Nifedipine Increases Cytochrome P4502C Expression and Endothelium-Derived Hyperpolarizing Factor–Mediated Responses in Coronary Arteries
Abstract—In addition to NO and prostacyclin, endothelial cells release a factor that elicits vasodilatation by hyperpolarizing the underlying vascular smooth muscle cells. In some vascular beds, this so-called endothelium-derived hyperpolarizing factor (EDHF) displays the characteristics of a cytochrome P450 (CYP)-derived arachidonic acid metabolite, such as an epoxyeicosatrienoic acid. Native porcine and cultured human coronary artery endothelial cells were screened for CYP epoxygenases, and CYP2B, CYP2C, and CYP2J were detected with reverse transcription–polymerase chain reaction. The CYP inducer β-naphthoflavone and the Ca2+ antagonist nifedipine significantly increased CYP2C mRNA but did not change the expression of CYP2J or CYP2B. To determine the relationship between CYP2C expression and EDHF production in native endothelial cells, we incubated porcine coronary arteries with nifedipine. Nifedipine enhanced endothelial CYP2C protein expression, as well as the generation of 11,12-epoxyeicosatrienoic acid. In organ bath experiments, pretreatment with nifedipine enhanced bradykinin-induced, EDHF-mediated relaxations as well as the concomitant hyperpolarization of smooth muscle cells. The specific CYP2C9 inhibitor sulfaphenazole, on the other hand, significantly attenuated EDHF-mediated hyperpolarization and relaxation. These results demonstrate that in porcine coronary arteries, the elevated expression of a CYP epoxygenase, homologous to CYP2C8/9, is associated with enhanced EDHF-mediated hyperpolarization in response to bradykinin. Therefore, we propose that an isozyme of CYP2C is the most likely candidate for the CYP-dependent EDHF synthase in porcine coronary arteries.
The existence of an NO/prostacyclin (prostaglandin [PG]I2)-independent component of endothelium-dependent relaxation has been convincingly demonstrated in various conduit and resistance arteries. Because this NO/PGI2-independent vasodilatation was associated with vascular smooth muscle hyperpolarization, the term “endothelium-derived hyperpolarizing factor” (EDHF) was coined.1
On the basis of the experimental data published to date, it is evident that several EDHFs exist in different species.1 However, the majority of reports that characterize EDHF in the coronary vascular bed support the concept that this EDHF is a cytochrome P450 (CYP)-related product. This hypothesis was based on the observation that CYP inhibitors and a number of anesthetic agents markedly attenuate EDHF-mediated responses.1 However, these conclusions are limited by the fact that these CYP inhibitors cannot discriminate between the different CYP isoforms and can directly interfere with Ca2+-dependent K+ channels,2 which are thought to be the main targets of EDHF in vascular cells.3 4
Therefore, the aim of the present study was to identify a candidate EDHF synthase in endothelial cells by screening for the expression of CYP epoxygenases and to determine the effects of clinically relevant antihypertensive agents (ie, ramiprilat, nifedipine) on the induction of CYP enzymes and CYP product formation, as well as the bradykinin-induced EDHF-mediated hyperpolarization and relaxation.
Cultured Endothelial Cells
Human umbilical vein endothelial cells were isolated and cultured as described previously,5 and human coronary artery endothelial cells (HCAECs) were obtained from CellSystems and cultured in SmGM-2 (CellSystems). In some experiments, cGMP levels were determined.5
Preparation of Porcine Coronary Arteries
Porcine epicardial artery segments (≈40 mm long, mean external diameter 2.4 to 2.8 mm) were excised, side branches were sealed with surgical clips, and the segment was cannulated at both ends and placed into vessel chambers. After equilibration, coronary arteries were perfused with MEM containing β-naphthoflavone (3 μmol/L, 48 hours), or nifedipine (0.1 μmol/L, 18 hours). Half of each segment was then cut into rings for organ chamber studies and precontracted with U46619 (0.1 to 1 μmol/L), and relaxation in response to bradykinin was determined in the presence of diclofenac (10 μmol/L) and in the absence and presence of the NO synthase (NOS) inhibitor Nω-nitro-l-arginine (300 μmol/L) as described previously.6 In some experiments, the endothelium was removed after pretreatment with either solvent or nifedipine, and relaxant responses to sodium nitroprusside (SNP) were determined.
Isolation of RNA and Protein
From the remaining coronary artery segment, endothelial cells or total RNA was isolated through intraluminal incubation with either dispase (2.4 U/mL) or guanidine isothiocyanate. Random hexanucleotide primers were used for reverse transcription (RT) of equal amounts of RNA, and the oligonucleotides used for polymerase chain reaction (PCR) were derived from a porcine CYP2C34 sequence (GenBank accession no. U35843) (upstream primer AGACAACGAGCACCACTCTG, downstream primer CCTTGGGGATGAGGTAGTTT) that exhibited a high homology to the human 2C8 sequence (GenBank accession no. Y00498) (CYP2B2 upstream primer TGGTGGAGGARCTGCGGAAATCC, downstream primer TGCCTTCGCCAAGACAAAYGCG [GenBank accession no. M34452]; CYP2J2 upstream primer CCCTCAYTTCAAGATCAACA, downstream primer GCAGATGAGGTTTTCTTCAT [GenBank accession no. U37143]; and elongation factor [EF] upstream primer GACATCACCAAGGGTGTGCAG, downstream primer GCGGTCAGCACACTGGCATA).
PCR products were separated on a 1.5% TAE agarose gel and visualized by staining with ethidium bromide. For the verification of the DNA fragment, the PCR products were transferred to nylon membranes and hybridized with 32P-labeled DNA fragments derived from a plasmid containing the coding sequence of CYP2C8, CYP2B, or CYP2J2. Phenol-soluble protein or microsomal fractions (100 000g pellets) prepared from isolated porcine coronary artery endothelial cells (PCAECs) were subjected to SDS–polyacrylamide gel electrophoresis (8%) and Western blotting with the use of 2 different CYP2C11 polyclonal antibodies (kindly provided by Dr E. Morgan [Atlanta, Ga] or purchased from Daiichi Pure Chemicals): an endothelial NOS (eNOS) polyclonal antibody (Transduction Laboratories) or a platelet-endothelial cell adhesion molecule 1 (PECAM-1) monoclonal antibody (Santa Cruz).
Analysis of CYP-Derived Products
Native PCAECs harvested with the use of dispase, as described earlier, were allowed to equilibrate in HEPES-modified Tyrode’s solution containing 300 μmol/L Nω-nitro-l-arginine and 10 μmol/L diclofenac, were stimulated with 100 nmol/L bradykinin (5 minutes), and were frozen in liquid nitrogen. Endothelial cells were homogenized, and CYP-derived eicosanoids were detected as described previously.7
After incubation with solvent or nifedipine, coronary artery rings were slit, mounted in a heated chamber, and maintained in HEPES-modified Tyrode solution containing 10 μmol/L diclofenac, 300 μmol/L Nω-nitro-l-arginine, and 1 μmol/L U46619 to mimic conditions in the organ chamber experiments. Smooth muscle membrane potential was recorded by impaling cells through the intima as described previously.8
Data are expressed as mean±SEM. Statistical evaluation was performed with Student’s t test for unpaired data, 1-way ANOVA followed by a Bonferroni t test, or ANOVA for repeated measures where appropriate. Values of P<0.05 were considered statistically significant.
CYP Expression in Endothelial Cells
To determine which CYP isoform could be implicated in the generation of EDHF-mediated responses, we used RT-PCR to screen native PCAECs, as well as cultured HCAECs. Of the CYP epoxygenases that we investigated, isozymes homologous to 2C8/9 and 2J2 were expressed in both HCAECs and PCAECs (Figure 1A⇓). Only the globally expressed CYP2C family fit the criteria of an EDHF synthase, because its expression was consistently enhanced by β-naphthoflavone in all of the endothelial cell types studied. To determine whether CYP2C expression and EDHF-mediated responses could be modulated by antihypertensive agents, we tested the effects of the Ca2+ antagonist nifedipine and the ACE inhibitor ramiprilat. While ramiprilat was without effect, 0.1 μmol/L nifedipine (18 hours) enhanced CYP2C RNA expression in native PCAECs (Figure 1B⇓).
CYP Induction and Epoxyeicosatrienoic Acid Production
To establish a functional link between the expression of CYP2C and the production of EDHF, experiments were performed with isolated endothelium-intact porcine coronary arteries, and epoxyeicosatrienoic acid (EET) levels were measured in endothelial cell homogenates. Intraluminal incubation with nifedipine enhanced the expression of CYP2C protein in endothelial cells (Figure 1C⇑). In endothelial cells harvested from coronary arteries maintained under control conditions for 18 hours, detectable amounts of 11,12-EET were generated in response to bradykinin (Figure 2⇓). A >3-fold increase in endothelial EET generation was detected after the incubation of coronary segments with either β-naphthoflavone or nifedipine (Figure 2⇓).
CYP Induction and EDHF-Mediated Hyperpolarization
Induction of CYP2C potentiated the bradykinin-induced, EDHF-mediated hyperpolarization of porcine coronary arteries (Figures 3A⇓ and 3B⇓) measured in the continuous presence of NOS and cyclooxygenase inhibitors. Nifedipine (0.1 μmol/L, 18 hours) did not affect the resting membrane potential of smooth muscle cells (−50.5±1.8 mV in control versus −49.0±2.2 mV in nifedipine-treated arteries) or the hyperpolarization elicited by 11,12-EET (the 1 μmol/L EET–induced hyperpolarization was 8.6±0.9 mV in control arteries versus 8.7±2.3 mV in arteries treated with nifedipine, n=4, NS). Sulfaphenazole, which selectively inhibits CYP2C9 in the human liver,9 significantly attenuated the EDHF-mediated hyperpolarization of coronary artery rings (hyperpolarization was reduced from 16.5±2.9 mV in the absence of sulfaphenazole to 6.9±2.2 mV in the presence of sulfaphenazole, n=11, P<0.05). The EDHF-mediated hyperpolarization of the porcine coronary artery, unlike EDHF-mediated responses in other arteries, was sensitive to iberiotoxin as well as to the combination of charybdotoxin and apamin (Figure 3C⇓).
In cultured cells, the expression of CYP isozymes is known to decrease with time. A similar effect is observed in native endothelial cells maintained for prolonged periods in culture medium. Indeed, although bradykinin-induced hyperpolarization of porcine coronary arteries is detectable for up to 60 hours, the amplitude of the hyperpolarization decreases in a time-dependent manner (compare Figures 3B⇑ and 3C⇑), and there is a rightward shift in the bradykinin-induced EDHF-mediated relaxation of porcine coronary arteries incubated for various times.8
CYP Induction and EDHF-Mediated Relaxation
In accordance with the effects of nifedipine on the agonist-induced, endothelium-dependent hyperpolarization of porcine coronary artery smooth muscle cells, nifedipine induced a leftward shift in the EDHF-mediated concentration-relaxation curve to bradykinin (EC50=−7.66±0.12 mol/L in control versus −8.35±0.15 mol/L in nifedipine-treated rings, P<0.001, n=6; Figure 4A⇓). The CYP2C9 inhibitor sulfaphenazole did not affect basal tone (data not shown) but significantly attenuated the bradykinin-induced EDHF-mediated relaxation of coronary artery rings (Figure 4B⇓). The potentiating effect of nifedipine was restricted to the EDHF-mediated relaxation as the NO-mediated relaxation of nifedipine-treated segments was attenuated (EC50=−8.35±0.09 mol/L in control versus −8.21±0.16 mol/L in nifedipine-treated rings, P<0.05, n=6; Figure 4C⇓). Nifedipine altered neither the Ca2+ response of cultured endothelial cells to bradykinin (data not shown) nor the responsiveness of vascular smooth muscle cells to nitrovasodilators. Relaxation responses to SNP in endothelium-denuded segments were identical in the solvent- and nifedipine-treated groups (EC50=−7.79±0.05 mol/L in control versus −7.83±0.05 mol/L in nifedipine-treated rings, NS, n=14; Figure 4D⇓).
Role of NO and Protein Kinase C in Nifedipine-Mediated CYP Induction
The nifedipine-induced attenuation of NO-mediated relaxation could be correlated with a slight decrease in the expression of eNOS. In cultured endothelial cells, 0.1 μmol/L nifedipine (18 hours) decreased the expression of eNOS (Figure 5⇓) and attenuated the basal production of cGMP (1.2±0.5 versus 0.7±0.2 pmol/well in the absence and presence of nifedipine, P=0.38, n=8, NS), as well as the bradykinin-stimulated increase in cGMP levels (15.5±1.4 versus 12.3±0.3 pmol/well in the absence and presence of nifedipine, n=8, P<0.05).
To determine whether the effect of nifedipine on CYP2C expression was a consequence of its effect on eNOS expression, experiments were performed with Nω-nitro-l-arginine and the NO donor (Z)-1-[2-(2-Aminoethyl)-N-(2-ammonioethyl)amino]diazen-1-ium-1,2-diolate (DETA-NO). In 3 of 6 experiments, 300 μmol/L Nω-nitro-l-arginine (18 hours) enhanced CYP2C expression, whereas 500 μmol/L DETA-NO (18 hours) consistently attenuated CYP2C expression. Moreover, in the presence of DETA-NO, nifedipine did not increase CYP2C expression (Figure 5⇑).
Because nifedipine has been suggested to inhibit PKC in endothelial cells,10 we compared the effects of the Ca2+ antagonist with those of an inhibitor of Ca2+-dependent and -independent protein kinase C (PKC) isoforms. Ro-31-8220 (100 nmol/L, 18 hours) slightly depressed the expression of CYP2C in human endothelial cells and completely prevented its induction by nifedipine (Figure 5⇑).
The results presented here demonstrate that EDHF-mediated hyperpolarization and relaxation elicited by bradykinin in porcine coronary arteries can be modulated by enhancement of the expression of CYP2C with nifedipine or by inhibition of CYP2C9 activity with the use of sulfaphenazole. This supports the concept that a CYP-dependent process is crucial for the generation of EDHF-mediated responses in porcine coronary arteries.
Previously, indirect evidence has suggested that a CYP epoxygenase-derived product mediates NO/PGI2-independent relaxation in the human,11 porcine,3 bovine,12 canine,13 and rat14 coronary circulations. One hindrance to the identification of the CYP isozyme or isozymes that may generate a hyperpolarizing factor in endothelial cells has been the fact that these enzymes are expressed at extremely low levels in cultured cells. Thus, it was necessary to screen both cultured and native coronary endothelial cells to determine which CYP enzymes may be involved in the generation of EDHF. Previous reports demonstrated that 2 CYP epoxygenases (CYP2C8/915 and CYP2J216 ) can be expressed in human endothelial cells, and we were able to detect homologs of both enzymes in the endothelium removed from porcine coronary arteries. Because we previously reported that a CYP-derived hyperpolarizing factor can be physically transferred from cultured endothelial to smooth muscle cells after the induction of CYP enzymes with β-naphthoflavone,3 we took the sensitivity to this inducer as a criterion for the identification of the putative coronary EDHF synthase. Only the CYP2C enzymes were consistently induced after incubation with β-naphthoflavone. Moreover, the results of the present investigation, which show the generation of EETs and the amplitude of EDHF-mediated smooth muscle hyperpolarization and relaxation were potentiated after the induction of CYP2C, suggest that a homolog of CYP2C8/9 is involved in the generation of EDHF in porcine coronary arteries. Indeed, we have recently shown that antisense oligonucleotides against CYP2C selectively attenuate the EDHF-mediated hyperpolarization and relaxation of coronary arteries.8
To demonstrate that CYP2C expression and EDHF-mediated responses could be induced by a completely different class of compounds to β-naphthoflavone, experiments were performed with the Ca2+ antagonist nifedipine. From a theoretical point of view, there was no reason to expect that nifedipine would be able to induce CYP2C, because this compound is known to be metabolized in the liver by CYP3A4.17 Moreover, the nifedipine-specific element, described in the CYP3A4 promoter,18 does not appear to be present in the CYP2C gene family. Because nifedipine has been reported to inhibit PKC in endothelial cells,10 we compared the effects of nifedipine on CYP2C expression with those of Ro-31-8220, which inhibits Ca2+-dependent and -independent isoforms of PKC. However, in contrast to nifedipine, Ro-31-8220 attenuated endothelial CYP2C expression, suggesting that although the activity of PKC may modulate endothelial CYP2C expression, the mechanism that underlies the upregulation of CYP2C expression by nifedipine is not related to its inhibitory action on PKC.
The prolonged incubation of cultured HCAECs with nifedipine downregulated eNOS protein levels. We therefore speculated that NO may regulate CYP/EDHF synthase expression. Indeed, in some experiments, the inhibition of basal NO production enhanced CYP2C expression. The inconsistency in this observation may be related to variations in basal NO production in different cell batches. High concentrations of NO, on the other hand, consistently decreased the expression of CYP2C. It is therefore tempting to suggest that the expression and activity of the EDHF synthase are attenuated by NO. Caution should be exerted in the interpretation of data that relate to the induction of CYP expression by pharmacological substances because these compounds may be partially metabolized by endothelial cells and thus induce the expression of CYP enzymes independent of any effects related to the production of NO.
The list of candidate EDHFs is relatively long, and in most vascular preparations in which EDHF clearly is not a CYP metabolite (eg, the guinea pig carotid artery19 and the rat hepatic artery20 ), EDHF-mediated hyperpolarizations are insensitive to iberiotoxin but sensitive to the combination of charybdotoxin and apamin. In the porcine and canine coronary arteries, as well as the rat renal artery, in which EDHF also appears to be a CYP metabolite, EDHF-mediated hyperpolarizations are equally sensitive to iberiotoxin21 and charybdotoxin/apamin.22 23 24 Thus, because the CYP2C product 11,12-EET, which is generated by coronary artery endothelial cells, activates iberiotoxin-sensitive K+Ca channels,25 the evidence accumulated to date strongly suggests that a CYP-derived product plays a crucial role in the bradykinin-induced generation of EDHF-mediated responses in the coronary artery. The reason for the caution in stating that the coronary EDHF is an EET is that it is currently unclear how endothelium-derived EETs are able to access smooth muscle cells to elicit hyperpolarization. EETs are by nature lipophilic, and simple diffusion from the endothelium to smooth muscle cells down a concentration gradient would be too slow to account for the rapidly initiated EDHF-mediated hyperpolarization and relaxation. One possibility that would also link many of the observations made in different vessels is that EETs are able to either pass through gap junctions or influence gap junctional communication of a second hyperpolarizing agent.
Taken together, the results of the present study demonstrate that nifedipine enhances the bradykinin-induced, EDHF-mediated hyperpolarization and relaxation of porcine coronary artery segments. These effects appear to be directly attributable to the upregulation of an enzyme homologous to CYP2C8/9 and the generation of the CYP metabolite 11,12-EET.
This work was supported by the Deutsche Forschungsgemeinschaft (SFB 553, B1). The authors are indebted to Isabel Winter, Michaela Stächele, and Stergiani Hauk for expert technical assistance.
- Received November 22, 1999.
- Revision received December 28, 1999.
- Accepted March 2, 2000.
Alvarez J, Montero M, García-Sancho J. Cytochrome P450 may link intracellular Ca2+ stores with plasma membrane Ca2+ influx. Biochem J. 1991;274:193–197.
Doughty JM, Plane F, Langton PD. Charybdotoxin and apamin block EDHF in rat mesenteric artery if selectively applied to the endothelium. Am J Physiol. 1999;276:H1107–H1112.
Bauersachs J, Popp R, Hecker M, Sauer E, Fleming I, Busse R. Nitric oxide attenuates the release of endothelium-derived hyperpolarizing factor. Circulation. 1996;94:3341–3347.
Kiss L, Bieniek E, Weissmann N, Schutte H, Sibelius U, Gunther A, Bier J, Mayer K, Henneking K, Padberg W, Grimm H, Seeger W, Grimminger F. Simultaneous analysis of 4- and 5-series lipoxygenase and cytochrome P450 products from different biological sources by reversed-phase high-performance liquid chromatographic technique. Anal Biochem. 1998;261:16–28.
Mancy A, Dijols S, Poli S, Guengerich P, Mansuy D. Interaction of sulfaphenazole derivatives with human liver cytochromes P450 2C: molecular origin of the specific inhibitory effects of sulfaphenazole on CYP 2C9 and consequences for the substrate binding site topology of CYP 2C9. Biochemistry. 1996;35:16205–16212.
Hempel A, Lindschau C, Maasch C, Mahn M, Bychkov R, Noll T, Luft FC, Haller H. Calcium antagonists ameliorate ischemia-induced endothelial permeability by inhibiting protein kinase C. Circulation. 1999;99:2523–2529.
Miura H, Liu Y, Gutterman DD. Human coronary arteriolar dilation to bradykinin depends on membrane hyperpolarization: contribution of nitric oxide and Ca2+-activated K+ channels. Circulation. 1999;99:3132–3138.
Oltman CL, Weintraub NL, VanRollins M, Dellsperger KC. Epoxyeicosatrienoic acids and dihydroxyeicosatrienoic acids are potent vasodilators in the canine coronary microcirculation. Circ Res. 1998;83:932–939.
Node K, Huo Y, Ruan X, Yang B, Spiecker M, Ley K, Zeldin DC, Liao JK. Anti-inflammatory properties of cytochrome P450 epoxygenase-derived eicosanoids. Science. 1999;285:1276–1279.
Guengerich FP, Martin MV, Beaune PH, Kremers P, Wolff T, Waxman DJ. Characterization of rat and human liver microsomal cytochrome P-450 forms involved in nifedipine oxidation, a prototype for genetic polymorphism in oxidative drug metabolism. J Biol Chem. 1986;261:5051–5060.
Frieden M, Sollini M, Bény J-L. Substance P and bradykinin activate different types of Kca currents to hyperpolarize cultured porcine coronary artery endothelial cells. J Physiol (Lond). 1999;519:2:361–371.
Nishikawa Y, Stepp DW, Chilian WM. In vivo location and mechanism of EDHF-mediated vasodilation in canine coronary microcirculation. Am J Physiol. 1999;277:H1252–H1259.
Zou AP, Fleming JT, Falck JR, Jacobs ER, Gebremedhin D, Harder DR, Roman RJ. Stereospecific effects of epoxyeicosatrienoic acids on renal vascular tone and K+-channel activity. Am J Physiol. 1996;270:F822–F832.
Baron A, Frieden M, Bény J-L. Epoxyeicosatrienoic acids activate a high-conductance, Ca2+-dependent K+ channel on pig coronary artery endothelial cells. J Physiol (Lond). 1997;504:3:537–543.