| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
INVITED REVIEW
Departments of 1Biochemistry, 2Internal Medicine, and 3Pediatrics, Carver College of Medicine, University of Iowa, Iowa City, Iowa
 |
ABSTRACT |
|---|
(PPAR) and PPAR
, suggesting that some functional effects may result from PPAR activation. EETs are metabolized primarily by conversion to dihydroxyeicosatrienoic acids (DHETs), a reaction catalyzed by soluble epoxide hydrolase (sEH). Many potentially beneficial actions of EETs are attenuated upon conversion to DHETs, which do not appear to be essential under routine conditions. Therefore, sEH is considered a potential therapeutic target for enhancing the beneficial functions of EETs. soluble epoxide hydrolase; eicosanoids; dihydroxyeicosatrienoic acids; cytochrome P-450; peroxisome proliferator-activated receptor
-oxidases (123). Early studies indicated that the EETs produce important biological effects, particularly in the vascular and renal systems (52, 68, 103), but there was only limited interest in these compounds until the middle 1990s, when they were shown to be synthesized in the endothelium and to function as an endothelium-dependent hyperpolarizing factor (EDHF) under certain conditions in the coronary circulation (7, 51). Subsequent studies indicated that deletion of soluble epoxide hydrolase (sEH), the enzyme that converts EETs to dihydroxyeicosatrienoic acids (DHETs), decreased blood pressure in male mice (143), and treatment with a selective sEH inhibitor decreased blood pressure in hypertensive rats (171). These results suggested that inhibition of EET conversion to DHET might be a new therapeutic approach for hypertension. Interest was further heightened by the observations that EETs have anti-inflammatory effects in the endothelium (120), stimulate angiogenesis (105, 115), and prevent arterial smooth muscle migration (151). These findings are described in detail in a number of recent reviews (77, 131, 172). Signal transduction pathways and transcriptional mechanisms involved in EET function have been identified (11, 56, 89, 150), and attempts are being made to isolate an EET membrane receptor that mediates these effects (144, 164). However, cells also rapidly take up EETs and incorporate them into phospholipids (4, 10, 149), suggesting the possibility of an intracellular mechanism of action. In this regard, heart fatty acid binding protein (H-FABP) binds EETs with Kd values that are only slightly higher than the Kd for arachidonic acid (161), implying that EETs may bind to other intracellular proteins, including transcription factors such as peroxisome proliferator-activated receptors (PPAR) that contain fatty acid binding sites. These recent advances are summarized in this review, with emphasis on the cellular mechanism of EET action, effects of sEH inhibition on these processes, and the potential role of EETs and their metabolites on PPAR-mediated gene expression.
|
EET PRODUCTION |
|---|
|
The functional effectiveness of two enantiomers also can differ. As an illustration, 11(R),12(S)-EET relaxes small renal arteries preconstricted with phenylephrine, but 11(S),12(R)-EET is inactive. Likewise, 11(R),12(S)-EET, but not the S/R enantiomer, increases the activity of the large-conductance Ca2+-activated K+ (BKCa) channels in cell-attached patches of renal vascular smooth muscle cells (183). 14(R),15(S)-EET also is a better ligand than 14(S),15(R)-EET for binding to guinea pig mononuclear cells (164), but the other enantiomer, 14(S),15(R)-EET, is more potent in activating smooth muscle BKCa channels and dilating bovine coronary arteries (9). In contrast, no stereoselectivity was observed for EET-mediated dilation of canine or porcine microvessels (177). Another consideration regarding stereoselectivity is that the two enantiomers may have different functions. CYP2J2 expressed in human kidney forms equal amounts of both 11,12-EET enantiomers (167), but only 11(R),12(S)-EET produces relaxation of small renal arteries (183). However, in addition to vasorelaxation, EETs have anti-inflammatory and natriuretic effects in the kidney (77, 180), and it is possible that the S/R enantiomer may contribute to these effects. Thus stereoselectivity is a very complex issue that complicates the investigation of some, but not all, aspects of EET function.
EET in Phospholipids
Small amounts of 8,9-, 11,12-, and 14,15-EET are present in the plasma, liver, and kidney, with 14,15-EET being the most abundant regioisomer (83–85). More than 90% of the EET contained in rat plasma is present in phospholipids, mostly in the low-density lipoproteins. The EETs in human kidney cortex and rat liver are contained almost entirely in the sn-2 position of phosphatidylcholine (PC), phosphatidylethanolamine (PE), and phosphatidylinositol (PI). The main enantiomers are 8(S),9(R)-EET, 11(S),12(R)-EET, and 14(R),15(S)-EET, which is similar to the distribution of R/S enantiomeric forms synthesized in the kidney and liver (83).
The presence of EETs in tissue phospholipids suggests that membrane lipid structural effects may be involved in some EET functions (10). Consistent with this possibility, PC containing 11,12-EET in the sn-2 position inhibits the open probability of the cardiac L-type Ca2+ channel reconstituted in a planar lipid bilayer (13). However, based on the values reported for rat liver phospholipids (84), EETs comprise only about 0.011% of the total fatty acyl chains in PC, 0.013% in PE, and 0.016% in PI. Although transient increases probably occur when cells are exposed to a bolus of EETs, it seems unlikely that the increase will be sufficient to have a generalized effect on membrane physical properties. On the other hand, perhaps the lipid microenvironment in localized domains may be perturbed sufficiently to produce a functional change, such as is observed when the reconstituted L-type Ca2+ channel is exposed to PC containing 11,12-EET.
|
EET METABOLISM |
|---|
|
-oxidation. A 22-carbon product is formed from 11,12- and 14,15-EET by chain elongation. However, -oxidation and chain elongation are prominent metabolic pathways only in cells with low inherent sEH activity or when a sEH inhibitor is added (39, 46). A methyl-terminal hydroxyl group can be inserted into 8,9-, 11,12-, and 14,15-EET by CYP -oxidases (22). Figure 2B shows the structures of these four classes of EET metabolites, DHET, the 16- and 22-carbon epoxides, and the -hydroxy derivative, using 14,15-EET and its products as the example. EETs can form glutathione conjugates (146). However, the functional significance of this reaction is questionable because the Km for 14,15-EET, the best substrate for glutathione-S-transferase, is 10 µM.
Only 5,6- and 8,9-EET are substrates for cyclooxygenase. 5,6-EET is converted to a prostaglandin analog, 5,6-epoxy-prostaglandin (PG) E1, which functions as a renal vasodilator (12). 8(S),9(R)-EET can undergo only a partial cyclooxygenase reaction and is converted to 11-hydroxy-8,9-EET, a renal vasoconstrictor and mitogen for glomerular mesangial cells (73, 178).
EET Incorporation Into Cell Lipids
Incorporation of EETs into phospholipids occurs through a coenzyme A (CoA)-dependent process (84, 158). The largest amount of EET is incorporated into PC, but in most cases PI contains a higher percentage of the 14,15-EET uptake than any of the other regioisomers (142, 155). Most of the radiolabeled EET is incorporated without chemical modification. However, small amounts of 11,12-DHET, 14,15-DHET, and a 22-carbon chain elongation product of 14,15-EET, 16,17-epoxy-
6,9,12-docosatrienoic acid (16,17-EDT), have been detected in the phospholipids (39, 155, 158). A small amount of 14,15-EET also is incorporated into endothelial and astrocyte triglycerides, and some unesterified 14,15-EET is present in astrocytes and endothelial cells (142, 155, 158). Likewise, a small amount of the 8,9-EET that is incorporated into arterial smooth muscle cells remains in unesterified form (44). The presence of intracellular unesterified EET suggests that EET binding to cytosolic FABP, which has been observed in vitro (161), may occur in the intact cell. Modeling of in vitro data suggests that binding to FABP may modulate the intracellular metabolism of EETs (162), as illustrated in Fig. 2A. The presence of unesterified EET also suggests the possibility that binding may occur to other intracellular proteins that contain fatty acid binding sites.
Incubations of endothelial cells with 14,15-EET and smooth muscle cells with 11,12-EET indicate that after these EETs initially accumulate in the cell lipids, they are continuously and progressively hydrolyzed and released into the extracellular fluid as DHETs (42, 155). This occurs under basal conditions, suggesting that any perturbation of membrane structural domains or signaling properties that might occur as a result of EET incorporation into phospholipids is only transient. A much larger and faster EET efflux occurs when the calcium ionophore A-23187 is added to endothelial cells containing 14,15-EET (158, 159). The material released into the extracellular fluid remains largely as 14,15-EET if a sEH inhibitor is present (39). These results suggest that EETs may be temporarily stored in endothelial phospholipids and rapidly released as a bolus when the endothelium is exposed to an agonist (149). Such a mechanism could explain the potentiation of bradykinin-stimulated vasorelaxation produced by EETs (158).
-Oxidation of EETs
As opposed to porcine cells that convert EETs almost entirely to DHETs (44, 155), cultured human endothelial cells, human vascular smooth muscle cells, and human skin fibroblasts convert EETs mostly to chain-shortened -oxidation products (40, 45, 46). This is consistent with the finding that cultured human coronary endothelial cells contain only one-thirtieth the sEH activity of porcine coronary endothelium (46). Studies with mutant fibroblasts indicate that EET -oxidation occurs in the peroxisomes (40). Although 18- and 14-carbon epoxy-fatty acids are formed, the most abundant -oxidation product contains 16 carbons (46). As illustrated in Fig. 2B, 14,15-EET is converted primarily to 10,11-epoxy-4,7-hexadecadienoic acid (10,11-EHD) by -oxidation. Similarly, 11,12-EET is converted to 7,8-epoxy-4,10-hexadecadienoic acid (40). The chain-shortened EET metabolites were not detected in earlier studies done with [1-14C]EETs, because the radiolabeled carboxyl carbon is removed in the first -oxidation cycle. Detection of these metabolites required incubations with [3H]EETs synthesized from [5,6,8,9,11,12,14,15-3H]arachidonic acid so that radioactivity remained in the products even though several carbons were removed from the carboxyl end of the EET. The 16-carbon intermediates probably accumulate because they contain a 4-cis-double bond (see Fig. 2B for 10,11-EHD structure). Two additional enzymes, 2,4-dienoyl-CoA reductase and 3,2-enoyl-CoA isomerase, are necessary for -oxidation to proceed through an intermediate that contains a 4-cis-double bond (91), and it appears that these enzymes are rate limiting for continued -oxidation of 11,12- and 14,15-EET in the cells that so far have been studied (40, 46).
As opposed to cultured human endothelial and vascular smooth muscle cells, where EET -oxidation is a prominent process, the physiological role of -oxidation in human vascular tissue is highly questionable in view of recent findings with surgical specimens (45). Human coronary artery and aortic segments converted 14,15-EET entirely to 14,15-DHET. Likewise, human saphenous vein segments converted 14,15-EET entirely to 14,15-DHET, whereas 10,11-EHD was the main product formed by endothelial and smooth muscle cells cultured from the saphenous vein. Western blots showed that freshly isolated human saphenous vein segments contain substantial amounts of sEH protein, whereas detectable amounts were not observed in cultured saphenous vein endothelial and smooth muscle cells. These data indicate that sEH expression decreases markedly when these human cells are grown in culture. Therefore, the high level of EET -oxidation observed in human endothelial and vascular smooth muscle cells probably is an artifact of the cell culture conditions.
Porcine coronary artery endothelial cells, which contain high levels of sEH and ordinarily form DHETs, convert 11,12- and 14,15-EET to -oxidation products when the cells are incubated with a selective sEH inhibitor (39). This provides additional evidence that -oxidation becomes prominent only when the sEH activity is deficient. Although -oxidation appears to be an alternate pathway, there is some evidence that the 16-carbon products that accumulate have bioactivity. For example, 10,11-EHD is almost as potent as 14,15-EET in relaxing isolated constricted porcine coronary arterioles and inhibiting cytokine-stimulated interleukin-8 production in human coronary endothelial cultures (46). However, 10,11-EHD does not retain the biological activity of EETs in all systems. For example, it is much less potent than 14,15-EET in dilating bovine coronary artery rings (35).
11,12-DHET also is catabolized by -oxidation when it accumulates in smooth muscle cells. As in the case of the EETs, the main -oxidation product formed is 7,8-DHHD, the corresponding 16-carbon dihydroxy metabolite that contains a 4-cis double bond (42). Although 7,8-DHHD can relax porcine coronary artery rings, it is less potent than 11,12-EET. Therefore, the main function of this -oxidation pathway appears to be removal of any residual DHET that is retained in the smooth muscle cells.
|
MECHANISM OF EET ACTION |
|---|
In addition to the details of EET formation, the initial mechanistic steps that mediate the autocrine and paracrine effects of EETs remain uncertain. One possibility is that EETs bind to a membrane receptor linked to an intracellular signal transduction pathway that initiates the functional response. The other is an intracellular mechanism in which EETs or phospholipids containing newly incorporated EETs directly interact with and activate ion channels, signal transduction components, or transcription factors that produce the functional response. It is likely that the actions of EETs are mediated by both mechanisms, thus accounting for their diverse effects.
Membrane Receptor Mechanism
Figure 3 is a schematic illustration of the postulated membrane receptor mechanism. The key element is that the functional response is initiated by EET binding to a plasma membrane EET receptor. This activates signal transduction pathways that regulate ion channels or gene expression, producing a change in cell properties and function. Evidence supporting this mechanism was obtained from studies with human U937 cells, which contain a cell surface protein that functions as a high-affinity stereoselective binding site for 14(R),15(S)-EET (165). Guinea pig mononuclear cells have a similar high-affinity 14,15-EET binding protein that sediments with the particulate material of the cell homogenate (164, 166). 14,15-EET binding increases the intracellular adenosine 3',5'-cyclic monophosphate (cAMP) content and activates protein kinase A (PKA), resulting in downregulation of the putative receptor (164, 165). Additional evidence for a mechanism involving a G protein-coupled receptor (i.e., a 7-transmembrane receptor) is provided by the observation that 11,12-EET induced activation of the BKCa channel and tissue plasminogen activator (tPA) expression is mediated by the Gs component of a heterotrimeric GTP binding protein (59, 95, 96, 121). Angiogenesis initiated by 11,12-EET also involves a cAMP-PKA signaling pathway that induces cyclooxygenase (COX)-2 expression (107).
|
In addition to the Gs-cAMP-PKA pathway, a number of other signal transduction mechanisms, shown in Fig. 3, have been found to be active in EET functional responses under various conditions. Activation of tyrosine kinase cascade, Src kinase, mitogen-activated protein kinase (MAPK), and phosphatidylinositol 3-kinase (PI-3K)/Akt pathways mediate actions of EETs in endothelial cells, arterial smooth muscle cells, glomerular mesangial cells, renal tubular epithelial cells, and myocardium (14, 15, 54, 72, 127, 140, 157). In addition, the anti-inflammatory effect produced by 11,12-EET in the endothelium is due to inhibition of cytokine-activated nuclear factor-
B (NF-B)-mediated transcription. This occurs by inhibition of IKK phosphorylation of IB (120, 150). The fact that other agonists typically activate these pathways through membrane receptor mechanisms provides indirect support for an EET receptor mechanism similar to that illustrated in Fig. 3, but so far the putative EET receptor has not been conclusively identified.
Intracellular Mechanism
An alternative possibility is that EET enters the cell and produces functional effects by directly interacting with intracellular effectors as shown in Fig. 4. According to this proposed mechanism, the EET is present intracellularly as a result of uptake, hydrolysis of phospholipids that contain EET, or synthesis from arachidonic acid by a CYP epoxygenase. The intracellular EET directly interacts with FABP, ion channels, or transcription factors that produce functional responses, or it is present in phospholipids that interact with membrane proteins or phospholipid-mediated signal transduction pathways.
|
(97, 161). These findings support the possibility that EETs act through an intracellular mechanism as depicted in Fig. 4, but there is no conclusive evidence indicating that this actually occurs in vivo. |
EET ACTIONS |
|---|
|
Vasodilation is the most extensively studied EET function. The most potent effects of EETs occur in small resistance vessels; for example, 14,15-EET has been observed to produce relaxation of isolated coronary microvessels at concentrations as low as 10 pM (39). This occurs through hyperpolarization and suggests that the EETs function as an EDHF in a number of vascular beds, including the coronary circulation (7, 8, 51). A proposed mechanism is the EET is released by the endothelium and produces hyperpolarization by acting on the vascular smooth muscle (149). Two recent observations in coronary preparations support this mechanism. One is the finding that the EDHF response in bovine coronary arteries is inhibited by the EET antagonist 14,15-epoxyeicosa-5(Z)-enoic acid (61); the other is that EET is the transferable mediator of vasorelaxation in a perfused system consisting of donor and detector coronary arteries (62). An alternative possibility is that the EETs hyperpolarize the endothelium and that this is transmitted to the smooth muscle by electrical coupling through myoendothelial junctions or by release of K+ into the intercellular space (50, 55). EDHF mechanisms that do not involve EETs also have been proposed, including the release of lipoxygenase products or hydrogen peroxide from the endothelium (50, 182). In addition to the uncertainty regarding mechanism, there also is a question as to whether the EDHF effect is functionally important under normal physiological conditions or only when endothelial nitric oxide and prostacyclin production are compromised.
Ion Channel Activation by EETs
EETs increase the open probability of the BKCa channel (75). This causes hyperpolarization of the vascular smooth muscle, producing vasorelaxation. An alternative mechanism proposed to underlie vasorelaxation is that EETs activate the transient receptor potential (TRPV4) Ca2+ channel, leading to hyperpolarization and vasorelaxation by forming a Ca2+ signaling complex (33, 55). Endothelial TRPV4 Ca2+ channels also are activated by 5,6-EET and 8,9-EET (156), which may explain the finding that 5,6-EET is a second messenger for Ca2+ entry into endothelial cells (65).
Studies in the bovine coronary artery with 11,12-EET indicate that activation of the BKCa channel is mediated by Gs protein in a process that involves ADP-ribosylation (95, 96). Likewise, 14,15-EET stimulates ADP-ribosylation in the liver (137), but the functional significance of this process is not known. Additional studies have indicated that the Gs protein also mediates BKCa activation by 11,12-EET in human kidney cells (59). Based on these observations, the receptor-mediated BKCa activation mechanism illustrated in Fig. 3 includes Gs, but the possibility that EETs directly interact with the BKCa channel as shown in Fig. 4 cannot be excluded.
EETs activate BKCa channels in other tissues. This process occurs in platelets, decreasing platelet adhesion to the endothelium (90), and in airway smooth muscle, producing bronchodilation through hyperpolarization. Inhibition of smooth muscle Cl– channels also is involved in the mechanism through which EETs produce relaxation of the airway smooth muscle (3, 31, 133, 134).
EETs are reported to affect other ion channels, including the KATP, Na+, and L-type Ca2+ channels. EETs bind to the myocardial KATP channel and thereby reduce its sensitivity to ATP by an allosteric alteration of the ATP binding site (99, 100). Activation of the mitochondrial KATP channels protects the myocardium against ischemia-reperfusion injury (140), suggesting that myocardial preconditioning may occur through a direct interaction between EETs and the channel. However, activation of the p42/p44 MAPK pathway also appears to be involved in myocardial preconditioning (140), and studies in mice with targeted deletion of the sEH support a mechanism involving EET-mediated activation of the PI3K signaling pathways and K+ channels (139). In addition to activating the myocardial KATP channel, EETs inhibit the myocardial Na+ channel by decreasing the probability of channel opening (94). Likewise, the cardiac L-type Ca2+ channel reconstituted in a planar phospholipid bilayer is inhibited when PC containing 11,12-EET in the sn-2 position is present in the bilayer (13).
Anti-inflammatory Effects of EETs
EETs produce an anti-inflammatory effect on the endothelium by inhibiting cytokine-induced NF-B transcription (120). 11,12-EET produces the most potent effect in bovine aortic endothelial cells. It inhibits IKK-mediated phosphorylation of IB, maintaining NF-B in an inactive state (150). 11,12-EET also enhances fibrinolysis by activating tPA gene expression through a cAMP-driven promoter. This involves a Gs protein-mediated signal transduction mechanism (121). Likewise, a cAMP-PKA signaling pathway mediates the inhibitory effect of 11,12-EET on rat aortic smooth muscle cell migration (151).
Angiogenesis
There is increasing evidence that EETs stimulate angiogenesis (55, 100, 107, 109). However, the signaling pathway that mediates this process appears to differ depending on the species, type of endothelium, and the EET regioisomer that initiates the process. A pathway involving activation of MAPK phosphatase-1 that inactivates c-Jun NH2-terminal kinase (JNK), leading to upregulation of cyclin D1 and proliferation, occurs in human umbilical vein endothelial cells that overexpress CYP2C9 or are incubated with 11,12-EET (55, 117). However, other studies with 11,12-EET have indicated that the angiogenic process is initiated by PI3K/Akt-dependent phosphorylation and inactivation of the forkhead transcription factors FOXO1 and FOXO3a, which decreases the cyclin-dependent kinase inhibitor p27Kip1 (125). This pathway is activated by phosphorylation of the epidermal growth factor (EGF) receptor (108). Still another angiogenic signal transduction pathway has been reported for human umbilical vein endothelial cells, this one involving cAMP-PKA activation, COX-2 induction, and PGI2 synthesis (101, 106).
A pathway involving ERK1/2 phosphorylation has been observed in porcine coronary endothelial cells incubated with 11,12-EET (54). This is consistent with the previous finding that 11,12-EET activates tyrosine kinase activity in porcine aortic endothelial cells (72).
A pathway involving MAPK, PI3K, and Akt also mediates the angiogenic response in bovine aortic endothelial cells either engineered to overexpress CYP epoxygenases or treated with EETs (157). Likewise, these signaling pathways are involved in the angiogenic response in murine pulmonary endothelial cells, but the results are more complicated. 8,9-EET and 11,12-EET stimulate proliferation of the pulmonary endothelial cells through the p38 MAPK pathway, whereas the response to 5,6-EET and 11,12-EET occurs through PI3K activation (127). To complicate things further, only 5,6-EET and 8,9-EET promote endothelial cell migration, tube formation, and in vivo neovascularization in mice (127), and although 11,12-EET is effective in stimulating the angiogenic response in most of the endothelial culture systems that have been studied, there is no evidence that it is active in vivo.
In summary, three signaling pathways appear to play a role in EET-mediated angiogenesis. This is summarized in Table 1 and illustrated schematically in Fig. 3. One is a cAMP-dependent pathway that activates the cAMP response-element binding protein (CREBP) and COX-2 expression. This pathway is activated by EETs produced by CYP2C9, especially 11,12-EET (106). The second pathway that also is activated by EETs produced by CYP2C9 involves PI3K and Akt, leading to an increase in cyclin D1 expression (55). Tyrosine phosphorylation of the EGF receptor is associated with this mechanism (108), as well as decreased expression of the cyclin D1 inhibitor p27Kip1 due to Akt-mediated phosphorylation of the forkhead transcription factors FOXO1 and FOXO3a (55). Another contributing factor to the increase in cyclin D1 expression is activation of MAPK phosphatase-1, which decreases JNK activity (55, 126). The third is a p38 MAPK pathway that is activated by 8,9-EET and 11,12-EET (127). Which of these pathways is operative probably depends on the species, type of endothelium, and EET regioisomers produced by the CYP epoxygenase. Furthermore, it is not known whether each of these pathways is activated by EET binding to the putative EET receptor, as depicted in Fig. 3, by direct interaction with EETs or membrane phospholipids as shown in Fig. 4, or as a secondary response to another effect of EET on the endothelium.
Mitogenesis
EETs stimulate mitogenesis of renal epithelial cells through a complex signal transduction mechanism. The most potent regioisomer is 14,15-EET. It activates cleavage of heparin-binding EGF-like growth factor (HB-EGF), which is a ligand for the EGF receptor (13, 16). This activates a tyrosine kinase signaling cascade initiated by Src kinase (14, 17), and metalloproteinases also are activated (16). In addition to stimulating the proliferation of renal cells, 14,15-EET appears to reinforce this response through an inhibitory effect on apoptosis in the renal epithelium. This is indicated by studies with LLCKPc14 cells transfected with a mutant bacterial CYP epoxygenase that produces only 14,15-EET. This prevented apoptosis of the LLCKPc14 cells through a mechanism involving activation of the PI3K/Akt signaling pathway (15). 14,15-EET and 8,9-EET also stimulate mitogenesis in cultured rat glomerular mesangial cells. However, in these cells, the mechanism involves activation of Na+/H+ exchange (69).
Other Functions
EETs stimulate the secretion of several polypeptide hormones. 5,6-EET and 14,15-EET stimulates growth hormone release from somatotrophs (145), and 8,9-EET increases dopamine-stimulated somatostatin release from hypothalamic neurons (81). CYP2J2 and endogenous EETs are present in the endocrine pancreas, suggesting that EETs also may be involved in pancreatic hormone secretion (173). However, the pathways that are involved in producing these effects have not been determined.
|
FUNCTION OF -3 EET ANALOGS
|
|---|
-3 analog of arachidonic acid (147), is converted to an epoxide derivative by human recombinant CYP2C epoxygenases with a catalytic efficiency similar to that of arachidonic acid (2). The main EPA epoxide derivative that is formed, 17(R),18(S)-epoxyeicosaquatraenoic acid (17,18-EEQ), is a potent activator of BKCa channels in arterial smooth muscle cells (93). Likewise, chemically synthesized EEQ regioisomers dilate canine and porcine coronary microvessels with EC50 values in the same range as those for the corresponding EET regioisomers (177). Furthermore, 11,12-EEQ was more potent than 11,12-EET in activating the cardiac KATP channel (102). These findings suggest that some functional effects of -3 fatty acids may be due to EEQ synthesis or, alternatively, to a reduction in EET synthesis because of competition between EPA and arachidonic acid for CYP epoxygenases. At present, however, there is no evidence that EEQs are produced in vivo or that -3 fatty acid supplementation has any effect on EET formation in either an intact cell or an experimental animal.
Epoxides have been chemically synthesized from docosahexaenoic acid (DHA), a 22-carbon -3 fatty acid (168). DHA is the most abundant -3 fatty acid in many tissues and accumulates to high levels in the brain (128, 147, 148, 163). The epoxydocosapentaenoic acids (EDPs) synthesized from DHA activate BKCa channels and dilate preconstricted porcine coronary arterioles (168). The EC50 range for vasodilation by the EDP regioisomers was 0.5–24 pM, and 13,14-EDP was 100 times more potent than 11,12-EET in activating the BKCa channels in coronary smooth muscle cells. Another epoxide derivative of DHA, 16,17-epoxy-docosatriene, is an intermediate in the pathway that produces neuroprotectin D1 in human ARPE-19 cells (113). However, 16,17-epoxy-docosatriene is formed as a result of a lipoxygenase reaction, not a CYP reaction, and there is no information as to whether it has vasoactive properties similar to an EDP.
|
SOLUBLE EPOXIDE HYDROLASE |
|---|
Linear rates of DHET formation were obtained with recombinant mouse sEH during incubations with racemic 14,15-EET and 11,12-EET for 1 and 4 min, respectively (162). The kinetic data were well fit by a Michaelis-Menten model with Km = 2.5 µM and Vmax = 38 µmol/min for 14,15-EET, and Km = 0.45 µM and Vmax = 9.2 µmol/min for 11,12-EET. Km values in the range of 3–5 µM also have been reported for 14,15-, 11,12-, and 8,9-EETs (176), and the calculated catalytic efficiencies for 11,12-EET in these studies varied from 0.3 to 21 µM–1·s–1 (162, 176). The higher values were obtained in medium containing 100 µg/ml phospholipid vesicles to solubilize the EETs (176), compared with medium containing 30 nM bovine serum albumin (162). Porcine coronary endothelial cells converted 60% of the available EET to DHET during a 1-h incubation with 2 µM 14,15-EET (39). During incubation of porcine aortic endothelial cells with 0.5 µM 14,15-EET, 50% of the DHET formed was produced in the first 10 min (155). Although not all of the available EET was converted to DHET, the amount of 14,15-DHET produced was linear in a 2-h incubation with 0.25–5 µM 14,15-EET (155). Endogenous EET concentrations have not been accurately determined, but the intracellular concentration of EET after exposure of platelets to thrombin has been estimated to reach levels as high as 1 µM (181). Based on this estimate, a simulated analysis performed using DynaFit with the production of 1 µM EET at a rate of 0.01 s–1 indicated complete conversion to DHET in 6 min (162). This simulation suggests that the sEH activity is sufficient to rapidly hydrate the amounts of EET likely to be generated under physiological conditions.
The addition of H-FABP or liver (L)-FABP to incubations containing recombinant sEH reduces the amount of 11,12- or 14,15-EET converted to DHETs, implying that binding to FABP may protect these EETs from catabolism and thereby prolong their intracellular action (162). In this regard, the simulation described above indicated that 
35% of the released EET would remain as EET after 10 min if the intracellular H-FABP concentration was 1 µM (162). This suggests that FABP binding also may regulate the availability of EETs to the other intracellular metabolic pathways as illustrated in Fig. 2A, but no information presently is available to indicate that binding of EETs to FABPs actually occurs in an intact cell.
Effect of Selective sEH Inhibitors on EET Metabolism and Function
Many potentially beneficial actions of EETs are attenuated when EETs are converted to DHETs. Therefore, as illustrated in Fig. 5, inhibiting sEH causes EETs to accumulate and be retained for longer periods after they are formed (39), presumably enhancing their beneficial autocrine and paracrine effects. Because DHETs have little or no activity compared with the corresponding EETs in producing a number of functional effects (6, 17, 43, 48, 94, 100, 120, 137, 144, 161), it is generally assumed that inhibiting DHET formation should not impair any vital physiological processes. Consistent with this view, no toxicity was observed in sEH gene-deleted mice (143), and none was reported in hypertensive rats treated with several different sEH inhibitors (78, 171, 180). Pharmacological inhibitors targeted to reduce DHET formation must be selective for sEH, because mammals contain four other epoxide hydrolases that have important metabolic and protective actions (117). However, this presents no difficulty, because potent selective inhibitors are available (110), and they have been structurally refined to increase water solubility so that they can be easily utilized for biochemical and animal experiments (86, 111).
|
-oxidation products. A DCU derivative, 1-cyclohexyl-3-dodecylurea (CDU), potentiated vasodilation produced by 14,15-EET in human coronary arterioles (92). CDU also protected the kidney vasculature and glomerulus from hypertensive injury in angiotensin-induced hypertension (180), which is consistent with the finding that renal sEH is located primarily in the vasculature (170). Another derivative, 1-cyclohexyl-3-dodecanoylurea (CUDA), inhibited the conversion of 11,12- and 14,15-EET to DHETs in surgical specimens of human saphenous vein, coronary artery, and aorta (45). A CUDA derivative in which the cyclohexyl group is replaced by an adamantanyl group, 1-adamantanyl-3-dodecanoylurea (AUDA), lowered blood pressure, increased the urinary EET/DHET ratio, and decreased macrophage infiltration in the kidneys of rats with salt-sensitive hypertension (78). AUDA also lowered blood pressure and increased urinary salt and water excretion in angiotensin-induced hypertension (80), and it decreased cerebral infarct size in spontaneously hypertensive rats following occlusion of the middle cerebral artery (29). Furthermore, AUDA augmented the anti-inflammatory effect of EETs in endothelial cells, probably by increasing EET-induced PPAR transcriptional activity (97), and AUDA-butyl ester reduced the inflammatory response produced by lipopolysaccharide in mice (135). An AUDA analog, 1-admantanyl-3-cyclohexylurea (AUC), increased the response of the TRPV4 channel to 5,6- and 11,12-EET in mouse aortic endothelial cells (156). Thus a number of compounds that are effective sEH inhibitors in intact cells, tissue specimens, and experimental animals are now available for use in investigational studies.
|
Recent data indicate that these substituted urea derivatives also produce effects through mechanisms other than sEH inhibition. AUDA has been observed to relax rat mesenteric resistance arteries through a direct action on the vascular smooth muscle that is dependent on the admantanyl group (122). CDU inhibits human aortic smooth muscle cell proliferation through a direct action that is independent of its inhibitory effect on sEH (25, 26). In addition, AUDA and CUDA activate mouse PPAR in a COS-7 cell expression system by a mechanism that is unrelated to sEH inhibition (37).
Functions of DHETs
The perception that DHETs have no vital biological function is supported by the findings that DHETs are either inactive or only minimally active compared with the corresponding EETs in mediating the following functions: ADP-ribosylation (137), inhibition of PGE2 production in aortic smooth muscle cells (43), mitogenesis in kidney tubular cells (17), VCAM-1 expression in endothelial cells (120), inhibition of myocardial Na+ channels (94), Ca2+ entry into aortic smooth muscle cells (48), activation of myocardial KATP channels (100), binding to H-FABP (161), and inhibition of cAMP-induced aromatase activity in aortic smooth muscle cells (144).
In opposition to these negative results, DHETs activity has been observed in a number of other biological systems. Bovine endothelial cells took up small amounts of DHETs available in the extracellular fluid and incorporated them into phospholipids, especially PC and PI (154). Furthermore, DHETs at a concentration of 1 µM inhibited the hydroosmotic effect of vasopressin (71), and at concentrations between 1 and 5 µM they produced relaxation of porcine coronary artery rings constricted with a thromboxane mimetic (42, 158). Although 14,15-DHET also produced relaxation of bovine coronary artery rings, it was only 20% as potent as 14,15-EET in this preparation (6). In contrast, DHETs produced relaxation of a canine coronary arteriole preparation with EC50 values in the range of 0.1 pM (124), and DHETs activated coronary smooth muscle BKCa channels at concentrations of 1–100 nM (101). DHETs, especially 11,12-DHET, produced relaxation in human coronary arterioles through a hyperpolarization mechanism (92). 14,15-DHET at concentrations between 3 and 10 µM activated PPAR-mediated gene expression in transfected COS-7 cells (38), and all of the DHET isomers at a concentration of 5 µM inhibited the activation of PPAR by rosiglitazone in transfected endothelial cells (97). Based on these findings, the general perception that DHET are inactive metabolites is incorrect, and it is possible that they may have important effects on vascular tone under conditions where EETs are rapidly converted to DHETs (41, 42, 101, 124, 158). However, there is no evidence that any of these DHET effects are essential for normal physiological function in vivo.
|
EFFECTS OF EETS AND RELATED CYP PRODUCTS ON PPAR-MEDIATED GENE EXPRESSION |
|---|
(NR1C1), which is expressed primarily in liver, heart, skeletal muscle, and kidney, regulates lipid utilization. PPAR
(NR1C2, also called PPAR) is expressed in many tissues and functions in the control of fatty acid oxidation and energy uncoupling. PPAR (NR1C3), which is expressed mainly in adipose tissue, intestine, and macrophages, regulates adipocyte differentiation, lipid storage, and insulin sensitivity. In addition, each of the PPARs has specific anti-inflammatory properties when they are activated (49). Both PPAR and PPAR are expressed in endothelial cells and blood vessels (27, 28), indicating that they have a role in vascular function.
Polyunsaturated fatty acids, including the most abundant members of the -6 and -3 classes of essential fatty acids, activate each of the three types of PPARs (30, 32, 58, 64, 88, 116, 169). Saturated fatty acyl-CoA derivatives activate PPAR (32, 74). Conjugated linoleic acid also activates PPAR (112), but it can act both as an agonist and antagonist for PPAR depending on the experimental context (66, 129). The synthetic sulfur-containing fatty acid analog tetradecylthioacetic acid activates human PPARs in the order of PPAR > PPAR > PPAR (160). Expression of either adipocyte-FABP or acyl-CoA binding protein in CV-1 cells decreased tetradecylthioacetic acid-induced PPAR transactivation, indicating that these binding proteins modulate the access of fatty acids to PPARs (70).
Polyunsaturated fatty acid metabolites produced by the cyclooxygenase and lipoxygenase pathways also function as PPAR ligands and activators. For example, 15-deoxy-12,14-PGJ2, a derivative of PGD2, activates PPAR (57, 87), and 8(S)-HETE is a potent activator of PPAR (32, 58, 169). Likewise, the arachidonic acid lipoxygenase products 12-HETE and 15-HETE, and the linoleic acid lipoxygenase products 9-hydroxyoctadecadienoic acid (HODE) and 13-HODE, activate PPAR (32). In addition, nitrolinoleic acid, which is formed by reaction of linoleic acid with nitric acid, activates all three PPARs, with the most potent effect being on PPAR (136). The importance of these polyunsaturated fatty acid products in the activation of the PPARs under physiological or pathological conditions is debated.
Because the PPARs bind an assortment of natural lipid products, it has been suggested that they serve as "generic" sensors for fatty acids and related products. However, there is evidence for the existence of high-affinity, as yet unidentified endogenous activators that are essential to some biological processes involving PPARs, such as for PPAR-mediated adipocyte differentiation (153). Thus additional naturally occurring lipid metabolites likely function as endogenous PPAR activators. Although EETs are a logical possibility considering their structural similarity to HETEs and HODEs, they largely have been ignored because of the finding that 8,9-EET is only 13% as effective in activating PPAR as pirinixic acid (Wy-14643), the widely used fibrate agonist (58). However, the possibility that EETs may function as endogenous PPAR activators should be reconsidered in view of recent results demonstrating that EETs and several EET metabolites bind to the isolated ligand-binding domain of PPARs and activate PPAR-mediated gene expression in cultured cell systems (22, 38, 97).
PPAR Activation by EETs
PPAR is activated when bovine aortic endothelial cells are exposed to laminar flow through in a process that is dependent on phospholipase A2 and CYP epoxygenases (98). This results in suppression of cytokine-induced NF-B activation and intercellular adhesion molecule (ICAM)-1 expression. A lipid extract of the flow medium also activated PPAR and suppressed NF-B activation and ICAM-1 expression. Subsequent results indicated that laminar flow caused a substantial increase in 8,9-, 11,12-, and 14,15-EET in the endothelial cells within 15 min (97), suggesting that EETs may be the active component of the lipid extract. Furthermore, addition of the selective sEH inhibitor AUDA to the perfusion medium enhanced PPAR activity stimulated by laminar flow, whereas overexpression of sEH reduced PPAR activity. AUDA also enhanced the inhibitory effect of EETs on TNF- mediated IB degradation (97), which explains the decrease in NF-B-stimulated expression of ICAM-1. Furthermore, a PPAR antagonist blocked the anti-inflammatory effect of laminar flow to inhibit TNF--mediated IB degradation. Additional studies demonstrated that the ligand-binding domain of PPAR binds 8,9-, 11,12-, and 14,15-EET with Kd values between 1.1 and 1.8 µM (97). Together, these findings have been interpreted to indicate that EETs mediate the anti-inflammatory effect of laminar flow on endothelial cells and that the mechanism may involve EET binding and activation of PPAR. They also suggest that selective sEH inhibitors will potentiate the anti-inflammatory effect in the endothelial cells, presumably by increasing the retention of 11,12- and 14,15-EET so that PPAR activation is prolonged.
PPAR Activation by EETs, EET Derivatives, and Other CYP Products
Although EETs are weak activators of PPAR, the -hydroxylated derivatives of 11,12- and 14,15-EET are potent activators (22). These EETs derivatives are produced by CYP -oxidases, another class of CYP monooxygenases that utilize fatty acids as substrates. These enzymes insert a hydroxyl group at or near the methyl-terminal end of the fatty acid chain in a NADPH-dependent reaction (10).
8,9-, 11,12-, and 14,15-EET are good substrates for CYP4A1 and CYP4A2 and are converted to 20-OH-EETs by these enzymes (22). The conversion of 14,15-EET to 20-OH-14,15-EET by a CYP -oxidase is illustrated in Fig. 2B. In a parinaric acid displacement assay used to measure the relative affinities of various compounds for the ligand-binding domain of PPAR, the Ki values for the EETs were between 22 and 46 nM. In contrast, the Ki value for 20-OH-14,15-EET was only 3 nM (22). Furthermore, in RK13 cells that overexpress either the human or mouse PPAR gene, 20-OH-14,15-EET increased PPAR-mediated gene expression to the same extent as Wy-14643, and 20-OH-11,12-EET also increased PPAR-mediated gene expression in these cells (22).
The DHET derivatives of EETs also activate PPAR. Studies in transiently transfected COS-7 cells containing a luciferase expression system demonstrated that 14,15-DHET at concentrations between 3 and 10 µM was as potent as Wy-14643 in activating mouse PPAR (38). The kinetics of activation produced by 14,15-DHET and Wy-14643 were similar. A fourfold increase in luciferase activity occurred after 3 h, and this increased to ninefold after 6 h. 14,15-DHET was three to four times more potent than any of the other DHET isomers, and like 20-OH-14,15-EE
TOP
ABSTRACT
EET PRODUCTION
