Soluble Epoxide Hydrolase Plays an Essential Role in Angiotensin II-Induced Cardiac Hypertrophy

Soluble epoxide hydrolase plays an essential role in angiotensin II-induced cardiac hypertrophy

Ding Aia,c, Wei Panga, Nan Lia, Ming Xub, Paul D. Jonesc,d, Jun Yangc, Youyi Zhangb, Nipavan Chiamvimonvate, John Y.-J. Shyyf, Bruce D. Hammockc,1, and Yi Zhua,1

aDepartment of Physiology and Pathophysiology and bInstitute of Vascular Medicine, Peking University Third Hospital and Key Laboratory of Molecular Cardiovascular Sciences, Health Science Center, Ministry of Education, Peking University, Beijing 100083, China; cDepartment of Entomology and Cancer Research Center and eDepartment of Cardiovascular Medicine, University of California Medical Center, University of California, Davis, CA 95616; dInternational Flavors and Fragrances Inc., 1515 State Highway 36, Union Beach, NJ 07735; and fDivision of Biomedical Sciences, University of California, Riverside, CA 92521

Contributed by Bruce D. Hammock, November 13, 2008 (sent for review August 13, 2008) Pathophysiological cardiac hypertrophy is one of the most common large conductance of the Ca2ϩ-activated Kϩ channel. EETs also causes of heart failure. Epoxyeicosatrienoic acids, hydrolyzed and cause vasodilation, which suggests that they contribute to the degraded by soluble epoxide hydrolase (sEH), can function as action of the endothelium-derived hyperpolarizing factor (6). endothelium-derived hyperpolarizing factors to induce dilation of Soluble epoxide hydrolase (sEH, Ephx2) catalyzes the con- coronary arteries and thus are cardioprotective. In this study, we version of EETs to the corresponding dihydroxyeicosatrienoic investigated the role of sEH in two rodent models of angiotensin acids (DHETs) (7, 8). An early study demonstrated that a II (Ang II)-induced cardiac hypertrophy. The protein level of sEH selective sEH inhibitor reversed the hypertensive phenotype of was elevated in the heart of both spontaneously hypertensive rats the spontaneously hypertensive rat (SHR) (7). Angiotensin II and Ang II-infused Wistar rats. Blocking the Ang II type 1 receptor (Ang II), a potent vessel constrictor, elevates blood pressure by with losartan could abolish this induction. Administration of a acting on several tissue types. I.p. injection of sEH-selective potent sEH inhibitor (sEHI) prevented the pathogenesis of the Ang inhibitors in Ang II-infused hypertensive rats greatly increased II-induced hypertrophy, as demonstrated by decreased left- the level of EETs and lowered systolic blood pressure (9). Thus, ventricular hypertrophy assessed by echocardiography, reduced increased production of EETs by CYP 450s or decreased hy- cardiomyocyte size, and attenuated expression of hypertrophy drolysis by sEH inhibition seems to reduce blood pressure in markers, including atrial natriuretic factor and ␤-myosin heavy vivo. A preponderance of evidence demonstrates that sEH is also chain. Because sEH elevation was not observed in exercise- or involved in cardiac hypertrophy. Recent studies of pressure norepinephrine-induced hypertrophy, the sEH induction was overload-induced hypertrophy in mice showed that sEH inhibi- closely associated with Ang II-induced hypertrophy. In vitro, Ang II tion prevented cardiac enlargement (10). Thus, the enzyme upregulated sEH and hypertrophy markers in neonatal cardiomy- activity of sEH appears to be involved in the pathogenesis of ocytes isolated from rat and mouse. Expression of these marker cardiac hypertrophy and hypertension. genes was elevated with adenovirus-mediated sEH overexpression Previous work indicated that the use of sEH inhibitors to but decreased with sEHI treatment. These results were supported increase the level of lipid epoxides may reduce cardiac hyper- by studies in neonatal cardiomyocytes from sEH؊/؊ mice. Our trophy by blocking NF-␬B activation (10). We also found that results suggest that sEH is speciﬁcally upregulated by Ang II, which activator protein 1 (AP-1) is involved in the transcriptional directly mediates Ang II-induced cardiac hypertrophy. Thus, phar- upregulation of sEH by Ang II in vascular endothelial cells macological inhibition of sEH would be a useful approach to (ECs), which may contribute to Ang II-induced hypertension prevent and treat Ang II-induced cardiac hypertrophy. (11). Given the potential role of sEH in the Ang II-induced cardiac hypertrophy, we studied whether sEH is upregulated in epoxyeicosatrienoic acid ͉ cardiomyocyte ͉ activator protein 1 the hypertrophic heart of the SHR and the Ang II-infused rat. Our results showed a specific upregulation of sEH by Ang II, ardiac hypertrophy is a major pathological event leading to which is necessary and sufficient for the induction of cardiac Cheart failure (1). Both animal models and human diseases hypertrophy. Moreover, a newly developed sEH inhibitor, 1-(1- show hypertrophic impairments associated with an increased methanesulfonyl-piperidin-4-yl)-3-(4-trifluoromethoxy-phenyl)- induction of the fetal gene program and repression of some adult urea (TUPS), could largely attenuate the Ang II-induced cardiac cardiac genes (2). This pathophysiological condition is charac- hypertrophy. terized by increased cardiomyocyte size, augmented protein Results synthesis, and altered gene expression. An increase in wall tension resulting from prolonged cardiac hypertrophy switches Ang II Upregulates sEH in the Hypertrophic Heart. To investigate the the myosin heavy chain (MHC) from the ␣-tothe␤-isoform and role of sEH in Ang II-induced cardiac hypertrophy, we used rat increases the expression of atrial natriuretic factor (ANF). Related to compensatory attenuation of wall stress, this change Author contributions: D.A., W.P., N.C., B.D.H., and Y. Zhu designed research; D.A., W.P., in gene expression is manifested in pathological cardiac hyper- N.L., M.X., and J.Y. performed research; D.A., W.P., and Y. Zhu analyzed data; P.D.J., Y. trophy (3, 4), the leading cause of cardiac remodeling and heart Zhang, J.Y.-J.S., B.D.H., and Y. Zhu contributed new reagents/analytic tools; and D.A., failure (5). J.Y.-J.S., B.D.H., and Y. Zhu wrote the paper. Arachidonic acid, derived from membrane phospholipids, can Conﬂict of interest statement: P.D.J., N.C., and B.D.H. authored University of California be converted to eicosanoids via three major enzymatic pathways, patents in this area. B. D. Hammock founded Arete Therapeutics to evaulate inhibitors of the soluble epoxide hydrolase as therapeutic agents. Arete has licensed UC IP in this area. namely, cyclooxygenase, lipoxygenase, and CYP 450 epoxygen- None of the authors received funding for this work from Arete Therepeutics, and Arete ase. In addition to generating ␻- and ␻-1 hydroxylation products, Therapeutics scientists did not contribute to these studies. CYP 450 epoxygenase produces 4 epoxyeicosatrienoic acid 1To whom correspondence should be addressed. E-mail: [email protected] or (EET) regioisomer metabolites: 5,6-, 8,9-, 11,12-, and 14,15-EET [email protected]. (6). EETs exert membrane potential-independent effects and This article contains supporting information online at www.pnas.org/cgi/content/full/ modulate several signaling cascades that affect cell function by 0811022106/DCSupplemental. ϩ increasing intracellular Ca2 concentration and activating a © 2009 by The National Academy of Sciences of the USA

564–569 ͉ PNAS ͉ January 13, 2009 ͉ vol. 106 ͉ no. 2 www.pnas.org͞cgi͞doi͞10.1073͞pnas.0811022106 Downloaded by guest on September 29, 2021 Fig. 2. Expression of sEH protein in the heart in three rat models. (A) Wistar rats (180–280 g, male) received losartan (Los, 25 mg/kg/d) or PBS (Ctrl) by oral gavage for 6 days. A minipump was then implanted in the dorsal region to deliver Ang II (A2) at 450 ng/kg/min (A2) or PBS for 3 days. (B) Rats were infused with norepinephrine (NE, 4.8 mg/kg/d) or vehicle (0.01% wt/vol Vit C) for 7 days. (C) Rats were trained with swimming (150 min/day, 5 days/week) for 8 weeks. n ϭ 6 in each group. The ratios of whole heart weight to body weight MEDICAL SCIENCES (HW/BW) and/or left ventricular weight to body weight (LVW/BW) were measured after the animals were killed. Western blot and real-time PCR analysis was performed and the results shown are representative of 6 rats in Fig. 1. sEH protein level is elevated in the heart of SHR rats and Ang II-infused each group. Wistar rats. (A–D) Wistar rats and SHR rats (180–280 g, male) had free access to tap water supplemented with 2% NaCl for 14 days. (E–G) Sprague–Dawley rats received a minipump implantation in the dorsal region to deliver Ang II at Ang II [see online supporting information (SI) Fig. S1A]. We 450 ng/kg/min (A2) or PBS for 14 days. Western blot analysis was performed further tested the effect of Ang II on NCMs isolated from mouse with GAPDH level as a control. The ratio of whole heart weight to body weight and rat. Quantitative real-time PCR revealed that 100 nM Ang (HW/BW) and plasma Ang II (A2) levels were measured after the animals were II significantly increased the mRNA levels of sEH, ANF, and killed. The cross-sections of the rat left ventricle underwent immunohisto- ␤ chemical staining with anti-sEH antibody in C. The sections were counter- -MHC in both cell types (Fig. S1B). Furthermore, the sEH stained with hematoxylin and the average cell area of myocytes was measured activity was increased significantly in Ang II-treated mouse on confocal microscopy in F. The results are presented as mean Ϯ SD from 6 rats NCMs (Fig. S1C). We previously reported that Ang II transcrip- in each group (*, P Ͻ 0.05; **, P Ͻ 0.01). tionally upregulated sEH in ECs through AP-1 transcription factor (11). To explore whether the upregulated sEH occurs at the transcriptional level, NCMs were transiently transfected with models with a high level of Ang II and the associated systemic sEH promoter constructs and then treated with Ang II. Ang II hypertension and cardiac hypertrophy. A high plasma level of activated the sEH promoter. However, a mutation of the AP-1 Ang II was induced in SHRs by feeding animals with saline site at Ϫ446 abolished the Ang II-induced promoter activities containing 2% NaCl for 14 days (11). The plasma level of Ang II was significantly higher in SHRs with hypertension than in (Fig. S1D). This result suggests a similar mechanism of sEH control Wistar rats (Fig. 1A). Cardiac hypertrophy was evident regulation in ECs and cardiomyocytes. from an increase in the ratio of heart weight to body weight in SHRs (Fig. 1B). Immunohistochemical analysis revealed in- sEH Upregulation in the Hypertrophic Heart is Ang II-Specific. To creased expression of sEH in the left ventricle of saline-treated ascertain the specificity of the sEH upregulation by Ang II in SHRs (Fig. 1C), and Western blotting showed an increase in sEH cardiac hypertrophy, we treated Wistar rats with losartan, an and ANF protein level in the heart of these rats (Fig. 1D). Wistar Ang II receptor 1 blocker, for 9 days [6 days before and 3 days rats infused with Ang II (450 ng/kg/min) through an implanted during Ang II infusion (450 ng/kg/min)]. With losartan admin- minipump showed a significantly higher ratio of heart weight to istration, systolic blood pressure in these animals was signifi- body weight than those without Ang II treatment (Fig. 1E). Cells cantly lower than with Ang II infusion alone (11). Western blot in the left ventricle were larger than those in controls (Fig. 1F), analysis revealed the level of sEH in the left ventricle to be and the sEH level was significantly higher in the myocardium of significantly higher with Ang II infusion alone than with vehicle Ang II-infused rats than in controls (Fig. 1G). infusion, losartan alone, or losartan plus Ang II (Fig. 2A and Fig. S2A). In contrast, the sEH level was unchanged in the myocar- Ang II Induces sEH Expression and Cardiac Hypertrophy in Vitro. To dium of rats with hypertrophy induced by norepinephrine (NE) investigate the association of Ang II, sEH expression, and infusion for 7 days or exercise training for 8 weeks; even ANF hypertrophy, we treated rat neonatal cardiomyocytes (NCMs) and ␤-MHC in both hypertrophy models were significantly with Ang II and assayed the level of sEH in cell lysates by increased (Fig. 2 B and C and Fig. S2 B and C). Thus, sEH Western blotting. Ang II increased the protein sEH level in upregulation was observed only in Ang II-induced cardiac NCMs in a dose-dependent manner, the level peaking at 100 nM hypertrophy, but was not seen in other types of physiological

Ai et al. PNAS ͉ January 13, 2009 ͉ vol. 106 ͉ no. 2 ͉ 565 Downloaded by guest on September 29, 2021 Fig. 4. Role of sEH inhibitor in Ang II-induced cardiomyocyte hypertrophy: rat (A and C) or mouse (B). NCMs were treated with 100 nM Ang II (A2) with or without TUPS (1 ␮M) for 24 h. (A) Western blot analysis was performed with GAPDH level as a control. (B) The DHETs and EETs in cytosolic extracts were separated on reverse-phase HPLC and then quantiﬁed by use of a 4000 QTRAP tandem mass spectrometer. For cell area, cells were stained with ␣-actin, and the cell area of myocytes was measured on confocal microscopy. Data are means Ϯ SD of 100 cell measurements from 3 independent experiments. (C) Quantitative RT–PCR analysis of mRNA level of ANF and ␤-MHC, the relative mRNA normalized to that of 18S (*, P Ͻ 0.05). Fig. 3. sEH inhibitor attenuates Ang II-induced cardiac hypertrophy in Sprague–Dawley rats. Sprague–Dawley rats(180–280 g, male) received TUPS (0.65 mg/kg/day) or PEG400 by oral gavage for 21 days. Seven days after of the left-ventricular wall at end systole/diastole, another indi- beginning TUPS administration, a minipump was implanted in the dorsal cation of cardiac hypertrophy, in Ang II-infused rats; the cardiac region to deliver Ang II (A2) for 14 days. PBS treatment was used as a control hypertrophy was significantly reduced with sEH activity inhib- (Ctrl). (A) Systolic blood pressure (BP) was measured every 2 days after im- ited by TUPS (Fig. S3). Ejection fraction, a surrogate of systolic plantation. (B) Heart weight to body weight (HW/BW) was measured after rats function, was increased in rats with Ang II infusion alone, which were killed. (C) Cross-sections of rat left ventricles histochemically stained with suggests compensatory responses. This effect was attenuated picric-sirius red for ﬁbrosis. (D) Real-time RT–PCR analysis of ANF and ␤-MHC with TUPS treatment as well. The cell size and fibrosis seen in mRNA level. Data are means Ϯ SD of the relative mRNA normalized to that of histological sections of the left ventricle from rats with Ang II 18S from 6 rats in each group. (E) Western blot analysis of sEH and GAPDH infusion alone were greater than those in control specimens, proteins was performed. The results are presented as mean Ϯ SD from 6 rats Ͻ Ͻ which confirmed the heart hypertrophy and remodeling, but in each group (*, P 0.05; **, P 0.01). were attenuated with TUPS treatment (Fig. 3C). Further, the mRNA levels of ANF and ␤-MHC were higher in extracts isolated from hearts with Ang II infusion alone but lowered with (exercise) or pathophysiological (NE) cardiac hypertrophy we TUPS treatment (Fig. 3D). The level of sEH protein was tested. increased in the Ang II infusion group, which was not significantly decreased by TUPS (Fig. 3E). TUPS administration had sEH Inhibition Blocks Ang II-Induced Cardiac Hypertrophy in Vivo and no effect on the nonhypertrophic untreated group, which indi- in Vitro. To decipher whether the Ang II-induced hypertrophy cates that sEH inhibition had little effect on the endogenous was due to sEH upregulation, we infused SpragueϪDawley rats reninϪangiotensin system. without or with Ang II (450 ng/kg/min) for 14 days. Half of the To define the role of sEH in hypertrophic cardiomyocytes, we rats from each group were given the sEH inhibitor TUPS (0.65 applied TUPS to cultured NCMs. Although the sEH expression mg/kg/day) or polyethylene glycol 400 (PEG400) as a control. was unaffected (Fig. 4A), TUPS (1 ␮M) markedly inhibited the Systolic blood pressure (SBP) in control rats and those treated basal and Ang II-induced sEH activity and cellular size of with TUPS alone remained within the normotensive range cardiomyocytes (Fig. 4B and Fig. S3). Consistent with results from in vivo experiments, the increase in ANF and ␤-MHC throughout the experimental period, but Ang II administration mRNA in cultured cells induced by Ang II was blocked by TUPS resulted in severe hypertension with a large increase in ratio of treatment (Fig. 4C). heart weight to body weight (Fig. 3 A and B). Importantly, the hypertension and cardiac hypertrophy were significantly atten- Direct Effect of Gene Manipulation of sEH on Hypertrophy in Cultured uated in the TUPS-treated hypertensive rats. Echocardiography Cardiomyocytes. We next studied whether adenoviral overexpres- performed at the end of experiment revealed increased thickness sion of sEH in cultured NCMs to mimic the induction by Ang II

566 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.0811022106 Ai et al. Downloaded by guest on September 29, 2021 rats. The sEH upregulation by Ang II could be blocked by losartan, a blocker of the AT1 receptor. Moreover, sEH upregulation occurred with Ang II-induced hypertrophy but not with NE-induced hypertrophy or that induced after swim training. At the molecular level, Ang II induces ANF expression through the extracellular signal-related kinase (ERK) cascade in cultured neonatal rat ventricular myocytes (15). A catalytically inactive mutant of ERK has been shown to inhibit the Ang II-induced ␤-MHC promoter activity (16). Ang II-induced hypertension, cardiac hypertrophy, and reactive oxygen species production are possibly mediated by p38 mitogen-activated protein kinase (MAPK) (17, 18). Ang II can also activate Jun N-terminal kinase (JNK), which leads to the activation of the transcription factor AP-1 in vitro and in vivo (19–21). ERK, p38, and JNK constitute the MAPK family signaling that activates the downstream AP-1 transcription factor (22, 23). AP-1/DNA interaction was prevented by the AT1 receptor antagonist in transgenic rats harboring both renin and angiotensinogen genes (24). Our previous work indicated that AP-1 activation is involved in the transcriptional upregulation of sEH by Ang II in ECs (11). We previously reported that basal promoter activity of sEH varied in different types of cells, of which a proximal Sp1 Fig. 5. Role of sEH overexpression in cultured NCMs. (A and B) Rat NCMs binding site was critical (25). During the pathophysiological were infected with Ad-sEH or Ad-GFP for 24 or 48 h. Western blot analysis of onset of hypertension, Ang II likely induces sEH via AP-1 sEH protein level in A and quantitative RT–PCR analysis of mRNA level of ANF activation in adult ventricular cardiomyocytes (11). In the cur- and ␤-MHC in B were measured. (C and D) Murine NCMs were infected with rent study, we found that the AP-1 site in the promoter region Ad-sEH or Ad-GFP for 24 h and sEH activity and the size of the myocytes were Ϯ was crucial for the Ang II activation of sEH promoter activity, determined as described in Fig. 4. Data are means SD from 3 independent since the mutation of the AP-1 site at Ϫ446 inhibited Ang

experiments (*, P Ͻ 0.05) . MEDICAL SCIENCES II-induced promoter activities (Fig. S1D). However, as compared with the induction of protein and mRNA levels, luciferase would induce cadiac hypertrophy. The protein level and activity activity driven by the sEH promoter was not markedly increased, of sEH were increased significantly in adenovirus (Ad)-sEH- which suggests the involvement of other factors, including other infected cells as compared with untreated or Ad-GFP-infected transcriptional regulation and possible posttranscriptional and controls (Fig. 5 A and C and Fig. S4). Of note, the cell area and posttranslational modification. expression of hypertrophic markers were increased in cells The elevated level of sEH would increase the hydrolysis of overexpressing sEH (Fig. 5 B and D). In contrast, Ang II-induced EETs and other epoxylipids, but the administration of the sEH inhibitor TUPS should offset the increased sEH expression. We cellular responses in wild-type cells were attenuated in cardio- found that in vivo administration of TUPS prevented the Ang myocytes isolated from Ephx2 gene-disrupted mice (12), as II-induced enlargement of the heart (Fig. 3). Because the shown in Fig. 6. These results suggest that sEH not only functions enzyme activity of sEH was greatly reduced in cardiomyocytes as a mediator for the hypertrophic effect of Ang II but also is (Fig. 4), we hypothesize TUPS exerts its cardioprotective effect, sufficient for inducing cardiac hypertrophy. at least in part, through the increased EET activity. ATP- Discussion sensitive potassium channels (KATP) have been suggested to be important for cardiac protection (26). The protective mecha- sEH inhibition with the ensuing decrease in EET degradation nisms contributed by these channels include depolarizing intra- has been suggested to lower blood pressure in several models of mitochondrial membrane, altering reactive oxygen species hypertension, including one induced by Ang II (9, 13, 14). production, and increasing mitochondrial Kϩ uptake with Administration of sEH inhibitors also attenuates cardiohyper- attendant reduction of Ca2ϩ overload (27–32). Although Ang trophy in murine models with transverse aortic constriction II inhibits KATP activity in ventricular myocytes (33), EETs can (TAC) (10). However, the involvement of sEH in Ang II-induced increase the opening for both sarcolemmal and mitochondrial cardiac hypertrophy is unclear. In the current study, we found KATP (sarc KATP and mito KATP). Cytochrome P450 CYP 2J2 sEH upregulated in the heart of SHRs and Ang II-infused Wistar overexpression in cardiomyocytes, presumably augmenting EET production, increases flavoprotein fluorescence, which is indicative of enhanced mito KATP activity (34). Indeed, treat- ing cardiomyocytes with a physiological level of EETs increased flavoprotein fluorescence (35). Through inhibiting sEH, TUPS treatment thus is likely to increase the cardiac level of EETs and other epoxylipids, which prevents the KATP inhibition by Ang II. Because MAPKs are involved in the upregulation of ANF and ␤-MHC, the increased EET activity may also inhibit KATP and MAPK to abolish the induction of these hypertrophic markers. The first generation of competitive sEH inhibitors such as Fig. 6. Role of sEH knockout on Ang II-induced hypertrophy in cultured dicyclohexyl urea (DCU) and 1-adamantyl-3-cyclohexyl urea mouse NCMs. Murine NCMs were isolated from sEHϪ/Ϫ mice or their wild-type (ACU) were powerful transition-state mimics based on the littermates (WT). Cell area and mRNA expression of ␤-MHC and ANF were urea moiety as a central pharmacophore, but they were very determined on treatment with or without 100 nM Ang II (A2). Data are difficult to formulate (36–38). A previous study of TAC- means Ϯ SD from 3 independent experiments (*, P Ͻ 0.05). induced cardiac hypertrophy involved 12-(3-adamantan-1-yl-

Ai et al. PNAS ͉ January 13, 2009 ͉ vol. 106 ͉ no. 2 ͉ 567 Downloaded by guest on September 29, 2021 uredio) dodecanoic acid (AUDA) and 1-adamantan-1-yl-3-{5- Plasmid of Human sEH Promoter Transient Transfection. The plasmids of the [2[(2-ethoxy)ethoxy]pentyl}urea (AEPU) (36, 39); AUDA human sEH promoter sEH-1091-Luc (11) and a mutation construct of the sEH showed improved solubility but limited oral availability, and promoter, sEH-1091M-luc with the AP-1 site at Ϫ446 mutated, were used for AEPU was even more soluble with excellent bioavailability, transient transfection. Plasmid DNA was transfected into rat NCMs by use of ␤ but both compounds were metabolized rapidly in rodents. the Superfect method (QIAGEN). CMV- -gal was cotransfected as a transfec- TUPS was synthesized for this study on the basis of recently tion control. After treatment, NCMs were lysed and the cell lysates were collected for luciferase activity assays. reported pipridine chemistry (40). Conformationally restricted TUPS retains the highly potent urea pharmacophore but has Adenovirus Construction and Infection. Ad-sEH, a recombinant adenovirus excellent oral availability and pharmacokinetic properties, so expressing human Ephx2, was generated by subcloning the cDNA encoding the compound is easier to use for testing the involvement of Ephx2 (NM࿝001979) into an adenoviral vector (Adeno-X Expression System2, sEH in physiological processes (41). Clontech). Confluent cardiomyocytes were infected with recombinant adeno- Although cardiac hypertrophy can result from physiological viruses at the indicated multiplicity of infection (47) and incubated for 24 h exercise, mechanical overload, and/or neurohormonal factors, before experimentation. Ad-GFP was used as a standard control. the mechanisms leading to hypertrophy during pathological and physiological states are distinct. Growth factors, such as insulin- Western Blot Analysis, Quantitative Real-Time RT–PCR, and Immunohistochem- like growth factor 1, are the most important cardiac factors istry. See method in online-only SI. involved in physiological hypertrophy, with phosphoinositide 3-kinase (PI3K) playing a critical role (42, 43). Mice with cardiac Animal Experiments. All animal experimental protocols were approved by the deletion of PI3K subunits exhibited attenuated Akt signaling in Peking University Institutional Animal Care and Use Committee. Male SHRs the heart, reduced heart size, and exercise-induced cardiac and Wistar and Sprague–Dawley rats (all were within 180–280 g) (Peking hypertrophy (42, 43). In pathophysiological hypertrophy, neu- University Health Science Center Animal Department) were kept in a 12-h rohormones, including NE and Ang II, play an important role to light/12-h dark cycle at a controlled room temperature and had free access to stimulate stress-mediated or -reactive cardiac hypertrophy and standard chow and tap water. To induce hypertension, different models were used as described previously (48). For the induction of Ang II-infused hyper- contribute to the progression to heart failure. However, the tension with sEH inhibitor TUPS, Sprague–Dawley rats were divided into 4 molecular events that signal hypertrophy are very complex and experimental groups: 1 group received sham surgery, a second group Ang II the signaling pathways for each neurohormone vs. infusion (450 ng/kg/min for 14 days), the third group Ang II and TUPS (oral Ϫ renin angiotensin in the cardiac hypertrophy response are gavage, 0.65 mg/kg/day, 7 days before surgery), and the fourth group only different. Interestingly, sEH expression was upregulated only in TUPS. TUPS (0.65 mg) was put in 1 ml PEG400 and vortexed until it completely the Ang II-induced, but not in the NE-induced pathologic state dissolved. Then deionized water (3 ml) was added to the solution. The result- in this study. We previously reported that the protein level of ing solution (1 ml) was given to a 250-g rat. Ang II was infused at a continuous sEH did not significantly change in TAC-induced hypertrophy in rate via an osmotic minipump (Alzet 1002). Then the rats were euthanized, mice, but sEH inhibitor could prevent cardiac enlargement. The and hearts were removed, blotted, and weighed to determine the ratio of mechanisms leading to TAC-induced hypertrophy are more heart weight to body weight. Systolic blood pressure (SBP) was determined complicated, in which multiple pathways, including Ang II- and every other day, beginning 1 day before the implantation of the minipumps NE-mediated pathways, may involved. It seems AT1-mediated for the duration of the study, by a computerized tail-cuff system (BP-2000, Ang II signaling is not crucial since TAC could still induce Visitech Systems). cardiac hypertrophy in AT1 receptor knockout mice (44). NF-␬B activation was inhibited by sEHI in the TAC model (10), whereas Analysis of Cardiac Function by Echocardiography. See method in online-only Fig. S3. the transcriptional upregulation of sEH by Ang II was exclusively via AP-1, which contributes to the pathological outcome of the sEH Activity Assays. The cells were scraped in 100 ␮l extract buffer (0.1 M Ang II-induced cardiac hypertrophy. The different response of Na HPO , pH 7.4, with 1 mM EDTA, 1 mM DTT, 1 mM PMSF, and 0.1% phox 2 4 cardiac Nox isforms gp91 and Nox4 to Ang II vs. TAC also Tween-20). The cytosolic supernatants were centrifuged at 1000 rpm for 5 suggested the difference between these two models (45). It is min. 14,15-EET was added to a final concentration of 50 ␮M. After incubation possible that TAC-induced hypertrophy is similar to induction by at 30 °C for 30 min, the reactions were terminated by adding 400 ␮l methanol. NE but not to that by Ang II (46, 47). After centrifugation, the DHETs and EETs in extracts were separated by In summary, our findings define a mechanism of Ang II- reverse-phase HPLC on an ACQUITY UPLC BEH C18 column, 2.1 ϫ 100 mm, 1.7 induced hypertrophy that differs from physiological and other ␮m (Waters). DHETs and EETs were quantified by use of a 4000 QTRAP tandem pathological hypertrophic models: sEH activity can be specifi- mass spectrometer (Applied Biosciences) with negative-mode electrospray cally upregulated by Ang II in cardiomyocytes in vitro and in ionization and multiple reaction monitoring. vivo. sEH and its substracts/products, namely EETs/DHETs, potentially play a regulatory role in Ang II-induced maladaptive Statistical Analysis. Results are expressed as mean Ϯ SD from at least 3 hypertrophy. Importantly, a potent sEH inhibitor could prevent independent experiments. The significance of variability was determined by the pathogenesis of cardiac hypertrophy. Thus, our findings may an unpaired 2-tailed Student’s t-test or ANOVA. Each experiment included have clinical significance for the treatment of cardiac hypertrophy. triplicate measurements for each condition tested, unless indicated other- wise. P Ͻ 0.05 was considered statistically significant. Materials and Methods ACKNOWLEDGMENTS. This study was supported in part by Grants 30570713 NCMs in Culture. Primary cultures of rat NCMs were generated from 1- to and 30630032 from the National Natural Science Foundation of China; by the 3-day-old Harlan Sprague–Dawley rat neonates and cultured as described Major National Basic Research Grant of China 2006CB503802 (to Y.Z.); by the Ϫ/Ϫ previously (48). Murine NCMs were isolated from 3-day-old C57BL/6J or sEH National Institute of Environmental Health Sciences Grants R37, ES02710, P42, neonatal mice as previously described (49). The detailed methods of NCM isola- ES04699, and ES05707 (to B.H.); by National Institutes of Health Grants HL tion are shown in the online-only SI. Cardiomyocytes were maintained in serum- 85727, P42, and ES04699 (to N.C.); and by National Heart, Lung, and Blood free DMEM for 24 h before adenoviral infection and/or treatments. Institute Ruth L. Kirschstein National Service Award F32 HL078096 (to P.D.J.).

1. Katz AM (1995) The cardiomyopathy of overload: an unnatural growth response. Eur 4. de Bold AJ, et al. (2001) The physiological and pathophysiological modulation of the Heart J 16(Suppl O):110–114. endocrine function of the heart. Can J Physiol Pharmacol 79(8):705–714. 2. Olson EN, Schneider MD (2003) Sizing up the heart: development redux in disease. 5. Mann DL (2004) Heart Failure: A Companion to Braunwald’s Heart Disease (Saunders, Genes Dev 17(16):1937–1956. Philadelphia), pp 10–40. 3. Morkin E (1993) Regulation of myosin heavy chain genes in the heart. Circulation 6. Spector AA, Fang X, Snyder GD, Weintraub NL (2004) Epoxyeicosatrienoic acids (EETs): 87(5):1451–1460. metabolism and biochemical function. Prog Lipid Res 43(1):55–90.

568 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.0811022106 Ai et al. Downloaded by guest on September 29, 2021 7. Yu Z, et al. (2000) Soluble epoxide hydrolase regulates hydrolysis of vasoactive ep- 28. Xu W, et al. (2002) Cytoprotective role of Ca2ϩ- activated Kϩ channels in the cardiac oxyeicosatrienoic acids. Circ Res 87(11):992–998. inner mitochondrial membrane. Science 298(5595):1029–1033. 8. Zeldin DC, et al. (1996) Biochemical characterization of the human liver cytochrome 29. Gross GJ, Peart JN (2003) KATP channels and myocardial preconditioning: an update. P450 arachidonic acid epoxygenase pathway. Arch Biochem Biophys 330(1):87– Am J Physiol Heart Circ Physiol 285(3):H921–930. 96. 30. Gross GJ, Fryer RM (1999) Sarcolemmal versus mitochondrial ATP-sensitive Kϩ channels 9. Imig JD, Zhao X, Capdevila JH, Morisseau C, Hammock BD (2002) Soluble epoxide and myocardial preconditioning. Circ Res 84(9):973–979. hydrolase inhibition lowers arterial blood pressure in angiotensin II hypertension. 31. O’Rourke B (2004) Evidence for mitochondrial Kϩ channels and their role in cardio- Hypertension 39(2 Pt 2):690–694. protection. Circ Res 94(4):420–432. 10. Xu D, et al. (2006) Prevention and reversal of cardiac hypertrophy by soluble epoxide 32. Siemen D, Loupatatzis C, Borecky J, Gulbins E, Lang F (1999) Ca2ϩ-activated K channel hydrolase inhibitors. Proc Natl Acad Sci USA 103(49):18733–18738. of the BK-type in the inner mitochondrial membrane of a human glioma cell line. 11. Ai D, et al. (2007) Angiotensin II up-regulates soluble epoxide hydrolase in vascular Biochem Biophys Res Commun 257(2):549–554. endothelium in vitro and in vivo. Proc Natl Acad Sci USA 104(21):9018–9023. 33. Tsuchiya K, et al. (1997) Functional compartmentalization of ATP is involved in angio- 12. Luria A, et al. (2007) Compensatory mechanism for homeostatic blood pressure regu- tensin II-mediated closure of cardiac ATP-sensitive Kϩ channels. Circulation lation in Ephx2 gene-disrupted mice. J Biol Chem 282(5):2891–2898. 96(9):3129–3135. 13. Loch D, Hoey A, Morisseau C, Hammock BO, Brown L (2007) Prevention of hypertension 34. Liu Y, Sato T, O’Rourke B, Marban E (1998) Mitochondrial ATP-dependent potassium in DOCA-salt rats by an inhibitor of soluble epoxide hydrolase. Cell Biochem Biophys channels: Novel effectors of cardioprotection? Circulation 97(24):2463–2469. 47(1):87–98. 35. Seubert JM, et al. (2006) Role of soluble epoxide hydrolase in postischemic recovery of 14. Imig JD, et al. (2005) An orally active epoxide hydrolase inhibitor lowers blood pressure and heart contractile function. Circ Res 99(4):442–450. provides renal protection in salt-sensitive hypertension. Hypertension 46(4):975–981. 36. Morisseau C, et al. (2002) Structural refinement of inhibitors of urea-based soluble 15. Aoki H, Richmond M, Izumo S, Sadoshima J (2000) Specific role of the extracellular epoxide hydrolases. Biochem Pharmacol 63(9):1599–1608. signal-regulated kinase pathway in angiotensin II-induced cardiac hypertrophy in 37. Morisseau C, et al. (1999) Potent urea and carbamate inhibitors of soluble epoxide vitro. Biochem J 347(Pt 1):275–284. hydrolases. Proc Natl Acad Sci USA 96(16):8849–8854. 16. Shih NL, et al. (2001) Reactive oxygen species modulate angiotensin II-induced beta- 38. McElroy NR, Jurs PC, Morisseau C, Hammock BD (2003) QSAR and classification of myosin heavy chain gene expression via Ras/Raf/extracellular signal-regulated kinase murine and human soluble epoxide hydrolase inhibition by urea-like compounds. pathway in neonatal rat cardiomyocytes. Biochem Biophys Res Commun 283(1):143–148. J Med Chem 46(6):1066–1080. 17. Bao W, et al. (2007) Effects of p38 MAPK Inhibitor on angiotensin II-dependent 39. Kim IH, Morisseau C, Watanabe T, Hammock BD (2004) Design, synthesis, and biological hypertension, organ damage, and superoxide anion production. J Cardiovasc Phar- activity of 1,3-disubstituted ureas as potent inhibitors of the soluble epoxide hydrolase macol 49(6):362–368. of increased water solubility. J Med Chem 47(8):2110–2122. 18. Wenzel S, Taimor G, Piper HM, Schluter KD (2001) Redox-sensitive intermediates 40. Jones PD, Tsai HJ, Do ZN, Morisseau C, Hammock BD (2006) Synthesis and SAR of mediate angiotensin II-induced p38 MAP kinase activation, AP-1 binding activity, and conformationally restricted inhibitors of soluble epoxide hydrolase. Bioorg Med Chem TGF-beta expression in adult ventricular cardiomyocytes. FASEB J 15(12):2291–2293. Lett 16(19):5212–5216. 19. Kudoh S, et al. (1997) Angiotensin II stimulates c-Jun NH2-terminal kinase in cultured 41. Chiamvimonvat N, Ho CM, Tsai HJ, Hammock BD (2007) The soluble epoxide hydrolase cardiac myocytes of neonatal rats. Circ Res 80(1):139–146. as a pharmaceutical target for hypertension. J Cardiovasc Pharmacol 50(3):225–237. 20. Yano M, Kim S, Izumi Y, Yamanaka S, Iwao H (1998) Differential activation of cardiac c-jun amino-terminal kinase and extracellular signal-regulated kinase in angiotensin 42. McMullen JR, et al. (2007) Protective effects of exercise and phosphoinositide II-mediated hypertension. Circ Res 83(7):752–760. 3-kinase(p110alpha) signaling in dilated and hypertrophic cardiomyopathy. Proc Natl MEDICAL SCIENCES 21. Izumi Y, et al. (2000) Important role of angiotensin II-mediated c-Jun NH(2)-terminal Acad Sci USA 104(2):612–617. kinase activation in cardiac hypertrophy in hypertensive rats. Hypertension 36(4):511–516. 43. Luo J, et al. (2005) Class IA phosphoinositide 3-kinase regulates heart size and physi- 22. Taimor G, Rakow A, Piper HM (2001) Transcription activator protein 1 (AP-1) mediates ological cardiac hypertrophy. Mol Cell Biol 25(21):9491–9502. NO-induced apoptosis of adult cardiomyocytes. FASEB J 15(13):2518–2520. 44. Harada K, et al. (1998) Pressure overload induces cardiac hypertrophy in angiotensin 23. Tu VC, Bahl JJ, Chen QM (2003) Distinct roles of p42/p44(ERK) and p38 MAPK in II type 1A receptor knockout mice. Circulation 97:1952–1959. oxidant-induced AP-1 activation and cardiomyocyte hypertrophy. Cardiovasc Toxicol 45. Byrne JA, et al. (2003) Contrasting roles of NADPH oxidase isoforms in pressure- 3(2):119–133. overload versus angiotensin II–induced cardiac hypertrophy. Circ Res 93:802. 24. Fiebeler A, et al. (2001) Mineralocorticoid receptor affects AP-1 and nuclear factor- 46. Rapacciuolo A, et al. (2001) Important role of endogenous norepinephrine and epi- kappab activation in angiotensin II-induced cardiac injury. Hypertension 37(2 Pt nephrine in the development of in vivo pressure-overload cardiac hypertrophy. JAm 2):787–793. Coll Cardiol 38(3):876–882. 25. Tanaka H, et al. (2008) Transcriptional regulation of the human soluble epoxide 47. Liao Y, et al. (2003) Activation of adenosine A1 receptor attenuates cardiac hypertro- hydrolase gene EPHX2. Biochim Biophys Acta 1779(1):17–27. phy and prevents heart failure in murine left ventricular pressure-overload model. Circ 26. Yamada S, et al. (2006) Protection conferred by myocardial ATP-sensitive Kϩ channels Res 93(8):759–766. in pressure overload-induced congestive heart failure revealed in KCNJ11 Kir6.2-null 48. Taigen T, De Windt LJ, Lim HW, Molkentin JD (2000) Targeted inhibition of calcineurin mutant. J Physiol 577(Pt 3):1053–1065. prevents agonist-induced cardiomyocyte hypertrophy. Proc Natl Acad Sci USA 27. Hanley PJ, Daut J (2005) K(ATP) channels and preconditioning: a re-examination of the 97(3):1196–1201. role of mitochondrial K(ATP) channels and an overview of alternative mechanisms. J 49. Xu Y, et al. (2005) The effects of intracellular Ca2ϩ on cardiac Kϩ channel expression Mol Cell Cardiol 39(1):17–50. and activity: novel insights from genetically altered mice. J Physiol 562(Pt 3):745–758.

Ai et al. PNAS ͉ January 13, 2009 ͉ vol. 106 ͉ no. 2 ͉ 569 Downloaded by guest on September 29, 2021