and labeled with anti-PECAM-1 antibody, followed by Alexa 488-goat anti-mouse IgG. Confocal microscopy images of two representative fields (a-c) and (d-f) are shown. The labeling pattern of the endocytic compartment bearing transferrin and its receptor (middle column) is very different in appearance from the junctional staining pattern of PECAM (left column). This is confirmed in the merged images on the right. PECAM shown in green; transferrin shown in red. Bars = 10 m.
Fig 2S. PECAM-1 does not localize with caveolin-1. HUVEC cells were incubated with Fab fragments of P1.1 antibody against PECAM-1 for 1h at 37C, then fixed, permeabilzed for 10 min with 0.2% Triton X-100 and incubated with rabbit anti-caveolin-1 antibody followed by Alexa Fluor 488 goat anti-mouse IgG (against P1.1) and Alexa Fluor 546 goat anti-rabbit IgG (against anti-caveolin antibody). Images were taken by confocal microscopy. Representative single planes (a-c) and Z-series confocal projections (d-f) are shown. Bar = 10 m.
Fig 3S. PECAM-1 and Caveolin-1 do not co-migrate on sucrose gradients. Homogenates of HUVEC monolayers were centrifuged on a discontinuous sucrose gradient as described in Methods. Fractions were analyzed by Western blotting for caveolin-1 and PECAM-1. The majority of caveolin 1 (fraction 1) was recovered in an opalescent band just below the 15-25% interface. Other samples shown on the blot are the 15-25% interface (2), 25-35% interface (3), and 40% sucrose (4) fractions. Essentially all PECAM reactivity was recovered in fraction 4.
Fig 4S. PECAM-1 is completely extracted with cold Triton X-100. Confluent cultured monolayers of HUVEC were fixed (a-c) or extracted with cold triton then fixed (d-f) prior to staining for PECAM-1 (a,d), ICAM-1 (b,e), or DAF (c,f). Note that cells in (a) were permeabilized after fixation but before staining to enable labeling of total cellular PECAM. Representative images acquired by widefield microscopy show that PECAM is completely extracted by cold triton treatment. Inset in (d) shows fluorescence of matched culture extracted as in (d-f) but exposed to secondary antibody only. Internal controls show that, as expected, ICAM-1, which does not reside in detergent-resistant microdomains is similarly extracted, while DAF, a known component of these microdomains, is resistant to cold triton extraction. Bar = 20 m.
Fig. 5S. Schematic diagram of PECAM recycling constitutively and during transmigration. Panels a and b represent side views of the endothelial monolayer with the apical surface facing upward. Panels a’ and b’ – b’’’ represent en face sections of the monolayer at the level of the horizontal black line in a and b, respectively. Groups of interconnected 50 nm vesicle-like structures (the “subjunctional reticulum”) are located just beneath the plasma membrane at the intercellular borders. These are connected at intervals to the cell surface at the border. PECAM-1 is concentrated along the plasma membrane at the intercellular border as well as in the subjunctional reticulum (red shading). At 37C PECAM is constitutively cycling into this compartment from the surface membrane (black arrows in a and a’) and back out of the compartment (white arrows) evenly along the intercellular borders. The total amount of PECAM on the endothelial surface and in the surface-connected reticulum remains relatively constant. During diapedesis (b, b’) recycling of PECAM is targeted so that it is now directed (black arrows) to the sector of the endothelial cell border where the migrating leukocyte (WBC) is advancing. Panels b’’ and b’’’ represent hypothetical events a few seconds after b’. Recycling PECAM is delivered to the interface between the leukocyte and endothelial cell, so that recycling PECAM is enriched in this zone while the subjunctional reticulum is depleted in adjacent areas of the endothelial cell. Note that for clarity this diagram depicts PECAM recycling in conjunction with membrane of the subjunctional reticulum. However, our data do not rule out the possibility that the membrane of this compartment remains relatively stable while the PECAM protein itself recycles by moving within the plane of the continuous membrane.
References for Supplementary Material
1. Muller, W. A., Weigl, S. A., Deng, X. & Phillips, D. M. PECAM-1 is required for transendothelial migration of leukocytes. Journal of Experimental Medicine 178, 449-460 (1993).
2. Yamashiro, D. J., Tycko, B., Fluss, S. R. & Maxfield, F. R. Segregation of transferrin to a mildly acidic (pH 6.5) para-Golgi compartment in the recycling pathway. Cell 37, 789-800 (1984).
3. Ghosh, R. N., Mallet, W. G., Soe, T. T., McGraw, T. E. & Maxfield, F. R. An endocytosed TGN38 chimeric protein is delivered to the TGN after trafficking through the endocytic recycling compartment in CHO cells. J. Cell Biol. 142, 923-936 (1998).
4. Salzman, N. H. & Maxfield, F. R. Fusion accessibility of endocytic compartments along the recycling and lysosomal endocytic pathways in intact cells. J. Cell Biol. 109, 2097-2104 (1989).
5. Graham, R. C., Jr. & Karnovsky, M. J. The early stages of absorption of injected horseradish peroxidase in the proximal tubules of mouse kidney: ultrastructural cytochemistry by a new technique. J. Histochem. Cytochem. 14, 291-302 (1966).
6. Williams, M. A. in Quantitative Methods in Biology (ed. Glauert, A. M.) 5-84 (Elsevier/North Holland Biomedical Press, Amsterdam, 1976).
7. Lisanti, M. P., Scherer, P. E., Vidugiriene, J., Tang, Z., Hermanowski-Vosatka, A., Tu, Y., Cook, R. F. & Sargiacomo, M. Characterization of caveolin-rich membrane domains isolated from an endothelial-rich source: Implications for human disease. J. Cell Biol. 126, 111-126 (1994).
8. Seveau, S., Eddy, R. J., Maxfield, F. R. & Pierini, L. M. Cytoskeleton-dependent membrane domain segregation during neutrophil polarization. Mol Biol Cell 12, 3550-3560 (2001).