(Stroke. 1995;26:265-270.)
© 1995 American Heart Association, Inc.
Articles |
Contraction of Human Brain Endothelial Cells Induced by Thrombogenic and Fibrinolytic Factors
An In Vitro Cell Culture Model
From the Stroke Center (Z.N., E.C., M.P.) and the Department of Biochemistry II (K.K., R.M.), Semmelweis University of Medicine, Budapest, Hungary.
Correspondence to Dr Zoltán Nagy, Semmelweis University of Medicine, National Stroke Center, H-1021 Budapest, Hûvösvölgyi út 116, Hungary.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Background and Purpose Vasogenic brain edema is a frequent complication of ischemic stroke. The mechanism of the blood-brain barrier opening that underlies the edema formation is poorly understood. In the present study we examined the response of endothelial cells cultured from adult human brain to thrombogenic and fibrinolytic factors that possibly accumulate in the occluded vascular segments in ischemic stroke.
Methods The changes in the morphology of cultured human brain microvascular endothelial cells were observed by phase-contrast light microscopy and quantified with computerized morphometry.
Results Active proteases (eg, thrombin, plasmin, urokinase) as well as heparin and protamine, but not fibrinogen and antithrombin III, produced significant changes in endothelial cell morphology. Two shape patterns of contraction were observed: protamine treatment resulted in rounded cells with a decrease in both cell perimeter and area, whereas all other agents induced spiderlike cell morphology with increased perimeter and reduced area. The rate of contraction was dose dependent, and at comparable enzyme concentrations plasmin produced faster contraction than thrombin. The observed changes were reversed 3 hours after abrogating the treatment.
Conclusions In an in vitro model we have demonstrated that factors involved in thrombus formation and dissolution induce endothelial cell contraction, which could affect focally the permeability of the blood-brain barrier by opening paracellular avenues between endothelial cells in vivo. Thus, the genesis of brain edema in thromboembolic stroke or occasionally during fibrinolytic therapy can be attributed in part to the contact of these factors with the microvascular endothelium.
Key Words: fibrinolysis • endothelial cells • brain edema • hemostasis
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The blood-brain barrier (BBB) is interposed between the circulating blood and the central nervous system and exists at the level of endothelial cells in vertebrates and in humans. In the brain, endothelial cells of microvessels are joined together by elaborate interendothelial tight junctions and form a monolayer that is devoid of fenestrae or wide gaps and contains very few endocytotic invaginations or pits. The endothelium regulates the ultrastable brain microenvironment as a permeability barrier, controlling different transport processes.1 The opening of interendothelial tight junctions of cerebral endothelium to macromolecules as a result of cell shrinkage after hyperosmotic treatment has been visualized.2 The opening of clefts by overstretching capillary walls has been demonstrated in a pressure pulse model and in acute hypertension.3 4
The normal endothelial cell layer provides a thromboresistant surface that prevents platelet or leukocyte adhesion and activation of the intrinsic and extrinsic coagulation system. The various factors of the coagulation-fibrinolytic system and the endothelial cells determine the thrombus formation.5
Vasogenic brain edema is reported to be a frequent complication of both ischemic stroke and fibrinolytic therapy.6 7 8 A number of investigations provide evidence for barrier disturbances in brain ischemia,9 but the basic mechanisms of their genesis are far from being fully elucidated. Because thrombogenic and fibrinolytic enzymes are known to interact with endothelial cells,10 11 12 13 14 15 we assume that they play a role in the opening of the BBB.
The culture of brain microvessel endothelial cells provides a system to study the major component controlling the permeability of the BBB. The interactions of a number of thrombogenic and fibrinolytic factors with noncerebral endothelial cells have been characterized.14 15 16 Because of the phenotypic diversity of endothelial cells, specialized according to their functions in different tissues, it is reasonable to examine the interaction of these factors with human brain capillary endothelial cells.
In the present study we measured the changes in the perimeter and surface area of brain capillary endothelial cells on time-lapse serial photography in the course of treatment with different thrombogenic and fibrinolytic factors. The changes of cell morphology allow us to hypothesize a novel mechanism operating in vasogenic brain edema induced by ischemic stroke and occasionally during thrombolysis.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Isolation and Culture of Human Brain Capillary Endothelial Cells
Capillaries were isolated from the brain of a 42-year-old subject autopsied because of sudden death due to cardiac arrest. The time interval between death and postmortem examination was approximately 4 hours. The brain was immediately transferred to the laboratory in cold tissue culture media. The leptomeninges and accompanying large vessels were carefully removed, and the brain cortex was chopped into 1-mm pieces. The isolation procedure thereafter was carried out according to the method for isolation of bovine brain microvessels published previously.17 18 Pieces were transferred into the conical tube rinsed with Dulbecco's minimal essential medium and were centrifuged for 10 minutes at a rate of 700 rpm. After the supernatant was removed, the tissue was suspended in 2 mL 0.5% collagenase solution per 1 g of tissue. After a 30-minute collagenase treatment at 37°C with use of a shaker, the samples were washed three times. After centrifugation for 7 minutes at 700 rpm at room temperature, the supernatant was aspirated and the pellets resuspended in 50 mL phosphate-buffered saline (PBS). The final pellet was resuspended in 30 mL 15% dextran per 1 g of tissue in one tube and centrifuged at 7200 rpm for 30 minutes at 4°C. The pellets were resuspended in 45% Percoll in Dulbecco's PBS, 25 mL/g of tissue, and centrifuged with 15 000 rpm for 25 minutes at 10°C. In the tube distinct bands could be differentiated. At the top of the tube below a wide band of myelin residuum, a narrow but well-defined dense band contained capillaries and single endothelial cells (the second band on the bottom of the tube contained red blood cells, occasionally neural cells). The harvested upper band was diluted to 50 mL with Dulbecco's PBS and washed three times, followed by centrifugation at 1000 rpm for 5 minutes at room temperature. Finally, the pellets were suspended in tissue culture medium (Dulbecco's modified essential medium with high glucose and F12 content [50%/50%], 30% heat-inactivated fetal calf serum, 10 U/mL heparin, 50 µg/mL endothelial cell growth mitogen, and antibiotics: 2.5 µg/mL amphotericin B, 50 µg/mL polymyxin B sulfate, 100 µg/mL penicillin, and 100 µg/mL streptomycin). Endothelial cells were seeded onto plastic wells or plastic flagons at 37°C in the tissue culture chamber. The medium was renewed regularly twice a week. Subcultured cells from the second to the sixth passages were used in the present study. Endothelial cells were characterized with 1,1'-diocatadecyl-3,3,3',3'-tetramethyl indocarbocyanine perchlorate–labeled acetylated low-density lipoprotein (DiI acet-LDL, Biomedical Technologies)19 and immunoperoxidase reaction with rabbit anti-human von Willebrand factor (Sigma).20
The freshly isolated suspension of human brain microvessel endothelium
and capillary segments attached to the substrate within 6 to 12 hours
after plating. Cells started migrating from the clumps 16 to 24 hours
after plating. Small discrete colonies of closely associated
spindle-shaped cells formed by days 4 to 7. By days 10 to 14 the small
colonies were becoming confluent, although complete confluence could
never be observed (Fig 1
). Lumen formation within the
colonies was quite a common phenomenon. The vast majority of the
endothelial cells (90% to 95%) bound the
endothelium-specific marker DiI acet-LDL and anti-human
von Willebrand factor antibodies (Fig 2).
|
|
Thrombogenic and Fibrinolytic Factors
Human plasma was collected from healthy volunteers.
Streptokinase, fluorescein isothiocyanate (FITC) on Celite, and porcine
pancreatic elastase were from Calbiochem. Human fibrinogen was
purchased from Kabi Diagnostica. Urokinase was from Abbott
Laboratories. p-Nitrophenyl p'-guanidinobenzoate
(NPGB) was obtained from Sigma. Sephadex G-25, lysine-sepharose 4B, and
heparin-sepharose 4B were the products of Pharmacia. All other reagents
were from Reanal.
Thrombin, plasminogen, and antithrombin III were prepared from freshly frozen human plasma.21 22 23 Miniplasminogen was prepared by digestion of plasminogen with porcine pancreatic elastase (20 U elastase per 1 mg plasminogen for 2 hours at room temperature. It was followed by inhibition of elastase with phenylmethylsulfonyl fluoride. Further steps were separation of the digest products on lysine-sepharose, and extensive dialysis against 25 mmol/L Na2HPO4/NaH2PO4 and 0.15 mol/L NaCl at pH 7.4). All protein concentrations were measured according to the method of Lowry et al.24 The active enzyme concentrations were determined by NPGB titration.25
Plasmin was prepared from the zymogen by activation with streptokinase (1000 U streptokinase per 1 mg precursor) for 20 minutes at room temperature.
Antithrombin III–enzyme complexes were prepared by incubating the enzyme with a twofold molar excess of antithrombin III for 30 minutes at room temperature.
Fibrin was prepared on the surface of endothelial cells. Fibrinogen was added to the cells in culture at the indicated concentrations and afterward was clotted with thrombin (final concentration 0.1 µmol/L, which clots fibrinogen in less than 20 seconds).
Visualization of Protein Binding to Endothelial Cells With a
Fluorescein Probe
The protein labeling with FITC was performed according to the
procedure of Rinderknecht26 with certain modifications.
The enzyme solution (protein concentration approximately 1 to 2 g/L)
was mixed with solid NaHCO3 to bring the concentration of
this reagent to 1%. The labeling reaction was initiated by the
addition of 2 mg of 10% FITC on Celite. After the incubation of the
reaction mixture for 1 hour at 4°C was accomplished under intensive
shaking, the mixture was centrifuged for 3 minutes and the supernatant
allowed to flow into a 1x5-cm column of Sephadex G-25, equilibrated
with 0.02 mol/L sodium phosphate buffer, pH 7.4. In the course of the
elution performed with the same buffer the labeled protein was
separated from the inert dye.
Morphometric Analysis
Morphometric analysis of individual endothelial cells was
performed by sequential measurement of the cell perimeter and surface
area in the course of treatment with different thrombogenic and
thrombolytic factors with use of AUTOCAD computer
software (Autodesk Inc). Ratios of perimeter to area were
calculated.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
FITC-labeled thrombin and plasmin uniformly decorated the endothelial cell membrane in both formaldehyde-fixed and nonfixed samples (Fig 2
). Cell contraction was observed in the presence of proteolytic enzymes in a dose-dependent fashion, whereas endothelial cells preserved their shape and morphology during the 120-minute observation period in tissue culture medium without additives. Thrombin, plasmin, urokinase, and pancreatic elastase induced marked contraction of endothelial cells, resulting in increased perimeter and reduced cell area.
After a 30- to 40-minute period of slow contraction the endothelial
cells, treated with thrombin, reacted rapidly, resulting in spiderlike
cell morphology (Fig 3). The contracted cells finally
had narrow flaps around the round cell body. The thrombin used at
different concentrations (50 nmol/L, 0.44 µmol/L, and 5.5 µmol/L)
resulted in the same pattern of cell contraction with dose-dependent
rates. The contraction was reversible: cells restored their morphology
3 hours after the removal of the thrombin solution.
|
The maximal cell contraction induced by plasmin was observed within 30
minutes (Fig 4). The different concentrations
(Table) induced similar effects on the cells. In the
presence of 2 mmol/L 6-aminohexanoate, the effect of plasmin was
essentially reversed.
|
|
The formation of a fibrin layer over the endothelial cells resulted in
slow cell shrinkage. The presence of fibrin markedly reduced the effect
of plasmin on the cells, and plasmin reduced the effect of fibrin
(Table).
Heparin at 2 U/mL and 0.2 U/mL concentrations induced endothelial cell contraction after 40 to 50 minutes of treatment, whereas a lower concentration of heparin (0.02 U/mL) did not affect the ratio of cell perimeter to surface area.
Protamine sulfate induced endothelial cell contraction in the initial 30 minutes. The cells became rounded, and both the perimeter and the area decreased.
Fibrinogen, antithrombin III, and streptokinase did not affect the cell morphology in the tissue culture during the 120-minute observation period.
In the groups of combined treatment (fibrin plus plasmin, plasmin plus
6-aminohexanoate, antithrombin III plus fibrin, and antithrombin III
complexes with thrombin or plasmin), no or moderate cell contraction
was detected (Table).
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
If an occluding thrombus is formed in the brain blood vessels, the massive thrombin generated on its surface may overcome the protective effect of the plasma protease inhibitors.27 Thus, downstream in the vascular bed active thrombin can interact with the capillary endothelial cells. As evidenced by our study, thrombin induces contraction in human brain capillary endothelial cells in culture, probably accompanied in vivo by opening of intercellular tight junctions (Fig 3
). The number of thrombin receptors in human arterial
endothelial cells is reported to be relatively low,28 and
no data are available for brain capillary endothelial cells concerning
thrombomodulin or heparan sulfate as potential thrombin binding sites.
Thus, the elucidation of the mechanism underlying the observed thrombin
interactions requires further experimentation. The interactions of plasmin with endothelial cells of diverse origin are well characterized in terms of binding constants.12 13 In our study endothelial cells from human brain capillaries contracted in the course of plasmin treatment. This response of the endothelium may be the basis of brain edema after thrombolytic treatment in ischemic stroke. If the administered plasminogen activators (streptokinase, urokinase, or tissue-type plasminogen activator) generate high concentrations of plasmin, the enzyme will bind not only to its substrate (fibrin) and its plasma inhibitors but also to the capillary endothelium. The consequent cell contraction would increase the permeability of the BBB. 6-Aminohexanoate is clinically used as an antifibrinolytic drug because of its well-known property of blocking the interactions of plasmin kringle domains with lysine residues on fibrin.29 30 In the presence of 6-aminohexanoate the plasmin-induced cell contraction is substantially diminished. This observation implies a role for the kringle domains in the interactions of plasmin and endothelial cells.
Of the two plasminogen activators (urokinase, streptokinase), only urokinase induces cell contraction. Thus, all of the proteases used here contract the brain endothelial cells, whereas proteins that possess no proteolytic activity (eg, fibrinogen, antithrombin, and streptokinase [the latter is frequently used in fibrinolytic therapy]) have no effect on the cell morphology. When the active center of the enzymes is blocked by a protease inhibitor (antithrombin III), cell contraction is essentially negligible, suggesting the involvement of the active center in the effects of the enzymes on the endothelial cells.
Fibrin is known to cause changes in the morphology of endothelial
cells31 and possibly contributes to the increase in the
permeability of the pulmonary microvasculature in thromboembolic
injury.32 These observations are in agreement with our
findings in human brain microvessel endothelial cell culture. The fact
that the lowest concentration of fibrin produces the most pronounced
effect strongly suggests that thrombin (present in all fibrin gels
at the same concentration and potentially released from its substrate)
is responsible for the cell contraction rather than the fibrin itself.
The elucidation of this question requires further experiments. On the
other hand, fibrin reduces the effects of plasmin on the endothelial
cells (Table), presumably because of the higher affinity of plasmin to
its substrate (fibrin) than to the cell receptors. Simultaneously
plasmin digests fibrin, and as a result the effect of fibrin and
plasmin together is less than that of either agent alone.
The opening of the BBB to blood-borne tracer (horseradish peroxidase)33 by perfusing a bolus of the polyanion heparin via the internal carotid artery may be due to endothelial cell contraction demonstrated in our present study. It has been suggested that heparin affects the endothelium through its Ca2+ chelating properties.34 The integrity of tight junctioned fibrils has been shown to depend on Ca2+.35 36 Similar mechanisms can be suggested in the heparin-induced cell contraction, but this hypothesis should be experimentally studied.
The polycation protamine (used clinically as a heparin neutralizer) in a boluslike perfusion through the carotid artery results in disassembly of tight junctions in the microvessel endothelium.33 Protamine, as a cation, binds to negatively charged glycoproteins and phospholipids of the endothelial cell membrane, but the exact mechanism of the charge-related cell contraction, induced by polycations and polyanions, is far from being understood.
The contraction of brain microvessel endothelial cells, observed on the action of a number of thrombogenic and fibrinolytic factors, is a phenomenon to be considered as a possible factor in focal opening of the BBB in cases of brain thrombosis and fibrinolysis. These observations allow us to suggest a novel concept of the mechanism of vasogenic brain edema in ischemic stroke.
|
Acknowledgments |
|---|
This study was supported in part by Hungarian grants OTKA 1066, ETT 1-2/1991, ETK 2/4-185, and United States–Hungarian Joint Fund grant 081/91.
Received April 15, 1994; revision received November 7, 1994; accepted November 7, 1994.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
1. Nagy Z. Blood-brain barrier and the cerebral endothelium. In: Johanson BB, Owman Ch, Widner H, eds. Pathophysiology of the Blood-Brain Barrier. Amsterdam, Netherlands: Elsevier Science Publishing Co; 1990:11-29.
2. Nagy Z, Peters H, Hüttner I. Fracture faces of cell junctions in cerebral endothelium during normal and hyperosmotic conditions. Lab Invest. 1984;50:313-322. [Medline] [Order article via Infotrieve]
3. Nagy Z, Mathieson G, Hüttner I. Opening of tight junctions in cerebral endothelium, II: effect of pressure-pulse induced acute arterial hypertension. J Comp Neurol. 1979;185:579-586. [Medline] [Order article via Infotrieve]
4. Nagy Z, Mathieson G, Hüttner I. Blood-brain barrier opening to horseradish peroxidase in acute arterial hypertension. Acta Neuropathol (Berl). 1979;48:45-53. [Medline] [Order article via Infotrieve]
5. Machovich R. Hemostasis. In: Machovich R, ed. Blood Vessel Wall and Thrombosis. Boca Raton, Fla: CRC Press; 1988:1-80.
6.
Wardlaw JM, Warlow CP. Thrombolysis in acute ischemic stroke:
does it work? Stroke. 1992;23:1826-1839.
7. Levine SR, Brott TG. Thrombolytic therapy in cerebrovascular diseases. Prog Cardiovasc Dis. 1992;34:235-262. [Medline] [Order article via Infotrieve]
8. Gumerlock MK. Cerebrovascular disease and the blood-brain barrier. In: Neuwelt EA, ed. Implications of the Blood-Brain Barrier and Its Manipulation. New York, NY: Plenum Medical Book Co; 1988:495-565.
9. Nagashima G, Nowak TS, Joo F, Ikeda J, Ruetzler C, Lohr J, Klatzo I. The role of the blood-brain barrier in ischemic brain lesions. In: Johanson BB, Owman Ch, Widner H, eds. Pathophysiology of the Blood-Brain Barrier. Amsterdam, Netherlands: Elsevier Science Publishing Co; 1990:311-321.
10.
Awbrey BJ, Hoak JC, Owen WG. Binding of human thrombin to
cultured human endothelial cells. J Biol Chem. 1979;254:4092-4096.
11.
Lollar P, Hoak JC, Owen WG. Binding of thrombin to cultured
human endothelial cells. J Biol Chem. 1980;255:10279-10283.
12. Bauer PI, Machovich R, Büki KG, Csonka É, Koch SA, Horváth I. Interaction of plasmin with endothelial cells. Biochem J. 1984; 218:119-124.
13.
Hall SW, Humphries JE, Gonias SL. Inhibition of cell surface
receptor bound plasmin by 
2-antiplasmin and
2-macroglobulin. J Biol Chem. 1991;266:12329-12336.
14. Goldal KS, Evensen SA, Brosstad F. Effects of thrombin on the integrity of monolayers of cultured human endothelial cells. Thromb Res. 1982;27:575-584. [Medline] [Order article via Infotrieve]
15.
Bauer PI, Machovich R, Arányi P, Büki KG, Csonka
É, Horváth I. Mechanism of thrombin binding to endothelial
cells. Blood. 1983;61:368-372.
16. Sago H, Iinuma K. Cell shape change and cytosolic Ca2+ in human umbilical-vein endothelial cells stimulated with thrombin. Thromb Haemost. 1992;67:331-334. [Medline] [Order article via Infotrieve]
17. Tao-Cheng JH, Nagy Z, Brightman MW. Tight junctions of brain endothelium in vitro are enhanced by astroglia. J Neurosci. 1987;7:3293-3299. [Abstract]
18. Bowman PD, Ennis SR, Rarey KE, Betz AL, Goldstein GW. Brain microvessel endothelial cells in tissue culture: a model for study of blood-brain barrier permeability. Ann Neurol. 1983;14:396-402. [Medline] [Order article via Infotrieve]
19.
Voyta JC, Via DP, Butterfield CE, Zetter BR. Identification
and isolation of endothelial cells based on their increased uptake of
acetylated low-density lipoprotein. J Cell Biol. 1984;99:2034-2040.
20. Dorovini-Zis K, Prameya R, Bowman PD. Culture and characterization of microvascular endothelial cells derived from human brain. Lab Invest. 1991;64:425-436. [Medline] [Order article via Infotrieve]
21.
Fenton JW, Fasco MJ, Stacknow AB. Human thrombins. J
Biol Chem. 1977;252:3587-3598.
22. Castellino FJ, Powell JR. Human plasminogen. Methods Enzymol. 1981;80:365-378.
23. Damus PS, Rosenberg RD. Antithrombin-heparin cofactor. Methods Enzymol. 1976;45:653-669. [Medline] [Order article via Infotrieve]
24. Lowry OH, Rosenbrough NJ, Farr AL, Randall RJ. Protein measurement with the folin phenol reagent. J Biol Chem. 1951; 193:265-275.
25. Chase T, Shaw E. Titration of trypsin, plasmin and thrombin with p-nitrophenyl p'-guanidinobenzoate. Methods Enzymol. 1970; 19:20-27.
26. Rinderknecht H. Ultra-rapid fluorescent labelling of proteins. Nature. 1962;193:167-168. [Medline] [Order article via Infotrieve]
27. Travis J, Salvensen GS. Human plasma proteinase inhibitors. Ann Rev Biochem. 1983;52:655-709. [Medline] [Order article via Infotrieve]
28. Nelken N, Soifer S, O'Keefe J, Vu TK, Charo J, Coughlin S. Thrombin receptor expression in normal and atherosclerotic human arteries. J Clin Invest. 1992;90:1614-1621.
29.
Bickford AF, Taylor FB, Sheena R. Inhibition of the
fibrinogen- plasmin reaction by 
-aminocarboxylic acids and
alkylamines. Biochim Biophys Acta. 1964;92:328-333. [Medline]
[Order article via Infotrieve]
30. Morris JP, Blatt S, Powell JR, Strickland DK, Castellino FJ. Role of lysine binding regions in the kinetic properties of human plasmin. Biochemistry. 1981;20:4811-4816. [Medline] [Order article via Infotrieve]
31. Kadish JL, Butterfield CE, Folkman J. The effect of fibrin on cultured vascular endothelial cells. Tissue Cell. 1979;11:99-108. [Medline] [Order article via Infotrieve]
32. Johnson A, Tahamont MV, Malik AB. Thrombin-induced lung vascular injury: role of fibrinogen and fibrinolysis. Am Rev Respir Dis. 1983;128:38-44. [Medline] [Order article via Infotrieve]
33. Nagy Z, Peters H, Hüttner I. Charge-related alterations of the cerebral endothelium. Lab Invest. 1983;49:662-671. [Medline] [Order article via Infotrieve]
34.
Husdan H, Leung M, Orepoulos D, Rapoport A. Measurement of
serum and plasma ionic calcium with the `space stat 20 ionized calcium
analyzer.' Clin Chem. 1977;23:1775-1777.
35. Meldoleshi J, Castiglioni G, Parma R, Nassivera N, De Camilli P. Ca++-dependent disassembly and reassembly of occluding junctions in guinea pig acinar cells: effects of drugs. J Cell Biol. 1978; 79:156-172.
36. Nagy Z, Goehlert VG, Wolfe LS, Hüttner I. Ca2+ depletion induced disconnection of tight junctions in isolated rat brain microvessels. Acta Neuropathol (Berl). 1985;68:48-52. [Medline] [Order article via Infotrieve]
This article has been cited by other articles:
 |
|
 |
|
  L. Denes, Z. Bori, E. Csonka, L. Entz, and Z. Nagy Reverse Regulation of Endothelial Cells and Myointimal Hyperplasia on Cell Proliferation by a Heatshock Protein-Coinducer After Hypoxia Stroke, March 1, 2008; 39(3): 1022 - 1024. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Mirza, V. A. Schmidt, C. K. Derian, J. Jesty, and W. F. Bahou Mitogenic Responses Mediated Through the Proteinase-Activated Receptor-2 Are Induced by Expressed Forms of Mast Cell alpha - or beta -Tryptases Blood, November 15, 1997; 90(10): 3914 - 3922. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||