Erythrocytes serve as reservoirs for cis- and trans-epoxyeicosatrienoic acids (EETs). Incubation of rat red blood cells (RBCs) with cis- and trans-EETs produces threo- and erythro-dihydroxyeicosatrienoic acids, respectively. The Vmax of EET hydrolysis by rat intact RBCs (2.35 ± 0.24 pmol/min/108 RBCs for 14,15-trans-EET) decreased by approximately 2 to 3-fold sequentially from 14,15-, 11,12- to 8,9-EETs for both cis- and trans-isomers. The Vmax of trans-EET hydrolysis by RBCs is approximately 2 to 3 times that of the corresponding cis-EETs. Incubation of EETs with recombinant murine soluble epoxide hydrolase (sEH) yielded the same geometric and regio preferences of EET hydrolysis as with rat intact RBCs. The principal epoxide hydrolase activity for EET hydrolysis (approximately 90%) is present in the erythrocyte cytosol. Western blots of sEH suggested a concentration of sEH protein to be approximately 2 μg/mg protein or 0.4 μg/109 RBCs. The apparent Km values of EETs were between 1 and 2 μM, close to the Km for purified sEH as reported. Erythrocyte hydration of cis- and trans-EETs was blocked by sEH inhibitors, 1,3-dicyclohexylurea and 4-[4-(3-adamantan-1-ylureido)cyclohexyloxy]benzoic acid. Erythrocyte sEH activity was inhibited more than 80% by 0.2% bovine serum albumin in the buffer. Preferred hydrolysis of 14,15-EETs and trans-epoxides characterizes sEH activity in RBCs that regulates the hydrolysis and release of cis- and trans-EETs in the circulation. Inhibition of sEH has produced antihypertensive and antiinflammatory effects. Because plasma trans-EETs would increase more than cis-EETs with sEH inhibition, the potential roles of trans-EETs and erythrocyte sEH in terms of circulatory regulation deserve attention.
Epoxyeicosatrienoic acids (EETs) are arachidonic acid (AA)-derived vasoactive, antithrombotic, antiproliferative, and anti-inflammatory lipid mediators (Larsen et al., 2006; Fleming, 2007) that are inactivated by soluble epoxide hydrolase (sEH), producing dihydroxyeicosatrienoic acids (DHETs) (Inceoglu et al., 2007). Formation of EETs is catalyzed by cytochrome P450 epoxygenases (Capdevila et al., 1991), hemoglobin (Jiang et al., 2007), and lipid peroxidation (Nakamura et al., 1997). The substrate preferences for cis- and trans-epoxides by sEH vary according to cell type (Seidegård et al., 1984) and species (Morisseau and Hammock, 2005). sEH is a therapeutic target for control of blood pressure (Sinal et al., 2000; Loch et al., 2007; Zhang et al., 2007), vascular inflammation (Davis et al., 2002; Schmelzer et al., 2005; Smith et al., 2005), and cancer progression (Morisseau and Hammock, 2005), as well as for cardiac (Seubert et al., 2006) and renal protection (Imig, 2005) by augmenting the levels of EETs in vivo. Inhibition of sEH also increases EET incorporation into phospholipids, “thereby modulating endothelial function in the coronary vasculature” (Weintraub et al., 1999). Increases in sEH activity have been associated with diabetes (Rodriguez and Clare-Salzler, 2006) and hypertension (Imig et al., 2002; Ai et al., 2007).
Epoxide hydrolases convert cis-EETs to threo-DHETs and trans-EETs to erythro-DHETs (Fig. 1). Cytochrome P450 epoxygenases generate cis-EETs, whereas EETs in vivo include both cis- and trans-EETs (Jiang et al., 2005). The 5,6-erythro-DHET was more potent in dilating preconstricted renal interlobar arteries than the 5,6-cis-EET, whereas, unlike cis- and trans-EETs, it did not inhibit platelet aggregation (Jiang et al., 2004). Red blood cells (RBCs) are reservoirs for both cis- and trans-EETs that can be released by ATP stimulation of erythrocyte P2X7 receptors (Jiang et al., 2007). The hydrolysis of EETs by erythrocytes may represent an important mechanism involved in circulatory regulation.
Human erythrocytes were originally discovered to possess a sEH that hydrates leukotriene A4 (LTA4) and, to a much lesser degree, hydrates cis-EETs (McGee and Fitzpatrick, 1985). Because epoxide hydrolases occur in multiple forms and possess individual substrate specificities (Ota and Hammock, 1980; Thomas et al., 1990), an additional form of sEH for the hydrolysis of EETs in RBCs could not be excluded (McGee and Fitzpatrick, 1985). It was a surprise that human erythrocyte cytosol was not found to hydrolyze stilbene oxide in a previous study, whereas the erythrocyte membrane has epoxide hydrolase activities at less than 1% of that of the granulocyte in human blood (Seidegård et al., 1984). In our studies of erythrocyte-derived EETs (Jiang et al., 2007), we found significant hydrolysis of cis-/trans-EETs by rat RBCs, which gives rise to questions regarding the nature of the epoxide hydrolase present in erythrocytes responsible for cis-/trans-EET hydrolysis.
To investigate the existence of sEH and to compare hydrolysis of cis- and trans-EETs in RBCs, we analyzed kinetics of rat RBCs in hydrolyzing cis- and trans-EETs and compared the regioselectivity and geometric selectivity with recombinant murine sEH. The comparable Km and geometric selectivity between rat erythrocyte cytosol and purified mouse liver sEH in hydrolyzing EETs suggested the presence of a typical sEH in RBCs as in hepatocytes. Rat sEH demonstrated preferred hydrolysis for trans- over cis-EETs with similar regioselectivity for both cis- and trans-EETs in a decreasing order from 14,15-, 11,12-, 8,9- to 5,6-EETs.
Materials and Methods
Animals. All animal procedures were approved by the Institutional Animal Care and Use Committee of New York Medical College and conformed to the Institute of Laboratory Animal Resources (1996). Eight-week-old male Sprague-Dawley rats were purchased from Charles River Laboratories (Wilmington, MA). Rats were maintained at 22°C with alternate light/dark cycles and fed ad libitum with standard rat chow and water.
Blood Cell Preparation. Sprague-Dawley rats (9–12 weeks old) were anesthetized with pentobarbital (65 mg/kg i.p.), and 10 ml of blood was drawn from the inferior vena cava after midline laparotomy using heparin-rinsed syringes and transferred to Vacuette heparin tubes (Thermo Fisher Scientific, Waltham, MA). After inverting four to six times, the blood was centrifuged at 800g at 4°C for 10 min. The supernatant was removed by aspiration, and the buffy layer was collected in some experiments. Packed RBCs were washed four times in an ice-cold physiological salt solution (PSS) with centrifugation at 400g for 10 min, and the washing buffer and any residual buffy layer were discarded after each wash. The PSS contained 5.0 mM KCl, 1.2 mM CaCl2, 1.2 mM MgSO4, 128 mM NaCl, 20 mM HEPES, 5 mM dextrose, and 2 mg/ml bovine serum albumin (BSA), and pH was adjusted to 7.4. Washed RBCs were suspended in PSS, examined, and counted using hemacytometers under the microscope.
Incubation and Eicosanoid Extraction. Given a total of approximately 40 ng/ml EETs in rat plasma, less than 5% is free, whereas over 95% is esterified in phospholipids (Karara et al., 1992; Jiang et al., 2005). To test hydrolysis of EETs by RBCs, 6 ng of cis- and/or trans-EETs was added to 2 ml of prewarmed RBCs (2 × 109 RBCs/ml) in PSS at 37°C for up to 30 min with shaking around a 3-mm orbit at 600 rpm in a VWR Incubating Mini Shaker (VWR, West Chester, PA). The nonenzymatic conversion of 5,6-EETs to DHETs was tested with control incubations using 2 ml of PSS buffer under the same conditions.
To determine the proportion of EET hydrolysis by RBCs in blood, 100 μl of plasma, the buffy layer diluted in phosphate-buffered saline (PBS), and diluted RBCs in PBS were incubated with trans-EETs (1 μM) for 5 min at 37°C, respectively. EET and DHET extraction for electrospray ionization (ESI)-liquid chromatography (LC)/mass spectrometry (MS) analyses were performed as described previously (Jiang et al., 2004).
Separation of Membrane and Soluble Fractions. Washed rat erythrocytes (4 ml) were lysed hypotonically in ice-cold sterile water and vortexed for 2 min. After restoration to isotonic buffer conditions, the crude lysate was centrifuged at 100,000g for 1 h with a Beckman Ultracentrifuge (Beckman Coulter, Fullerton, CA) at 4°C. The supernatant was filtered through a 0.45-μm nylon syringe filter to obtain the cytosol fraction of RBCs. Both the pellet and the cytosol were diluted in PSS buffer without BSA to correspond to a cellular concentration of 2 × 109 RBCs/ml.
Western Blots of Erythrocyte sEH. Total protein concentration was quantified with the Pierce BCA assay (Pierce, Rockford, IL), using Fraction V BSA as the calibrating standard. After thawing the frozen erythrocyte cytosol, sEH activity was measured using racemic [3H]trans-1,3-diphenylpropene oxide as described previously (Morisseau and Hammock, 2007). For each cytosolic sample, 50 μg of protein was loaded on a 12% SDS-polyacrylamide gel electrophoresis. The separated proteins were then transferred onto a polyvinylidene difluoride membrane, and sEH was detected using a rabbit serum raised against recombinant mouse sEH (Davis et al., 2002). Detection was done using a goat antibody raised against rabbit IgG labeled with horseradish peroxidase. Bands were revealed using the ECL kit (Amersham, Piscataway, NJ). Recombinant purified mouse sEH (500 ng) was used as a positive control.
Kinetic Studies of EET Hydrolysis. Because BSA was found to have inhibitory effects on EET hydrolysis, PSS buffer without BSA was used in the last wash of rat RBCs, and RBCs were diluted to 2 × 109 RBCs/ml in PSS without BSA. To achieve substrate saturation, 14,15-, 11,12-, and 8,9-cis- or trans-EETs (0.1–4 μM) in 50 μl of RBCs or cellular fractions were hydrolyzed at 37°C for 10 min. Endogenous EETs and DHETs released from 1 × 108 RBCs in a buffer without BSA were negligible (Jiang et al., 2007), which was confirmed by the insignificant alternative EET/DHET peaks in LC/MS analyses. Freshly prepared rat intact RBCs were used for the incubations because freeze-thawing of RBCs reduced EET hydrolysis by 50% as tested. After incubation, 2 ml of ice-cold ethyl acetate/hexanes (1:1) was added to each tube to stop the reaction, the mixtures of which were extracted after adding internal standards of EET-d8 and DHET-d8 and adjusting pH to 4 with 10% acetic acid. The kinetics of 5,6-EET were not measured because of nonenzymatic conversion to 5,6-DHET (Jiang et al., 2004).
IC50 Assay. IC50 of sEH inhibitors were determined based on National Institutes of Health Chemical Genomics Center enzymatic assay guidance manual. Rat erythrocyte cytosol (corresponding to 5 × 108 RBCs) was incubated with cis- and trans-4-[4-(3-adamantan-1-ylureido)cyclohexyloxy]benzoic acid (AUCB) (Hwang et al., 2007) as well as 1,3-dicyclohexylurea (DCU), respectively, for 5 min in PBS, pH 7.4, at 37°C before substrates of 1 μM 14,15-, 11,12-, and 8,9-trans-EETs were added. Activity of sEH was assessed by analyzing erythro-DHETs formed with LC/MS after ethyl acetate extraction. Assays were performed in triplicate. IC50 and S.E. were determined by regression of at least eight data with a minimum of two points in the linear region of the curve on either side of the IC50.
ESI LC/Tandem Mass Spectrometry Analyses. ESI LC/tandem mass spectrometry (MS/MS) analyses of EETs and DHETs were carried out as described previously (Jiang et al., 2005). In brief, a Finnigan LCQ Advantage quadrupole ion-trap mass spectrometer (Thermo Fisher Scientific) equipped with ESI source run by Xcalibur software was used. Reversed-phase high-performance liquid chromatography (HPLC) was run with a Luna C18(2) 250 × 2.0-mm column (Phenomenex, Torrance, CA) maintained at 30°C with an isocratic eluent of acetonitrile/water/methanol/acetic acid (60:30:10:0.05) at a flow rate of 0.30 ml/min. For EET hydrolysis kinetic studies, the isocratic eluent was kept for 5 min and followed by a gradient to a final composition of acetonitrile/water/methanol/acetic acid (75:15: 10:0.05) in 15 min. ESI was carried out at an ion transfer tube temperature of 260°C, a spray voltage of 4.5 kV, a sheath gas flow of 34 units, and an auxiliary gas flow of 20 units (units refer to arbitrary values set by the LCQ software). MS/MS breakdown for m/z 337 was at an energy level of 30% set by the instrument, and a 7-point Gaussian smoothing was applied in the mass data processing.
Statistical Analysis. Results are presented as mean ± S.E.M. Parameters of Vmax, Km, and IC50 were analyzed with GraphPad Prism 5 software (GraphPad Software Inc., San Diego, CA) using Michaelis-Menten kinetics. One-way analysis of variance followed by a Bonferroni test for selected groups were used to analyze for differences. A P value less than 0.05 was regarded as statistically significant.
Reagents. Standard cis-EETs and LTA4 hydrolase were purchased from Cayman Chemical (Ann Arbor, MI). Standard racemic trans-EETs were synthesized by Dr. John R. Falck (Falck et al., 2003; Jiang et al., 2004). Potent sEH inhibitors, cis- and trans-AUCB, were prepared as described previously (Hwang et al., 2007). SC22716, DCU, and AA were ordered from Sigma-Aldrich (St. Louis, MO). AA was used immediately after HPLC purification. Fatty acid-free BSA and HPLC grade organic solvents were obtained from Thermo Fisher Scientific, and EET-d8 standards were from Biomol Research Laboratories (Plymouth Meeting, PA). Recombinant murine sEH was produced in a baculovirus expression system (Grant et al., 1993) and purified by affinity chromatography. The preparations were at least 97% pure as judged by SDS-polyacrylamide gel electrophoresis and scanning densitometry. No detectable esterase or glutathione transferase activity, which can interfere with this sEH assay, was observed.
Hydrolysis of EETs by Rat RBCs in PSS. Incubation of 6 ng of each cis- and trans-EETs with 4 × 109 rat RBCs in 2 ml of PSS at 37°C for 2 min revealed conversion of EETs to corresponding DHETs (Fig. 2). LC/MS chromatography resulted in complete separation of individual EETs and DHETs when cis- or trans-EETs were incubated separately (Fig. 2, A and B). cis-EETs were converted to threo-DHETs (Fig. 2A), whereas trans-EETs were converted to erythro-DHETs (Fig. 2B). When eight EET isomers were incubated together with RBCs, the chromatogram was similar to a combination of individual incubations (Fig. 2C), indicating no inhibition of hydrolysis by EET isomers at 10 nM. The hydrolysis of cis- and trans-EETs by rat RBCs can be blocked by 10 μM DCU (Fig. 2D), a traditional sEH inhibitor, as well as by 100 nM cis- and trans-AUCB (similar to Fig. 2D), novel potent sEH inhibitors. A cell-permeable LTA4 hydrolase inhibitor, SC22716, did not inhibit EET hydrolysis by rat RBCs, and commercial LTA4 hydrolase (25 μg/ml) did not hydrolyze cis- or trans-EETs when tested under similar conditions of incubation.
Selectivity ofcis- andtrans-EET Hydrolysis by Recombinant Murine sEH. When incubating recombinant murine sEH (0.5 μg/ml) with cis- or trans-EETs, preferred hydrolysis for trans- over cis-EETs and for 14,15-EETs over other regioisomers similar to the hydrolysis of RBCs (Fig. 3) was also observed. Comparative hydration rates are 121.4 ± 18.2, 55.6 ± 9.6, and 29.6 ± 4.7 nmol/min/mg protein for 14,15-, 11,12-, and 8,9-cis-EET, respectively, and 230.6 ± 35.1, 174.3 ± 31.4, and 85.8 ± 13.7 nmol/min/mg protein for 14,15-, 11,12-, and 8,9-trans-EET, respectively, for 1 μM EETs in PBS at 37°C as averaged from at least three incubations.
Active Formation and Hydrolysis of EETs by Rat RBCs. The presence of erythro-DHETs in vivo was expected as trans-EETs were identified in plasma and phospholipids (Jiang et al., 2005). LC/MS/MS spectra of erythro-DHETs produced with the hydrolysis of trans-EETs by RBCs (Fig. 4) provided the starting point for individual identification of erythro-DHETs. Incubation of AA (20 μM) with 4 × 109 RBCs in 2 ml of PSS produced both threo- and erythro-DHETs as well as hydroxyeicosatetraenoic acids (HETEs) and cis-/trans-EETs (Fig. 5), suggesting that erythrocytes function as a potential source of AA-derived eicosanoids in plasma. Identification of individual erythro-DHETs by LC/MS/MS is shown in Fig. 5, whereas identification of HETEs and cis-/trans-EETs was as reported previously (Jiang et al., 2007). Breakdown of m/z 337 to m/z 237 is ideal for the MS/MS identification of 14,15-DHETs (Fig. 4). The selection of m/z 207 for the identification of 14,15-DHETs (Fig. 5) is based on the observation that MS/MS product ions with weaker abundances, such as m/z 237 for 14,15-DHETs and m/z 207 for 11,12-DHETs, do not show up in the analysis of barely detectable amounts of DHETs. Using the same method of analysis, both free erythro- and threo-DHETs were identified in rat plasma with the total amount at 1.0 ± 0.2 ng/ml; erythro-DHETs accounted for approximately one fourth of the total free DHETs (n = 3).
EET Hydrolysis by Rat RBCs Compared with Other Blood Fractions. Comparison of EET hydrolysis by rat plasma, buffy layer, and RBCs revealed predominant EET hydrolysis by RBCs in blood. EETs were not hydrolyzed by plasma, particularly when there was no hemolysis. The buffy layer containing most of the leukocytes and platelets in blood hydrolyzed EETs, approximating 9.1 ± 2.2% (n = 4) of the hydrolysis by rat RBCs as calculated from the hydrolysis of 14,15-, 11,12-, and 8,9-cis- or trans-EETs (1 μM) for 5 min at 37°C.
Localization of Epoxide Hydrolases in Rat RBCs. Localizing the specific epoxide hydrolase for EET hydrolysis in rat RBCs disclosed its major existence in the cytosol of RBCs. The rat RBC membrane residue accounted for 11.9 ± 5.1 and 9.4 ± 4.5% (n = 4) of the total hydrolysis of trans- and cis-EETs by RBCs, respectively, as identified by the conversion of 1 μM 14,15-EETs to their respective DHETs. When 0.2% BSA was included in the buffer for the study, the specific activity of rat RBCs or cytosol to hydrolyze EETs at 1 μM fell to an approximately 10 to 20% level (n = 4), indicating considerable inhibition of EET hydrolysis by BSA binding of EETs.
Protein Analysis of Erythrocyte sEH. Total protein concentration of the prepared cytosol of rat RBCs corresponding to 2 or 4 × 109 RBCs/ml was 4.8 ± 0.6 mg/109 RBCs (n = 6), indicating less than 50% recovery of hemoglobin in the prepared cytosol after centrifugation and filtration. Whereas a dense band was detected for the positive control in Western blots of sEH (Fig. 6), weaker bands were observed for the samples, suggesting a concentration of sEH protein to be approximately 2 μg/mg protein or 0.4 μg/109 RBCs based on densitometry comparisons. The last rat erythrocyte cytosol (sample 3) was freshly prepared and frozen, whereas samples 1 and 2 had been stored and frozen at –20°C for 5 months, which may explain the lighter sEH band than the one observed for sample 3.
Kinetics of Erythrocyte sEH on EET Hydrolysis. Kinetic studies of the sEH activities of rat RBCs were tested using 50 μl of RBCs (2 × 109 RBCs/ml) in PSS without BSA for 0.1 to 4 μM 14,15-, 11,12-, and 8,9-cis- or trans-EETs (Fig. 7). The specific EET hydrolysis at 4 μM were excluded in the kinetic analysis (Table 1) because of distortion of the kinetic parameters caused by apparent inhibition of the hydrolysis of 8,9-EETs by 14,15-EETs that have a greater affinity for the sEH in RBCs. The Vmax of rat RBCs for trans-EET hydrolysis was greater by 3-fold or more than the Vmax for cis-EETs.
Erythrocyte EETs were mostly esterified in RBC membranes; the presence of cytosolic EETs and DHETs was minimal. Kinetic analyses of individual EETs were carried out using 50 μl of RBC cytosol corresponding to 1 × 108 RBCs to exclude factors affecting EET uptake and DHET release, as well as endogenous EETs. A more accurate Vmax and Km for the cis-/trans-EET hydrolysis was obtained using erythrocyte cytosol (Table 2) instead of intact RBCs. The exception is 14,15-cis-EET that does not conform to the Michaelis-Menten kinetics when the substrate concentration is at 4 μM, possibly revealing the participation of LTA4 hydrolase in hydrolyzing 14,15-cis-EET (McGee and Fitzpatrick, 1985).
IC50 of sEH Inhibitors. IC50 of cis-AUCB, trans-AUCB, and DCU on trans-EET hydrolysis by rat erythrocyte cytosol was determined as 12.5 ± 1.5, 5.1 ± 1.3, and 156.5 ± 1.3 nM, respectively (Fig. 8). Substrate selection and experimental conditions may have contributed to the different relative potencies of cis- and trans-AUCB compared with the IC50 reported for cis-AUCB (0.89 nM) and trans-AUCB (1.3 nM) (Hwang et al., 2007). The IC50 of DCU is comparable with the reported IC50 of 160 nM on recombinant human sEH and 90 nM on recombinant murine sEH (McElroy et al., 2003). However, nanosuspension of DCU greatly enhanced the potency of DCU (Ghosh et al., 2008).
The preferred hydrolysis of 14,15-EETs and trans-epoxides characterizes sEH activities in rat RBCs. Incubation of rat RBCs with 10 nM EETs in PSS demonstrated that hydration of trans-EETs by RBCs was at rates that differ by more than three folds from hydration rates of the corresponding cis-EETs (Fig. 3). The rates of EET hydrolysis by rat RBCs decreased at approximately 2 to 3-fold sequentially from 14,15-, 11,12-, and 8,9- to 5,6-EETs for both the cis- and trans-isomers. The results are in accord with the report that 14,15-cis-EET was found to be hydrolyzed faster than other cis-regioisomers by sEH (Zeldin et al., 1995). 5,6-EETs are hydrolyzed the least by rat RBCs, consistent with the inability of the 5,6-EET to serve as a suitable substrate for liver sEH (Chacos et al., 1983).
The 3 to 5-fold greater specific hydrolysis of trans-relative to cis-EET by rat RBCs at 10 nM (Fig. 2 and 3) presumably is a combined result of sEH, membrane epoxide hydrolase, membrane transport, inhibition by albumin, as well as contributions of endogenous EETs. Nevertheless, the result may reflect de facto EET hydrolysis by RBCs in vivo. The RBC LTA4 hydrolase that has a Km of 20 μM for 14,15-cis-EET (McGee and Fitzpatrick, 1985) will probably not play a significant role in hydrolyzing EETs at low nanomolar concentrations. Hydrolysis of EETs by RBCs was immediate (Fig. 2), indicating rapid uptake of EETs and rapid release of DHETs by RBCs. The uptake of trans-EETs by RBCs may be faster than that of cis-EETs, considering that trans-EETs are more hydrophobic than cis-EETs as was also evidenced by the later elution of trans-EETs than cis-EETs, respectively, in reversed-phase HPLC separations (Fig. 2). We have reported that the cystic fibrosis transmembrane conductance regulator and pannexin-1 are involved in the secretion of EETs from rat RBCs (Jiang et al., 2007). Fatty acid-binding proteins are known to inhibit hydrolysis of EETs by sEH (Widstrom et al., 2003), as was manifested by inclusion of 0.2% fatty acid-free BSA in the incubations that resulted in over 80% inhibition of the rate of EET hydrolysis.
The Vmax for trans-EET hydrolysis is approximately 3-fold or greater than that of cis-EETs, respectively, when testing incubations of 0.1 to 2 μM of the three cis- or trans-EETs together with rat intact RBCs (Table 1). The diminution of the hydrolysis of 8,9-EETs when incubating the three cis- or trans-EETs together at 4 μM, respectively, suggested substrate saturation and greater affinity of 14,15- and 11,12-EETs than 8,9-EETs for the sEH, as was also confirmed by individual EET kinetics obtained with the erythrocyte cytosol (Table 2). Nonconformance to the Michaelis-Menten kinetics for the hydrolysis of 14,15-cis-EET at concentrations of 4 μM is probably caused by involvement of LTA4 hydrolase in the erythrocyte cytosol (McGee and Fitzpatrick, 1985).
The highly similar regio and geometric selectivity of EET hydrolysis by recombinant murine sEH and rat erythrocyte cytosol suggests either identity or close similarity of erythrocyte and hepatic sEH. Western blots of sEH suggested that a concentration of sEH protein is approximately 2 μg/mg protein or 0.4 μg/109 RBCs (Fig. 6). The apparent Km of 1 to 2 μM EETs for rat RBCs is comparable with the Km of 3 to 5 μM EETs for the purified mouse liver sEH (Zeldin et al., 1995). The apparent Km and Vmax of rat erythrocyte for the hydrolysis of cis-/trans-EETs approximate those of the sEH in human leukocytes for the hydrolysis of cis-stilbene oxide (Seidegård et al., 1984). Purification of enzymes involves procedures that can increase or decrease specific activities of an enzyme. Thus, it may not be appropriate to compare the Vmax of erythrocyte cytosol with Vmax of purified sEH.
The divergent abilities of LTA4 hydrolase to hydrolyze LTA4 and cis-EETs seem to preclude erythrocyte LTA4 hydrolase as an effective sEH that hydrolyzes cis-EETs (McGee and Fitzpatrick, 1985). This is consistent with the inability of the LTA4 hydrolase inhibitor, SC22716, to affect EET hydrolysis as well as the inability of commercial LTA4 hydrolase to hydrolyze EETs as tested in the present study. The LTA4 sEH purified from human erythrocytes hydrolyzes 14,15-cis-EET with a Km of 20 μM (McGee and Fitzpatrick, 1985), which is 20-fold greater than the Km of sEH for 14,15-cis-EET in our study. Despite LTA4 and trans-EETs sharing the trans-epoxide configuration, sEH demonstrated extraordinary specificity for substrate selection. The presence of sEH in RBCs is further supported by inhibition of EET hydrolysis by sEH inhibitors, cis- and trans-AUCB (Fig. 8).
This study mainly addressed the comparison of added cis- and trans-EET hydration by rat erythrocytes because comparable amounts of cis- and trans-EETs are present in plasma and erythrocyte phospholipids (Jiang et al., 2005). The conversion of AA to EETs and HETEs by rat RBCs (Fig. 5) is a function of hemoglobin-activating oxygen in a monooxygenase-like fashion (Starke et al., 1984). Peroxy radicals of AA may mediate the formation of more trans- than cis-EETs in RBCs (Jiang et al., 2004). EETs in vivo demonstrate chiral prevalences (Wei et al., 2006), and regio- and enantioselectivity have been identified for the enzymatic hydration of cis-EET enantiomers (Zeldin et al., 1993, 1995). Analysis of the chirality of trans-EET isomers and their enantioselectivity for hydration has yet to be carried out.
It is a challenge to estimate the relative contributions to the hydrolysis of EETs by RBCs and by specific organs. However, the role of erythrocyte sEH in the regulation of circulating EETs may be particularly significant when considering potential effects of EETs on the rheological and hemodynamic determinants of the circulation, such as in cardiovascular and hematological diseases, as well as physiological implications for regulating blood flow in the microcirculation. Release of EETs from RBCs into the circulation in response to ATP stimulation of the erythrocyte P2X7 receptor (Jiang et al., 2007) will contribute to activation of endothelial peroxisome proliferator-activated receptor γ transcription by EETs (Liu et al., 2005), which is magnified through elevating EETs at the blood-endothelial interface in response to inhibition of sEH in RBCs. Activation of peroxisome proliferator-activated receptor γ inhibits nuclear factor κB-mediated expression of adhesion molecules and endothelin that promotes vascular wall damage and atherogenesis (Liu et al., 2005). Furthermore, localization of sEH in the RBC and the increase of EETs resulting from its inhibition presumably contributes to elevating the positive effects of EETs on regional blood flows to a greater degree than sEH localized “in the smooth muscle layers of the arterial wall” (Yu et al., 2004).
This study revealed preferential hydrolysis of trans- over cis-EETs and the presence of sEH in rat RBCs. Angiotensin II up-regulates sEH expression in the vascular endothelium (Ai et al., 2007); sEH inhibition greatly lowered systolic blood pressure in angiotensin II-induced hypertensive rats (Imig et al., 2002; Ghosh et al., 2008). Consistent with the positive effects of EETs on the circulation, sEH has been proposed as a novel therapeutic target for control of blood pressure (Sinal et al., 2000), prevention of renal damage (Imig, 2005) and stroke (Zhang et al., 2007), as well as amelioration of inflammation (Schmelzer et al., 2005). Because plasma trans-EET levels would increase more than cis-EETs with sEH inhibition, the potential functional roles of trans-EETs and erythrocyte sEH in terms of contributing to the regulation of the circulation deserve attention.
We thank Gail Anderson for assistance in preparation of the manuscript, Dr. Sung Hee Hwang for preparation of AUCB, and Dr. John Quilley for assistance in revision of the manuscript.
This work was supported by grants from Philip Morris USA Inc. and Philip Morris International (to H.J.) and by National Institutes of Health Grants PPG 34300 (to J.C.M.) and HL-25394 (to J.C.M.). Support was also obtained from National Institute of Environmental Health Sciences Grant R37 ES02710 (to B.D.H.) and the National Institute of Environmental Health Sciences Superfund Basic Research Program (P42 ES004699; to B.D.H.).
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
ABBREVIATIONS: EET, epoxyeicosatrienoic acid; AA, arachidonic acid; sEH, soluble epoxide hydrolase; DHET, dihydroxyeicosatrienoic acid; RBC, red blood cell; LTA4, leukotriene A4; PSS, physiological salt solution; BSA, bovine serum albumin; PBS, phosphate-buffered saline; ESI, electrospray ionization; LC, liquid chromatography; MS, mass spectrometry; AUCB, 4-[4-(3-adamantan-1-ylureido)cyclohexyloxy]benzoic acid; DCU, 1,3-dicyclohexylurea; MS/MS, tandem mass spectrometry; HPLC, high-performance liquid chromatography; HETE, hydroxyeicosatetraenoic acid; SC27716, 1-[2-(4-phenylphenoxy)ethyl]pyrrolidine.
- The American Society for Pharmacology and Experimental Therapeutics