Abstract

Regulated P-selectin surface expression provides a rapid measure for endothelial transition to a proinflammatory phenotype. In general, P-selectin surface expression results from Weibel-Palade body (WPb) exocytosis. Yet, it is unclear whether pulmonary capillary endothelium possesses WPbs or regulated P-selectin surface expression and, if so, how inflammatory stimuli initiate exocytosis. We used immunohistochemistry, immunofluorescence labeling, ultrastructural assessment, and an isolated perfused lung model to demonstrate that capillary endothelium lacks WPbs but possesses P-selectin. Thrombin stimulated P-selectin surface expression in both extra-alveolar vessel and alveolar capillary endothelium. Only in capillaries was the thrombin-stimulated P-selectin surface expression considerably mitigated by pharmacologic blockade of the T-type channel or genetic knockout of the T-type channel α1G-subunit. Depolarization of endothelial plasma membrane via high K+ perfusion capable of eliciting cytosolic Ca2+ transients also provoked P-selectin surface expression in alveolar capillaries that was abolished by T-type channel blockade or α1G knockout. Our findings reveal an intracellular WPb-independent P-selectin pool in pulmonary capillary endothelium, where the regulated P-selectin surface expression is triggered by Ca2+ transients evoked through activation of the α1G T-type channel.

  • Weibel-Palade body
  • von Willebrand factor
  • thrombin
  • inflammation

the endothelium plays a critical role in inflammation by controlling leukocyte recruitment to the site of infection (46). As a component of the innate defense system, leukocyte recruitment in the lung is important for eradicating infections, repairing tissue injury, and restoring efficient gas exchange following bacterial pneumonia. Encumbering this process exacerbates the lung injury associated with bacterial pneumonia. Conversely, excessive leukocyte recruitment in the lung, such as that caused by an uncontrolled systemic inflammatory response following septic shock, contributes to the pathogenesis of multiple disorders, including acute lung injury and its most severe form, acute respiratory distress syndrome (48). Intracapillary leukocyte sequestration, activation, and emigration in the lung proceed in a sequence of overlapping events (39). In contrast to the well-characterized multistep adhesion cascade of leukocyte recruitment in the systemic circulation, whereby leukocytes first tether and roll on endothelial selectins, become activated, and then adhere firmly on activated integrins (4, 28, 29, 44), leukocyte rolling does not occur in the alveolar capillaries (26), which constitute the major site for leukocyte sequestration in the lung (7, 9, 10, 15, 26). To date, insufficient attention has been given to endothelial selectins with regard to their role in leukocyte trafficking in the lung, as even their expression in capillaries remains incompletely defined.

Endothelial P-selectin is intimately involved in the inflammatory response. Regulated P-selectin surface expression provides a rapid measure for endothelial activation and transition to a proinflammatory phenotype (16, 30, 45). P-selectin surface expression results from the exocytosis of the endothelial cell-specific secretory organelle, the Weibel-Palade body (WPb; Refs. 2, 20). In the systemic circulation, rapid P-selectin translocation to the endothelial cell membrane via WPb exocytosis promotes adhesive interactions with leukocytes (22). Yet, in the pulmonary circulation, the role of P-selectin in intracapillary leukocyte sequestration is controversial. On one hand, various studies have postulated that lung capillaries do not possess WPbs (14) and do not express P-selectin (12, 23, 37, 38, 52). On the other hand, acute lung inflammation is attenuated using P-selectin-deficient mice or anti-P-selectin antibodies (3, 8, 13, 34, 35), suggesting the presence of P-selectin. Thus it remains unclear as to whether pulmonary capillary endothelium possesses WPbs or demonstrates regulated P-selectin surface expression.

The exocytosis of WPbs evoked by naturally occurring Gq-linked inflammatory agonists is a Ca2+-dependent process that requires Ca2+ entry from the extracellular space (19). We previously demonstrated that pulmonary microvascular endothelial cells specifically express a voltage-gated, α1G (CaV3.1)-subtype T-type Ca2+ channel (50, 56). Thrombin-induced transitions in membrane potential activate this channel, leading to a rise in cytosolic Ca2+ ([Ca2+]i). Blockade of the T-type channel attenuates sickle cell retention in the inflamed pulmonary circulation (50, 56), a process that has been linked to endothelial P-selectin (47). In an in vitro study, we showed that release of von Willebrand factor (vWF), generally considered to colocalize with P-selectin in WPbs, is differentially controlled in pulmonary macro- and microvascular endothelial cells, where the α1G T-type channel mediates regulated vWF release exclusively in pulmonary microvascular endothelial cells (55). Moreover, our recent work (51) demonstrated that α1G-mediated cytosolic Ca2+ transients lead to endothelial P-selectin surface expression in alveolar septal capillaries, while having no impact on endothelial barrier integrity. Altogether, these findings support a notion that Ca2+ entry through the α1G T-type channel is implicated in the endothelial proinflammatory phenotype transition.

Here, we conducted a systematic study to assess the distribution of WPbs and the regulated P-selectin surface expression in pulmonary microvascular endothelium, with special emphasis on alveolar septal capillary endothelium. We resolved that in capillary endothelium the subcellular localization of P-selectin is WPb independent, and the surface expression of P-selectin is regulated by mechanism(s) distinct from WPb exocytosis. Specifically, Ca2+ entry through the endothelial site-specific α1G T-type channel controls the P-selectin surface expression in alveolar capillary endothelium.

MATERIALS AND METHODS

Animals.

Experimental protocols in rodents were approved by the Institutional Animal Care and Use Committee of the University of South Alabama. Male Sprague-Dawley rats were purchased from Charles River Laboratories (Wilmington, MA or Sulzfeld, Germany). The α1G-deficient mutation was originally described by Kim et al. (24). Mice homozygous for α1G-deficient (α1G−/−) and their wild-type littermates (α1G+/+) were generated by heterozygote mating. Male or female mice ranging 8–12 wk in age and 18–35 g in weight were used in the study.

Human specimens.

Archived blocks of normal human lung tissue, obtained by surgical excision or biopsy at the University of South Alabama Medical Center, were used in the transmission electron microscopy study. The use of human archived tissues for this study was approved by the University of South Alabama Medical Center Institutional Review Board.

Isolated lung preparation and experimental protocols.

Lungs were isolated for ex vivo perfusion as previously described (18, 51). Thrombin (Gq-linked agonist), mibefradil, and the antibody used for probing endothelial surface P-selectin expression were added to the perfusate, i.e., Earle's buffered salt solution containing 4% bovine serum albumin (4%BSA/EBSS), in the reservoir while the lungs were perfused. Mibefradil (10 μmol/l) was applied at least 10 min before the application of other agents. The lungs were perfused with thrombin (5 U/ml) for 10 min or high K+ (40 mmol/l) for 35 min, each alone or together with mibefradil. For immunohistochemistry, the lungs were rinsed with 4%BSA/EBSS, fixed in 4% paraformaldehyde in PBS by vascular perfusion, immersed in fixative, embedded in paraffin, and sectioned at 5 μm. For in situ confocal fluorescence microscopy, the lungs were rinsed with 4%BSA/EBSS and then perfused with an affinity purified FITC-conjugated rat anti-mouse P-selectin monoclonal antibody (35 μg/ml; BD Biosciences, San Jose, CA) in 4%BSA/EBSS for 30 min. Subsequently, viable lung slices were prepared using the method reported by Bergner and Sanderson (1). Briefly, following a 10-min flush with 4%BSA/EBSS via vascular perfusion, the pulmonary vessels were infused with 6% gelatin and the airways with 2% agarose (type VII-A, low gelling temperature; Sigma, St. Louis, MO) in HBSS at 37°C. The lungs were immediately placed at 4°C for 30 min to allow the gelatin and agarose to gel and then sectioned into ∼200 μm slices using a Vibratome 1000Plus manual tissue sectioning system (Technical Products International, St. Louis, MO). The lung slices were mounted in an Attofluor cell chamber (Molecular Probes, Eugene, OR), moisturized with a drop of 4%BSA/EBSS, and placed on the stage of a PerkinElmer UltraView RS confocal microscope (PerkinElmer Life and Analytical Sciences, Boston, MA). A ×60, 1.20 numerical aperture water immersion objective along with 488-nm excitation and 525-nm emission filter were used for FITC imaging. A set of serial optical sections (Z-stacks) was taken at 0.5-μm intervals.

Immunohistochemistry.

Immunohistochemistry for endothelial surface P-selectin was performed as we recently described (51). Immunohistochemistry was performed similarly for endothelial vWF and platelet glycoprotein GPIIb/IIIa. The primary antibodies and the dilutions used in these studies are as follows: 5 μg/ml goat anti-mouse P-selectin polyclonal antibody (R&D Systems, Minneapolis, MN); 1:250 rabbit anti-mouse vWF antibody (Dako, Glostrup, Denmark); 1:250 rabbit anti-human vWF antibody (Dako); and 5 μg/ml rat-anti-mouse integrin α2b (platelet glycoprotein IIb of IIb/IIIa complex) antibody (R&D Systems). Negative controls were obtained by omission of the primary antibody.

Double immunofluorescence labeling of vWF and P-selectin.

The 5-μm paraffin-embedded mouse lung sections were deparaffinized in xylene and rehydrated in 100% and 95% ethanol. Next, the sections were fixed in ice-cold methanol for 5 min, heated in antigen retrieval buffer (Dako) at 95°C for 20 min, permeabilized with 0.25% Triton X-100/PBS (PBST) for 10 min, and blocked with 10% normal goat serum in PBST, For double labeling, sections were first incubated overnight with a rabbit anti-vWF antibody (Dako; 1:200) in 5% goat serum/PBST at 4°C, washed with PBS, and then incubated for 1 h with DyLight 488-conjugated goat-anti-rabbit (Jackson ImmunoResearch Laboratories, West Grove, PA; 1:600) in 5% goat serum/PBST at room temperature in the dark and washed with PBS. The sections were second incubated overnight with a goat anti-P-selectin antibody (R&D Systems; 5 μg/ml) and then incubated with biotinylated goat anti-rabbit antibody (1:600; Jackson ImmunoResearch Laboratories) and cyanine 3-conjugated streptavidin, each for 1.5 h. Lastly, the sections were counterstained with DAPI (Sigma; 1 μg/ml) for nuclear morphology, washed, mounted with a fluorescent mounting medium (Vector Laboratories), and examined with a Nikon A1 confocal laser microscope system. A ×60, 1.20 numerical aperture water immersion objective was used along with the following filter settings: for DyLight 488, 493-nm excitation and 518-nm emission; for cyanine 3, 550-nm excitation and 570-nm emission; and for DAPI, 345-nm excitation and 455-nm emission. A set of serial optical sections in both the fluorescence and the differential interference contrast mode (Z-stacks) was taken at 0.2- to 0.3-μm intervals.

Quantum dot immunolabeling of P-selectin.

Quantum dot immunolabeling of P-selectin was performed in mouse lung cryostat sections according to the method described by Giepmans et al. (17). Antibodies used were goat anti-mouse P-selectin polyclonal antibody (the same antibody used for immunohistochemistry) and Qdot 655 rabbit F(ab′)2 anti-goat IgG conjugate (Invitrogen, Carlsbad, CA). At the completion of labeling, tissue sections were further processed for transmission electron microscopy.

Transmission electron microscopy and sample preparation.

Human lung tissue blocks were from five individuals (n = 5). Rodent lungs were isolated from anesthetized adult Sprague-Dawley rats (n = 4) and adult wild-type mice (n = 5), fixed in 3% glutaraldehyde in cacodylate buffer by vascular perfusion, and then immersed in fresh fixative. The specimens were rinsed in cacodylate buffer, postfixed for 1 h with 1% aqueous osmium tetroxide, dehydrated with graded alcohol series, and embedded in PolyBed 812 epoxy resin (Polysciences, Warrington, PA). Thick (1 μm) sections were cut with glass knives and stained with 1% toluidine blue and examined via light microscopy for “structure orientation.” Thin (80 nm) sections cut from the same block with diamond knives were stained with uranyl acetate and Reynold's lead citrate and examined and photographed with a Philips CM 100 transmission electron microscope (FEI, Hillsboro, OR), and measurements were made from the micrographs.

In situ real-time fluorescence microscopy.

In situ real-time fluorescence imaging of the [Ca2+]i in lung capillary endothelial cells was performed as previously described (27). In brief, isolated lungs of male Sprague-Dawley rats (385 ± 22 g body wt) were perfused with autologous blood at constant flow of 14 ml/min, a left atrial pressure of 5 cmH2O, and a pulmonary arterial pressure of 10 ± 1 cmH2O. Lungs were positioned on a custom-built microscope stage and superfused with normal saline at 37°C to prevent drying. Membrane-permeable fura 2-AM (5 μmol/l; Molecular Probes, Eugene, OR), which deesterifies intracellularly to impermeant Ca2+ dye fura 2, was loaded to lung capillary endothelial cells for 20 min via a wedged microcatheter. Capillary endothelial cells were imaged during capillary perfusion with 2% dextran (70 kDa; Sigma); 1% FBS (Gemini BioProducts, Calabasas, CA) HEPES solution containing 150 mmol/l Na+, 5 mmol/l K+, 1.5 mmol/l Ca2+, and 20 mmol/l HEPES (Serva, Heidelberg, Germany) at pH 7.4 and osmolarity of 295 mosM (baseline); or an equimolar HEPES solution of 40 mmol/l K+ in the presence or absence of the T-type channel blocker mibefradil (10 μmol/l). Endothelial fura 2 fluorescence was excited by a near monochromatic beam from a digitally controled galvanometric scanner (Polychrome IV; T.I.L.L. Photonics, Martinsried, Germany) at 340, 360, and 380 nm. Fluorescence emission was collected through an upright intravital microscope (AxiotechVario 100 HD; Zeiss, Jena, Germany) equipped with an apochromat objective (UAPO 40 × W2/340; Olympus, Hamburg, Germany) and appropriate dichroic and emission filters (FT 425 and BP 505–530; all Zeiss) by a CCD camera (Sensicam; PCO, Kelheim, Germany) and subjected to digital image analysis (TILLvisION 4.01; T.I.L.L. Photonics). Endothelial [Ca2+]i was calculated from the 340/380 ratio based on a Kd of 224 nmol/l and appropriate calibration parameters.

Morphometric assessment of P-selectin expression.

The extent of P-selectin expression in the alveolar capillary endothelium was assessed and scored in a blinded manner by two different investigators using a morphometric approach (51). Briefly, images were visualized in Adobe Photoshop CS3 (Version 10.0.1; Adobe Systems, San Jose, CA) with a grid overlay. The P-selectin volume fraction in the alveolar capillary endothelium for each experiment was determined using a point counting method. A total of 3–5 images per lung from 3 lungs in each treatment group were analyzed, yielding 82 total images and an average of 134 total septal points per image. The P-selectin volume fraction for each lung was calculated as the ratio of P-selectin-positive points relative to total points landing on the alveolar septal wall. The volume fraction data were then averaged for each treatment group.

Data analysis.

Numerical data are reported as means ± SE. One-way ANOVA was used to evaluate differences between experimental groups, with a Student-Newman-Keuls post hoc test as appropriate. Significance was considered P < 0.05.

RESULTS

Heterogeneous distribution of WPbs in pulmonary vascular endothelium.

To determine the expression of WPbs in alveolar capillary endothelium, we performed cross-species ultrastructural assessment of WPb distribution in pulmonary vascular endothelium, with a focus on microvascular segment in human, rat, and mouse lungs (n = 5, 4, and 5, respectively). WPbs, with characteristic elongated structure and striated interior, were found with great ease in extra-alveolar vessel endothelium in humans, rats, and mice (Fig. 1). At the junction between extra-alveolar microvessels and the alveolar capillaries, WPbs were consistently observed in endothelial cells situated at the extra-alveolar side of the junction (Fig. 1, C-E). However, WPbs were not found in any alveolar capillary endothelial cells (Fig. 2). A quantitative analysis of the WPb distribution indicated that WPbs were always observed in vessels >10 μm in diameter (i.e., extra-alveolar vessels) and that the endothelium of vessels <10 μm in diameter (i.e., capillaries) lacked expression of WPbs in any of the three species (Fig. 3). These data support the notion that WPbs are not present in alveolar capillary endothelium.

Fig. 1.

Heterogeneous distribution of Weibel-Palade bodies (WPbs) in pulmonary vascular endothelium. Representative transmission electron micrographs of cross sections of extra-alveolar vessels in human (A and B), rat (CE), and mouse (F) lung. A: low-resolution micrograph; arrow-directed area is exhibited in higher resolution in B. WPbs are indicated with arrowheads. C: low-resolution micrograph taken at the junction (Jct) where an extra-alveolar vessel (EAV) connects to an alveolar capillary (Cap); higher resolution of the boxed area is shown in D and highlighted in E. Note that the endothelial cell (EC) exhibiting WPbs (arrowheads) is at the extra-alveolar side of the junction. F: higher resolution micrograph demonstrating a cluster of WPbs (arrowheads).

Fig. 2.

Heterogeneous distribution of WPbs in pulmonary vascular endothelium. Micrographs of a cross section of an alveolar capillary in human lung. A: low-resolution micrograph; higher resolution of the boxed areas are shown in B and C; no WPbs were detected, while detailed cell ultrastructure clearly was revealed. N, nucleus; BM, basement membrane. *Endothelial caveolae: note that their ∼70-nm outer diameter is consistent with that previously described (42).

Fig. 3.

Heterogeneous distribution of WPbs in pulmonary vascular endothelium. Summarized data from a total of 188 measurements (human 47, rat 51, and mouse 90) of the diameter of the vessels where the WPbs were or were not detected, i.e., +WPb or −WPb, respectively, obtained from human, rat, and mouse lungs (n = 5, 4, and 5, respectively). Symbols, scatter plot; gray lines, median with range. Note the abrupt transitions of the vessel calibers from the vessels possessing WPbs to vessels possessing no WPbs in all 3 species examined.

vWF and P-selectin are expressed but not colocalized in alveolar capillary endothelium.

Since our recent work (51) documented the regulated P-selectin surface expression in the alveolar septal network, we sought to determine whether vWF is also expressed and, if so, colocalized with P-selectin in this vascular compartment. Immunohistochemical staining first revealed the expression of vWF in alveolar capillary endothelium in human and mouse lungs (Fig. 4, A-C). The immunoreactivity to vWF in capillary endothelium appeared to be weak and irregular in mouse while prominent in human, whereas the extra-alveolar vessel endothelium displayed a remarkable immunoreactivity to vWF in both species. Colocalization of endothelial vWF and P-selectin was subsequently examined using double immunofluorescence labeling and confocal microscopy. vWF and P-selectin exhibited quite distinct distribution in alveolar capillary endothelium, where their colocalization was rarely observed (Fig. 4D). In stark contrast, both proteins were strongly expressed and extensively colocalized in extra-alveolar endothelium (Fig. 4E). Although these results demonstrate the expression of vWF and P-selectin in alveolar capillary endothelium, they indicate that both proteins unlikely reside within the same vesicular structure, consistent with evidence documenting the absence of WPbs in this vascular segment.

Fig. 4.

von Willebrand factor (vWF) is expressed but not colocalized with P-selectin in alveolar capillary endothelium. AC: immunohistochemistry for vWF in human and mouse lung sections. A: human alveolar capillary (∧) and extra-alveolar vessel (*) endothelium regularly displayed considerable immunoreactivity to vWF. B and C: mouse pulmonary endothelium displayed heterogeneous immunoreactivity to vWF: weak and irregular in capillary endothelium (∧) and prominent in extra-alveolar vessel endothelium (*). Images acquired at 21°C on a Nikon Eclipse E600W upright optical microscope with a ×20, 0.50 numerical aperture objective (for A) and a ×60, 0.85 numerical aperture objective (for B and C) (replicated 3 times). D and E: double immunofluorescence labeling of vWF and P-selectin in mouse lung sections. Representative projected images of all parts of Z-stacks combining both fluorescence and differential interference contrast mode taken at 0.2-μm intervals from the areas of an alveolus (D) and an extra-alveolar vessel (E). Immunofluorescence was performed with an anti-vWF antibody followed by a DyLight 488-conjugated secondary antibody (green) and an anti-P-selectin antibody followed by a biotinylated secondary antibody with cyanine 3-conjugated streptavidin (red) and nuclear fluorescence staining with DAPI (blue). Colocalization of vWF and P-selectin is evidenced by the yellow color. Images acquired at 21°C on a Nikon A1 confocal laser microscope with a ×60, 1.20 numerical aperture water immersion objective (replicated 3 times).

Subcellular localization of P-selectin in alveolar capillary endothelium is WPb independent.

We next sought to determine the subcellular localization of P-selectin in this vascular compartment. Quantum dot immunolabeling of P-selectin was performed in mouse lung sections, and the localization of the dots was subsequently examined with transmission electron microscopy. As expected, quantum dots were detected in extra-alveolar vessel endothelium within the confinement of WPbs (Fig. 5, A and B). Conversely, in alveolar septal capillaries, quantum dots were detected in clusters in the endothelial cytosol, yet interestingly they did not appear to associate with any discernible subcellular structure (Fig. 5, C and D). Quantum dots were also occasionally seen at the surface of alveolar capillary endothelium (Fig. 5D). These results confirm the presence of a subcellular WPb-independent P-selectin pool in alveolar capillary endothelium.

Fig. 5.

Subcellular localization of P-selectin is WPb independent in alveolar capillary endothelium. Representative transmission electron micrographs of mouse lung cryostat sections immunolabeled for P-selectin and visualized with Qdot 655. A: low-resolution micrograph of a cross section of an extra-alveolar vessel; arrow-directed area is shown in higher resolution in B. Quantum dots (arrowheads in B) are seen within the confinement of a tangentially sectioned WPb. C: low-resolution micrograph of a cross section of an alveolar capillary; arrow-directed area is shown in higher resolution in D. Quantum dots are seen in clusters in the endothelial cytosol (circle and arrowhead in D) and reveal no association with any intracellular structures. Some quantum dots are seen at the endothelial surface (ellipse in D).

Regulated P-selectin surface expression occurs in both extra-alveolar vessel and alveolar capillary endothelium.

To determine whether a regulated P-selectin surface expression occurs in alveolar capillary endothelium under inflammatory conditions, immunohistochemistry was performed in nonpermeablized tissue sections prepared from isolated mouse lungs perfused with or without thrombin (5 U/ml). In lungs not stimulated with thrombin, the immunoreactivity to P-selectin was continuous and intense in extra-alveolar endothelium while sporadic in alveolar septal endothelium (Fig. 6A). The immunoreactivity to P-selectin in such nonstimulated lungs was presumably due to exposing intracellular P-selectin via tissue sectioning rather than expressing P-selectin at the endothelial surface. In lungs stimulated with thrombin, extensive focal immunoreactivity to P-selectin was detected in alveolar septal endothelium (Fig. 6, B and C). These results clearly demonstrate the occurrence of regulated P-selectin surface expression in alveolar capillary endothelium in lungs challenged with thrombin.

Fig. 6.

Both alveolar capillary and extra-alveolar vessel endothelium exhibit regulated P-selectin surface expression. Representative micrographs of immunohistochemistry for surface P-selectin in tissue sections from mouse lungs that had been perfused with regular perfusate (4%BSA/EBSS) either alone (A) or with thrombin (5 U/ml; B and C) for 10 min. A: in control lung, extra-alveolar endothelium exhibited continuous and intense immunoreactivity to P-selectin (arrows); alveolar septal endothelium exhibited only sporadic immunoreactivity to P-selectin (arrowheads). B: in thrombin-stimulated lung, extra-alveolar endothelium exhibited the same immunoreactivity to P-selectin (arrows) as in A; alveolar septal endothelium exhibited extensive focal immunoreactivity to P-selectin (note the dark brown staining in boxed area, also denoted by arrowheads in C). C: enlarged boxed area in B. Images acquired at 21°C on a Nikon Eclipse E600W upright optical microscope with a ×40, 0.75 numerical aperture objective (replicated 3 times each). DF: immunofluorescence imaging of endothelial surface P-selectin in extra-alveolar vessels. Representative merged FITC-fluorescence images of a serial optical sections taken at 0.5-μm intervals in ∼200-μm viable slices of the mouse lung. Lungs were perfused with an FITC-conjugated P-selectin antibody alone (D), thrombin (5 U/ml) followed by an FITC-conjugated isotype IgG (E), or thrombin followed by the FITC-conjugated P-selectin antibody (F). Only minimal fluorescent signal was present in D and E. Note the appearance of substantial FITC fluorescence on the luminal surface of an extra-alveolar vessel in F. Margins for the extra-alveolar vessel (denoted by EAV) are depicted by blue dotted lines in D and E. Images acquired at 21°C on a PerkinElmer UltraView RS confocal microscope with a ×60, 1.20 numerical aperture water immersion objective (replicated 3 times each).

Despite these observations in alveolar endothelium, immunohistochemistry proved inadequate to differentiate changes in surface P-selectin in extra-alveolar endothelium, as the immunoreactivity to P-selectin was readily detected regardless of thrombin challenge (Fig. 6, A and B). Thus the regulated P-selectin surface expression in extra-alveolar endothelium was examined with an alternative approach, where confocal fluorescence microscopy was performed in lung slices (∼200-μm thick) maintaining both tissue viability and in situ pulmonary vascular architecture. The slices were prepared from lungs following sequential perfusion with thrombin and an FITC-conjugated P-selectin antibody. Only minimal fluorescent signals were detected in control lungs using either the FITC-conjugated P-selectin antibody without thrombin stimulation (Fig. 6D) or an FITC-conjugated isotype IgG following thrombin stimulation (Fig. 6E). In contrast, a significant increase in FITC fluorescence in extra-alveolar endothelium was revealed in lungs following thrombin stimulation (Fig. 6F). Note that this method was insufficiently sensitive to detect P-selectin surface expression in alveolar septal capillaries. Collectively, these findings confirm that regulated P-selectin surface expression occurs in both extra-alveolar vessel and alveolar capillary endothelium.

Pharmacologic blockade of the T-type channel or genetic knockout of the α1G-subunit mitigates thrombin-stimulated P-selectin surface expression in alveolar capillary endothelium.

Our recent work (51) has clearly linked the α1G-mediated cytosolic Ca2+ transients to the regulated P-selectin surface expression in alveolar capillary endothelium. The current study sought to put this background into a physiological context. Specifically, we determined the role of the α1G T-type channel in thrombin-stimulated surface expression of endothelial P-selectin in alveolar septal capillaries. The immunohistochemistry in conjunction with the in situ confocal fluorescence microscopy was performed in sections from the lungs challenged with thrombin via vascular perfusion in the presence and absence of the T-type Ca2+ channel blocker mibefradil (50, 56). As demonstrated in Fig. 7, the thrombin-stimulated P-selectin surface expression was significantly attenuated with mibefradil in alveolar capillary endothelium (Fig. 7, A and B) but was not altered in extra-alveolar vessel endothelium (Fig. 7, D and E). A parallel result was obtained when the study was replicated in lungs deficient for the α1G-subunit, where the thrombin-stimulated P-selectin surface expression was nearly abolished in alveolar septal endothelium (Fig. 7C) but remained unchanged in extra-alveolar endothelium (Fig. 7F). Collectively, these results identify a critical role of the site-specific endothelial α1G T-type channel in mediating the agonist-stimulated P-selectin surface expression in alveolar capillary endothelium.

Fig. 7.

Pharmacologic T-type channel blockade or genetic α1G knockout mitigates thrombin-stimulated P-selectin surface expression in alveolar capillary endothelium. AC, immunohistochemistry for surface P-selectin in tissue sections from wild-type (α1G+/+) lungs stimulated with thrombin alone (A) or together with mibefradil (B) or from α1G-deficient (α1G−/−) lungs stimulated with thrombin (C). Thrombin-stimulated P-selectin surface expression in alveolar septal endothelium in wild-type lungs (note the dark brown staining in A) but not in mibefradil-administered or α1G-deficient lungs (B and C). Replicated 3 times each. DF: immunofluorescence imaging of surface P-selectin in extra-alveolar endothelium from lungs stimulated with thrombin alone (D) or together with mibefradil (E), or from α1G-deficient lungs stimulated with thrombin (F). Thrombin-stimulated P-selectin surface expression in extra-alveolar endothelium was not altered by mibefradil or α1G knockout. Replicated 3 times each.

α1G T-type channel-mediated Ca2+ entry is sufficient to control regulated P-selectin surface expression in alveolar capillary endothelium.

We have recently demonstrated that alveolar capillary endothelium in mice possesses a functional T-type Ca2+ channel dominated by the α1G-subtype. In mouse lung, depolarization induces a Ca2+ entry in alveolar capillary endothelium, which is completely abolished by pharmacologic blockade of the T-type channel or genetic knockout of the α1G-subunit (51). In the present study, we confirmed this effect of depolarization in isolated perfused lungs of rats. Depolarizing the endothelial membrane to the window current range of voltages of the endothelial α1G T-type channel, i.e., −60 to −30 mV (50, 56) via exposing vascular endothelium to high K+ (40 mmol/l) containing perfusate, induced a slowly developing and sustained increase in [Ca2+]i, from 101 ± 2 nmol/l to 138 ± 4 nmol/l, in alveolar capillary and small caliber microvessel endothelium. This depolarization-induced [Ca2+]i transient was completely blocked by mibefradil (Fig. 8).

Fig. 8.

Pulmonary microvessel endothelium possesses a functional T-type Ca2+ channel: depolarization-induced endothelial cytosolic Ca2+ ([Ca2+]i) response in pulmonary microvessels. A: fura 2-loaded endothelial cells of lung microvessels were imaged in situ by real-time fluorescence microscopy. Representative images of the 340-nm/380-nm ratio pseudocolor coded for [Ca2+]i at baseline (left) and after high K+ concentration (40 mmol/l) perfusion in control lungs (top) and lungs pretreated with mibefradil (10 μmol/l; bottom). White lines, vessel margins; n = 5 each (number of animals studied). B: group data of endothelial [Ca2+]i in situ. [Ca2+]i was determined at 5-min intervals at baseline as well as during high K+ concentration perfusion in control lungs (○) and lungs pretreated with the T-type Ca2+ channel blocker mibefradil (●). Data are means ± SE. *P < 0.05 vs. control, Mann-Whitney U-test.

The thrombin-stimulated cytosolic Ca2+ transients involve activation of multiple signaling pathways, including store- and receptor-operated Ca2+ channels, along with the voltage-gated α1G T-type Ca2+ channel, in pulmonary microvascular endothelial cells (50, 55, 56). Next, we sought to examine whether α1G-mediated Ca2+ entry is sufficient to trigger endothelial P-selectin surface expression in alveolar septal capillaries. Immunohistochemistry for P-selectin revealed that exposure of endothelium to the high K+ (40 mmol/l)-containing perfusate resulted in P-selectin surface expression in capillary endothelium, with an intensity and expression pattern virtually identical with that induced by thrombin (Fig. 9A). Such depolarization-coupled P-selectin surface expression was abolished by mibefradil (Fig. 9B). Consistently, depolarization failed to induce P-selectin surface expression in septal capillary endothelium in lungs deficient for the α1G-subunit (Fig. 9C). Altogether, these results suggest that the Ca2+ transient through the activated α1G T-type channel is sufficient to control regulated P-selectin surface expression in alveolar capillary endothelium.

Fig. 9.

Pharmacologic T-type channel blockade or genetic α1G knockout abolishes depolarization-induced P-selectin surface expression in alveolar capillary endothelium. Immunohistochemistry for surface P-selectin in tissue sections from wild-type (α1G+/+) lungs perfused with high K+ (40 mmol/l) alone (A) or together with mibefradil (10 μmol/l; B) or from α1G-deficient (α1G−/−) lungs perfused with high K+ (C). Depolarizing endothelial membrane evoked P-selectin surface expression in alveolar septal endothelium in wild-type lungs (note the dark brown staining in A) but not in mibefradil-administered or α1G-deficient lungs (B and C) (replicated 3 times each).

To verify the subjective immunohistochemical assessment of the alveolar capillary endothelial surface P-selectin expression, a morphometric approach was employed to determine a volume fraction for P-selectin in the alveolar septum in each experimental group. The results of this quantitative assessment (Fig. 10) validate the prominent increase in P-selectin volume fraction in lungs challenged with thrombin or high K+ (P < 0.001) and the abrogation of such an increase via pharmacologic T-type channel blockade or genetic α1G knockout (P < 0.001).

Fig. 10.

Morphometric assessment of alveolar endothelium P-selectin expression using a point-counting strategy, yielding a measure of the P-selectin volume fraction. P-selectin volume fraction increased significantly in wild-type (α1G+/+) lungs stimulated with thrombin or challenged with high K+. Such increase in P-selectin volume fraction was not observed in wild-type lungs with mibefradil treatment (α1G+/+/Mib) or in lungs deficient for α1G1G−/−). Data are means ± SE. #P < 0.001, compared with all other groups. *P < 0.001, compared with the group of α1G+/+ with thrombin or with high K+, by one-way ANOVA with Student-Newman-Keuls post hoc test.

To confirm that the surface P-selectin in septal capillary endothelium was of endothelial origin and not derived from adherent platelets, we examined the immunoreactivity of the alveolar capillaries for a platelet-specific glycoprotein, GPIIb/IIIa. Such an immunohistochemistry approach first revealed a prominent immunoreactivity of the alveolar capillaries for GPIIb/IIIa in blood-perfused lungs (data not shown). However, the study only detected a minimal level of GPIIb/IIIa in the capillaries of isolated lung preparations exposed to thrombin in both the wild-type lungs and the lungs deficient for the α1G-subunit (n = 3 each). The volume fraction for GPIIb/IIIa in the septal capillaries, averaged 0.010, was comparable between the two groups yet did not demonstrate correlation with the volume fraction for P-selectin in the same lung preparations (Supplemental Fig. 1; supplemental data for this article are available online at the Am J Physiol Lung Cell Mol Physiol website.). These findings further attribute the surface P-selectin to the septal capillary endothelium.

DISCUSSION

The present study provides compelling evidence that lung alveolar septal capillary endothelium lacks expression of WPbs yet still possesses regulated P-selectin surface expression. Thrombin elicits P-selectin surface expression occurs in both extra-alveolar vessel and alveolar capillary endothelium. However, only the regulated P-selectin surface expression in alveolar capillary endothelium is controlled by Ca2+ entry through the α1G T-type channel.

Lung alveolar capillary endothelium differs considerably in structure and function from endothelium in extra-alveolar vessels. Consistent with this the notion, we previously identified that the cultured pulmonary microvascular endothelial cells express a functional voltage-gated α1G T-type Ca2+ channel, whereas endothelial cells of extra-alveolar conduit vessels, e.g., pulmonary arteries, do not express this channel (50, 51, 55, 56). The present study further advances our understanding that alveolar capillary endothelium is highly specialized. We found that although endothelial cells in this compartment lack WPbs, they nonetheless possess a discrete subcellular P-selectin pool and an α1G T-type channel regulated P-selectin surface expression.

The lack of WPbs in pulmonary capillary endothelium was initially proposed by Fuchs and Weibel, who estimated 0.0005–0.001 (0.05–0.1%) volumetric fraction of WPbs in pulmonary capillaries vs. 0.005 to 0.03 (0.5–3%) in pulmonary conduit vessels in rats (14). The issue of whether lung capillaries constitutively possess WPbs has been left unsettled for decades, due to a paucity of detailed systematic investigations. Hence, one of the primary goals in the present study was to acquire evidence that would close this knowledge gap. Our cross-species ultrastructural assessment indicates clearly that WPbs are present only in pulmonary vessels of >10 μm in diameter, i.e., in vessels outside the alveolar septal compartment. The segmental difference in the distribution of WPbs along the pulmonary vascular tree is clearly demarcated by the junction between the extra-alveolar vessel and the alveolar septal capillary. Such distribution of WPbs is consistent among the three species examined. These findings, in agreement with the previous report by Fuchs and Weibel (14), strongly support the conclusion that alveolar capillary endothelium lacks expression of WPbs.

Despite the lack of WPbs in alveolar capillary endothelium, expression of vWF and P-selectin, two major components of WPbs, has previously been documented in this vascular segment (11, 12, 21, 23, 33, 36, 53). Utilizing quantum dot immunolabeling technique, we detected a subcellular P-selectin pool in alveolar capillary endothelium that is not related to WPbs. This observation is intriguing, as endothelial P-selectin is generally thought to colocalize with vWF within WPbs. The present work demonstrates the expression of vWF in alveolar capillary endothelium in human and mouse lungs. Unfortunately, there is no information currently available that documents the subcellular localization of vWF in pulmonary capillary endothelium. In general, P-selectin is synthesized in the same cell as vWF (the endothelial cell and megakaryocyte) and sorted by vWF to the same secretory organelles, i.e., endothelial WPbs and platelet α-granules. The molecular basis for this distinct P-selectin pool in capillary endothelium is unknown. Also, the present study did not resolve the precise subcellular localization of P-selectin. Although ultrastructurally we show that subcellular P-selectin does not seem to associate with any discernible structures, it appears that, with the available instrument resolution, the P-selectin molecules localize to an electron dense area of the cell. It is reasonable to speculate that, in the absence of WPbs, P-selectin may be localized in endosomes, lysosomes, or some unidentified vesicles, as occurred when biogenesis of WPbs was disrupted in mice lacking vWF (6, 31), and that P-selectin surface expression may mostly be achieved by vesicular trafficking between these structures and the plasma membrane. Nonetheless, the lack of WPbs and the presence of a subcellular P-selectin pool in alveolar capillary endothelium support the idea that the regulated P-selectin surface expression may also be distinct from the process of WPb exocytosis.

Thrombin has been established to exert prominent edemagenic and proinflammatory effects on vascular endothelium. Interestingly, endothelial responses to thrombin exhibit a segment-specific heterogeneity in the pulmonary circulation, e.g., barrier disruption in conduit vessels vs. barrier enhancement in alveolar capillaries (43). In the present study, we demonstrated that P-selectin surface expression is stimulated by thrombin in both extra-alveolar vessel and alveolar capillary endothelium. We believe that the surface P-selectin in the alveolar septum must derive from local endothelium and not from adherence of P-selectin shed from the platelet and/or extra-alveolar endothelium. Our conclusion is based on the observation that, first, P-selectin was not found in the alveolar septum in nonstimulated lungs; second, there was little to no P-selectin expression in alveolar septum when mibefradil was present despite extensive P-selectin expression on extra-alveolar endothelial surface in thrombin-stimulated lungs; and third, minimal immunoreactivity to GPIIb/IIIa was detected in the capillaries of isolated perfused lung preparations. Further, as the alveolar capillary endothelium lacks expression of WPbs, the regulated capillary endothelial P-selectin surface expression could not be due to the exocytosis of WPbs but rather to a rapid translocation of P-selectin from its cytosolic location to the apical surface membrane. Such rapid translocation event in the absence of an exocytotic event is not understood. Indeed, the precise subcellular localization of P-selectin in capillary endothelium is left unresolved in the present study. However, the extensive focal expression of P-selectin directly following thrombin stimulation suggests the presence of an immediate releasable intracellular P-selectin pool in alveolar capillary endothelium. Of note, soluble P-selectin that may be derived from extra-alveolar vessels and deposited on the capillary endothelium, if any, has not been entirely ruled out as a possibility for the source of alveolar endothelium surface P-selectin. The presence of the platelet glycoprotein Ib/IX/V (GPIb/IX/V) complex, a counterreceptor for P-selectin (5, 40), in certain endothelial membranes (e.g., HUVECs and aortic endothelium; Ref. 49) could raise the concern that the GPIb/IX/V complex mediates the binding of soluble P-selectin to the alveolar endothelium. Although this is not likely the scenario in alveolar endothelium, future studies will be required to directly determine whether such a mechanism contributes to our present observations.

Our findings of depolarization-activated, low threshold, mibefradil-inhibitable, and α1G-dependent cytosolic Ca2+ transient in alveolar capillary endothelium in rat lungs, together with our recent parallel findings in mouse lungs (51), provide compelling evidence that alveolar capillary endothelium expresses a functional T-type channel dominated by the α1G-subtype. We have previously determined that the α1G T-type channel controls agonist-induced, regulated vWF release in cultured pulmonary microvascular endothelial cells (55) and that the α1G-mediated cytosolic Ca2+ transients are dedicated to regulation of P-selectin surface expression in alveolar capillary endothelium (51). The current study establishes a physiologic context for this channel in that thrombin-induced transitions in plasma membrane potential activate the α1G T-type channel in alveolar capillary endothelium. Importantly, pharmacologic blockade of the T-type channel or genetic knockout of the α1G-subunit mitigated the effect of thrombin on P-selectin surface expression exclusively in alveolar capillary endothelium. This evidence collectively points to a functional role of the α1G T-type channel in promoting an endothelial proinflammatory phenotype.

In summary, our findings support a novel physiologic and pathophysiologic paradigm in pulmonary capillary endothelium in that both the subcellular localization and regulated surface expression of the leukocyte adhesion molecule P-selectin are distinct from those in conduit vessel endothelium. Further, our work supports the notion that P-selectin surface expression in alveolar capillary endothelium is uniquely coupled to α1G-mediated Ca2+ entry. We propose that in vivo, Gq-linked inflammatory agonists (54), excessive airway distention (as occurs in mechanical ventilation), and ischemia/reperfusion (41) could induce endothelial plasmalemmal depolarization to the extent sufficient to activate the α1G T-type channel (56). For instance, ischemia/reperfusion causes lung injury by P-selectin-dependent mechanisms (32), which could be a relevant scenario for the T-type channel activation. We speculate that the P-selectin translocated to the luminal surface of the capillary endothelium could capture circulating leukocytes, leading to their intracapillary sequestration and ultimate migration into the interstitium and air spaces. Notably, a nonspecific selectin inhibitor (fucoidin) has been shown to accelerate leukocyte passage through alveolar capillaries, suggesting that selectins delay capillary leukocyte transit (25). For decades, there has been a conceptual gap in the lung biology field between the notions that leukocyte trafficking mainly occurs in alveolar capillaries; that leukocyte trafficking in certain models of acute lung injury takes place in a P-selectin-dependent manner (3, 8, 13, 34, 35); and that alveolar capillary endothelium lacks expression of WPbs. Understanding the mechanism(s) underlying the distinct subcellular localization and regulated surface expression of P-selectin, as well as the functional complexity of the surface P-selectin in alveolar capillary endothelium, is of importance in closing this gap and developing novel and specific anti-inflammatory strategies.

GRANTS

This work was supported by National Heart, Lung, and Blood Institute Grants HL-74116 (to S. Wu) and HL-66299 (to M. I. Townsley and S. Wu).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).

ACKNOWLEDGEMENTS

We thank Drs. Troy Stevens and Jack W. Olson for numerous invaluable discussions, Dr. Donna L. Cioffi for critical reading of the manuscript, and Freda K. McDonald for excellent technical assistance with transmission electron microscopy study.

Present address of J. A. King: Department of Pathology, Charleston Area Medical Center, Charleston, WV.

REFERENCES

  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
  29. 29.
  30. 30.
  31. 31.
  32. 32.
  33. 33.
  34. 34.
  35. 35.
  36. 36.
  37. 37.
  38. 38.
  39. 39.
  40. 40.
  41. 41.
  42. 42.
  43. 43.
  44. 44.
  45. 45.
  46. 46.
  47. 47.
  48. 48.
  49. 49.
  50. 50.
  51. 51.
  52. 52.
  53. 53.
  54. 54.
  55. 55.
  56. 56.
View Abstract