Activated Mast Cells Increase the Level of Endothelin-1 mRNA in Cocultured Endothelial Cells and Degrade the Secreted Peptide
Subendothelial mast cells have been implicated in the pathogenesis of allergic inflammation, in atherosclerosis, and in the regulation of vascular tone. Because endothelin-1 (ET-1) is an important regulator of vascular tone and has also been implicated in the pathogenesis of atherosclerosis, we studied the role of mast cells in the metabolism of endothelial cell-derived ET-1. In mast cell-endothelial cell cocultures, activation of the mast cells with ensuing degranulation was accompanied by the increased expression of ET-1 mRNA in the endothelial cells, yet the immunoreactive ET-1 protein in the coculture medium disappeared almost completely during the 24-hour coculture. Activation of the mast cells with the ensuing degranulation resulted in proteolytic degradation of ET-1 by the 2 neutral proteases, chymase and carboxypeptidase A, of the exocytosed mast cell granules. With synthetic ET-1 and purified mast cell granule enzymes, efficient degradation of ET-1 by chymase and carboxypeptidase A was verified. These in vitro results imply a novel role for mast cell-derived neutral proteases in ET-1 metabolism and suggest that activated subendothelial mast cells are important local regulators of ET-1 metabolism.
Mast cells (MCs) reside in connective tissue and in mucosal surfaces, often in association with the microvascular endothelium.1 Activated MCs can release various potent proinflammatory mediators, such as histamine, proteases, and several cytokines, such as tumor necrosis factor-α (TNF-α).2 Activation of the MCs close to the endothelium has been shown to result in rolling, adhesion, and extravasation of leukocytes.3 Recently, activated MCs have also been found subendothelially in human atherosclerotic plaques.4 Importantly, the large number of activated MCs present in eroded atherosclerotic plaques suggests that the MCs are also involved in the development of the thrombotic complications of atherosclerosis.5
Endothelial cell (EC)-derived endothelin (ET)-1 is a potent vasoactive peptide with multiple functions, notably, vasoconstriction and growth-inducing properties.6 ET-1 can also induce rolling and adhesion of leukocytes via ET type A receptors,7 and ET-1 has been shown to be capable of inducing degranulation of MCs.8 Several findings suggest that ET-1 may also participate in the development of atherosclerosis: (1) immunoreactive ET-1 and increased ET-1 mRNA levels have been found in atherosclerotic plaques,9,10⇓ (2) a correlation has been reported between increased plasma levels of ET-1 and the severity of atherosclerosis,11 and (3) oxidized LDL has been shown to increase ET-1 levels in the endothelium in vitro.12 Interestingly, ET type A receptor-dependent vasoconstriction in atherosclerotic coronary arteries of hypercholesterolemic mice was found to lead to total coronary occlusion with ensuing myocardial infarction.13 Thus, ET-1 appears to be one of the compounds leading to increased vasoreactivity of the culprit lesions in acute coronary events.
Because the ECs of atherosclerotic plaques express ET-1 and are also exposed to the products released from activated subendothelial MCs, we established a coculture system to study the effect of MCs and MC activation on the ET-1 metabolism of ECs.
Male Wistar rats weighing 200 to 250 g were used. For 2 weeks before the studies, the rats were allowed free access to tap water and standard rat chow. The studies were approved by the institutional (Minerva Institute) standing committee on animal research.
Isolation of Rat Peritoneal MCs and Purification of Granule Chymase and Carboxypeptidase A
MCs (85% to 95% purity) were isolated from the peritoneal and pleural cavities of male Wistar rats as described.14,15⇓ The animals were killed by cervical dislocation under CO2 anesthesia, and the study was approved by the institutional (Minerva Institute) standing committee on animal research. The purity of the MCs was estimated by staining the isolated cells with Moore-James stain and counting the toluidine blue-positive cells.
To isolate MC granule remnants, the cells were stimulated to degranulate with compound 48/80 (Sigma), and the expelled granules (granule remnants), containing 2 neutral proteases, chymase and carboxypeptidase A (CPA), which were tightly bound to the heparin-proteoglycan matrix, were isolated as described by Lindstedt et al.15 The chymase in the granule remnants was inactivated by pretreatment with phenylmethylsulfonyl fluoride (PMSF), and inactivation was verified by measuring chymase activity spectrophotometrically with N-benzoyl-l-tyrosine ethyl ester as substrate.15 The activity of CPA was inhibited by adding potato carboxypeptidase inhibitor (CPA inhibitor, 0.4 mg/mL) to the incubation medium.
Purification of Chymase and CPA From MC Granule Remnants
To dissociate chymase and CPA from the heparin proteoglycans of the granule remnants, the remnants were incubated in high salt buffer consisting of 10 mmol/mL phosphate buffer supplemented with 2 mol/L KCl, pH 7.0.16 The mixture was then applied to a Sephacryl S-200 HR column and eluted with the same high salt buffer. Fractions containing chymase or CPA activity were collected separately and further purified on a HiTrap heparin-Sepharose column. A single band with an apparent molecular mass of 28 kDa was detected in purified chymase solution, and a single band with a molecular mass of 35 kDa was detected in purified CPA solution after SDS-PAGE and silver staining. Analysis of the proteolytic products of angiotensin I with reverse-phase high performance liquid chromatography (HPLC) revealed that no CPA activity was present in the purified chymase solution and, correspondingly, that no chymase activity was present in the purified CPA solution.
Degradation of ET-1 by Purified Rat Chymase and CPA
Synthetic ET-1 was obtained from the Peptide Institute. Two hundred forty-six nanograms (4 N-benzoyl-l-tyrosine ethyl ester units) of purified chymase or 100 ng of purified CPA was incubated with 120 pg of ET-1 (1 mg/mL of albumin in PBS) in a total volume of 120 μL, for 4 hours at 37°C. PMSF (0.25 mg/mL, final concentration) was added to inhibit chymase and CPA inhibitor (0.4 mg/mL, final concentration) to inhibit CPA. After 4 hours, degradation was stopped by adding PMSF and CPA inhibitor to all samples, which were stored at −20°C until the ET-1 levels were determined, as described below.
Human aortic ECs (CC-2535) were purchased from Clonetics. Cells in the third passage, 500 000 per cryovial in liquid N2, were rapidly thawed under warm water. One milliliter of cell suspension (500 000 cells) was mixed with 24 mL of cell culture medium (M-199 supplied with 1 mmol/L sodium pyruvate, 4 mmol/L l-glutamine, 100 IU/mL penicillin, 100 mg/mL streptomycin, and 20% FCS; all were from GIBCO Laboratories) and seeded into five 25-cm2 cell culture flasks precoated with 0.2% gelatin. Thereafter, ECs were allowed to grow to confluence in the flasks, removed by trypsinization, seeded into 35-mm wells, allowed to grow to near confluence, and finally used for the experiments.
In the second passage, confluent ECs were washed with PBS, removed by trypsinization (0.05% trypsin-0.02% EDTA), and seeded into 6-multiwell plates at 1×105 cells per well in M-199 (GIBCO), with 20% FCS. Freshly isolated MCs (1×105 cells per well) were then added to the wells, and then 5 μg/mL MC activator, substance 48/80, or vehicle was added. The number of MCs used in the coculture experiments was based on the estimated density of MCs in the human arterial wall.17 The cells were incubated at 37°C in humidified 95% air/5% CO2 for an additional 24 hours. After incubation, the media were collected and centrifuged at low speed to sediment the MCs, and then the incubation media were collected and stored at −20°C until they were assayed for ET-1. ECs were washed twice with ice-cold PBS, followed by isolation of total RNA.
In another set of experiments, 1×105 rat peritoneal MCs per milliliter were incubated with substance 48/80 or vehicle at 37°C. The MCs were then removed by centrifugation, and the supernatants were collected. Thereafter, volumes of supernatants corresponding to the material released from 5×103, 2.5×104, or 5×104 MCs were added to wells containing 1×106 ECs per well. The ECs were incubated for 24 hours, and the culture medium was assayed for ET-1.
Radioimmunoassay of ET-1
ET-1 radioimmunoassay was performed as described.18 ET antiserum was generated in rabbits. The sensitivity of the assay was 0.8 pg per tube, and recovery of the synthetic ET-1 added was 80%. The antiserum cross-reacted completely with human ET-2 and ET-3 (Peninsula). However, it has been previously shown that ECs produce only ET-1 but not ET-2 or ET-3.19 Cross-reactions were <0.1% with the following substances: 20 to 50, 74 to 91, and 171 to 201 sequences of preproendothelin (Peptide Institute); human big endothelins 1 to 38 and 22 to 38 (Peninsula); human atrial natriuretic peptide (Peninsula); angiotensin II (Bachem); and Arg8-vasopressin (Bachem).
RNase Protection Assay
ET-1 and actin probes were generated by reverse transcription-PCR with the use of human EC total RNA.20 32P-labeled riboprobes were transcribed by using a Maxiscript in vitro transcription kit (Ambion Inc), and a solution hybridization RNase protection assay was carried out by using the RPA II Ribonuclease Protection Assay Kit (Ambion) according to the manufacturer’s instructions. The protected RNA was resolved by electrophoresis on 5% polyacrylamide-8 mol/L urea gels, and bands were visualized by autoradiography. The results were quantified by densitometry and normalized for the amount of β-actin.
Degradation of Radioiodinated ET-1 in Coculture of ECs and Activated MCs
Radioiodinated ET-1 (15 000 cpm) was added to wells containing ECs and activated MCs. Culture medium was collected and centrifuged after 24 hours, and the supernatant was subjected to HPLC as described by Sirviö et al.21 The radioactivity in the fractions was then counted in an LKB γ-counter.
Effect of Isolated MC Granule Remnants on ET-1 Levels in EC-Conditioned Culture Medium
The effect of isolated granule remnants on the ET-1 content of human aortic EC-conditioned culture medium, referred to as EC-conditioned medium, was tested. Different concentrations of granule remnants (0.008 to 5 μg of protein per milliliter) or PMSF-treated granule remnants (5 μg/mL) were incubated for 24 hours with EC-conditioned medium, followed by measurement of ET-1. In some experiments, we studied the effect of the CPA inhibitor (0.4 mg/mL) on the effect of granule remnants on ET-1 in EC-conditioned culture medium.
HPLC Analysis of ET-1 Degradation by MC Granule Remnants
Untreated granule remnants, PMSF-treated granule remnants (chymase-inhibited remnants), and granule remnants in the presence of 0.4 mg/mL CPA inhibitor (CPA-inhibited remnants; each was 30 μg of granule remnant total protein per milliliter) were incubated for 1 hour at 37°C with HPLC-grade synthetic ET-1 (60 μg/mL, The Peptide Laboratory). After incubation, the samples were precipitated with ethanol, and the supernatants were evaporated and diluted with 105 μL HPLC running buffer (water/0.037% trifluoroacetic acid). Reverse-phase HPLC was performed with a VYDAC 218 TP52 column (0.21 cm×25 cm, VYDAC Corp) equilibrated with 0.05% trifluoroacetic acid, with a linear gradient of acetonitrile (0% to 70% in 30 minutes, flow rate of 0.15 mL/min) on Beckman System Gold 126 HPLC equipment (Beckman Coulter Inc). The detection was set to 220 nm.
Results are expressed as mean±SEM. Statistical analyses of the raw data were performed with 1-way ANOVA, followed by post hoc tests (Student t test) for comparison between groups. Statistical significance was accepted at a value of P<0.05.
Effect of MCs on ET-1 Levels in EC Cultures
The concentration of ET-1 in human aortic EC culture medium was 600±60 pg/mL (mean±SEM, n=3) after 24 hours of incubation (Figure 1A), reflecting secretion of ET-1 into the medium during incubation. When ECs and MCs were incubated together and a noncytotoxic MC activator, compound 48/80, was added to the culture medium, the level of ET-1 in the culture medium decreased dramatically, being only 35±6 pg/mL 24 hours after stimulation of the MCs. Compound 48/80, when added to EC cultures in the absence of MCs, did not have any effect on ET-1 levels (Figure 1A, middle bar).
We next studied the effect of MC-preconditioned media on ET-1 levels in EC cultures. As shown in Figure 1B, addition of the preconditioned media derived from activated MCs decreased the level of ET-1, whereas that derived from nonactivated MCs did not. The extent of the decrease did depend on the quantity of added medium, and a profound decrease of ET-1 levels was observed when 200 μL of MC-preconditioned medium containing material released from 5×104 MCs was added into the EC culture medium. These results show that direct contact between MCs and ECs was not required for the observed ET-1-lowering effect.
Effects of MCs on ET-1 mRNA Levels in ECs
The ET-1 mRNA level in control ECs at 24 hours was 0.53±0.03 arbitrary units (Figure 2). If ECs were cocultured with MCs that were stimulated with compound 48/80 (5 μg/mL), the ET-1 mRNA level in the ECs increased significantly (to 0.85±0.02 arbitrary units, bar a versus bar b of Figure 2; P<0.001). In contrast, coculture of the ECs with nonactivated MCs resulted in only a modest and insignificant increase of ET-1 mRNA levels (bar a versus bar c of Figure 2, P>0.05). This small increase of ET-1 mRNA in the presence of nonactivated MCs was most likely caused by the slow spontaneous degranulation of MCs during the 24-hour incubation, which also takes place in the absence of any activators. The addition of MC activator 48/80 in the absence of MCs did not have any significant effect on the ET-1 mRNA levels (bar d of Figure 2).
Degradation of Radioiodinated ET-1 by Activated MCs
To explore why MC activation resulted in decreased ET-1 levels in the EC culture medium even though there was an increase in ET-1 mRNA levels, we studied the fate of the ET-1 in the coculture system. For this purpose, radioiodinated 125I-ET-1 was added to wells containing ECs either alone or in coculture with activated MCs, and after incubation for 24 hours, we analyzed the HPLC fractions for radioactivity. Pure 125I-ET-1 produced only 1 peak of radioactivity. After 24 hours of incubation, 3 major peaks of radioactivity (fractions 42 to 50) were observed, reflecting partial degradation of 125I-ET-1 in the culture medium (not shown). However, incubation of 125I-ET-1 with ECs and activated MCs resulted in a profound shift in the distribution of radioactivity, yielding a major peak that eluted in fractions 31 to 34 (not shown). Thus, the 125I-ET-1 had been degraded, and this degradation depended on stimulation of MCs in the EC-MC coculture system.
Degradation of Released ET-1 by Isolated MC Granule Remnants
The results presented above suggest that activated MCs release material capable of degrading ET-1. On activation, rat peritoneal MCs exocytose their granules, which contain 2 neutral serine proteases, chymase and CPA, which are tightly bound to the granule proteoglycan matrix.22 Therefore, we next isolated the exocytosed granules, ie, granule remnants, from activated MCs and studied their effect on ET-1 levels in EC-conditioned medium. Figure 3 shows that granule remnants decrease the ET-1 levels in EC-conditioned medium dose-dependently.
To determine which of the 2 granule remnant-associated enzymes, chymase or CPA, was responsible for the observed degradation of ET-1 in EC-conditioned medium, we selectively inhibited these enzymes (Figure 4A). First, the activity of granule chymase was inhibited with the serine protease inhibitor PMSF (leaving CPA active). Treatment of the granule remnants with PMSF resulted in almost complete inhibition of the chymase activity of the granule remnants, from 2.700 to 0.003 U/μg total granule remnant protein. However, in spite of the inhibition of chymase activity, the granule remnants (5 μg/mL) efficiently degraded ET-1 in the medium. Indeed, at the concentration of granule remnants used, CPA alone could degrade most (95%) of the ET-1 present in the incubation medium during the 24-hour incubation period. When the CPA activity of the granule remnants was selectively inhibited with a CPA inhibitor (leaving chymase active), the concentration of immunoreactive ET-1 in the incubation medium still decreased, indicating that chymase can also degrade ET-1. However, the granule remnants possessing only chymase activity degraded ET-1 less effectively (50%) than those with only CPA activity (95%), suggesting that chymase is less effective in degrading ET-1. This conclusion is supported by the results of a time-course experiment in which, after 4 hours of incubation, fully active granule remnants had degraded ≈70% of ET-1, but those with mere chymase activity had degraded only 10% of ET-1 (Figure 4B). Similar to the results shown in Figure 4A, the granule remnants with mere chymase activity had degraded ≈50% of the ET-1 after 24 hours of incubation.
Degradation of Synthetic ET-1 by Isolated MC Granule Proteases
In the coculture experiments, we measured the degradation of the EC-derived ET-1 present in the EC culture medium. Accordingly, it was not possible to deduce whether CPA and chymase degraded ET-1 directly or indirectly by activating some other ET-1-degrading enzyme(s) present in the EC-conditioned medium. Therefore, we next isolated and purified CPA and chymase from the exocytosed MC granules and studied their ability to degrade synthetic ET-1 in buffer solution. These results are shown in Figure 5. No spontaneous degradation of ET-1 was observed during the 4-hour incubation period. However, the addition of either chymase or CPA resulted in a significant decrease in the amount of immunoreactive ET-1. When these 2 enzymes were added together, the degradation of ET-1 was even more profound. The degradation of ET-1 by chymase could be inhibited by adding PMSF, and the degradation by CPA could be inhibited by adding the CPA inhibitor.
Finally, to verify that the loss of immunoreactivity of ET-1 by CPA or chymase was caused by proteolytic degradation of the peptide, we analyzed the HPLC pattern of ET-1 after incubation with chymase and/or CPA (Figure 6). For this purpose, ET-1 was incubated for 1 hour at 37°C with granule remnants in which either chymase (Figure 6B), CPA (Figure 6C), or both chymase and CPA (Figure 6D) were active. Intact ET-1 showed a single peak eluting at 28 minutes (Figure 6A). Degradation of ET-1 with chymase resulted in a significant decrease in the size of the ET-1 peak and in the appearance of several smaller, faster eluting peaks (Figure 6B). Degradation of ET-1 with CPA resulted in complete disappearance of the ET-1 peak and appearance of the 2 smaller, faster eluting peaks at 27 and 26 minutes (Figure 6C). Identical results were also observed when purified CPA from a commercial source (Sigma) was used instead of MC granules (data not shown). In the presence of both chymase and CPA, the ET-1 peak disappeared completely, and several minor faster eluting peaks appeared (Figure 6D). To study the sites of cleavage of ET-1 by CPA, we analyzed the HPLC fragments of the 2 major degradation products of ET-1 by using mass spectroscopic analysis (ESI-MS, Esquire-LC, Bruker-Daltonics). This analysis revealed the generation of 2 truncated peptides with molecular masses of 2078 and 2190 Da, ie, masses corresponding to ET-1 (1-18) and ET-1 (1-19), respectively. Thus, the exopeptidase CPA had cleaved the C-terminal amino acids tryptophan and 2 isoleucines from the C-terminus of ET-1 (data not shown). Taken together, the experiments demonstrate that chymase and CPA can degrade ET-1.
A close relationship seems to exist between ECs and MCs regarding the metabolism of EC-derived ET-1. In cocultures, activated MCs increased the mRNA levels of ET-1 in ECs. However, in contrast to this induction of ET-1 mRNA expression, an increased rate of degradation of secreted ET-1 by the activated MCs was observed. The degradation of ET-1 was shown to be caused by exocytosed MC granules. Both of the 2 granule-associated neutral proteases, chymase and CPA, were involved in ET-1 degradation. This was shown by using selective enzyme inhibitors. The findings were further confirmed by studies in which synthetic ET-1 and purified enzymes were used and in which the resulting degradation products were analyzed by using HPLC. All these studies showed conclusively that chymase and CPA are capable of degrading ET-1. Human chymase has been shown previously to be capable of degrading ET-1,23 but to our knowledge, this is the first report to show that CPA can also degrade ET-1. This peptidolytic function could represent an important novel role for CPA in vivo in the pericellular region of activated MCs. This suggestion is supported by the fact that no endogenous inhibitors are known at present for the MC carboxypeptidase in rodents or in humans. The possibility that CPA has a significant role in ET-1 metabolism, at least in some tissues of the human body, is also supported by a recent study showing that a markedly increased immunoreactivity and an abnormal distribution of ET-1 are observed in the brains of patients affected with galactosialidosis.24 This disease results from a deficiency of lysosomal protective protein/cathepsin A, a protein with serine carboxypeptidase activity that can also degrade ET-1 in vitro.
The present findings raise an obvious question: Is this effect of activated MCs on ET-1 degradation also significant in vivo? Approximately one third of the ET-1 synthesized by ECs is released luminally, and two thirds is released abluminally.25 Luminally released ET-1 is rapidly taken up by the lungs and the kidneys,26 whereas the fate of abluminally released ET-1 is not known. Given the localization of MCs subintimally in close contact with ECs in the vessel wall in the microvasculature27 and macrovasculature,28 we speculate that MCs may, at least in some circumstances after being activated, participate in ET-1 regulation by decreasing its subendothelial concentration locally. Further studies are needed to clarify this issue.
MC granules also contain TNF-α,29 a known stimulator of ET-1 production.30 Accordingly, the induction of ET-1 mRNA expression could be due to TNF-α released by the activated MCs. By releasing ET-1, endothelial cells, again, could stimulate MCs, although it appears that ET-1 cannot activate all types of MCs.31,32⇓ Thus, there seems to be a close relationship between ECs and MCs regarding the local metabolism of ET-1. Inasmuch as ET-1 can activate MCs and activated MCs can increase the cellular ET-1 mRNA level, a positive-feedback loop exists. This loop will then be counteracted by a negative-feedback system consisting of degradation and inactivation of ET-1 by activated MCs. When MCs are stimulated excessively by compounds other than ET-1, such as allergens or anaphylatoxins, at least relative depletion of ET-1 could ensue. This depletion of ET-1 could be an additional factor favoring vasodilation in conditions such as anaphylactic shock.
ET-1 induces proliferation in smooth muscle cells and counteracts the vasodilatory and antiproliferative actions of NO and prostacyclin,6 thus potentially contributing to atherogenesis. Indeed, observations of ET-1 gene expression and ET-1 peptide in atherosclerotic plaques,9,10⇓ as well as increased plasma levels of ET-1 in atherosclerotic subjects,11 suggest a role for ET-1 in the atherosclerotic process. One trigger of ET-1 production may be oxidized LDL, which increases ET-1 synthesis by ECs in vitro.12 ET-1 could then be involved in MC activation in atherosclerotic plaques. Because MC activation has been shown to result in leukocyte rolling, adhesion, and extravasation3 and because ET-1 can induce leukocyte rolling and adhesion, the increased ET-1 expression in atherosclerotic lesions could contribute to the inflammatory reaction not only directly but also by activating the MCs in the lesions. However, degradation of ET-1 by the activated MCs would tend to limit the inflammatory reaction and bring the system into balance again.
In summary, we have shown that activated MCs degrade secreted endothelial ET-1 and that at the same time they stimulate ET-1 mRNA expression, suggesting the presence of a biological feedback system between ECs and MCs. Therefore, we propose that activated MCs participate in the metabolism of abluminally released ET-1 at sites where MCs are located in the subendothelial space. The potential clinical implications of this novel EC-MC relationship include derangements in endothelial function in circulatory failure and in atherogenesis.
Financial support was provided by the Sigrid Juselius, Magnus Ehrnrooth, and Finnish Allergy Foundations, the Finnish Foundation for Cardiovascular Research, and the Finska Läkaresällskapet. We thank Päivi Ihamuotila, BSc, and Ahmet Pekiner, BSc, for expert technical assistance.
Received September 18, 2001; revision accepted October 10, 2001.
Schwartz LB. The mast cell.In: Kelley WN, Harris ED Jr, Ruddy S, Sledge CB, eds. Textbook of Rheumatology. 4th ed. Philadelphia, Pa/London, UK/Toronto, Canada/Montreal, Canada/Sydney, Australia/Tokyo, Japan: WB Saunders Co; 1993: 304–318.
Kaartinen M, Penttilä A, Kovanen PT. Accumulation of mast cells in the shoulder region of human coronary atheroma, the predilection site of atheromatous rupture. Circulation. 1994; 90: 1669–1678.
Kovanen PT, Kaartinen M, Paavonen T. Infiltrates of activated mast cells at the site of coronary atheromatous erosion or rupture in myocardial infarction. Circulation. 1995; 92: 1084–1088.
Zeiher AM, Goebel H, Schächinger V, Ihling C. Tissue endothelin-1 immunoreactivity in the active coronary atherosclerotic plaque: a clue to the mechanism of increased vasoreactivity of the culprit lesion in unstable angina. Circulation. 1995; 91: 941–947.
Boulanger CM, Tanner FC, Bea ML, Hahn AWA, Werner A, Lüscher T. Oxidized low density lipoproteins induce mRNA expression and release of endothelin from human and porcine endothelium. Circ Res. 1992; 70: 1191–1197.
Caligiuri G, Levy B, Pernow J, Thorén P, Hansson GK. Myocardial infarction mediated by endothelin receptor signaling in hypercholesterolemic mice. Proc Natl Acad Sci U S A. 1999; 96: 6920–6924.
Kokkonen JO, Kovanen PT. Accumulation of low density lipoproteins in stimulated rat serosal mast cells during recovery from degranulation. J Lipid Res. 1989; 30: 1341–1348.
Lindstedt KA, Kokkonen JO, Kovanen PT. Soluble heparin proteoglycans released from stimulated mast cells induce uptake of low density lipoproteins by macrophages via scavenger receptor-mediated phagocytosis. J Lipid Res. 1992; 33: 65–75.
Wang Y, Kovanen PT. Heparin proteoglycans released from rat serosal mast cells inhibit proliferation of rat aortic smooth muscle cells in culture. Circ Res. 1999; 84: 74–83.
Itoh Y, Yanagisawa M, Ohkubo S, Kimura C, Kosaka T, Inoue A, Ishida N, Mitsui Y, Onda H, Fujino M. Cloning and sequence analysis of cDNA encoding the precursor of a human endothelium-derived vasoconstrictor peptide, endothelin: identity of human and porcine endothelin. FEBS Lett. 1988; 231: 440–444.
Eklund KK, Stevens RL. Mast cell proteases.In: Avile F, ed. Innovations in Proteases and Their Inhibitors. Berlin, Germany: Walter de Gruyter & Co; 1993: 241–257.
Nakano A, Kishi F, Minami K, Wakabayashi H, Nakaya Y, Kido H. Selective conversion of big endothelins to tracheal smooth muscle-constricting 31-amino acid-length endothelins by chymase from human mast cells. J Immunol. 1997; 159: 1987–1992.
Wagner OF, Christ G, Wojta J, Vierhapper H, Parzer S, Nowotny PJ, Schneider B, Waldhausl W, Binder BR. Polar secretion of endothelin-1 by cultured endothelial cells. J Biol Chem. 1992; 267: 16066–16068.
Sirviö M-L, Metsärinne K, Saijonmaa O, Fyhrquist F. Tissue distribution and half-life of 125I-ET-1 in the rat. Biochem Biophys Res Commun. 1990; 67: 1191–1195.
Kaartinen M, Penttilä A, Kovanen PT. Mast cells of two types differing in neutral protease composition in the human aortic intima: demonstration of tryptase- and tryptase/chymase-containing mast cells in normal intimas, fatty streaks and the shoulder region of atheromas. Arterioscler Thromb. 1994; 14: 966–972.
Kaartinen M, Penttilä A, Kovanen PT. Mast cells in rupture-prone areas of human coronary atheromas produce and store TNF-α. Circulation. 1996; 94: 2787–9227.