The calcium antagonist verapamil is commonly prescribed for arrhythmias, antianginal therapy, and myocardial ischemia, but suffers from extensive first-pass metabolism (McAllister et al., 1986). This results in low drug bioavailability and considerable variability of therapeutic plasma levels (Piovan et al., 1995). Metabolism of verapamil leads to pharmacological inactivation and, thus, patients require frequent dosage of this particular drug. Although the therapeutic benefit of verapamil is well documented (Weaver, 1986), the tissue-specific metabolism may impact its mode of action in cardiovascular diseases. Indeed, verapamil is initially metabolized into two break-down products, namely norverapamil and D-617¹. These metabolites are subject to further metabolism by the cytochrome P450 system to form additional secondary metabolites. Tracy et al. (1999) reported several cytochrome P450 isoforms to be involved in the metabolism of verapamil, and this includes CYP3A4, CYP3A5, CYP2C8, CYP2C18, CYP2D6, and CYP2E1.

Verapamil acts primarily on heart tissue and is transported via coronary vessels to heart muscle cells. An important barrier to overcome is the endothelium, and it is highly interesting to explore the contribution of human coronary arterial endothelial cells in the heart-specific metabolism of verapamil.

Materials and Methods

Cell Culture. Primary human coronary arterial endothelial cells were obtained from CellSystems (St. Katharinen, Germany). Cells were cultured (fourth passage) in six wells in endothelial growth medium-2 (CellSystems) until confluence. Cells were treated with 2 μM verapamil for 24 h and supernatant was harvested thereafter.

Quality Assurance. Expression of the endothelial-specific surface protein PECAM-1 was determined by fluorescence-activated flow cytometry as described previously (Thum et al., 2000).

RNA and cDNA. RNA was isolated from endothelial cells using a total RNA Isolation kit (Macherey-Nagel, Duren, Germany) according to the manufacturer's recommendation. Quality and quantity of isolated RNA were checked using capillary electrophoresis (Agilent Bioanalyzer 2100) following the manufacturer's instructions. Total RNA (2 μg) from each sample was used for reverse transcription, as described previously (Thum and Borlak, 2000). The resulting cDNA was frozen at–20°C until further experimentation.

Semiquantitative Reverse Transcription-PCR. PCRs were carried out with a thermal cycler (T3; Biometra, Göttingen, Germany) using the following PCR conditions: denaturation at 95°C for 45 s, annealing at 72°C for 60 s, and extension at 72°C for 60s. A total of 30–36 PCR cycles were carried out. Detailed oligonucleotide sequence information was published previously (Thum and Borlak, 2000) and can be obtained from the authors. DNA contamination was checked for by direct amplification of RNA extracts prior to conversion to cDNA. Contamination of RNA extracts with genomic DNA could be excluded. PCRs were done within the linear range of amplification, and amplification products were separated using a 1.5% agarose gel and stained with ethidium bromide. Gels were photographed on a transilluminator (Kodak Image Station 440; see Fig. 1).

• Download figure
• Open in new tab
• Download powerpoint

Fig. 1.

Gene and protein expression of major cytochrome P450 monooxygenases in primary human coronary arterial endothelial cells.

a, ethidium bromide-stained 1.5% agarose gel. Results after reverse transcription-PCR experiments as described under Materials and Methods. CYC, cyclophilin A; bp, base pairs. b, Western immunoblotting of CYP2C and CYP2E1 in total protein extracts of cultured endothelial cells (EAhy926) (e1–e3) and microsomal extracts of human liver (HL).

Protein Expression. Western immunoblotting was done as follows. Total protein extracts (100 μg) from a cultured endothelial cell line (EAhy926) and microsomal extracts from human liver tissue (50 μg) were denaturated at 95°C for 5 min, followed by SDS-polyacrylamide gel electrophoresis on 12% polyacrylamide gels, and blotted onto a polyvinylidene difluoride membrane [PerkinElmer Life Sciences (Germany) GmbH, Rodgau-Jügesheim, Germany] at 350 mA for 2 h in a buffer containing 400 mM glycine, 50 mM Tris (pH 8.3). Nonspecific binding sites were blocked with Rotiblock (Roth, Karlsruhe, Germany) in 1× TBS buffer. After electroblotting of proteins, membranes were incubated with polyclonal antibodies for CYP2C and CYP2E1 (Chemicon, Hofheim, Germany; dilution 1:1,000–1:2,000) for 1 h and washed three times with 1× TBS buffer containing 0.1% Tween 20 (Roth). Subsequently, the membranes were incubated with a 1:10,000 diluted anti α-sheep antibody (Chemicon) for 1 h at room temperature followed by three successive washes with 1× TBS buffer containing 0.1% Tween 20 (Roth). Immunoreactive proteins were visualized with a chemiluminescence reagent kit [PerkinElmer Life Sciences (Germany) GmbH] according to the manufacturer's instructions, and bands were scanned with the Kodak Image Station CF 440 and analyzed using the Kodak 1D 3.5 imaging software (Eastman Kodak, Rochester, NY).

Chemicals. Chemicals were obtained from the following companies: AcCN (Malinckrodt-Baker, Deventer, Holland), MeOH (Mallinckrodt Baker, Deventer, Holland), NH₄Ac (Merck, Darmstadt, Germany), acetic acid (Fluka, Buchs, Switzerland), norverapamil [5-N-(3,4-dimethoxyphenethyl)amino-2-(3′,4′-dimethoxyphenyl)-2-isopropyl-valeronitrile] (Sigma/RBI, Natick, MA), and verapamil [5-N-(3,4-dimethoxyphenethyl)methylamino-2-(3′,4′-dimethoxyphenyl)-2-isopropyl-valeronitrile] (lot 56H0925; Sigma-Aldrich, Steinheim, Germany). O-Demethylverapamil [5-N-(3,4-dimethoxyphenethyl)methylamino-2-(3′-methoxy-4′-hydroxyphenyl)-2-isopropylvaleronitrile, D-703] was a kind gift of W. L. Nelson (University of Washington, Seattle, WA).

LC/MS Analysis. For solid phase extraction, a lipophilic cartridge (RP8 Select B; Merck) was used and conditioned with methanol, followed by equilibration with water. The sample was loaded onto the cartridge without any organic solvent and washed with 3% (v/v) methanol to separate any sample matrix. Verapamil and its basic metabolites were eluted with methanol. Eluants were evaporated to dryness and reconstituted in 200 μl of acetonitrile/ammonium acetate (0.01 M, pH 6.0, 50:50, v/v). Aliquots of 20 μl were injected onto the LC/MS system.

LC/MS analyses were done on a Waters (Milford, MA) LC instrument (pumps 590) coupled to an ion trap mass spectrometer (Esquire from Bruker Daltonics, Bremen, Germany) operated with positive ion electrospray conditions in the full scan and, when possible, in the MSⁿ mode. The nebulizer pressure was set to 40 psi and the dry gas temperature to 350°C, while +3 kV were applied to the nebulizing capillary. Full mass spectra were acquired by scanning the mass range of m/z 100 to 700. Collision-induced dissociation spectra were obtained from the protonated molecules (M + H)⁺. Highperformance liquid chromatographic analysis was carried out with a gradient elution of ammonium acetate buffer (0.01 M, pH 6.0) and acetonitrile using a flow of 0.2 ml/min. The gradient started with 25% acetonitrile and was raised to 50% within 15 min, and further, to 75% acetonitrile from 15 to 30 min. From 30 to 55 min, delivery of acetonitrile remained constant and was decreased to 25% from 55 to 60 min. Separation of verapamil and its metabolites was achieved on a 250 × 2 mm RP select B column with a particle size of 4 μm (Merck).

Recovery. Recovery experiments were done in quadruplicate. Solid phase extraction was done as described above, and the resultant eluent was reduced in volume to 200 μl. Measurement of these extracts was done as described above, and recoveries of 85 ± 4% and 81 ± 5% for verapamil and norverapamil, respectively, were estimated.

Quantification. Abundances were calculated from the peak ratios of the metabolites relative to verapamil (see Table 1). This allows a semiquantitative estimate of metabolites.

Results

Cell Culture and Gene/Protein Expression Studies. On the basis of phase contrast light microscopic examination, neither morphological changes nor signs of abnormalities were observed in control or verapamil-treated cell cultures. In addition, >95% of cultured human coronary arterial endothelial cells expressed PECAM-1, an endothelial-specific adhesion molecule well documented for its utility as a differentiation marker (Thum et al., 2000). As shown in Fig. 1a, we found CYP1A1, CYP2A6/7, CYP2A13, CYP2B6/7, CYP2C8, CYP2E1, CYP2J2, and cyclophilin (housekeeping gene) to be expressed in cultures of human coronary arterial endothelial cells, but transcript levels of CYP1A2, CYP2C9, CYP2C18, CYP2C19, CYP2D6, CYP2F1, CYP2S1, CYP3A4, CYP3A5, CYP3A7, and CYP4B1 were below the limit of detection. Protein expression of CYP2C and CYP2E1 was evidenced by Western immunoblotting. As shown in Fig. 1b, the polyclonal antibodies recognized epitopes of CYP2C and CYP2E1 in extracts of cultures of the endothelial cell line EAhy926 and microsomal extracts of human liver (positive control). Noticeably, the polyclonal antibody used to probe for CYP2C detected major isoforms of this protein (CYP2C8/9/18/19).

Metabolite Identification. Identification of metabolites was done essentially as described previously (Walles et al., 2001). Analysis of extracts from culture supernatants of primary human coronary arterial endothelial cells are highly suggestive for metabolism of verapamil to proceed via production of D-617 (1.4%) and norverapamil (1%; see Table 1). Next to the O-demethylated products minor amounts of D-702 (0.2%), D-703 (0.2%), and D-620 (0.05%) were produced additionally.

Metabolite M1 (D-620). In the full-scan MS mode, a protonated molecule, (M + H)⁺ = 277, was observed, which points to oxidative dealkylation of the lower substituted phenylalkyl moiety with simultaneous demethylation (M = 178 Da less than verapamil). Further fragmentation of m/z = 277 in the MS² mode produced two abundant ions of m/z = 234 (loss of the isopropyl group) and m/z = 260 (loss of ammonia), thus demonstrating oxidative N-demethylation (see Table 1).

Metabolite M3 (D-617). The (M + H)⁺ of m/z = 291 provided evidence for oxidative dealkylation of the lower substituted phenylalkylamine moiety (M = 164 Da less than verapamil). Fragmentation in the MS²-mode (see Table 1) produced two abundant ions of m/z = 248 (loss of the isopropyl group) and m/z = 260 (N-C cleavage of the higher substituted moiety), which supported the structure of M3 shown in Fig. 2.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 2.

Metabolic pathway of verapamil in primary human coronary endothelial cells.

Metabolite M6 (norverapamil). The protonated molecule (M+H)⁺ of m/z = 441 pointed to oxidative dealkylation. Fragmentation in the MS² mode (see Table 1) produced an abundant ion of m/z = 165 and additional ions at m/z = 398, 289, and 151, which are typical for oxidative N-demethylation. The prominent ion of m/z = 165 can be explained by N-alkyl cleavage with charge transfer, whereas the m/z = 289 evidenced C-C cleavage in the α-position and subsequent proton transfer. The fragment of m/z = 398 is formed by loss of isopropyl, m/z = 151, by C-C cleavage in the α-position to the nitrogen and charge transfer. The structure of metabolite M6 was confirmed with a synthetic reference compound.

Metabolite M7 (D-702). The protonated molecule (M + H)⁺ of m/z = 441 was produced by oxidative desmethylation of a methoxy group. Fragmentation in the MS² mode (see Table 1) produced an abundant ion of m/z = 291. The prominent ion at m/z = 291 stems from a N-alkyl cleavage with proton transfer. The other fragment ions of m/z = 303 (C-C cleavage in α-position to the nitrogen), m/z = 248 and 260 (loss of the isopropyl group or methylamine from m/z 291) confirmed O-demethylation in position 21 or 23. It was shown previously that O-demethylation of D-702 occurred in position 21 (Nelson et al., 1988).

Metabolite M9 (D-703). The (M + H)⁺ of m/z = 441 is the result of oxidative dealkylation, e.g., loss of a methyl group. Fragmentation in the MS² mode (see Table 1) led to abundant ions of m/z = 289, 165, and 151, which pointed to oxidative demethylation in position C-31 or C-33. The prominent ions of m/z = 289 and 151 are C-C cleavages in the α-position to the nitrogen (the former accompanied by proton transfer). The ions of m/z = 165 are in support of oxidative demethylation at position 31 or 33. It was shown previously that O-demethylation of D-703 occurs in position 31 (Nelson et al., 1988). The structure of M9 was confirmed with a synthetic reference compound.

Discussion

The endothelium consists of about 10 × 10¹² cells and weights about 1.5 kg. Little is known about its ability to oxidize drugs. The few investigations published so far focused on the bioactivation of endogenous compounds, and this included studies with arginine (Palmer et al., 1987) and arachidonic acid (Fisslthaler et al., 1999). Additionally, Salvemini et al. (1993) showed endothelial cells to metabolize glyceryl trinitrate to nitric oxide, and Brittebo and Brandt (1994) demonstrated metabolic activation of the food mutagen 3-amino-1,4-dimethyl-5H-pyrido-[4,3-b]indole in cultures of endothelial cells of mice.

Recently, we showed rat aortic endothelial cells to express several genes that code for drug-metabolizing enzymes (Thum et al., 2000), but there is limited information on the expression and activity of P450 monooxygenases in human endothelium. We now demonstrate human coronary arterial endothelial cells to contribute to verapamil's bio-transformation and link gene and protein expression of two major cytochrome P450 monooxygenases, viz. CYP2C8 and CYP2E1, to verapamil's disposition. Importantly, we demonstrate the metabolism of verapamil to proceed via N-desalkylation, i.e., production of D-617 and norverapamil. These metabolites were previously identified in assays with human or rat liver microsomes (Nelson et al., 1988). The latter investigators also reported that less than 3% of verapamil is excreted as unchanged drug in humans. It is well established that verapamil undergoes extensive first-pass metabolism (McEvoy, 2001). Its intrinsic hepatic clearance was also studied (Iwatsubo et al., 1997), and from the above-cited studies, we conclude that human coronary arterial endothelial cells may only contribute to about 3% of the overall metabolism of verapamil. Consequently, it is unlikely that coronary arterial endothelial cells play a major role in the overall systemic biotransformation of verapamil, but tissue-specific metabolism of verapamil may be important for local drug response. Further studies should investigate the intra- and extracellular distribution of metabolites.

Tracy et al. (1999) demonstrated CYP3A4, CYP3A5, and CYP2C8 to be key players in the metabolism of verapamil, and our observation of CYP2C8 and CYP2E1 to be expressed in cultures of primary human coronary arterial endothelial cells fits well to the overall production of metabolites reported in this study. Nonetheless, expression of CYP3A4 and CYP3A5 was below the limit of detection; this illustrates, further, tissue specificity in the expression of isoforms of P450 monooxygenases and their contribution toward the oxidation of drugs and other xenobiotics. Some of the P450 isoforms identified by us are key players in the metabolic activation of polycyclic aromatic hydrocarbons and certain components of cigarette smoke. It will therefore be of interest to investigate, further, whether activity of these isoforms (CYP1A1, CYP2A6/7, CYP2E1) can be causally related to the inflammatory reactions and cardiovascular disease frequently observed in exposed animals (Zhang et al., 2001). In addition, CYP2C8 is highly implicated in the metabolic turnover and production of vasodilative agents including 11,12-epoxyeicosatrienic acid, and further studies are now on the way to determining the relevance of CYP2C8 genetic polymorphisms in the overall control of vascular tonus. Next to verapamil, other cardiovascular drugs are substrates for endothelial CYP2E1 or CYP2C, including the β-blocking agent carvedilol (Oldham and Clarke, 1997) and the 3-hydroxy-3-methyl-glutaryl coenzyme A reductase inhibitor, fluvastatin (Scripture and Pieper, 2001).

In conclusion, endothelium is a metabolically active tissue and should be investigated, particularly when tissue-specific metabolism and/or metabolic inactivation is being considered.