Effects of Bradykinin on Prostaglandin I2 Synthesis in Human Vascular Endothelial Cells
Abstract—The effects of bradykinin on the regulatory mechanisms of prostacyclin synthesis in endothelial cells were investigated in association with intracellular Ca2+ kinetics, cytosolic phospholipase A2 (cPLA2) activity, and mRNA expression of cPLA2 and prostaglandin H synthase (PGHS) isoforms. Bradykinin enhanced prostacyclin release from endothelial cells time-dependently, but pretreatment with EGTA H-7 or HOE 140 inhibited bradykinin-induced prostacyclin release. Bradykinin increased both the influx of extracellular Ca2+ and Ca2+ release from the intracellular Ca2+ storage sites. These reactions occurred within 5 minutes after bradykinin stimulation. Within 15 minutes, bradykinin activated cPLA2 to 1.3-fold the control level. The constitutive expressions of mRNA of cPLA2, PGHS-1, and PGHS-2 was 87, 562, and 47 amol/μg RNA, respectively. With the stimulation of bradykinin, cPLA2 mRNA increased to 746 amol/μg RNA in 15 minutes, PGHS-1 mRNA increased to 10 608 amol/μg RNA, and PGHS-2 mRNA increased to 22 400 amol/μg RNA in 180 minutes. Pretreatment with cycloheximide superinduced cPLA2 and PGHS-2 mRNA expression but almost completely inhibited PGHS-1. Pretreatment with EGTA had effects similar to pretreatment with cycloheximide in the case of cPLA2 and PGHS-1 but did not affect PGHS-2. These findings suggest that the elevation of cPLA2 activity caused by the increase of intracellular Ca2+ concentration is important in the early phase of bradykinin-induced prostacyclin synthesis and that the mechanisms regulating cPLA2 are different from those regulating PGHS isoforms in endothelial cells.
- kinetics, calcium
- phospholipases A
- prostaglandin H synthase
- polymerase chain reaction
Bradykinin (BK) has blood pressure–lowering effects that can be induced by either the direct action of BK on target organs or the secretion of secondary agents. BK in plasma mainly acts on endothelial cells and stimulates them to release various agents associated with vascular tonus and blood coagulation, such as prostaglandins, endothelium-derived relaxation factor, catecholamines, and cytokines, resulting in the maintenance of circulation homeostasis. BK is resolved into inactive peptides by 2 kininases, carboxypeptidase N (kininase I) and dipeptide hydrase (kininase II). Kininase II is the same enzyme as angiotensin I–converting enzyme (ACE) and is secreted by endothelial cells. The clinical observation that ACE inhibitors lower blood pressure, even under low renin conditions,1 suggests a renin-angiotensin system–independent mechanism for evoking their actions. From reports that the blood pressure–lowering effect of ACE inhibitors correlates with the increase of intravenous concentration of kinin and prostacyclin (prostaglandin I2, PGI2), it appears that this mechanism is due to the local accumulation of BK and BK-induced vasodilators, especially PGI2, rather than to the reduction of vasoconstrictor angiotensin II;2 however, the effects of ACE inhibition on BK and prostaglandins are unclear. PGI2 is synthesized in response to release of arachidonic acid (AA) from phospholipids followed by transformation of AA to PGH2 by prostaglandin H synthase (PGHS=cyclooxygenase; COX) and subsequent transformation of PGH2 by PGI2 synthase. Recent studies have revealed that PLA2 and PGHS have several isoforms and that many agents affect their expression at the level of genome transcription or protein translation. However, there have been few reports about the relation between BK-induced PGI2 synthesis and signal transduction mechanisms in endothelial cells. In the present study, we investigated the effects of BK on the regulatory mechanisms of PGI2 generation in human umbilical vein endothelial cells (HUVEC) with reference to intracellular Ca2+ kinetics, cytosolic PLA2 activity, and mRNA expression of cPLA2 and PGHS isozymes.
Culture of HUVEC
Endothelial cells were obtained from human umbilical cord vein and cultured according to our previously described method.3 We used primary cultured HUVEC that formed confluent monolayers in the following experiments. Cells were identified as vascular endothelial cells by electron microscopic findings of Weibel-Palade bodies in their cytoplasm and detection of ACE activity.
Assay of PGI2 and Inositol 1,4,5-Trisphosphate Concentration
The concentration of PGI2 and inositol 1,4,5-trisphosphate (IP3) were measured with the use of previously described methods.3 Briefly, the amount of PGI2 released from HUVEC was measured as the stable metabolite 6-keto prostaglandin F1α with the use of a [3H] 6-keto prostaglandin F1α radioimmunoassay kit (New England Nuclear), and the IP3 in HUVEC was measured with the use of the D-myo-inositol 1,4,5-trisphosphate [3H] Biotrack radioimmunoassay system (Amersham Pharmacia Biotech).
Measurement of cPLA2 Activity
After incubation in tissue culture medium (TCM) for 12 hours, the cells were scraped from the dish and collected into glass test tubes. The contents were centrifuged at 3000g for 5 minutes at 4°C. The cells were resuspended in TCM and centrifuged under the same conditions. The cells were finally suspended in TCM at a final concentration of 105 cells/mL and sonicated at 4°C for 3×10 seconds. The sonicated cells were centrifuged at 3000g for 10 minutes to remove unlysed cells. The resulting lysate was then centrifuged at 10 000g for 60 minutes. The supernatant (soluble fraction) was transferred to another container and used as a source for the assay of cPLA2. The activity of cPLA2 was measured by the method of Kramer,4 with some modification. Each substrate was dried under nitrogen gas and suspended in distilled water by sonication at room temperature. The standard incubation system for the assay of cPLA2 contained, in 0.25 mL, 50 mmol/L HEPES (pH 7.4), 150 mmol/L NaCl, 1 mmol/L CaCl2, 2 mmol/L 2-mercaptoethanol, 2×10−6 mol/L 1-palmitoyl-2-[1-14C]arachidonyl-phosphatidylcholine, 10−6 mol/L dioleoyl glycerol and enzyme preparation. The incubation was for 15 to 180 minutes at 37°C in the presence of BK (final concentration 10−5 mol/L). The reaction was stopped by adding 1.25 mL of Dole’s reagent. After 0.5 mL of H2O and 0.75 mL of n-heptane were added, the contents were spun in a vortex mixer for 5 minutes and centrifuged at 3000g for 5 minutes. An aliquot (0.8 mL) of the upper phase was pipetted into another tube containing 1 mL of heptane and 100 mg of silicic acid powder. After the tubes were shaken in a vortex mixer for 5 minutes and centrifuged as above, the heptane phase was transferred into scintillation vials. The radioactivity was measured by liquid scintillation spectrometry (TRI-CARB 460C Automatic Liquid Scintillation System, Packard Instrument Co Inc) with 10 mL of Atomlight (New England Nuclear) as a scintillator agent.
Measurement of Intracellular Ca2+ Concentration
Intracellular Ca2+ concentration ([Ca2+]i) was measured by the method of Grynkiewicz et al,5 with some modification. The cells were scraped from the dish and collected in centrifugation tubes. After being suspended in 3 mL of buffer A containing 1.0% BSA, the cells were incubated in the presence of 3.2×10−6 mol/L fura-2/AM at 37°C for 45 minutes. After resuspension in 1.0% BSA-containing solution (cell count 106/mL), they were centrifuged at 250g for 10 minutes. The fluorescence intensity from the endothelial cell suspensions then was recorded in UV-compatible cuvettes with a Ca2+ analyzer (CAF-100: Japan Spectroscopic Co, Ltd) at excitation wave lengths of 340 nm and 380 nm and at an emission wavelength of 510 nm.
After the cells were washed 3 times with buffer A, they were incubated with buffer A containing 1.85×105 Bq/mL of 45CaCl2 (7.1484×108 Bq/mg CaCl2, New England Nuclear) for 10 minutes. After incubation, the cells were washed 3 times with buffer A, and 1.0 mL of 6% trichloroacetic acid then was added to the dish. The cells were then rubbed off the dish with a rubber spatula, and the radioactivity of intracellular 45Ca was measured by means of liquid scintillation spectrometry.
Measurement of Ca2+ Release From Ca2+ Storage Sites
First, the cells were scraped from the dish and permeabilized with saponin according to our previously described method.3 The saponized HUVEC were subsequently incubated for 15 minutes after the addition of 45Ca2+, phosphocreatine (PC), CPK, succinic acid disodium salt (Suc), and adenosine 5′-triphosphate magnesium salt (ATP). After the supernatant was removed by centrifugation, the cells were resuspended in 3×10−8 mol/L Ca2+-EGTA buffer solution containing PC, CPK, Suc, and ATP and then were apportioned into several tubes (2×105 cells/200 μL). After stimulation of the cells with the test agent for a certain period, 200 μL of cell suspension was filtered on a filter by means of a diaphragm vacuum pump and 1225 Sampling Manifold (Millipore Co) with 30 mL of a rinsing buffer solution. The radioactivity of HUVEC on the vacuum pump filter was measured as residual 45Ca in the cells by means of liquid scintillation spectrometry. The Ca2+ release from Ca2+ storage sites induced by agents was calculated at 10 seconds and compared with the Ca2+ release induced by Ca2+ ionophore A23187.
Quantitative Measurement of mRNA
Synthesis of Primers for Genomic DNA Preparation, Reverse Transcription Reaction, and Competitive Polymerase Chain Reaction
Primers used for genomic DNA (gDNA) amplification, reverse transcription, and for competitive polymerase chain reaction (PCR) were synthesized on the Gene Assembler Plus (Pharmacia LKB Biochrom Ltd). The sequences of 5′- and 3′- flanking oligonucleotide primers used for PCR amplification of gene and mRNA transcription were 5′-GGAAGAAGCAGTTGCCAGATGCCCAGCTCC-3′ (PGHS-1-Fw), 5′-TTGTCTCCATAAATGTGGCCGAGGTCTACC-3′ (PGHS-1-Rv), 5′-GTGTTCTATTATGAGATTCTGAACTCCCCA-3′ (PLA2-Fw), 5′-TAGAATGAGGCAGACAAAAGTTGGAAGGCC-3′ (PLA2-Rv), 5′-ATGTTCCACCCGCAGTACAG-3′ (PGHS-2-Fw), and 5′-GCCCTTCACGTTATTGCAGAT-3′ (PGHS-2-Rv).
RNA Isolation and Reverse Transcription
Total cellular RNA was isolated from HUVEC and was reverse transcribed to complementary DNA (cDNA) by Moloney murine leukemia virus reverse transcriptase according to our previously published method.6 The concentrations of total RNA were determined by spectrophotometer at a 260-nm wave length.
Amplification of Genomic PGHS-1 DNA and Its Purification for Internal Standard of Competitive PCR
Genomic DNA samples were extracted from heparinized blood and amplified by PCR following the method of Saiki et al7 in the presence of 25 pmol of each primer (PGHS-1-Fw and -Rv), 100 μmol of each dNTP, 20 μg genomic DNA, 1×buffer (5 mmol/L Tris-HCl [pH 8.3], 20 mmol/L MgCl2, 50 mmol/L KCl, 0.01% gelatin), and 5 U of AmpliTaq (Takara Co), for a total volume of 200 μL. With the use of the Gene ATAQ controller (Pharmacia LKB Biochrom Ltd), 40 PCR cycles were performed as follows: denaturation at 94°C for 1 minute, annealing at 62°C for 2 minutes, and extension at 72°C for 3 minutes. Five microliters of the reaction mixture containing the PCR product was subjected to electrophoresis on 2% agarose gel for confirmation of the product size. The remaining 195 μL was loaded onto a 10% polyacrylamide gel (acrylamide/bis-acrylamide: vol/vol, 29:1) and electrophoresed for 40 minutes. The segment of the gel containing the DNA band was cut out by visualizing the ethidium-bromide–stained gels in long-wavelength ultraviolet light. The gel slice then was transferred to a microfuge tube containing elution buffer (10 mmol/L Tris-HCl, pH 8.0, 1 mmol/L EDTA, 0.2% SDS, 0.3 mol/L NaCl). After the tube was incubated at 37°C for 12 hours, the supernatant was transferred to another microfuge tube. Two volumes of 100% ethanol were added to the tube at 4°C, and the tube was stored at −20°C for 30 minutes. The DNA was recovered as a pellet by centrifugation at 12 000g for 10 minutes at 4°C in a microfuge. Then the pellet was rinsed twice with 70% ethanol and redissolved in 50 μL of dH2O.
Synthesis of Mutant-cDNA of cPLA2 and PGHS-2 by Site-Directed Mutagenesis With PCR Used for Internal Standard of Competitive PCR
Complementary DNA of a mutant-cDNA of cPLA2 and PGHS-2 were prepared by PCR-based, site-directed mutagenesis following the method of Higuchi et al.8 The mutagenic primer of sequence 5′-GTGTTCTATTATGAGATTCTGAACTCCCCAGA-GAAAGCTTGCTCTC-3′ (PLA2-mFw), 5′-ATGTTCCACCCGCAGTACAGAAAG-TATCACAAGCTT-3′ (PGHS-2-mFw) was synthesized on the Gene Assembler Plus. Amplification of PLA2- or PGHS-2–cDNA with each mutagenic primer and Rv-primer by PCR showed a single change, C to T (italicized in PLA2-mFw) or G to A (italicized in PGHS-2-mFw), which created a unique Hind III site. The amplification and purification then were performed.
Competitive PCR for Quantification of PGHS-1 or -2 and cPLA2 mRNA With Amplified Genomic DNA or Mutant cDNA Fragments Used as Internal Standard
In a series of tubes, PCR amplification was performed on reaction mixtures (10 μL) containing 1 μL of cDNA of PGHS-1, PGHS-2, or cPLA2 and an increasing concentration of amplified genomic DNA of PGHS-1, mutant-cDNA of PGHS-2 (0.01, 0.1, 1.0, 10, 100, 1000 amol), or mutant-cDNA of cPLA2 (0.25, 1.0, 16, 256, 4096 amol), 250×10−6 mol/L of each dNTP plus 3.7×105 Bq of [α-32P] dCTP (1.11×1014 Bq/mmol: New England Nuclear), 0.5×10-6 mol/L of each paired primer (Fw and Rv), 0.25 U of Taq DNA polymerase (AmpliTaq), in PCR buffer (100 mmol/L Tris-HCl [pH 8.3], 500 mmol/L KCl, 15 mmol/L MgCl2, 0.01% [wt/vol] gelatin). The PCR cycle was repeated 40 times following the protocol described above. Before loading the PCR products on gels for electrophoresis, cPLA2 or PGHS-2 PCR products were digested by 10 U of Hind III for 60 minutes at 37°C. The PCR products of mutant-cDNA of cPLA2 and PGHS-2 were divided into 2 fragments by Hind III digestion, but the PCR products of cDNA of native cPLA2 and PGHS-2 did not have the Hind III–digested site. In this manner, competitive PCR products of native and mutant-cDNA of cPLA2 and PGHS-2 were differentiated by digesting them with Hind III before electrophoresis. The samples then were loaded onto an 8% polyacrylamide gel (acrylamide/bis-acrylamide: vol/vol, 29:1) and analyzed by electrophoresis. The gel was stained with ethidium bromide (0.4 μg/mL) and photographed. The bands corresponding to the amplified products of the internal standard (gDNA or digested mutant c-DNA) and mRNA were cut out, and their radioactivity was counted. The data were analyzed by plotting the logarithm of the gDNA cpm/cDNA cpm ratio as a function of the logarithm of the gRNA concentration.
Data are expressed as mean±SEM. ANOVA was used to make comparisons. For F ratios significant at the level of ≤5%, Duncan’s multiple range test was applied to determine differences between any 2 groups. Differences of ≤5% (P<0.05) were considered statistically significant. Percent changes, for which a normal distribution cannot be assumed, were compared by means of the Kruskal-Wallis nonparametric method for ANOVA. Whenever the χ2 test results were significant, the Mann-Whitney test was used to determine the significance of the differences between pairs of means.
The addition of BK increased PGI2 release dose-dependently at concentrations from 10−10 mol/L through 10−4 mol/L, and the release reached a plateau at the concentration of 10−5 mol/L, so we used this concentration (10−5 mol/L) of BK in the following experiments. The stimulation of BK remarkably increased PGI2 generation within 15 minutes, and the effect was sustained through 180 minutes. This increase was not inhibited by pretreatment with the B1 receptor antagonist des-Arg9-BK(10−4 mol/L) or the PGHS-2 inhibitor NS-398 (10−6 mol/L) but was inhibited by pretreatment with the B2 receptor antagonist HOE 140 (10−4 mol/L) and was reduced to the constitutive level by pretreatment with the extracellular Ca2+ chelator EGTA (10−3 mol/L) or protein kinase C (PKC) inhibitor H-7 (10−5 mol/L). Pretreatment with a phospholipase C (PLC) inhibitor, U-73122 (10−5 mol/L), or a diacylglycerol (DAG) lipase inhibitor, RHC80267 (5×10−5 mol/L), reduced PGI2 release to the level under the constitutive generation and pretreatment with a PLA2 inhibitor, mepacrine (10−4 mol/L), or a PGHS inhibitor, aspirin (10−3 mol/L), reduced PGI2 release to almost zero in control as well as in BK stimulation experiments (Table⇓). The concentration of BK in incubation fluid measured by radioimmunoassay-dextran coated charcoal method at 15, 60, 180 minutes after BK addition was 1.12±0.06 (10−5 mol/L), 1.16±0.02 (10−5 mol/L), and 1.10±0.11 (10−5 mol/L), respectively. These data were not statistically different from each other.
Within 15 minutes of BK addition, the cytosolic PLA2 activity was increased to 132±10.3% of control. A further increase to 199.4±6.5% of control was found at 60 minutes, and this high level was maintained for 180 minutes.
The [Ca2+]i in HUVEC was 86.37±8.34 nmol/L in the steady state. BK caused a transitory increase in [Ca2+]i, with the peak at 30 seconds after addition and the return to basal level within 5 minutes (Figure 1A⇓). BK also increased 45Ca uptake to 145±5.3% that of control. Pretreatment with EGTA (10−3 mol/L) remarkably decreased the increase of both [Ca2+]i and 45Ca uptake induced by BK (Figure 1B⇓).
The generation of IP3 in HUVEC was increased by BK from 0.213±0.038 (pmol/2×105 cells) to 0.476±0.052 (pmol/2×105 cells). Pretreatment with EGTA (10−3 mol/L) inhibited the BK-induced IP3 generation (Figure 2A⇓). The addition of IP3 (10−5 mol/L) increased the 45Ca release from the intracellular calcium storage sites in saponized HUVEC. The increase of 45Ca release induced by IP3 was 35.7% less than that caused by the calcium ionophore A23187 (10−5 mol/L) (Figure 2B⇓).
The constitutive expression of cPLA2 mRNA was 86.8±6.8 amol/μg RNA in HUVEC. BK increased it to 746.5±43.3 amol/μg RNA at 15 minutes after stimulation, and the same level of mRNA expression was maintained for 180 minutes. The constitutive expression of PGHS-1 mRNA was 562.2±21.8 amol/μg RNA in HUVEC. Whereas BK had increased the expression only to 773.2±15.4 amol/μg RNA after 15 minutes of stimulation, it reached to 10 608±467.4 amol/μg RNA in 180 minutes. The constitutive expression of PGHS-2 mRNA was only 46.8±3.0 amol/μg RNA in HUVEC. BK did not affect its expression until 60 minutes after stimulation; however, the increase reached 22 400±2650 amol/μg RNA at 180 minutes (Figure 3⇓). Pretreatment with cycloheximide (5 μg/mL) for 60 minutes superinduced the mRNA expression of cPLA2 and PGHS-2 to 2183±193 and 6891±547 amol/μg RNA, respectively, but it suppressed the mRNA expression of PGHS-1 to 16.6±2.5 amol/μg RNA. Pretreatment with EGTA (10−3 mol/L) for 60 minutes superinduced the mRNA expression of cPLA2 to 2718±309 amol/μg RNA, but it suppressed that of PGHS-1 to 1.1±0.8. The mRNA expression of PGHS-2 was increased by EGTA pretreatment from 46.8±3.0 amol/μg RNA to 89.5±23.7 amol/μg RNA; however, the difference of PGHS-2 mRNA expression between control and EGTA pretreatment was not statistically significant (Figure 4⇓).
ACE inhibitors are the most widely used drugs in the treatment of hypertensive patients. Their pharmacological effects depend not only on the reduction of the vasoconstrictor angiotensin II but also the local accumulation of kinins and kinin-related PGI2 release, and the importance of the latter mechanisms has become apparent recently. Several kinds of BK-receptor subtypes have been identified and characterized according to their pharmacological and biochemical properties. Above all, the B2 receptor is widely distributed over human tissues, especially smooth muscle cells and neurons, and mediates most of the cellular actions of BK.9 In the present study, pretreatment with the B1-receptor antagonist des-Arg9-bradykinin did not inhibit the increase of BK-induced PGI2 generation in HUVEC, but pretreatment with B2-receptor antagonist HOE 140 inhibited the BK-induced increase in HUVEC. The increased effect of BK on PGI2 release was sustained for 3 hours after stimulation, but the increase rate of PGI2 release declined after 15 minutes of BK stimulation. Because the concentration of BK in incubation fluid did not change from the beginning of incubation to 180 minutes after BK addition, this suggests that the cause of the rate of decline of PGI2 release was not due to the reduction of BK concentration in incubation fluid but to other factors, for instance, downregulation of BK receptors induced by long-term exposure of BK. Pretreatment with the calcium chelator EGTA, PKC inhibitor H-7, PLC inhibitor U-73122, or DAG lipase inhibitor RHC80267 remarkably reduced BK-induced PGI2 release, but the effects of the latter 2 inhibitors were more suppressive. These findings suggest that PLC and DAG affect BK-induced PGI2 synthesis in HUVEC not only by PKC but also the PKC-independent pathway. From these results, we speculate that BK acts on HUVEC through B2 receptors and that extracellular Ca2+ and protein phosphorylation related to PKC, PLC, and DAG play important roles in the process of BK-induced PGI2 synthesis.
The release of AA from phospholipids in the cell membrane is the first step of PGI2 generation, and PLA2 acts mainly in this step. PLA2 has several isoforms. The genomic structures of type I and II of secretory PLA2 (sPLA2) and cytosolic PLA2 (cPLA2) have been determined. The release of AA from phospholipids of cell membranes is chiefly dependent on cPLA2 and partially on sPLA2.10 In the present study, BK elevated cPLA2 activity to 132% of control within 15 minutes after stimulation, and this level of activity was maintained for 180 minutes. It is well known that cPLA2 activity is affected by intracellular Ca2+ kinetics. BK increased [Ca2+]i transitorily to ≈240% of control. Pretreatment with the extracellular Ca2+ chelator EGTA reduced BK-induced Ca2+ influx and [Ca2+]i increase, but this BK-induced [Ca2+]i increase was not completely inhibited, and we could observe a small amount of increase in [Ca2+]i despite EGTA pretreatment. This increase in [Ca2+]i was thought to be mobilized from Ca2+ storage sites in cytosol. The intracellular Ca2+ storage sites consist of mitochondrial and nonmitochondrial sites, and the latter, especially for IP3-induced Ca2+ release, is a main source of [Ca2+]i. In our present study, BK significantly increased the concentration of cytosolic IP3 in HUVEC, and the addition of IP3 (10−5 mol/L) promoted the 45Ca release from the Ca2+ storage sites in saponized HUVEC. These results suggest that the BK-induced [Ca2+]i increase consists of not only Ca2+ influx but also IP3-induced Ca2+ release from the [Ca2+]i storage sites in HUVEC. This elevated [Ca2+]i in turn activates cPLA2. It was recently reported that cPLA2 contains the CaLB domain, a Ca2+-dependent translocation domain in its N-terminal,11 and that cPLA2 translocates from cytoplasm to cell membrane in the gradient of [Ca2+]i concentration.12 It was also reported that the phosphorylation by MAP kinase raised PLA2 activity to severalfold that of control.13 These mechanisms raised cPLA2 activity, thereby promoting the release of AA from membrane phospholipids and its subsequent conversion into PGI2 through the PGHS pathway.
The synthesis of PGI2 can also be controlled by altering the activity or expression of the enzyme. Because the increase effect of BK on PGI2 generation is sustained for 3 hours after stimulation, we believe that not only the raise of enzymatic activity but also the quantitative increase of cPLA2 or PGHS protein through mRNA transcription is involved in BK-induced PGI2 synthesis in HUVEC. PGHS is another rate-limiting enzyme in the biosynthesis of prostanoids. Recently, 2 isoforms of PGHS, PGHS-1 (COX-1) and PGHS-2 (COX-2), were detected, and their gene structures and proteins were examined. PGHS-1 is expressed constitutively in most tissues, but PGHS-2 is highly and rapidly induced in response to cell activation.14 In our present study, pretreatment with the PGHS-2 inhibitor NS-398 did not affect BK-induced PGI2 release until 3 hours after BK stimulation, and the mRNA of PGHS-1, quantified by competitive PCR, was more markedly expressed than that of cPLA2 or PGHS-2. This finding indicates that PGI2 generation of the constitutive phase and early phase of BK stimulation in HUVEC depends on PGHS-1, not PGHS-2. The importance of PGHS-1 in the constitutive generation of prostanoids has been reported in most tissues including endothelial cells,15 and our results support these findings.
The stimulation of BK provided interesting results. The mRNA expression of cPLA2 increased to ≈8-fold that of control at 15 minutes after BK stimulation and maintained this level for 180 minutes. In contrast, that of PGHS-1 was unchanged until 60 minutes after BK stimulation but then had increased remarkably by 180 minutes. Although many investigators have evaluated the effects of substances on PGHS-1 mRNA expression, an increase of PGHS-1 mRNA was reported in a few studies, and their ratios of increase in expression of PGHS-1 mRNA were only ≈2-fold to 4-fold that of control level.14 It was, therefore, assumed that the role of this enzyme was to produce prostaglandins that regulate normal cellular processes. Contrary to these former reports, in our study the level of PGHS-1 mRNA expression increased to ≈18-fold that of control at 180 minutes. Thus, it is possible that PGHS-1 acts as not only a constitutive enzyme but also an inducible enzyme in the regulation of PGI2 synthesis in BK-stimulated HUVEC. The expression of PGHS-2 mRNA was only 44 amol/μg RNA in an unstimulated HUVEC but had increased to ≈500-fold of control level at 180 minutes. The changes in PGHS-2 mRNA induced by BK were typical of an inducible enzyme. From our present findings, we speculate that there are 3 steps in the process of BK-induced PGI2 synthesis in HUVEC. The first step is an immediate activation of cPLA2 mediated by increase of intracellular Ca2+ and protein phosphorylation of the already-existing enzyme. The second is a new synthesis of cPLA2 protein by gene transcription. In these 2 steps, the main reaction is the supply of AA by cPLA2 and the synthesis of PGI2 by constitutively existing PGHS-1. The third step is the gene transcription and synthesis of new proteins of the PGHS isoforms. We observed no increase in PGHS-1 and PGHS-2 mRNA until 180 minutes. This time lag of BK-induced mRNA expression of cPLA2 and PGHS isoforms is notable. It seems likely that cPLA2 protein is generated by an induction of cPLA2 mRNA before PGHS and provides PGHS with AA and that PGHS, which is synthesized belatedly for cPLA2, in turn synthesizes PGI2 through the PGHS pathway. The newly synthesized cPLA2 and PGHS proteins apparently act mainly in the later phase (after 180 minutes of BK stimulation) and maintain the sustained phase of PGI2 generation. Their time course appears to be a reasonable reaction; however, we could not quantify the protein of cPLA2 and PGHS. Further examinations to determine the relation of mRNA expression and protein synthesis are necessary.
To investigate the mechanism of intracellular signal transduction in BK-stimulated HUVEC, we evaluated the effects of cycloheximide and EGTA on mRNA expression. As speculated, BK stimulation under the condition of cycloheximide pretreatment remarkably increased the level of cPLA2 and PGHS-2 mRNA expression in HUVEC. The cycloheximide-induced superinduction of cPLA2 mRNA was previously reported in rat mesangial cells incubated with epidermal growth factor.16 Cytosolic phospholipase A2 mRNA, similar to immediate early genes, contains an AU-rich sequence in its 3′-untranslated region, and cycloheximide superinduces immediate early genes by altering the stability of mRNA and inhibiting the synthesis of AU-binding factor, which increases the instability of mRNA. In the case of PGHS-2, cycloheximide also superinduces mRNA expression to about 90-fold that of the control level. Because PGHS-2 mRNA also contains an AU-rich sequence in its 3′-untranslated region,14 mechanisms similar to those of cPLA2 are thought to be involved in its cycloheximide-induced superinduction. Moreover, Srivastava et al17 detected a new protein induced by interleukin-1β in rat mesangial cells and reported that this protein stabilizes PGHS-2 mRNA by binding its AU-rich region. Interestingly, the finding was completely different for PGHS-1 mRNA expression. As its expression is almost completely inhibited by cycloheximide, we speculate that there exists an enhancer protein that is indispensable for the transcription of PGHS-1 mRNA. In our present study, EGTA pretreatment inhibited the Ca2+ influx by chelating extracellular Ca2+, resulting in a decrease in the BK-induced increase of [Ca2+]i by ≈50%. Although the importance of intracellular Ca2+ as a second messenger in intracellular signal transduction has been recognized, a detailed account of the relation between the second messenger, including intracellular Ca2+, and mRNA has not been provided. EGTA pretreatment increased cPLA2 mRNA expression to the same extent as pretreatment with cycloheximide but decreased PGHS-1 mRNA expression to almost zero. The finding that PGHS-2 mRNA expression was not affected by EGTA pretreatment is unexpected because protein phosphorylation by PKC18 or protein tyrosine kinase activation is causally linked to expression of mRNAs19 and because some kinds of PKC isoforms are Ca2+-sensitive. The report that the Ca2+ ionophore A23187 induces PGHS-2 mRNA expression in IL-1β–stimulated chondrocytes20 shows the important role of intracellular Ca2+ in mRNA expression. The elucidation of these phenomena must await further investigations of the intracellular Ca2+–dependent signaling mechanisms involved with cPLA2 and PGHS mRNA expression.
This is the first report to investigate the effects of BK on the regulatory mechanisms of PGI2 generation by cPLA2 activity and the mRNA expression of cPLA2 and PGHS isoforms in human vascular endothelial cells. We determined that the elevation of cPLA2 activity caused by the increase of intracellular Ca2+ concentration is an important event in the early phase of BK-induced PGI2 synthesis, and we conclude that the mechanisms for the regulation of cPLA2 and PGHS isoforms in endothelial cell are different. However, because we could not directly measure the protein and PGHS activity, further evaluations are needed to elucidate the regulatory mechanisms of BK-induced PGI2 synthesis in HUVEC.
S. Yamasaki and S. Komatsu contributed equally to this work.
- Received May 4, 1999.
- Revision received June 8, 1999.
- Accepted March 1, 2000.
Brunner HR, Gavras H, Waeber B, Kershow GR, Turini GA, Vukovich RA. Oral angiotensin-converting enzyme inhibitor in long-term treatment of hypertensive patients. Ann Intern Med. 1979;90:19–23.
Brown NJ, Ryder D, Gainer JV, Morrow JD, Nadeau J. Differential effects of angiotensin converting enzyme inhibitors on the vasodepressor and prostacyclin responses to bradykinin. J Pharmacol Exp Ther. 1996;279:703–712.
Kramer RM. The Ca2+-sensitive cytosolic phospholipase A2 is a 100-kDa protein in human monoblast U937 cells. J Biol Chem. 1991;266:5268–5272.
Grynkiewicz G, Poenie M, Tsien RY. A new generation of Ca2+ indicators with greatly improved fluorescence properties. J Biol Chem. 1985;260:3440–3450.
Toratani A, Sawada S, Kono Y, Higaki T, Imamura T, Tada Y, Yamasaki S, Sato T, Komatsu S, Akamatsu N, Tamagaki T, Nakagawa K, Tsuji H, Nakagawa M. Interleukin-1α stimulated prostacyclin release by increasing gene transcription of prostaglandin H synthase and phospholipase A2 in human vascular endothelial cells. J Cardiovasc Pharmacol. 1999;33:843–851.
Saiki R, Gelfand D, Stoffel S, Scharf S, Higuchi R, Horn G. Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science. 1988;239:487–491.
Higuchi R, Krummel B, Saiki RK. A general method of in vitro preparation and specific mutagenesis of DNA fragments: study of protein and DNA interactions. Nucleic Acids Res. 1988;16:7351–7367.
Channon JY, Leslie CC. A calcium-dependent mechanism for associating a soluble arachidonil-hydrolyzing phospholipase A2 with membrane in the macrophage cell line RAW264.7. J Biol Chem. 1990;265:5409–5413.
Hla T, Neilson K. Human cyclooxygenase-2 cDNA. Proc Natl Acad Sci U S A. 1992;89:7384–7388.
Kawaguchi H, Yavari R, Stover ML, Rowe DW, Raisz LG, Pilbeam CC. Measurement of interleukin-1 stimulated constitutive prostaglandin G/H synthase mRNA levels in osteoblastic MC3T3–E1 cells using competitive reverse transcriptase polymerase chain reaction. Endcr Res. 1994;20:219–223.
Maxwell AP, Goldberg HJ, Tay AH-N, Li Z-G, Arbus GS, Skorecki KL. Epidermal growth factor and phorbol myristate acetate increase expression of mRNA for cytosolic phospholipase A2 in glomerular mesangial cells. Biochem J. 1993;295:763–766.
Srivastava SK, Tetsuka T, Daphna-Ikan D, Morrison AR. IL-1β stabilizes COX-II mRNA in renal mesangial cells: role of 3′-untranslated region. Am J Physiol. 1994;267:F504–F508.
Hirai K, Takayama H, Tomo K, Okuma M. Protein-tyrosine-kinase-dependent expression of cyclooxygenase-1 and -2 mRNAs in human endothelial cells. Biochem J. 1997;322:373–377.
Geng Y, Blanco FJ, Cornerission M, Lotz M. Regulation of cyclooxygenase-2 expression in normal human articular chondrocytes. J Immunol. 1995;155:796–801.