help button home button Am J Pathol Angiogenesis Meeting
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Lagunoff, D.
Right arrow Articles by Rickard, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Lagunoff, D.
Right arrow Articles by Rickard, A.
(American Journal of Pathology. 1999;154:1591-1600.)
© 1999 American Society for Investigative Pathology

Regular Articles

Mast Cell Granule Heparin Proteoglycan Induces Lacunae in Confluent Endothelial Cell Monolayers

David Lagunoff and Alice Rickard

From the Department of Pathology, St. Louis University School of Medicine, St. Louis, Missouri


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
The addition of rat mast cell granules to confluent bovine pulmonary artery endothelial cell monolayers resulted in the formation of numerous lacunae in the cultures. Several lines of evidence identified heparin proteoglycan as the component of the granule matrix responsible for the effect: presence of the activity in the proteoglycan fraction after chromatography of granule extracts, inhibition of granule activity by digestion with heparinase I, the failure of proteolysis of the proteoglycan fraction with proteinase K to significantly diminish its activity, and the failure of chymase and carboxypeptidase inhibitors to inhibit granule activity. The onset of hole formation was delayed for several hours after granule addition to the culture, and maximal hole formation occurred between 8 and 16 hours and was sustained as long as 24 hours. The lacunae formed by the separation of motile endothelial cells within the monolayer and was not attributable to cell contractile activity or cell loss. Time-lapse video recording showed that the holes were dynamic, individual holes expanding and regressing over a period of hours. Formation of lacunae occurred on gelatin and fibronectin surfaces alike. The presence of active chymase in the granules prevented the action of the proteoglycan. Heparin glycosaminoglycan as distinct from the proteoglycan did not similarly affect the endothelial monolayers but did block the action of granules added subsequently, indicating the likelihood of a heparin-reactive receptor or binding site.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
The predilection of mast cells for associating with and affecting blood vessels is well known.1 Among the principal components of the secretory granules, histamine binds ionically to the granule matrix from which it is readily released on secretion2 whereupon it is capable of inducing increased permeability of the microcirculation3 and contraction of endothelial cells4 as well as relaxation of vascular smooth muscle and contraction of other smooth muscle.5 Histamine is a small, readily diffusible molecule with a transient action on its target cells. In contrast, the macromolecular constituents of the granule matrix, principally heparin proteoglycan and several proteases,1 are likely to persist in the local extracellular compartment with the possibility of inducing additional reactions in blood vessels, such as angiogenesis, or delayed increases in permeability, such as are seen in the late-phase response that accompanies mast cell degranulation in vivo. To explore these possibilities, we have examined the effects on endothelial cells of mast cell granules stripped of histamine and found that granule matrices induce the formation of lacunae in confluent endothelial monolayers by a noncontractile mechanism. The granule component responsible for the formation of lacunae is heparin proteoglycan.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Bovine pulmonary artery endothelial cells (BPAE), CCL 209 from the American Type Culture Collection (Rockville, MD), were maintained in Eagle's minimal essential medium (MEM) with 2 mmol/L glutamine, 10% fetal calf serum (Sigma Chemical Co., St. Louis, MO), 100 U/ml penicillin, and 100 µg/ml streptomycin, pH 7.3. Cells were plated on dishes or flasks (Corning/Costar, Cambridge, MA) and cultured in a gas-flow, humidified incubator at 37°C using 5% CO2 in air. Through passage 25, cultures were subcultured approximately every 7 days using 0.25% porcine trypsin (1:250; JRH Biosciences, Lenexa, KS) and plated at a 1:3 dilution.

Experiments evaluating formation of lacunae in confluent cultures were performed variously on cells grown in 35-mm dishes, Permanox eight-chamber slides (Nunc, Naperville, IL), or 24-well, flat-bottom, multiwell plates (Corning/Costar, Corning, NY). The plastic surfaces were standardly coated with gelatin (Difco Laboratories, Detroit, MI); bovine fibronectin (Sigma) and Pronectin F (Protein Polymer Technologies, San Diego, CA) were also tested. All cultures were grown to confluence before initiating an experiment unless specifically indicated to the contrary.

For measuring lacunar areas, the cell monolayers were rinsed with PBS to remove serum protein and fixed in 3% p-formaldehyde in PBS. Cultures were stained with 0.4% sulforhodamine B in 1% acetic acid. Images were captured by video camera, and an image analysis program (Optimas 5.0, Edmonds, WA, or ImagePro Plus, Media Cybernetics, Silver Springs, MD) were used to evaluate holes that formed in the monolayer. Thresholds for the lacunae were established interactively. The percentage of the area of the video frame occupied by the lacunae was determined for several fields per monolayer, and duplicate monolayers were assessed. Values are expressed as mean fractional area occupied by lacunae with standard error (SE) of the mean.

To evaluate cell movements, cultures were plated onto collagen-coated, 25-mm, 1 1/2 coverslips. Coverslips were coated with type 1 rat tail collagen (Upstate Biotechnology, Lake Placid, NY).6 Coverslips with adherent cells were placed either directly into a Leiden closed perfusion chamber (Medical Systems Corp., Greenvale, NY) or into a sterile Lexiglass holder for treatment before transfer to the perfusion chamber. The holder had 26-mm diameter wells 6 mm deep with a small eccentric semicircular depression at the periphery for ease of adding and removing coverslips or media.

The perfusion chamber was equilibrated at 37°C before inserting a coverslip with its cells. The assembled chamber was placed on the microscope stage within a Nikon enclosed plexiglass housing maintained at 37°C with forced heated air. Culture medium, pH 7.3, fortified with 20 mmol/L Hepes buffer, was pre-equilibrated with 5% CO2 in air and perfused through the chamber via R-3603 Tygon tubing at a constant rate of 0.019 ml/minute from a 60-ml syringe driven by an infusion pump (infusion/withdrawal pump, Harvard Apparatus, South Nautick, MA). Cell activity was recorded with a 40x phase contrast objective on a Nikon inverted microscope and recorded with a 960 M Sony CCD video camera connected to a S-VHS Panasonic AG-6730 time-lapse videocassette recorder. Routinely, a time lapse ratio of 120/1 was used.

Isometric tension generated by the cells was monitored according to the method of Kolodney and Wysolmerski.7 Granules were added to the monolayers in the apparatus after an equilibration period, and tension was monitored for 18 to 20 hours thereafter. At the completion of the measurements, thrombin and cytochalasin D were added in sequence to evaluate the status of the monolayer as described by Kolodney and Wysolmerski.7

Normal peritoneal mast cells were isolated from male Sprague-Dawley retired breeders (Harlan, Indianapolis, IN) by washing the peritoneal cavity of the rats, after anesthesia with CO2, decapitation, and exsanguination, with 10 ml of cold, heparinized balanced salt solution (BSS) containing 10 mmol/L phosphate, 154 mmol/L NaCl, 2.7 mmol/L KCl, and 0.68 mmol/L CaCl2, with 0.5% albumin and 10 U of heparin/ml (Elkins-Sinn, Cherry Hill, NJ). The pooled cell washes from 8 to 10 animals were centrifuged for 8 minutes at 130 x g at 4°C. The supernatants were removed, and cells from four to six animals were resuspended in 4 ml of BSS with albumin and layered onto a Percoll suspension (7.5 ml of Percoll (Pharmacia Biotech, Piscataway, NJ), 1 ml of salt (1.54 mol/L NaCl, 27 mmol/L KCl, 6.8 mmol/L CaCl2), 0.3 ml of water, and 0.14 ml of bovine albumin (Path-O-Cyte 4, ICN, Costa Mesa, CA)) adjusted to pH 7.2 with 0.1 mol/L KH2PO4, and centrifuged at 225 x g for 20 minutes at 4°C. The mast cells sedimented to the bottom of the gradient at >90% purity. After removal of other cells and supernatant Percoll, the mast cells were resuspended in 7.5 ml of BSS with albumin and centrifuged at 150 x g for 5 minutes. The sedimented cells were resuspended in 5 ml of BSS with albumin. Cells were counted in a hemacytometer after fixation and staining with acidic toluidine blue.8

Mast cell granule isolation was based on the method of Krüger et al.9 A total of 2 x 106 to 3 x 106 cells were suspended in 1 ml and sonicated using a 40-Hz water bath sonicator (Sonagen Ultrasonic Cleaner, Branson Instruments, Stanford, CT) for two 15-second bursts and checked for the extent of cell disruption; 95% loss of intact cells served as the criterion for adequate sonication. The disrupted cells were centrifuged twice at 130 x g for 8 minutes at 4°C. The granule-containing supernatants were pooled and carefully layered on top of a concentrated suspension of Percoll (4.5 ml of Percoll, 0.45 ml of salt solution (1.54 mol/L NaCl, 27 mmol/L KCl, 6.8 mmol/L CaCl2, 0.1 mol/L Hepes) and 0.063 ml of Path-O-Cyte 4) and centrifuged at 27,000 x g for 20 minutes at 4°C. Granules separated into two bands; those in the topmost band, designated A granules, had lost their membranes during sonication. The lower, denser band (C granules) consisted of granules that had retained their membranes and thus maintained their native high density. The two bands were removed from the gradient separately and diluted in PBS, and each was centrifuged at 1800 x g, removing much of the Percoll. After a second wash and centrifugation in a Beckman Microfuge B at 10,000 x g, membranes were stripped from the C granules by suspending them in 0.1% Triton X-100 in PBS for 10 minutes on ice. The small amount of tryptase present in the granules was released from the granules on removal of the membranes.10 Granule histamine and any residual tryptase were eluted from the granules with PBS. The stripped granules were centrifuged for 3.5 minutes at 10,000 x g, washed once in PBS, and resuspended in a final volume of 0.5 ml of PBS. Aliquots were removed for protein determination and assay of chymase activity. The remaining sample was treated for 30 minutes with phenylmethylsulfonyl fluoride (PMSF), 50 µg/ml, at 37°C to inhibit the chymase activity at least 95%, confirmed by assay of residual activity. The granules were stored at 4°C.

Heparin proteoglycan was obtained from isolated granules that were solubilized in 3 mol/L NaCl, 10 mmol/L Hepes, pH 7.2, on ice for 10 minutes and passed over a 30 x 1 cm Superose 6 or 12 HR column eluting with the same high salt solution using a Pharmacia FPLC system. Samples containing heparin were pooled, and 2 vol of 10 mmol/L Hepes, pH 7.5, were added, followed by 4 vol of ethanol. The precipitate was allowed to form overnight at 4°C and was collected by centrifugation at 27,000 x g. The ethanol was removed, and the pellet containing the heparin proteoglycan was dissolved in PBS.

Heparin proteoglycan was isolated from rat skin essentially according to the procedure of Horner.11 Subcutaneous tissue was removed by scraping the pelts with a scalpel blade. The pooled material was weighed and suspended in 3 vol of 3 mol/L KCl, 0.1 mol/L Hepes, pH 7.4 and homogenized in a Waring blender. The crude homogenate was stirred continuously for 1 hour at 40°C, large pieces of tissue were removed, and the remaining sample was centrifuged at 45,400 x g for 30 minutes at 4°C. The lipid pellicle at the surface was removed from the sample, and the remainder of the suspension was filtered through glass wool. The sample was diluted to 1 mol/L KCl with water, and a 10% volume of 3% cetylpyridinium chloride (CPC) in 1 mol/L KCl at room temperature was added. The mixture was stirred for 20 minutes and then centrifuged at 1800 x g for 10 minutes. The supernatant was removed, and the pellet was resuspended in 3% CPC in water, stirred for 20 minutes, and again centrifuged. Pellets were combined and washed several times at 37°C with ethanol saturated with sodium thiocyanate. The precipitate was dissolved in PBS followed by concentration of the sample in a Centricon 10 concentrator. The sample was sterilized by filtration before addition to the cell culture. Heparin concentration was determined with the N-sulfated hexosamine assay. The ratio of N-sulfated hexosamine to uronic acid of the proteoglycan so isolated was 0.85, identifying the principal glycosaminoglycan present as heparin. If heparan sulfate is assumed to be the principal contaminant contributing uronic acid to the preparation, it can be estimated that the heparin content is 70% to 80% depending on the N-sulfate content of the contaminating heparan sulfate.

The heparin preparation used in the experiments to test the efficacy of heparin glycosaminoglycan to induce the same effects as heparin proteoglycan and to test for possible blocking of heparin proteoglycan was obtained from Sigma. Analysis of several lots of the preparation (grade 1-A from porcine intestinal mucosa), including that currently available (H 3393), indicate a high N-sulfate to uronic acid ratio (>90% of theoretical), 3 to 3.3 anionic charges per disaccharide by both chemical analysis and acridine orange titration,12 and a protein content of less than 1%.

To evaluate possible loss of cells from the monolayer, BPAE cells were plated in MEM, 10% FCS containing [3H]thymidine (ICN Biomedical, Costa Mesa, CA) at 0.05 µCi/ml media. Cultures were maintained and re-fed in medium containing the radiolabeled thymidine for 6 days. Radioactive medium was then removed, and the monolayer was washed three times and placed in MEM, 10% FCS. Granules were added to the cultures, and 4 or 8 hours later the supernatant was removed, and the culture was washed with PBS. The wash was added to the supernatant and centrifuged at 225 x g for 5 minutes. The radioactivity of an aliquot of the supernatant was measured in Cytoscint scintillation fluid (ICN). The pellet was dissolved in 0.45 ml of 0.5 N NaOH and neutralized with HCl, and its radioactivity was measured. The remaining cell sheet on the plate was also solubilized in 1 ml of 0.5 N NaOH for 5 minutes, and the plate was washed with another 0.5 ml of NaOH. The combined sample was neutralized with HCl, and its radioactivity was measured.

Uronic acid was assayed by the method of Bitter and Muir.13 Hexosamine-N-sulfate was determined as described by Lagunoff and Warren14 or in the case of the heparin glycosaminoglycan by selective N-sulfate hydrolysis14 and assay of the resulting free amino groups with fluorescamine. Chymase activity was determined using the substrate N-succinyl-leu-leu-val-try-7-amido-4-methylcoumarin at 4 µg/ml in 0.01 mol/L Hepes, 10 mmol/L CaCl2, and 0.1% Triton X-100, pH 7.5.8 Tryptase was assayed as previously described.10 Carboxypeptidase activity was assayed based on the hydrolysis of hippuryl-DL-phenyllactic acid (hip-phe; Sigma). The substrate was 11.4 mmol/L in BSS, pH 7.5. Ten microliters of sample was added to 0.25 ml of the hip-phe for 10 minutes at 37°C. The reaction was stopped with 0.25 ml of 20% trichloracetic acid at room temperature. After a 10-minute centrifugation, 0.3 ml of supernatant was removed and added to 0.7 ml of 0.1 mol/L borate, pH 10.6. To this solution, 0.10 ml of freshly prepared fluorescamine reagent (3 mg of fluorescamine in 20 ml of acetone) was added. Fluorescence was measured using a Perkin-Elmer 650-10S fluorescence spectrophotometer at nominal excitation/emission wavelengths of 395/475 nm.

Lactic dehydrogenase (LDH) was assayed using the method described by Riley.15 Phenol red was omitted from the culture media used for these experiments as the dye interfered with the assay. Fluorescence was measured at 360/480 nm. Percent LDH release was calculated as the activity in the medium divided by the total activity in Triton-X-100-treated cultures.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
When mast cell granule matrices stripped of histamine were added to confluent monolayers of endothelial cells, lacunae appeared in the monolayer after a delay of several hours (Figure 1) . The fractional area of the monolayer occupied by lacunae was dependent on the quantity of granules added to the cultures and the length of exposure time (Figure 2) . Formation of lacunae induced by mast cell granules was not significantly different on plastic dishes coated with gelatin or fibronectin (Figure 3) . Recombinant mouse TNF-{alpha} when added to BPAE monolayers in 35-mm dishes at concentrations of 1, 10, and 100 U/ml failed to induce lacunae in 24 hours, and antibody to TNF-{alpha} did not inhibit the effect of the granules.



View larger version (153K):
[in this window]
[in a new window]
  		Figure 1. Phase micrographs of endothelial cells. A: Control cells, 13-day culture, after 7 hours of perfusion. B: Granule-treated cells, 8-day culture, 9 hours after the addition of granules (13.6 µg of granule heparin/2 x 105 ECs) and 7 hours of perfusion. Images are single frames captured from time-lapse video tape recordings (S-VHS). The mast cell granules are evident in B. Magnification, x490.

 


View larger version (15K):
[in this window]
[in a new window]
  		Figure 2. Effect of granule quantity and time on formation of lacunae. Cells were plated on eight-well Permanox slides coated with fibronectin. Cultures, 4 day, were fixed at intervals after exposure to different amounts of granules and stained, and the retraction fraction was measured. Values are means with SEs of the means for six separate determinations.

 


View larger version (13K):
[in this window]
[in a new window]
  		Figure 3. Comparison of lacunae formation in monolayers cultured on gelatin and fibronectin. Cells were plated on eight-well Permanox slides coated with fibronectin or gelatin, grown to confluency for 4 days, and treated with granules for 8 hours. Values are means with SEs of the means for six separate determinations.

 
Inhibition of granule chymase with PMSF did not modify the effect of granules on endothelial monolayers, nor did inhibition of granule carboxypeptidase with carboxypeptidase inhibitor from potato tuber (Sigma; Table 1 ). As tryptase10 and both ß-glucuronidase and ß-hexosaminidase10,16,17 are, like histamine, lost from the granule matrices under the conditions used in the preparation of the granules, these granule components were not studied.


View this table:
[in this window]
[in a new window]
  		Table 1. Effect of Protease Inhibitors on Activity of Mast Cell Granules

 
To test the possibility that granule binding of a serum factor was responsible for lacunae formation, experiments were conducted in culture medium with FCS replaced by serum albumin. Under these circumstances the effect of granules was lost. However, when granule chymase activity was inhibited with {alpha}1-antiprotease, the induction of lacunae could be demonstrated in the absence of whole serum (Table 1) , indicating that the effect of serum was to inhibit the chymase rather than provide an essential factor. Consistent with this interpretation, when granules were suspended in 10% FCS, chymase in the granules exhibited nearly complete loss of activity over a period of several hours (Figure 4) . All subsequent experiments were performed with granules treated with PMSF to inhibit the chymase and obviate any variability arising from different rates or extents of inactivation of the enzyme by serum.



View larger version (17K):
[in this window]
[in a new window]
  		Figure 4. Inhibition of granule chymase activity by fetal calf serum. Isolated mast cell granules were suspended in MEM containing different concentrations of fetal calf serum. Samples were removed at intervals and assayed for chymase activity.

 
With TNF-{alpha}, chymase, tryptase, and carboxypeptidase eliminated as the granule constituents responsible for the effects of granules on endothelial cells, attention was directed to heparin. A range of concentrations of heparin glycosaminoglycan (Sigma) did not reproduce the effects of granules on endothelial monolayers, but when a 3 mol/L salt extract of granules was chromatographed on Superose 12 (Figure 5) , activity on endothelial monolayers was found in the highest molecular weight fraction that contained heparin proteoglycan (Table 2) . Proteinase K digestion of the active fraction with re-chromatography did not significantly diminish the activity (Table 3) . Heparin proteoglycan isolated from rat skin also proved active in inducing lacunae in endothelial cell monolayers (Figure 6) . Additional evidence that the active component of the granules was heparin proteoglycan was provided by inhibition of activity on treatment of granules with heparinase I (Sigma H2519; Figure 7 ). Although heparin glycosaminoglycan did not induce lacunae, the glycosaminoglycan was able to prevent the formation of lacunae on the subsequent addition of granules (Table 4) .



View larger version (17K):
[in this window]
[in a new window]
  		Figure 5. Superose separation of a 3 mol/L salt extract of granules. Isolated mast cell granules were extracted in 3 mol/L NaCl and chromatographed on Superose 12, eluting in 3 mol/L salt. Column elution was monitored with absorbance at 280 nm. Fractions were analyzed for uronic acid and protein.

 

View this table:
[in this window]
[in a new window]
  		Table 2. Activity of Mast Cell Granule Fractions from Superose 12 Column

 

View this table:
[in this window]
[in a new window]
  		Table 3. Effect of Proteinase K Treatment on Granule Proteoglycan Activity

 


View larger version (15K):
[in this window]
[in a new window]
  		Figure 6. Formation of lacunae induced by rat skin heparin proteoglycan. Monolayers grown to confluence for 6 days in 24-well plates coated with gelatin were treated with granules or rat skin heparin proteoglycan. After 19 hours of incubation, cultures were fixed and stained, and the fraction of the monolayer area occupied by holes was measured. Values are means with SEs of the means for three measurements.

 


View larger version (13K):
[in this window]
[in a new window]
  		Figure 7. Effect of heparinase I on granule activity. Granules were treated with heparinase I as described in the text, and control and heparinase-treated granules were assessed for activity. Monolayers in 24-well plates coated with gelatin were grown to confluence for 6 days, and the retraction fraction was measured 4 hours after the addition of granules. Values are means with SEs of the means for six separate determinations.

 

View this table:
[in this window]
[in a new window]
  		Table 4. Inhibition of Granule Activity by Heparin Glycosaminoglycan

 
To characterize the process of formation of the lacunae, the morphology of endothelial cells in culture was examined using extended time-lapse video recording. This technique allowed continuous examination of cell behavior after addition of granules. In the recordings, there was no morphological evidence of an increased rate of cell death. Consistent with this observation there was no measurable increase in release of LDH into the medium. Cell loss from the monolayer was specifically examined by labeling cells with [3H]thymidine and then measuring both soluble and particulate radioactivity. No significant differences were found between the radioactivity in the fractions from cells exposed to granules and from control cells (Table 5) .


View this table:
[in this window]
[in a new window]
  		Table 5. Loss of DNA from Cultures Treated with Mast Cell Granules

 
Videos of monolayers after exposure to granules provided no evidence of contractile activity that might be pulling the cells apart, such as is the case with thrombin, and specific measurements of the isometric tension of the monolayer in three experiments demonstrated no increase in tension; in two of the three experiments, there was a slow, partial loss of tension over a period of several hours beginning before the time of appearance of obvious lacunae (Figure 8) . Endothelial cells exposed to granules and later treated with thrombin and cytochalasin D in sequence responded normally.



View larger version (10K):
[in this window]
[in a new window]
  		Figure 8. Contractile tension of the monolayer was measured according to Kolodney and Wysolmerski,7 in dynes, before and after the addition of granules (10 µg of heparin/105 ECs). After 20 hours, thrombin and cytochalasin D were added successively. Two of three experiments gave results like those shown. In one experiment, the granules had no apparent effect on the tension.

 
When ECs were cultured at subconfluent densities, the trypsinized cells initially rounded up; these condensed cells adhered to the collagen-coated glass, spread, and began to move about at a leisurely pace. From time to time the cells divided. ECs came together and parted during this phase, unlike epithelial cells, which after initial contact form stable adherens junctions resulting in tight cell clusters of immobile cells.17 After 7 days of culture following a 1/3 split, confluent endothelial cells continued to show movement in the time-lapse recordings; this activity increased after exposure to granules, and lacunae formed after a several-hour-long lag period. The lacunae appeared as microscopic gaps between cells. The gaps were dynamic, opening and closing at different locations in the field of observation over time (Figure 9) . The larger gaps often exceeded the area occupied by three or four cells. There was no evidence that any cells came loose from the substrate to create the lacunae. An occasional mitotic or apoptotic event led to loss of cell adherence, reversibly and irreversibly, respectively, in control and granule-treated cultures alike. Any gaps in the monolayer resulting from either of these two processes were promptly filled in by a combination of expansion of adjacent cells and cell movement.



View larger version (167K):
[in this window]
[in a new window]
  		Figure 9. Sequence of phase micrographs of a perfused endothelial cell monolayer. Granules were added at 11:40 and the cultures perfused from 14:00. Frames were captured at ~40-minute intervals as indicated by the video clock on the time-lapse recording. Holes in each image were painted black in Photoshop. The black arrows identify a shrinking hole and the white arrows an expanding hole. Magnification, x260.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
The studies we report here originated in speculation that mast cell granule matrix constituents in addition to histamine might have an effect on endothelium. When isolated mast cell granules depleted of histamine were added to confluent monolayers of bovine pulmonary artery endothelial cells, we observed that holes developed at multiple sites in the monolayer where endothelial cells separated from one another. This response differed morphologically from the contractile response of monolayers to vasoactive agents such as thrombin, and the cells did not exhibit the increase in contractile tension characteristic of the response to thrombin.7,19

Rat mast cell secretory granules contain, in addition to histamine and serotonin, high molecular weight heparin proteoglycan, chymase, carboxypeptidase A, tryptase, ß-hexosaminidase, ß-glucuronidase, TNF-{alpha},20 FGF-2,21,22 and VEGF.23 The mast cell granule matrix is formed by ionic bonding between basic proteins of modest molecular weight and heparin proteoglycan.2,16 The compact intracellular granule structure is transformed to a more diffuse organization during the process of secretion. This change can be mimicked by stripping isolated granules of their encompassing membranes with Triton X-100 or hypotonic solutions and then exposing them to 0.1 mol/L sodium chloride at pH 7.5.24 The structural change is accompanied by the loss of histamine, tryptase, ß-glucuronidase, and ß-hexosaminidase.

Our initial prime candidates for the granule component acting on endothelial cells were TNF-{alpha} and chymase. Both were readily eliminated as the active component of the granules, and it was further possible to demonstrate that the effects of granules in the absence of serum were expressed only when chymase was inhibited. Carboxypeptidase, the other known tightly bound component of the granule matrix, was excluded by demonstrating that extensive inhibition did not diminish the effect of the granules.

The next approach to the identification of the active component of the granule matrix was to solubilize granules in 3 mol/L NaCl and separate the constituents. Under these conditions, activity was found in the high molecular weight, heparin-containing fractions that appear at or near the void volume of Superose 12 or Superose 6 columns well separated from the granule proteins. The heparin in this fraction is present as a proteoglycan with a molecular weight estimated between 750 and 1000 kd.11,25 Heparin proteoglycan prepared from isolated mast cells contained ~1% protein, an amount consistent with the protein intrinsic to the proteoglycan. In an effort to exclude the possibility that traces of another protein or polypeptide avidly bound to heparin contributed to the activity, the heparin proteoglycan isolated from granules was treated with proteinase K and reisolated. The exposure to proteinase K neither reduced the size of the proteoglycan as estimated on Superose 6 nor significantly reduced its activity on endothelial cells. Partially purified heparin proteoglycan prepared from rat skin by differential CPC precipitation was also active on monolayers. Additional evidence supporting the identification of heparin as the critical component of granules was provided by the elimination of much of the granule activity with heparinase I.

Porcine intestinal heparin glycosaminoglycan was inactive in producing lacunae at concentrations up to 100 µg/ml, an order of magnitude greater than that of the effective concentrations of heparin proteoglycan. However, the glycosaminoglycan was able to block the effect of granules, indicative of a binding site for the proteoglycan that can be occupied by either heparin proteoglycan or heparin glycosaminoglycan but activated only by the proteoglycan.

Binding of heparin to cell surfaces has been reported from several laboratories,26-29 and a putative binding protein has been isolated from endothelial cells.30 The relationship of that protein to the binding site responsible for the effects of heparin proteoglycan is not known.

To examine the formation of the lacunae, we used time-lapse video recording. This technique revealed that the lacunae were transient. A hole would open between cells, enlarge, and then close, and another hole would open at another site. There was no measurable loss of cells from the monolayer to account for the holes, nor was there an increase in the isometric tension of the monolayer to account for the holes. The dynamic character of the lacunae suggested that their formation might be associated with increased motility of the endothelial cells. Direct observation of endothelial motility with time-lapse video recording demonstrated an obvious increase in motility of the cells after contact with the granules. Quantitative measurements of cell motility will be reported separately. Mast cell granules have previously been demonstrated by Azizkhan et al30 to increase motility of preconfluent endothelial cells. In their experiments, which used a different method of measuring motility, the effect of granules was mimicked by heparin glycosaminoglycan, in contrast to the negative findings with respect to lacunae in our studies using the glycosaminoglycan.

Exogenous heparin in several systems enhances the effects of FGF-2,32 and FGF-2 is known to enhance motility of preconfluent endothelial cells.33 The occurrence of lacunae in the absence of serum does not eliminate a role for FGF-2 as the endothelial cells themselves remain a potential source of the growth factor.34 The failure of heparin glycosaminoglycan to induce lacunae argues against such an effect, as most reported experiments on the cooperation of heparin with FGF-2 have used the glycosaminoglycan form. We have tested the effect of exogenous FGF-2 on confluent endothelial cells and found that it increases motility but does not induce the formation of lacunae.

Although the effects of mast cell granules on bovine pulmonary endothelial cells are striking and reproducible, there remain several important questions concerning the phenomenon. One such question relates to in vivo correlates of the lacunae. Although access of granule proteoglycan to endothelial cells could be severely constrained by the basal lamina that separates mast cells from the endothelium of even the smallest vessels, it is conceivable that the proteolytic armamentarium of the granules could allow their penetration of the basement membrane before the onset of inhibition by serum protease inhibitors necessary for the expression of the action of the heparin proteoglycan. Gaps such as those we observed in tissue culture have been described in vivo associated with the delayed increase in permeability seen in the late-phase reaction after mast cell secretion.35 Although the late-phase response is associated with and believed to be dependent on an influx of neutrophils, the mediator responsible for the formation of the interendothelial gaps associated with the increase in vascular leakage 4 to 6 hours after the initiation of the reaction has not been identified.34 Our studies suggest that heparin proteoglycan should be considered a candidate for that mediator. Another potential role for mast cell granule proteoglycan suggested by our studies is the reactivation of contact-inhibited endothelial cells as a first step in setting the stage for angiogenesis.

A second question concerns the mechanism of gap formation. As we have ruled out a concerted contractile activity pulling the cells apart, the most likely basis for the observed separation of ECs is a loosening of cell-cell junctions. Adherens junctions, which are well developed in confluent ECs are dependent on cadherin-5 (VE-cadherin).36,37,38 Preliminary immunofluorescence examination of the junctions in ECs exposed to granules using specific antibody to cadherin-5 revealed no loss of this protein from the apposing cell junctions except where the cells bordered lacunae. Detailed studies are in progress to evaluate the effects of heparin proteoglycan on EC adherens junctions with particular attention to the state of phosphorylation of the cadherin39 and associated proteins40 of the junctions. Examination of tight junctions may also be necessary.


   Acknowledgements
 
We thank Rob Wysolmerski for the measurements of endothelial cell isometric tension and John Grant for many of the measurements of endothelial cell lacunae.


   Footnotes
 
Address reprint requests to Dr. David Lagunoff, Department of Pathology, St. Louis University School of Medicine, 1402 S. Grand Blvd., St. Louis, MO 63104. E-mail: lagunofd{at}slucarel.sluh.edu

Supported by NIH grant HLBI 54245 and St. Louis University School of Medicine.

Accepted for publication February 12, 1999.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Lagunoff D: Mast cell and basophil. Bronchial Asthma: Mechanisms and Therapeutics, ed 3. Edited by Weiss EB, Stein M. Boston: Little, Brown and Co, 1993, pp 271–286
  2. Lagunoff D: Structural aspects of histamine binding: the mast cell granule. von Euler US Rosell S Uvnas B eds. Mechanisms of Release of Biogenic Amines. 1966, :pp 79-94 Pergamon, New York
  3. Lewis T: The Blood Vessels of the Human Skin and Their Responses. 1927 Shaw and Sons Ltd, London
  4. Majno G, Shea SM, Leventhal M: Endothelial contraction induced by histamine type mediators. J Cell Biol 1969, 42:647-672[Abstract/Free Full Text]
  5. Dale HH: Some chemical factors in the control of the circulation. Lancet 1929, 1:1179–1182, 1233–1237, 1285–1290
  6. Maples JA: A method for the covalent attachment of cells to glass slides for use in immunohistochemical assays. Am J Clin Pathol 1985, 83:356-363[Medline]
  7. Kolodney MS, Wysolmerski RB: Isometric contraction by fibroblasts and endothelial cells in tissue culture: a quantitative study. J Cell Biol 1992, 117:73-82[Abstract/Free Full Text]
  8. Lagunoff D, Rickard A: Methods for the study of rat peritoneal mast cell secretion. In Vitro Methods for Studying Secretion. Edited by Poisner AM, Trifaro JM. Amsterdam, Elsevier Science Publishers, 1987 pp 13–28
  9. Kruger P-G, Lagunoff D, Wan H: Isolation of rat mast cell granules with intact membranes. Exp Cell Res 1980, 129:83-93[Medline]
  10. Lagunoff D, Rickard A, Marquardt C: Rat mast cell tryptase. Arch Biochem Biophys 1991, 291:52-58[Medline]
  11. Horner AA: Macromolecular heparin from rat skin: isolation, characterization and depolymerization with ascorbate. J Biol Chem 1971, 246:231-239[Abstract/Free Full Text]
  12. Lagunoff D: Analysis of dye binding sites in mast cell granules. Biochemistry 1974, 13:3982-3986[Medline]
  13. Bitter T, Muir HM: A modified uronic acid carbazole reaction. Anal Biochem 1962, 4:330-334[Medline]
  14. Lagunoff D, Warren G: Determination of 2-deoxy-2-sulfoaminohexose content of mucopolysaccharides. Arch Biochem Biophys 1962, 99:396-400[Medline]
  15. Riley V: LDH in the normal and malignant state in mice and the influence of a benign enzyme-elevating virus. Busch H eds. Methods in Cancer Research. 1968, :pp 493-618 Academic Press, New York
  16. Lagunoff D, Pritzl P: Characterization of rat mast cell granule proteins. Arch Biochem Biophys 1975, 173:554-563
  17. Lagunoff D, Pritzl P, Mueller L: N-Acetyl-ß-glucosaminidase in rat mast cell granules. Exp Cell Res 1970, 61:129-132[Medline]
  18. McNeill H, Ryan TA, Smith SJ, Nelson WJ: Spatial and temporal dissection of immediate and early events following cadherin-mediated epithelial cell adhesion. J Cell Biol 1993, 120:1217-1226[Abstract/Free Full Text]
  19. Goeckeler ZM, Wysolmerski RB: Myosin light chain kinase-regulated endothelial cell contraction: the relationship between isometric tension, actin polymerization and myosin phosphorylation. J Cell Biol 1995, 130:613-627[Abstract/Free Full Text]
  20. Beil WJ, Login GR, Galli SJ, Dvorak AM: Ultrastructural immunogold localization of tumor necrosis factor-{alpha} to the cytoplasmic granules of rat peritoneal mast cells with rapid microwave fixation. J Allergy Clin Immunol 1994, 94:531-536[Medline]
  21. Reed JA, Albino AP, McNutt NS: Human cutaneous mast cells express basic fibroblast growth factor. Lab Invest 1995, 72:215-222[Medline]
  22. Qu Z, Liebler JM, Powers MR, Galey T, Ahmadi P, Huang X-N, Ansel JC, Batterfield JH, Planck SR, Rosenbaum JT: Mast cells are a major source of basic fibroblast growth factor in chronic inflammation and cutaneous hemangioma. Am J Pathol 1995, 147:564-573[Abstract]
  23. Grutzkaul A, Kruger-Krasagakes S, Baumeister H, Schwarz C, Kogel H, Welker P, Lippert U, Henz BM, Moller A: Synthesis, storage and release of vascular endothelial growth factor/vascular permeability factor (VEGF/VPF) by human mast cells: implications for the biological significance of VEGF 206. Mol Biol Cell 1998, 9:875-884[Abstract/Free Full Text]
  24. Lagunoff D: Contributions of electron microscopy to the study of mast cells. J Invest Dermatol 1972, 58:296-311[Medline]
  25. Yurt RW, Leid RW, Austen KF: Native heparin from rat peritoneal mast cells. J Biol Chem 1977, 252:518-521[Abstract/Free Full Text]
  26. Atkins FM, Friedman MM, Metcalfe DD: Biochemical and microscopic evidence for the internalization and degradation of heparin containing granules by bovine endothelial cells. Lab Invest 1985, 52:278-285[Medline]
  27. Barzu T, Molho P, Tobelem G, Petitou M, Caen J: Binding and endocytosis of heparin by human endothelial cells in culture. Biochim Biophys Acta 1985, 845:196-203[Medline]
  28. Glimelius B, Busch C, Hook M: Binding of heparin on the surface of cultured human endothelial cells. Thromb Res 1978, 12:773-782[Medline]
  29. Hiebert LM, McDuffie NM: The internalization and release of heparins by cultured endothelial cells: the process is cell source, heparin source, time and concentration dependent. Artery 1998, 17:107-108
  30. Patton WA, Granzow CA, Getts LA, Thomas SC, Zotter LM, Gunzel KA, Lowe-Krentz LJ: Identification of an endothelial heparin receptor using monoclonal antibodies that block heparin binding to porcine aortic endothelial cells. Biochem J 1995, 311:461-469
  31. Azizkhan RG, Azizkhan JC, Zetler BR, Folkman J: Mast cell heparin stimulates migration of capillary endothelial cells in vitro. J Exp Med 1980, 152:931-944[Abstract/Free Full Text]
  32. Gospodarowicz D, Ferrara N, Sweigerer L, Neufeld G: Structural characterization and biological functions of fibroblast growth factor. Endocr Rev 1987, 95–114
  33. Lee Y, McIntire LV, Zygourakis K: Analysis of endothelial cell locomotion: differential effects of motility and contact inhibition. Biotechnol Bioeng 1994, 43:622-634
  34. Ku P-T, D'Amore P: Regulation of basic fibroblast growth factor (bFGF) gene and protein expression following its release from sublethally injured endothelial cells. J Cell Biochem 1995, 58:328-343[Medline]
  35. Baluk P, Bolton P, Hirata A, Thurston G, McDonald D: Endothelial gaps and adherent leukocytes in allergen-induced early- and late-phase plasma leakage in rat airways. Am J Pathol 1998, 152:1463-1476[Abstract]
  36. Heimark RL, Degner M, Schwartz SM: Identification of a Ca2+-dependent cell-cell adhesion molecule in endothelial cells. J Cell Biol 1990, 110:1745-1756[Abstract/Free Full Text]
  37. Suzuki S, Sano K, Tanihara H: Diversity of the cadherin family: evidence for eight new cadherins in nervous tissue. Cell Regul 1991, 2:261-270[Medline]
  38. Lampugnani MG, Resnati M, Raiteri M, Pigott R, Pisacane A, Houen G, Ruicoh LP, Dejana E: A novel endothelial-specific membrane protein is a marker of cell-cell contacts. J Cell Biol 1992, 118:1511-1522[Abstract/Free Full Text]
  39. Esser S, Lampugnani MG, Corada M, Dejana E, Risau W: Vascular endothelial growth factor induces VE-cadherin tyrosine phosphorylation in endothelial cells. J Cell Sci 1998, 111:1853-1865[Abstract]
  40. Takeda H, Nagafuchi A, Yonemura S, Tsukita S, Behrens J, Birchmeier W, Tsukita S: V-serc kinase shifts the cadherin-based cell adhesion from the strong to the weak state and beta-catenin is not required for the shift. J Cell Biol 1995, 131:1839-1847[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 Arterioscler. Thromb. Vasc. Bio.Home page
J.-B. Michel
Anoikis in the Cardiovascular System: Known and Unknown Extracellular Mediators
Arterioscler. Thromb. Vasc. Biol., December 1, 2003; 23(12): 2146 - 2154.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Lagunoff, D.
Right arrow Articles by Rickard, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Lagunoff, D.
Right arrow Articles by Rickard, A.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS