INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

CYTOCHROME P-450 is one of three major groups of enzymes that can metabolize arachidonic acid to many biologically active eicosanoids (14, 19). Cytochrome P-450 metabolizes arachidonic acid to 5,6-, 8,9-, 11,12-, and 14,15-epoxyeicosatrienoic acids (EETs), their corresponding dihydroxyeicosatrienoic acids (DHETs), and several 5-hydroxyeicosatetraenoic acids (HETEs), including 19- and 20-HETE (7, 19). This enzymatic pathway was first described in the liver (6, 18); however, it is now clear that arachidonic acid can be metabolized by cytochrome P-450 in many tissues, including the pituitary gland, eye, kidney, adrenal gland, and blood vessels (11, 17, 22). The EETs have been shown to stimulate the release of insulin and glucagon from pancreatic islet cells (10), luteinizing hormone, growth hormone, oxytocin and vasopressin from the pituitary gland (20), somatostatin and luteinizing hormone-releasing hormone from the hypothalamus (5, 27), and catecholamines from the adrenal chromaffin cells (15). The EETs are synthesized by the vascular endothelium (25, 26), and they relax vascular smooth muscle by opening potassium channels and hyperpolarizing the smooth muscle cells (4, 13, 16). It has been proposed that the EETs represent endothelium-derived hyperpolarizing factors. Despite the importance of the EETs in a variety of physiological and pharmacological effects, there have been few methods developed to quantify EETs in biological samples. The most common method for determining EETs has been the method of gas chromatography-mass spectrometry (GC-MS) (8, 12, 26, 28-30). A negative ion chemical ionization GC-MS was used to measure the urinary excretion of the EETs and their hydrolysis products, the DHETs in normal subjects, pregnant women, and women with pregnancy-induced hypertension (8). EET excretion was elevated in pregnant women and further increased in women with pregnancy-induced hypertension. Under most circumstances, the EET production in biological systems is extremely low. We have developed a fluorescence polarization immunoassay to quantify 14,15-EET and 14,15-DHET in cell cultures (21). However, the method can detect only the 14,15-EET and 14,15-DHET regioisomers. Because of the need for a very sensitive and simplified method for measuring EETs, we have developed a microbore column high-performance liquid chromatography (HPLC) assay that utilizes fluorometric detection. The carboxylic acid group of the EETs was easily reacted with 2-(2,3-naphthalimino)ethyl trifluoromethanesulfonate (NT) to produce strong fluorescent derivatives (1, 3, 31). This method allows the measurement of endogenous EETs at very low concentrations in cell culture media.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Materials. Four regioisomers of EETs were synthesized and purified as previously described by Corey et al. (9) and modified by Rosolowsky and Campbell (25). NT was purchased from Molecular Probes, (Eugene, OR). Tridecanoic acid (TA), methacholine, and acetonitrile (<0.005% water) were obtained from Sigma Chemical (St. Louis, MO). C₁₈ Bond Elut solid-phase extraction (SPE) columns were purchased from Varian (Harbor City, CA). Bradykinin was obtained from Peninsula Laboratories (Belmont, CA). [³H₈]14,15-EET was purchased from NEN Life Science Products (Boston, MA). Tetraethylammonium carbonate (TEAC) was synthesized as previously described by Akasaka and co-workers (2). Other chemicals or reagents were of analytic or HPLC grades. Distilled deionized water was used throughout this study.

Culture and stimulation of bovine coronary artery endothelial cells. Bovine hearts were obtained from a local abattoir and used immediately. Endothelial cells (ECs) were isolated from bovine coronary arteries using a modification of the previously described technique (24). In general, the cells were grown in 75-cm² flasks and used within the third passage. The cultured media were removed from the flasks, and the cells were washed twice with protein-free culture media. HEPES buffer (15 ml) containing bradykinin (10⁶ M) or methacholine (10⁶ M) was then added and incubated at 37°C for 10 min. Cells incubated with 15 ml of HEPES buffer were used as the control. After the stimulation, miconazole (10⁵ M) was added and incubated for 10 min. The cells were then scraped and transferred to 50-ml tubes and sonicated briefly two times. A 50-µl aliquot of each sample was used to determine the protein concentration using Bio-Rad protein assay (Bio-Rad, Hercules, CA). The internal standard, 30 µl of 10⁶ M TA, was added to the remaining portions of the samples and mixed. Ethanol (5 ml) was then added to the samples, mixed, and centrifuged at 1,500 rpm for 3 min. The supernatants were transferred to reaction tubes and frozen at 80°C or extracted.

Solid-phase extraction. The frozen samples were allowed to thaw on ice and then applied to the C₁₈ Bond Elut SPE columns that had been preconditioned with 5 ml of ethanol and 15 ml of water. The columns were washed with 20 ml of water and allowed to run dry. The fatty acid metabolites were then eluted from the column with 5 ml of ethyl acetate. The ethyl acetate layer was removed from the water layer at the bottom of the reaction tubes. The water layer was then extracted twice with 1 ml of ethyl acetate. The ethyl acetate portions were combined for each sample and dried under the stream of nitrogen gas.

Fluorescent derivatization. The standards or samples were dissolved in 35 µl of acetonitrile. Ten microliters of freshly prepared TEAC (10³ M) in acetonitrile was added to the solution and vortexed lightly. Freshly prepared NT (5 µl, 10³ M) in acetonitrile was then added, vortexed lightly, purged with nitrogen gas, and capped. The reaction tubes were placed in the dark at room temperature for various times in the time course study or for 15 min in the remaining experiments. This reaction is very sensitive to moisture. Therefore, it must be performed in a dry environment. After the reaction, the samples were transferred to the inserts in the sample vials. They were then frozen at 80°C or analyzed by HPLC.

HPLC. The EET standards and samples were analyzed on the Hewlett-Packard 1090 Series II liquid chromatograph (Hewlett-Packard, Palo Alto, CA) using an autosampler. The typical injection volume was 2 µl. A microbore column Luna C₁₈ (2) (250 × 1.0 mm, 5 µm) (Phenomenex, Torrance, CA) with water-acetonitrile mobile phase was used for separation. The gradient started from 50% A (acetonitrile) and 50% B (water) and linearly increased to 75% A over 30 min, 75% A to 85% A over 70 min, 85% A to 100% A over 20 min, and held for 20 min. The flow rate was 60 µl/min. The fluorescence was detected with the Hewlett-Packard 1046A programmable fluorescence detector at the excitation wavelength 259 nm and the emission wavelength 395 nm with a 370-nm long-pass filter. The chromatograms were recorded and analyzed by a Hewlett-Packard Chem Station. A separate mobile phase system was also used to verify the identities of chromatographic peaks in the samples. Therefore, some samples were reanalyzed using an isocratic mobile phase of 70% A and 30% B.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Sample extraction and recovery. The extraction recovery of EETs using the C₁₈ SPE columns was investigated by extracting [³H₈]14,15-EET in HEPES buffer at various percentages of ethanol. The EET recoveries of 51%, 95%, and 73% were obtained at 15%, 25%, and 35% of ethanol content, respectively. Therefore, 25% ethanol was used in this study to optimize the recovery of EETs. Adjusting the ethanol concentration allowed good extraction of the EETs without the need for acidifying the sample thus reducing the hydrolysis of EETs to DHETs.

Fluorescent derivatization and reaction product identification. The reaction of EETs with NT using TEAC catalyst is shown in Fig. 1A. This reaction is extremely sensitive to moisture. The reagents and solvents must be moisture free, and the reaction must be performed in a dry environment. The reaction of NT with EETs gave the highest yield when the concentration of TEAC was two times greater than NT. This result was similar to the labeling of carboxylic acids with 2-(2,3-anthracenedicarboximido)ethyl trifluoromethanesulfonate as previously described (1). The ratio of TEAC to NT was kept at 2: 1 throughout these studies.

View larger version (12K): [in this window] [in a new window]   Fig. 1.   A: schematic diagram indicating the reaction of 14,15-epoxyeicosatrienoic acid (EET) and 2-(2,3-naphthalimino)ethyl trifluoromethanesulfonate (NT) using tetraethylammonium carbonate (TEAC) as a catalyst in acetonitrile at room temperature. B: positive ion mass spectrum of the reaction product, 14,15-EET-NT, using an atmospheric pressure chemical ionization liquid chromatography-ion trap mass spectrometry. Relative abundance of the ions is plotted against the mass-to-charge ratio (m/z).

View larger version (16K): [in this window] [in a new window]   Fig. 2.   Peak areas for 14,15-EET-NT with various reaction times. The reaction was performed with 100 pg of 14,15-EET in 35 µl of acetonitrile, 5 µl of NT (103 M), and 10 µl of TEAC (103 M) at room temperature.

Standard curve and detection limit. Typical standard curves for the four EET regioisomers are shown in Fig. 3. The curves of regioisomeric EETs are identical and linear from 5 to 200 pg. The limit of detection (signal-to-noise ratio = 3) of 2 pg was achieved for all four regioisomeric EETs. These measurements were separately performed with each EET standard. The 8,9-EET-NT and 5,6-EET-NT were not resolved in this study and were analyzed as the mixture of these two regioisomers.

View larger version (18K): [in this window] [in a new window]   Fig. 3.   Typical standard curves of 5,6-, 8,9-, 11,12-, and 14,15-EETs. Peak areas of EET-NTs were normalized to the area of the tridecanoic acid (TA) internal standard and plotted as a function of the amount of EET per injection.

HPLC separation. Separation of EETs and TA from more abundant contaminating fatty acids was achieved by using a gradient of acetonitrile and water. The fluorescent products of 14,15-EET, 11,12-EET, and TA were well resolved from the products of 5,6-EET and 8,9-EET. Table 1 shows the retention times of NT products of these four regioisomers of EET and TA. This HPLC system provides an ability to identify and quantify 14,15-EET, 11,12-EET, and the mixture of 8,9-EET and 5,6-EET. The EET concentrations and their identities were obtained by comparing their peak areas and their retention times to the known standards. In some cases, the identities of the peaks were also confirmed by adding known EET standards to the samples before extraction. Figure 4 shows the chromatograms of four regioisomers of EET-NTs (Fig. 4A) and the methacholine-stimulated sample-NT (Fig. 4B). The sample contained a peak that matched the 14,15-EET-NT at 70.42 min and a peak that matched the TA-NT at 92.03 min. It also had a peak at the similar retention time of 8,9- and 5,6-EET-NT (77.39 min). However, the control cells (without stimulation) also had the peak at 77.39 min.

View larger version (27K): [in this window] [in a new window]   Fig. 4.   Reverse-phase chromatograms of the 4 EET standards and a sample from endothelial cells stimulated with methacholine. The separation was performed on a C18 microbore column with water-acetonitrile mobile phase. A: 4 regioisomeric EET-NT peaks and TA-NT peak. B: sample with TA internal standard that has been reacted with NT. There are peaks that have identical retention times as 14,15-EET-NT (at 70.42 min) and TA-NT (at 92.03 min). The sample also contained an apparent peak with identical retention time of 8,9-EET-NT and 5,6-EET-NT at 77.39 min. C: isocratic chromatogram of the 4 EET standards. D: isocratic chromatogram of the same sample as in B. The separation was performed with the 70% acetonitrile in water mobile phase. The apparent peak with the same retention time of 8,9-EET-NT and 5,6-EET-NT in B did not comigrate with these EETs in the isocratic system. The sample peak eluted at 69.56 min compared with 67.02 min for 8,9- and 5,6-EET-NT. These peaks in the samples represent fatty acid interference. *Peaks of interest.

Production of EETs by endothelial cells. On stimulation of ECs with either bradykinin or methacholine, a marked increase in EETs was observed. Figure 5 shows the production of 14,15-EET by ECs stimulated with bradykinin and methacholine. The control ECs produced only small amounts of EETs. The major EET was identified as 14,15-EET. Other isomers of EET were present at a concentration too low for quantitation. Both bradykinin and methacholine at 10⁶ M stimulated 14,15-EET production. Methacholine increased the release of 14,15-EET by fivefold (P < 0.01), whereas bradykinin caused a twofold increase.

View larger version (21K): [in this window] [in a new window]   Fig. 5.   Production of endogenous EETs by bovine coronary artery endothelial cells. Cells were stimulated with either bradykinin (BK, 106 M) or methacholine (MeCh, 106 M) for 10 min at 37°C. The major EET product was identified as 14,15-EET. Each value represents the mean ± SE (n = 5-6 samples).

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The amount of ethanol in the sample before the extraction with SPE columns was critical for the recoveries of both EETs and TA. This finding is similar to previous reports for other eicosanoids (23). Too low an amount of ethanol may not completely dissolve EETs and too high an amount of ethanol may prevent adsorption of EETs to the SPE columns. The 25% ethanol content was used for a maximal recovery of EETs with a lower recovery of the more hydrophobic TA.

NT was chosen as the fluorescent conjugating probe because its reaction products have a relatively high stability, a high molar absorptivity, and a strong fluorescence. In addition, it has a short excitation wavelength (259 nm) and a long emission wavelength (395 nm). This allowed the use of wider detector slits to obtain higher sensitivity. TEAC was used as a catalyst as suggested by Akasaka and co-workers (1) for conjugating fatty acids with 2-(2,3-anthracenedicarboximido)ethyl trifluoromethanesulfonate. Because the reaction is moisture sensitive, it must be performed in a dry environment. The NT and TEAC reagents should be freshly prepared. Prolonged storage of these reagents resulted in the breakdown of NT and lower yields of the reaction products. The reacted samples should be kept at 80°C until analysis. A long storage of samples at room temperature resulted in more fluorescence background peaks.

The derivatization reaction took place very rapidly, and the maximal fluorescence signal was obtained after 10 min. Therefore, the samples were reacted with NT for 15 min and frozen at 80°C until analysis. There was no significant difference in the extent of derivatization between the four regioisomers of EET.

A C₁₈ reverse-phase microbore column with water-acetonitrile mobile phase greatly improved the resolution and sensitivity of these compounds. It is difficult to resolve the four regioisomers of EETs from each other and from other more abundant fatty acids using an analytic (4.6 × 250 mm) HPLC column. Despite the high resolution of the microbore column technique, contaminating fatty acids were detected in some samples that comigrated with 5,6- and 8,9-EET; thus, a second mobile phase was required to confirm the identities of the peaks.

The control ECs produced very low concentrations of EETs. However, the EET production significantly increased when ECs were stimulated with methacholine. Bradykinin also stimulated EET production but was less effective than methacholine. This may be due to the metabolism of bradykinin by ECs. The major EET released from ECs was identified as 14,15-EET, whereas other regioisomeric EETS were probably produced at too low concentrations for detection.

In conclusion, we demonstrate that very low concentrations of EETs can be determined in biological samples by using a microcolumn HPLC and fluorescence detection. The microbore column improves the sensitivity and separation of fatty acids. Despite the high interference from more abundant fatty acids, this method provides good separation in the chromatographic range for the EET peaks. This high resolution cannot be obtained from an analytic column. Some disadvantages are that the microcolumn requires a good pumping system at a low flow rate and long run time for reproducible retention time and sample resolution. However, it is a powerful tool for determination of the regulation of EET release from cells and tissues that other methods cannot achieve.