Phospholipase A2 (PLA2), a regulatory enzyme found in most mammalian cells, catalyzes the breakdown of membrane phospholipids to arachidonic acid. There are two major cytosolic types of the enzyme, calcium-dependent (cPLA2) and calcium-independent (iPLA2) PLA2. The present study investigated whether or not iPLA2 plays a role in the endothelium-dependent contractions of the aorta of the spontaneously hypertensive rat and its normotensive counterpart, the Wistar-Kyoto rat. The presence of iPLA2 in the endothelial cells was identified by using immunochemistry and immunoblotting. Aortic rings with and without the endothelium were suspended in organ chambers for isometric tension recording. The production of prostanoids was measured by using enzyme immunoassay kits. iPLA2 was densely distributed in endothelial cells of the aorta of both strains. At 3 × 10−6 M, the selective iPLA2 inhibitor, bromoenol lactone (BEL), abrogated endothelium-dependent contractions induced by acetylcholine but not those evoked by the calcium ionophore A-23187. The effects of BEL were similar in the aortae of Wistar-Kyoto and spontaneously hypertensive rats. The nonselective PLA2 inhibitor quinacrine abolished the contractions triggered by both acetylcholine and A-23187, whereas the store-operated calcium channel inhibitor SKF-96365 prevented only the acetylcholine-induced contraction. The acetylcholine- but not the A-23187-induced release of 6-keto prostaglandin F1α was inhibited by BEL. The release of thromboxane B2 by either acetylcholine or A-23187 was not affected by BEL. In conclusion, iPLA2 plays a substantial role in the generation of endothelium-derived contracting factor evoked by acetylcholine.
- endothelium-derived contracting factor
- thromboxane A2
the endothelium modulates the tone of the underlying vascular smooth muscle by releasing various vasoactive mediators. Some, including nitric oxide, cause relaxation and are referred to as endothelium-derived relaxing factors (10, 20). The endothelium can also release endothelium-derived contracting factors (EDCFs) (8, 10, 17, 42). In the rat aorta, endothelium-dependent contractions to acetylcholine are caused by the production of metabolites of arachidonic acid, mainly prostacyclin, which then activate thromboxane-prostanoid receptors of the vascular smooth muscle cells (11, 17, 18, 35, 41). The endothelium is dysfunctional in blood vessels of humans with various cardiovascular diseases (6, 28, 36, 37) and in animal models including the adult spontaneously hypertensive rat (SHR) (17). This dysfunction favors the production of EDCFs. During the production of EDCFs, membrane phospholipids are converted by cytosolic phospholipase A2 (PLA2) to arachidonic acid. The presumably membrane-bound cyclooxygenase-1 (COX-1) then catalyzes arachidonic acid into endoperoxides, the precursor of EDCF (32, 34). There are two major forms of PLA2, calcium-dependent (cPLA2) and calcium-independent (iPLA2) PLA2 (40). Earlier studies suggested that iPLA2 is involved in acetylcholine-induced, endothelium-dependent relaxations of the rat aorta (5, 23). The present study was designed to determine the role, if any, of iPLA2 in endothelium-dependent contractions.
Animals and tissue preparation.
This investigation was approved by the Committee on the Use of Laboratory Animals for Teaching and Research of the University of Hong Kong. Male adult Wistar-Kyoto (WKY) and SHRs (37 to 39 wk old) were housed in a room with standardized temperature (21 ± 1°C) and exposed to a 12-h:12-h light-dark cycle. They had free access to a standardized diet (LabDiet 5053, St. Louis, MO) and tap water. The rats were anesthetized with pentobarbital sodium (30 mg·ml−1·kg−1 ip). The mean arterial blood pressure was measured by means of a polyethylene cannula inserted into the left carotid artery and connected to a pressure transducer (Statham P 23 ID, Gould, Oxnard, CA). The mean blood pressures of SHRs and WKY rats were 192 ± 1.4 (n = 40) and 130 ± 1.9 mmHg (n = 40), respectively. After the blood pressure measurement, the rats were euthanized with pentobarbital sodium (70 mg·ml−1·kg−1). Their thoracic aortae were isolated and placed immediately in cold Krebs-Ringer buffer with the following composition: (in mM) 118 NaCl, 4.7 KCl, 2.5 CaCl2, 1.2 MgSO4, 1.2 KH2PO4, 25 NaHCO3, and 11.1 glucose (control solution). The adhering fat and connective tissues were removed. The aortae were cut into rings (length, ∼3 mm). In some rings, the endothelium was removed mechanically by gently rubbing the intimal surface of the rings with a syringe needle. Some of the rings were used to measure isometric force, whereas the others were used for measuring the release of 6-keto prostaglandin F1α and thromboxane B2 (the stable metabolite of prostacyclin and thromboxane A2, respectively), for immunohistochemistry or for immunoblotting.
Immunohistochemical and immunofluorescent labeling.
SHR and WKY aortic rings were used to make slices of 5 μm thick, which were then dehydrated with incremental concentrations of alcohol, followed by an exposure to xylene and paraffin. Antigen retrieval was carried out by incubating in citrate buffer (Dako, Glostrup, Denmark) for 20 min in a microwave oven (300 W). The sections were blocked with 5% dry milk in 0.05% Tween-Tris-buffered saline (TTBS) and then incubated with either milk buffer (negative control) or diluted primary rabbit anti-iPLA2 antibodies (1:200; LifeSpan Biosciences) overnight at 4°C. The sections were next incubated with either horseradish peroxidase-conjugated anti-rabbit antibody (1:1,000, Dako) or AlexaFluor488 goat anti-rabbit (1:200, Invitrogen) in TTBS for 45 min at room temperature. For immunohistochemical labeling, the signals were visualized by incubation with 0.01% diaminobenzidine (Dako) and 0.02% H2O2 and the sections were further counterstained with Mayers hematoxylin for 30 s and then inspected with an epifluorescence microscope (Olympus BX51, Ballerup, Denmark) using a 20× objective. For immunofluorescent labeling, the sections were incubated with 4,6-diamino-2-phenylindole (DAPI) to label nuclear DNA for 3 min and images were taken by a confocal laser-scanning fluorescence microscopy (Olympus FV1000, Hamburg, Germany). It was performed using a 20× (numerical aperture, 0.95) Olympus water immersion objective. Images were acquired at 1,024 × 1,024 pixels with and without internal zoom. DAPI and AlexaFluor488 were sequentially excited using 405- and 488-nm lasers, respectively, with fluorescence monitored through 425–475- and 500–530-nm band-pass filters (acousto-optical tunable filter), respectively. The intensity of the coloring was analyzed by using ImageJ (http://rsbweb.nih.gov/ij/). Rat kidney sections were used for positive control staining.
Protein extraction and immunoblotting.
Whole aortae were cut into small pieces and then homogenized in lysis buffer containing 20 mmol/l Tris·HCl, 1% Triton X-100, 150 mmol/l NaCl, 1 mmol/l EDTA, 1 mmol/l EGTA, 2.5 mmol/l sodium pyrophosphate, 1 mmol/l β-glycerophosphate, and 1 mmol/l sodium orthovanadate, supplemented with a cocktail of protease inhibitors containing 100 mmol/l phenylmethylsulfonyl fluoride, 10 μg/ml trypsin inhibitor, 1 mg/ml leupeptin, and 2 μg/ml pepstatin A. The mixture was centrifuged at 5,000 rpm at 4°C for 3 min, and the supernatant was kept at −80°C until use. For gel electrophoresis, 50 μg of tissue homogenate protein were used. The samples were mixed with 1× sample buffer (NuPAGE LDS sample buffer 4×, Invitrogen, Carlsbad, CA) and 1× reducing agent (10× reducing agent, Invitrogen) and diluted with ultrapure water to obtain 40 μl. The samples were boiled for 10 min at 95°C and subsequently separated by SDS-PAGE (10%) at 200 V, 500 mA for 1 h. The proteins were transferred electrophoretically onto nitrocellulose membranes. The blotting was performed at 1,000 V, 300 mA for 2 h. Subsequently, the membranes were blocked in TBS with 5% dry milk at room temperature for 2 h, washed in TTBS, and then incubated with primary anti-iPLA2 antibody (1:200) overnight at 4°C. The membranes were then incubated with horseradish peroxidase-conjugated anti-rabbit antibody (1:4,000 in milk, room temperature, for 2 h, Amersham Biosciences, Piscataway, NJ). Bound secondary antibody was detected by chemiluminescence (Amersham Biosciences) and exposed to X-ray film. To reprobe β-actin, the membranes were washed with TTBS and incubated with the monoclonal β-actin antibody (Sigma, St. Louis, MO). The optical densities of the protein bands were determined with the computerized program MultiAnalysis (Bio-Rad). Densitometric analysis was normalized to the immunoreactive β-actin band.
Isometric force measurement.
Rings were suspended in organ chambers that contained 5 ml of control solution (37°C) aerated with 95% O2-5% CO2l-arginine methyl ester (l-NAME; 10−4 M) for 30 min to optimize endothelium-dependent contractions (2, 30). Some rings were exposed to different concentrations of the selective, mechanism-based iPLA2 inhibitor bromoenol lactone [(BEL); 10−6 and 3 × 10−6 M, (1, 4, 9)], nonselective PLA2 inhibitor quinacrine (10−5 M), or with the store-operated calcium (SOC) channel inhibitor SKF-96365 (19) and then to cumulative concentrations of acetylcholine (10−8 to 10−5 M) or of the calcium ionophore A-23187 (10−8 to 10−6 M). Rings without endothelium were exposed to increasing concentrations of 15(S)-hydroxy-11, 9-(epoxymethano)prostadienoic acid (U-46619, a thromboxane mimetic, 10−10 to 10−6 M) to obtain endothelium-independent contractions.Nω-nitro-
Release of 6-keto prostaglandin F1α and thromboxane B2.
To measure the release of 6-keto prostaglandin F1α and thromboxane B2, the aortic rings of WKY and SHRs were placed in small chambers containing 1 ml of control solution at 37°C aerated with 95% O2-5% CO2. They were allowed to equilibrate for 90 min. Drugs, including BEL (10−6 or 3 × 10−6 M) and l-NAME (10−4 M), were given 30 min before exposing the rings to a single concentration of either acetylcholine (10−5 M, for 10 min) or A-23187 (10−5 M, for 30 min) to mimic the conditions of the organ chamber experiments. The rings were then removed and the solution was diluted 500 or 50 times before measuring the concentration of 6-keto prostaglandin F1α or thromboxane B2, respectively, using EIA kits purchased from Cayman Chemical (Ann Arbor, MI). The assays were performed according to the manufacturer's instructions. The release of 6-keto prostaglandin F1α and thromboxane B2 is expressed in picograms per millimeter ring per milliliter (12).
All contractions are expressed in percentages of the initial KCl reference contraction. Results are presented as means ± SE with n referring to the number of rats used. Statistical analysis was performed using Student's t-test for comparison of two groups or two-way ANOVA followed by the Bonferroni post hoc test for unpaired observations. All statistical comparisons were performed using Prism version 3a (GraphPad Software, San Diego, CA). Differences were considered to be statistically significant when P was <0.05.
Acetylcholine, A-23187, BEL, l-NAME, quinacrine, and SKF-96365 were purchased from Sigma Chemical. U-46619 was purchased from Biomol (St. Louis, MO). DMSO was purchased from Merck (Darmstadt, Germany). BEL, U-46619, and the calcium ionophore A-23187 were prepared daily as a stock solution in absolute DMSO, which was diluted further with control solution (0.1% of DMSO in the organ chamber). Concentrations are expressed as final molar concentrations.
Immunohistochemical and immunofluorescent labeling.
The iPLA2 enzyme, seen as a brown coloring in immunohistochemical labeling and green coloring in immunofluorescent labeling, was present densely in endothelial cells. Weak signals can be found in vascular smooth muscle cells. The intensity of the enzyme was comparable in aortae of SHRs [13.12 ± 0.98 arbitrary units (AU)] and WKY (12.26 ± 0.16 AU) rats with no significant difference (Fig. 1, A–F).
The iPLA2 protein was expressed in the aorta of both WKY and SHRs (Fig. 1G). The selective iPLA2 antibody detected one discrete protein band of around 80 kDa. The expression of iPLA2 was comparable in the WKY and SHR aorta and was significantly reduced after removal of the endothelium (Fig. 1G).
Isometric force measurements.
In rings without endothelium, 10−5 M but not 3 × 10−6 M BEL significantly inhibited the U-46619-induced contractions (Fig. 2). Further experiments were conducted with 3 × 10−6 M of the iPLA2 inhibitor.
In aortic rings of SHRs and WKY rats with endothelium, acetylcholine and A-23187 caused concentration-dependent, endothelium-dependent contractions. The response was significantly greater in preparations from SHRs (acetylcholine, EC50 = 2.7 × 10−7 M; and A-23187, EC50 = 2.1 × 10−7 M) than in those from WKY rats (acetylcholine, EC50 = 5.6 × 10−5 M; and A-23187, EC50 = 2.0 × 10−6 M). The acetylcholine- but not the A-23187-induced contractions were inhibited by 3 × 10−6 M BEL. No inhibitory effect of BEL was observed at 10−6 M. The nonselective PLA2 inhibitor quinacrine (10−5 M) abolished the endothelium-dependent contractions induced by both acetylcholine and A-23187.
SKF-96365 (10−5 M) abolished the endothelium-dependent contractions to acetylcholine but had no significant effect on the response to A-23187 in the SHR aorta (Fig. 3).
Release of prostanoids.
In the aortae of both strains, acetylcholine (10−5 M) and A-23187 (10−5 M) evoked the release of 6-keto prostaglandin F1α (the major metabolite of prostacyclin). In the presence of BEL (3 × 10−6 M), the acetylcholine- but not the A-23187-induced production of 6-keto prostaglandin F1α was reduced significantly. Lowering the concentration of BEL to 1 × 10−6 M had no significant effect on either the acetylcholine- or the A-23187-induced production of 6-keto prostaglandin F1α. The release of thromboxane B2 was minimal with acetylcholine but was significantly larger when induced by A-23187. No significant differences in the release of thromboxane B2 were observed between the controls and the rings incubated with BEL. The release of 6-keto prostaglandin F1α but not that of thromboxane B2 induced by acetylcholine was significantly greater in SHR aortae. However, in rings stimulated by A-23187, more thromboxane B2 was released in preparations from SHRs (Fig. 4).
The rat aorta was used in the present study, since it has been the standard preparation to measure EDCF-mediated responses (2, 13, 15, 17, 30) and allows the measure of the release of vasoconstrictor prostanoids (12, 13, 31). The available evidence on smaller arteries confirms the existence of cyclooxygenase-derived endothelium-dependent contracting factor in blood vessels relevant for the control of peripheral resistance and its prevalence in the hypertensive strain (16, 24, 25). Studies in the human forearm demonstrate that indomethacin improves the reduced endothelium-dependent vasodilatation to acetylcholine in aged and essential hypertensive patients, indicating that the production of EDCF contributes to the endothelial dysfunction with aging and hypertension (27, 29, 36). The levels of messenger RNA and protein of COX-1 are comparable in the aorta of 5- and 10-wk-old WKY and SHRs (21), but these levels are significantly higher in preparations of 36- (11) or 40-wk-old (21) SHRs than in those of age-matched WKY rats. Since the overexpression of COX-1 is the major factor that contributes to the overproduction of EDCF in the SHR, the present experiments were performed on the aortae of 37- to 39-wk-old rats.
Results from immunohistochemistry and immunofluorescent labeling suggest that iPLA2 is present in the endothelial cells, but weak signals were also detected in the smooth muscle cells. To obtain a more quantitative analysis, Western blot analysis was performed. The results showed that there was a significant decrease in expression level of the enzyme in aortae without endothelium, confirming that iPLA2 is present in the endothelium. Since the endothelium consists of a monolayer of cells, the substantial decrease in protein expression upon the removal of the endothelium confirms that iPLA2 is highly distributed in endothelial cells compared with the vascular smooth muscle. However, there is still iPLA2 present in preparations in which the endothelium has been removed. This could be due to either an incomplete removal of the endothelial cells or to the presence of the enzyme in the vascular smooth muscle. Actually, the involvement of iPLA2 in nicotine-induced contractions has been demonstrated in the rat basilar artery (14).
Results from immunohistochemistry and immunofluorescent labeling suggest that iPLA2 is present in the endothelial cells, and weak signals were also detected in the smooth muscle cells. To have a more quantified analysis, immunoblotting was performed and the results showed that there was a significant decrease in expression level of the enzyme in aortae without the endothelium, which confirmed that iPLA2 is present in the endothelium. Since the endothelium is a just a single layer of cells, the substantial decrease in protein expression upon removal of the endothelium suggests that iPLA2 is highly distributed in endothelial cells compared with the smooth muscle cells. However, there is still iPLA2 present in preparations in which the endothelium has been removed. This could be due to an incomplete removal of the endothelial cells or to the presence of the enzyme in the vascular smooth muscle.
In the present study, BEL was used as an irreversible, selective inhibitor of iPLA2. BEL has a 1,000-fold selectivity for inhibition of iPLA2 versus the cPLA2 (1, 4, 9). BEL, at 10−5 M, inhibited the contractions induced by U-46619 in rings without endothelium, demonstrating the inhibitory effect of this concentration is not related to endothelial function. The endothelium-dependent contractions to acetylcholine but not those to A-23187 were inhibited if the concentration of BEL was lowered to 3 × 10−6 M, a concentration that did not affect the response of the vascular smooth muscle to the activation of thromboxane-prostanoid receptors [the main mediators of the response to EDCF (3, 11, 17)]. Therefore, to study endothelium-related factors, we choose 3 × 10−6 M as the concentration for further experiments. The fact that BEL significantly reduced acetylcholine-induced contractions but had no effects on U-46619-induced endothelium-independent contractions indicates that iPLA2 is involved in the response to acetylcholine and does so by affecting the endothelial cells only. In confirmation of previous studies (15, 17), the contractions to acetylcholine were significantly greater in SHR than in WKY rats, which reflects the higher COX-1 and proscyclin synthase expression in the aorta of the SHRs (11, 21, 31). The effects of BEL were similar in the aortae of SHRs and its normotensive control WKY rats, demonstrating that the contribution of iPLA2 does not depend on the level of arterial pressure to which the vascular wall is exposed chronically, as also reflected by the comparable distribution of iPLA2 in both SHR and WKY aortae demonstrated by immunohistochemistry and immunoblotting.
The conclusion that the iPLA2 inhibitor acts only on the endothelial cells is supported by the measure of the release of 6-keto prostaglandin F1α, the metabolite of prostacyclin that is a major EDCF released by acetylcholine in the aorta of SHRs, in confirmation of earlier studies (12, 13, 15). In the present experiments, acetylcholine did not induce a significant increase in the release of thromboxane B2 over the basal levels, suggesting that thromboxane A2 plays a minor role in the acetylcholine-induced endothelium-dependent response compared with that to A-23187, again in confirmation of previous studies (12, 13, 15). The present observations provide no explanation for this discrepancy. The lack of effect of 3 × 10−6 M BEL on A-23187-induced contractions suggests that iPLA2 is not involved significantly in the endothelium-dependent contractions to the calcium ionophore. Since A-23187 uses a nonreceptor-mediated mechanism to bring calcium into the cell directly, the lack of the effect of BEL suggests that iPLA2 takes part in the upstream of the increase in the endothelial calcium concentration required to evoke the phenomenon (33). The present findings confirm that endothelium-dependent contractions to acetylcholine are prevented by quinacrine (17) and demonstrate that those induced by A-23187 are also abolished by the nonselective PLA2 inhibitor quinacrine. Endothelium-dependent contractions to both acetylcholine and the calcium ionophore are accompanied by increases in endothelial calcium concentration (33). Thus the present observations imply that cPLA2 is involved in the generation of EDCF-mediated contractions downstream of the increase in the cytosolic concentration of the activator ion and equally so irrespective of the initial trigger (acetylcholine or A-23187).
Endothelium-dependent contractions of the SHR aorta evoked by acetylcholine are due to the activation of muscarinic receptors (7). The activation of G protein-coupled muscarinic receptors leads to the displacement of inhibitory calmodulin from iPLA2 (22, 38, 39). The then-activated iPLA2 in turn produces lysophospholipids. The latter will activate the SOC channel, leading to an influx of extracellular calcium (26). In the present study, the SOC channel inhibitor SKF-96565 prevented the acetylcholine-induced, endothelium-dependent contraction but not those to A-23187. These observations are explained best if iPLA2 was involved in the endothelium-dependent contraction to acetylcholine by producing lysophospholipids, which in turn open SOC channels, permitting the influx of extracellular calcium and the subsequent activation of cPLA2 generating the arachidonic acid necessary for the production of EDCFs by cyclooxygenase. By contrast, the calcium ionophore A-23187 bypasses the cell membrane receptors and causes a direct increase in endothelial calcium that activates cPLA2 directly without involving iPLA2. This interpretation would explain why BEL, which inhibits iPLA2 only, did not affect A-23187-induced endothelium-dependent contraction, whereas quinacrine, which inhibits both forms of PLA2, abolishes the response to both agonists.
In conclusion, the present study demonstrates that iPLA2 is a key initial player in the process of generating EDCFs in the arterial wall in response to acetylcholine.
The work described in this study was supported partly by the Hong Kong Research Grant Council (University of Hong Kong-777507M), the Research Centre of Heart, Brain, Hormone and Healthy Aging, and the Center for D-receptor Activation Research (Boston, MA).
No conflicts of interest are declared by the author(s).
We thank Prof. Urs Ruegg, Boye Jensen, and Victoria Bolotina for most helpful suggestions.
- Copyright © 2010 the American Physiological Society