International Journal of Obesity (2015) 39, 755–761 © 2015 Macmillan Publishers Limited All rights reserved 0307-0565/15 www.nature.com/ijo

ORIGINAL ARTICLE Interaction of epoxyeicosatrienoic acids and adipocyte fatty acid-binding protein in the modulation of cardiomyocyte contractility

V Lamounier-Zepter1, C Look1, W-H Schunck2, I Schlottmann1, C Woischwill1, SR Bornstein1,AXu3 and I Morano2

BACKGROUND: Adipocyte fatty acid-binding protein (FABP4) is a member of a highly conserved family of cytosolic proteins that bind with high affinity to hydrophobic ligands, such as saturated and unsaturated long-chain fatty acids and . Recent evidence has supported a novel role for FABP4 in linking obesity with metabolic and cardiovascular disorders. In this context, we identified FABP4 as a main bioactive factor released from human adipose tissue that directly suppresses heart contraction in vitro. As FABP4 is known to be a transport protein, it cannot be excluded that lipid ligands are involved in the cardiodepressant effect as well, acting in an additional and/or synergistic way. OBJECTIVE: We investigated a possible involvement of lipid ligands in the negative inotropic effect of adipocyte factors in vitro. RESULTS: We verified that blocking the CYP pathway in adipocytes attenuates the inhibitory effect of adipocyte- conditioned medium (AM) on isolated adult rat cardiomyocytes, thus suggesting the participation of epoxyeicosatrienoic acids (EETs) in the cardiodepressant activity. Analysis of AM for EETs revealed the presence of 5,6-, 8,9-, 11,12- and 14,15-EET, whereas 5,6- EET represented about 45% of the total EET concentration in AM. Incubation of isolated cardiomyocytes with EETs in similar concentrations as found in AM showed that 5,6-EET directly suppresses cardiomyocyte contractility. Furthermore, after addition of 5,6-EET to FABP4, the negative inotropic effect of FABP4 was strongly potentiated in a concentration-dependent manner. CONCLUSIONS: These data suggest that adipocytes release 5,6-EET and FABP4 into the extracellular medium and that the interaction of these factors modulates cardiac function. Therefore elevated levels of FABP4 and 5,6-EET in obese patients may contribute to the development of heart dysfunction in these subjects. International Journal of Obesity (2015) 39, 755–761; doi:10.1038/ijo.2014.193

INTRODUCTION previous studies, we verified a direct effect of FABP4, independent Adipocytes are known to produce a wide variety of bioactive of its function as a transport protein, in the modulation of heart molecules, which are released into the bloodstream.1,2 The function function. We demonstrated that FABP4 was released from mature of these molecules is not completely understood; however, some of adipocytes into the extracellular medium and elicited a direct and 2+ them may partly contribute to the development of obesity-related acute Ca -dependent depression of cardiomyocyte contraction 5 metabolic and cardiovascular diseases. In this context, we have in vitro. Furthermore, we recently showed that FABP4 is recently shown that human mature adipocytes release substances associated with cardiac remodeling in overweight and obese 13 that strongly and acutely suppress the contraction of isolated women and that FABP4 serum levels are linked with left 14 perfused hearts3 andisolatedcardiomyocytesbyattenuating ventricular diastolic dysfunction in morbidly obese subjects. intracellular Ca2+ levels,4 suggesting a new direct role of adipose Thus the elevated levels of circulating FABP4 as observed in obese tissue in the pathogenesis of cardiac dysfunction. Based on this patients may be partially responsible for the development of heart initial work, we have further identified adipocyte fatty acid-binding dysfunction in these subjects. As FABP4 is known to be a transport protein (FABP4) as one of the cardiodepressant factors produced protein, it cannot be excluded that lipid ligands are involved in the and released by human adipocytes.5 cardiodepressant effect as well, acting in an additional and/or Fatty acid-binding proteins (FABPs) are members of a highly synergistic way. Therefore, in the present in vitro study we conserved family of cytosolic proteins, which are found in investigated a possible involvement of lipid ligands in the different cell types and show a high affinity for long-chain fatty negative inotropic effect of adipocyte factors in interaction acids and other hydrophobic ligands. FABPs have important roles with FABP4. in fatty acid solubilization, transfer and storage of fatty acids in eukaryotic cells.6 FABP4 is predominantly present in adipose tissue.7 It is also abundantly present in human serum, probably METHODS due to its release from adipocytes.5,8 Recent evidence has Substances supported a novel role for FABP4 in linking obesity with metabolic FABP4, the epoxyeicosatrienoic acids (EETs) 5,6-, 8,9-, 11,12- and 14,15 and – syndrome, mellitus and .9 12 In the the (CYP) epoxygenase inhibitor 2-(2-propynyloxy)-

1Medical Clinic III, University of Technology Dresden, Dresden, Germany; 2Max-Delbrück-Center for Molecular Medicine, Berlin-Buch, Germany and 3State Key Laboratory of Pharmaceutical Biotechnology and Department of Medicine, University of Hong Kong, Hong Kong, China. Correspondence: Dr V Lamounier-Zepter, Medical Clinic III, University of Technology Dresden, Fetscherstr. 74, Dresden 01307, Germany. E-mail: [email protected] Received 29 May 2014; revised 15 October 2014; accepted 27 October 2014; accepted article preview online 5 November 2014; advance online publication, 25 November 2014 Eicosanoids and FABP4 in heart function V Lamounier-Zepter et al 756 benzenehexanoic acid (PPOH) were purchased from Cayman Chemicals 27 min. After washing, increasing Ca2+-concentrations were applied to cells (Ann Arbor, MI, USA). 8,9-, 11,12- and 14,15-EETs were dissolved in ethanol, to obtain Ca2+-tolerant cardiomyocytes. Isolated cells were resuspended in and 5,6-EET was dissolved in acetonitrile. medium 199 complemented with 0.2% bovine serum albumin, 5% fetal The inhibitor nordihydroguaiaretic acid (NDGA), the bovine serum, 5 mmol l − 1 creatine and taurine, 2 mmol l − 1 carnitine, − 1 (COX) inhibitor indomethacin and the CYP epoxygenase 10 μmol l cytosine-D-arabinofuranoside and antibiotics. Cardiomyocytes inhibitor 17-octadecynoic acid (17-ODYA) (all from Sigma, Munich, were cultured in laminin-coated four-well chamber slides (Nunc, Wiesbaden- Germany) and PPOH were dissolved in dimethyl sulfoxide (DMSO). All Schierstein, Germany) for at least 2 h. substances were stored at − 20 °C, FABP4 at − 80 °C. Acetonitrile, ethanol (both from Roth, Karlsruhe, Germany) and DMSO were used as controls at appropriate concentrations. Incubation of adult rat cardiomyocytes Adult cardiomyocytes were electrically stimulated at 1 Hz until both shortening and fura-2 signals reached a steady level. Electric pacing was Human adipose tissue then switched off, and AM (1:6 dilution), EETs (from 0.01–30 nmol l − 1), Human white adipose tissue was obtained from healthy overweight and FABP4 (60 nmol l − 1) or the corresponding vehicle buffers were added obese women (aged 39–70 years, body mass index ranging from 24.1 to directly to the cardiomyocytes and incubated for 5 min. Subsequently, − 33.5 kg m 2, n = 7) undergoing elective surgical mammary and abdominal electrical pacing was restarted, and mechanical and fluorescence signals reduction. All women were otherwise free of metabolic and endocrine were collected again. diseases. Informed consent was obtained from all donors before the surgical procedure. The study was approved by the ethical committee of 2+ the Technical University of Dresden. Measurement of cell shortening and Ca transients Attached cardiomyocytes were loaded with fura-2-AM dissolved in Hank’s balanced salts solution buffered with 10 mmol l − 1 4-(2-hydroxyethyl)-1- Isolation of human adipocytes and preparation of adipocyte- piperazineethanesulfonic acid (HEPES) at pH 7.4 in the dark at room conditioned medium (AM) temperature for 15 min. Subsequently, cardiomyocytes were washed for − Adipocytes were isolated from subcutaneous human adipose tissue as 30 min with Hank’s balanced salts solution buffered with 10 mmol l 1 described before.4 Briefly, after surgical removal, adipose tissue samples of HEPES. Only cardiomyocytes of optically intact rod-shaped morphology 20–60 g wet weight were immediately transported to the laboratory in with clear cross striation were analyzed. We used an IonOptix Contractility Dulbecco’s modified Eagle’s medium/Nutrient Mix F12 (Life Technologies, and Fluorescence System (IonOptix, Milton, MA, USA) to measure cell Karlsruhe, Germany) with 100 U ml − 1 penicillin and 100 μgml− 1 streptomycin. shortening and Ca2+ transients. Cell shortening was measured using the − After removal of fibrous material and blood vessels, adipose tissue video-edge technique at a sampling rate of 240 s 1. Calcium transients was digested in Krebs–Ringer bicarbonate buffer supplemented with were monitored as a ratio of fluorescence emission at 510 nm obtained by 120 U ml − 1 collagenase type IV from Clostridium histolyticum (Serva, alternate excitation at 340 and 380 nm (340/380 ratio). Heidelberg, Germany). Isolated mature adipocytes were kept at 37 °C in fi a humidi ed atmosphere of 5% CO2 and cultured for 24 h in serum-free Statistics Dulbecco’s modified Eagle’s medium/Nutrient Mix F12 culture medium. After 24 h, the AM was collected and centrifuged for 10 min at 1000 r.p.m. Experiments with isolated cardiomyocytes were performed on at least 40 at 4 °C to remove cellular debris. Supernatants were stored at − 80 °C until individual heart preparations. Six-to-eight cardiomyocytes were used for further use. functional analysis in each heart preparation. Values are expressed as ’ fi In another set of experiments, isolated adipocytes were incubated with means ± s.e.m. We used paired Student s t-test for signi cance analysis. o fi inhibitors for CYP epoxygenase (PPOH and 17-ODYA), lipoxygenase A P-value of 0.05 was considered statistically signi cant. All curves and fi (NDGA) and COX (indomethacin) in a concentration of 10 μM or DMSO as calculations of signi cance were done using the GraphPad Prism version control for 24 h. 4.00 for Windows (GraphPad Software, San Diego, CA, USA).

Characterization of AM RESULTS fi AM from ve different preparations was analyzed for EETs and The production of cardiodepressant factors in AM is influenced by dihydroxyeicosatrienoic acids (DHETs) by Lipidomix GmbH (Lipidomix CYP epoxygenase GmbH, Berlin, Germany).15 The quantification of EETs and DHETs was performed by high-performance liquid chromatography/tandem mass We investigated whether (AA) metabolites are spectrometry (Agilent 1200 coupled to the triple quadrupole tandem mass involved in the cardiodepressant effect of the AM. Therefore we spectrometer Agilent 6410, Agilent Technologies Inc., Santa Clara, CA, USA) cultivated human primary adipocytes in the presence of different using a Rapid-Resolution-C18-column and acetonitrile/ammonium acetate inhibitors (PPOH, 17-ODYA, indomethacin and NDGA), each as a solvent. blocking one of the three major enzymatic pathways of AA metabolism. DMSO was used as a control in an appropriate Animals concentration. Subsequently, the acute effect of AM under Animal experiments were performed using 12-week-old male WKY rats blockage of the AA metabolism was tested on isolated weighing between 221.0 and 305.0 g (263.2 ± 3.1 g, mean ± s.e.m., n = 40). cardiomyocytes (Figure 1). Rats were kept on a 12-h light–dark cycle with 55% humidity at an ambient After incubation with AM, the shortening amplitude of isolated temperature of 23 ± 2 °C and given food and water ad libitum. The cardiomyocytes significantly decreased by 25.9 ± 4.1%. Incubation investigation conforms to the Guide for the Care and Use of Laboratory with AM/DMSO induced a similar reduction of the shortening Animals, published by the US National Institutes of Health (NIH Publication amplitude by 32.3 ± 8.1%. AM obtained from adipocytes, which No. 85-23, revised 1996). The study was approved by the Institutional were treated with the non-selective COX inhibitor indomethacin Animal Care in the state of Berlin, Germany. and the lipoxygenase inhibitor NDGA, had a smaller but still significant negative inotropic effect on the contraction of isolated Isolation of adult rat cardiomyocytes cardiomyocytes. After incubation with AM/NDGA and AM/ As previously described,5 rats were anesthetized with isoflurane followed indomethacin, the shortening amplitude decreased by by intraperitoneal injection of 8 mg of xylazine and 35 mg of ketamine. 15.3 ± 6.4% and 13.6 ± 7.4%, respectively. The cardiodepressive Hearts were rapidly removed and connected to a canula in a Langendorff effects of AM/NDGA and AM/indomethacin were not significantly perfusion system. In the first 3 min, perfusion was performed at 37 °C with Ca2+-free Krebs–Henseleit solution containing 10 mmol l − 1 butanedione different compared with the effect of AM without blockade of the AA metabolism. On the contrary, incubation with the CYP monoxime gassed with a mixture of 95% O2 and 5% CO2. Subsequently, 0.04% collagenase type 2 (Worthington Biochemical Corporation, Lakewood, epoxygenase inhibitor 17-ODYA significantly blunted the inhibi- NJ, USA) and 0.2% bovine serum albumin were added to the gassed tory effect of AM on cardiomyocytes to a shortening amplitude Krebs–Henseleit solution and digestion was carried out for the following decrease of 5.3 ± 8.2%. Incubation with a further CYP epoxygenase

International Journal of Obesity (2015) 755 – 761 © 2015 Macmillan Publishers Limited Eicosanoids and FABP4 in heart function V Lamounier-Zepter et al 757 Human adipocytes in cell culture

Arachidonic Acid

9 8 6 5 1 COOH 20

11 12 14 15 17-ODYA NDGA

Cytochrome P450 Cyclooxygenase Lipoxygenase

PPOH Indomethacin 5,6-EET 5-HETE 8,9-EET 8-HETE

11,12-EET A2 12-HETE 14,15-EET 15-HETE DHETs

Adipocyte-conditioned medium

Contractile analysis of isolated cardiomyocytes Figure 1. Schematic representation of experimental design for inhibition of arachidonic acid metabolism in human adipocytes. inhibitor PPOH showed a smaller reduction in shortening amplitude but not significantly different compared with the effect of AM (Figure 2). The average length of cardiomyocytes was 111.2 ± 2.3 μm. The acute exposure of AM, AM/DMSO, AM/indomethacin, AM/NDGA, AM/PPOH and AM/17-ODYA did not affect the resting cell length over the range of time tested.

Concentration of EETs in AM The CYP epoxygenase pathway forms four regioisomeric EETs (5,6-, 8,9-, 11,12- and 14,15-EET). To determine whether the negative inotropic effect of AM may be explained by one of these fatty acids, AM from adipose tissue of five different donors was analyzed for EETs. In fact, all four EETs were detectable in AM. Table 1 shows the mean concentration of EETs in AM and the Figure 2. Effect of control medium (CM, open bar), AM (filled bar calculated concentrations in the experiment. 5,6-EET was detected left) and AM (filled bars right) collected from adipocytes treated with in the highest concentration, representing about 45% of the total DMSO, NDGA, PPOH, 17-ODYA or indomethacin (INDO) on fractional EET concentration in AM, which is about 1.7-, 2.6- and 5.8-fold shortening of isolated cardiomyocytes. Inhibitors were used in a concentration of 10 μM. Values are expressed as the percentage higher than 11,12-, 8,9- and 14,15-EET, respectively. change of basal value (mean ± s.e.m.), n = 8–13 experiments for each condition, *Po0.05 in comparison to the effect of AM, ***Po0.001 5,6-EET directly suppresses shortening amplitude of isolated in comparison to the effect of CM. cardiomyocytes We tested a possible direct effect of the four EETs on with 5,6-EET (Figures 3b and c). Additionally, the increase rate and cardiomyocyte contraction. Incubation with 8,9-, 11,12- or decay rate of intracellular calcium significantly diminished upon 14,15-EET had no significant effect on cardiomyocyte contraction 5,6-EET treatment (Figures 3e and f). The treatment of cardio- or calcium transient. On the contrary, upon incubation with myocytes with 8,9- and 11,12-EET significantly suppressed the 5,6-EET the shortening of cardiomyocytes was significantly decay rate of calcium transient (Figure 3f). The contraction and suppressed by 39.6 ± 11.4%, associated with a significant reduc- relaxation velocity as well as the increase rate of intracellular tion of the calcium transient by 29.2 ± 5.4% (Figures 3a and d). calcium were not significantly altered by 8,9- and 11,12-EET. No Analysis of the kinetic parameters revealed a significant significant changes in any parameter investigated were observed decrease of contraction and relaxation velocity after treatment after treatment with 14,15-EET (Figure 3).

© 2015 Macmillan Publishers Limited International Journal of Obesity (2015) 755 – 761 Eicosanoids and FABP4 in heart function V Lamounier-Zepter et al 758 The average length of cardiomyocytes was 106.9 ± 2.1 μm. The the effect of 60 nmol l − 1 FABP4 without 5,6-EET. Further analysis acute exposure of cardiomyocytes with EETs and controls did not of the curve kinetics showed that addition of 5,6-EET strongly affect the resting cell length over the range of time tested. potentiated the inhibitory effect of FABP4 on contraction and relaxation velocity in a similar way (Figure 4). μ FABP4 and 5,6-EET act in a synergistic manner on cardiomyocytes The average cell length was 101.2 ± 2.2 m. The acute exposure with FABP4, 5,6-EET and control did not affect the resting cell We have previously shown that FABP4 released by human adipocytes directly and acutely suppresses cardiomyocyte length over the range of time tested. contraction.5 As the main role of FABP4 is to bind fatty acids and to regulate their biological function, we investigated whether DISCUSSION the presence of 5,6-EET may alter the cardiodepressive effect In this study, we observed that EETs, especially in interaction with of FABP4. FABP4, acutely suppress cardiomyocyte contraction in vitro. According to our previous results,5 treatment with FABP4 AA is metabolized to prostaglandins, and throm- reduced shortening amplitude of isolated cardiomyocytes by boxane by COX and to a series of hydroxyeicosatetraenoic acids 30.6 ± 7.3%. Simultaneous treatment of the cardiomyocytes with (5-, 8-, 12-, 15-HETEs) and leukotrienes by lipoxygenase. These FABP4 and 5,6-EET strongly potentiated the negative inotropic products are traditionally involved in the regulation of vascular effect of FABP4 in a concentration-dependent manner. At a − 1 − 1 tone, of renal and pulmonary function and in inflammatory concentration of 60 nmol l FABP4 and 30 nmol l 5,6-EET, the processes.16 A third pathway for the metabolism of AA was further shortening amplitude decreased by 63.1 ± 9.1%. At a concentra- fi 17 tion of 60 nmol l − 1 FABP4 and 1 nmol l − 1 5,6-EET, a 50.6 ± 14.3% identi ed, involving the CYP enzymes. The CYP epoxygenase reduction of shortening amplitude was achieved. Upon incubation pathway produces four regioisomeric EETs (5,6-, 8,9-, 11,2- and with 60 nmol l − 1 FABP4 and 0.01 nmol l − 1 5,6-EET, shortening 14,15-EET) and their hydration products, the DHETs. CYP enzymes amplitude was reduced by 29.4 ± 11.8%, which was comparable to of the P-450 1A, 2B, 2C, 2D, 2E, 2J and 4A families have all been reported to catalyze the formation of EETs in various tissues, including the liver, lung, kidney, brain, intestines, pancreas, heart Table 1. Concentration of EETs in adipocyte-conditioned medium and blood vessels.16,18 EETs function primarily as autocrine and paracrine mediators in the cardiovascular system, lung, kidney and Mean concentration Mean concentration in the 16,19 − − inflammatory response. In the cardiovascular system, EETs are in AM, nmol l 1 experiment (1:6 dilution), nmol l 1 important regulators of vascular tone and have direct effects on 5,6-EET 4.56 ± 1.03 0.76 endothelial cells, inducing growth, differentiation, promoting 18–22 8,9-EET 1.76 ± 0.44 0.29 and preventing vascular inflammation. Further- 11,12-EET 2.64 ± 0.64 0.44 more, EETs have been reported to act as protective agents against 14,15-EET 0.78 ± 0.34 0.13 in endothelial and epithelial cells.23,24 Regarding Total EET 9.42 ± 2.25 1.62 vascular tone, the effects of EETs differ according to organ, Abbreviations: AM, adipocyte-conditioned medium; EET, epoxyeicosatrie- species, to their regioisomers and stereoselectivity. In mesenteric, noic acid. Values are mean ± s.e.m., n = 5 adipocyte preparations (different cerebral and , EETs act mainly as donors). vasodilators.16,19,25 In human, bovine and canine coronary arteries, all EET regioisomers relax precontracted vessels with a similar

Figure 3. Effect of control (open bars) and 5,6-, 8,9-, 11,12- and 14,15-EET (filled bars; 30 nM each) on fractional shortening (a), contraction (b) and relaxation velocity (c) of cardiomyocyte contraction, calcium transient (d) and on increase rate (e) and decay rate (f) of calcium transient of isolated cardiomyocytes. Values are expressed as the percentage change of basal value (mean ± s.e.m.), n = 6–11 experiments for each concentration, *Po0.05, **Po0.01 in comparison to the effect of the corresponding control.

International Journal of Obesity (2015) 755 – 761 © 2015 Macmillan Publishers Limited Eicosanoids and FABP4 in heart function V Lamounier-Zepter et al 759 shortening of cardiomyocytes and led to a significant reduction of calcium transient. A systematic analysis of EET plasma concentrations in humans has not yet been performed. Levels of total plasma EETs between approximately 0.3–3.1 and 94 nM have been reported in healthy control subjects in clinical studies.33–36 Regarding 8,9-, 11,12- and 14,15-EET, concentrations of approximately 1.2-1.9 nM have been found in control subjects without metabolic syndrome, with a 37 total EET level of approximately 4.7 nM. Mean concentration of total EETs in AM was 9.4 nM, with concentrations for individual EETs ranging from 0.8 to 4.6 nM. 5,6-EET significantly reduced shortening amplitude and contraction and relaxation velocity at 30 nM. Together with FABP4, 5,6-EET in concentrations as low as 1nM led to a significant effect on cardiac parameters. Hence, EET concentrations in our experiments are within the spectrum of reported plasma levels in humans. The effect of 5,6-EET on isolated ventricular myocytes was also investigated by Moffat et al.38 Using guinea pig hearts, they observed that 5,6- and 11,12-EET significantly increased cell shortening and intracellular calcium concentrations.38 These results are different from our findings. Whether or not species or methodological differences are solely responsible for this is presently not clear. Interestingly, in the same study by Moffat et al.,38 these EETs had no effects on contractility in isolated perfused guinea pig hearts. In addition, 5,6- and 11,12-EET depressed recovery of contractile function after ischemia of the hearts.38 In another study in isolated rat hearts, 14,15-EET reduced the developed left ventricular pressure.39 Even though we have not analyzed the influence of EETs in Langendorff perfused isolated hearts, these studies are consistent with our results in so far as they show that EETs can have negative effects on cardiac parameters. Therefore, despite the fact that EETs are often described as having cardioprotective effects,32,40,41 they can also lead to unfavourable results. With respect to ion channels, it was demonstrated that 5,6-, 8,9-, 11,12- and 14,15-EET inhibit Na+ currents in rat myocytes.42 This can subsequently lead to reduction of cardiac contractility and, regarding 5,6-EET, would

− 1 be in line with our results. EETs have been shown to lead to Figure 4. Effect of control medium (open bar), 60 nmol l FABP4 varying effects in cardiac ion channels. ATP-sensitive K+ channels (striated bar) and FABP4/5,6-EET (filled bars) on fractional shortening 43 a b c were activated by 8,9- and 11,12-EET. Also, 11,12-EET enhanced ( ), contraction velocity ( ) and relaxation velocity ( ) of isolated 44 cardiomyocytes. Three different concentrations of 5,6-EET were L-type calcium current in isolated rat ventricular myocytes. On added to FABP4: 30, 1 and 0.01 nmol l − 1. Values are expressed as the the other hand, it has been demonstrated that 11,12- and percentage change of basal value (mean ± s.e.m.), n = 5–10 experi- 14,15-EET inhibit the open probability of L-type Ca2+ channels ments for each concentration, *Po0.05, **Po0.01, ***Po0.001 in from porcine cardiac muscle reconstituted into planar lipid comparison to the effect of control medium. bilayers.45 Numerous studies have shown that 5,6-EET is involved in the regulation of /.16,32 Our study is 26–28 potency. On the contrary, in the rat caudal arteries 5,6-EET one of the few investigations suggesting that 5,6-EET could have a 29 was the only active regioisomer inducing a vasodilation. more direct impact on cardiomyocytes. Altogether, these studies Interestingly, in rat renal microcirculation 11,12-EET induces a show that different EETs have a different impact on the vasodilatory response in interlobular and afferent arterioles, cardiovascular system, which is indicative of their physiological whereas 5,6- and 8,9-EET act as vasoconstrictors.30,31 In cerebral relevance. Considering the complexity of the mode of action of vasculature, the actions of the EETs also differ depending on EETs within the cardiovascular system, further studies are needed species and regioisomers involved but, in general, contribute to to elucidate these multiple effects in more detail. coupling neuronal activity to blood flow in localized brain FABP4 is a cytosolic protein predominantly expressed in regions.32 adipose tissue with a major function in intracellular transport In this study, we show that EETs participate in the cardiode- and targeting of lipid ligands. Recent data from our group and pressant effect of the AM in vitro. Inhibition of the CYP others suggest a new extracellular role of FABP4.5,46 We have epoxygenase pathway led to a significant reduction of the previously identified FABP4 as a main adipocyte-derived cardio- cardiodepressant effect of AM, whereas inhibition of the COX depressant factor.5 We demonstrated that FABP4 leads to a and lipoxygenase pathways did not diminish the inhibitory effect negative inotropic effect in isolated rat cardiomyocytes via a of AM on cardiomyocytes. These results indicate that the reduction of contractility parameters and intracellular calcium epoxygenase-dependent EETs are involved in the cardiodepres- transient. Interestingly, high amounts of FABP4 are found in sant activity of AM. All four regioisomeric EETs were present in AM, human plasma and plasma concentrations are directly linked with while 5,6-EET was found in the highest concentration. Accordingly, the metabolic syndrome.8 The cardiodepressive effect of FABP4 treatment with 8,9-, 11,12- or 14,15-EET had only a small, was further supported by clinical studies showing a correlation nonsignificant negative inotropic effect on cardiomyocyte con- between circulating FABP4 and cardiac changes in obese traction, while treatment with 5,6-EET significantly suppressed the subjects.13,14

© 2015 Macmillan Publishers Limited International Journal of Obesity (2015) 755 – 761 Eicosanoids and FABP4 in heart function V Lamounier-Zepter et al 760 Several lines of evidence suggest that the cardiodepressive 7 Baxa CA, Sha RS, Buelt MK, Smith AJ, Matarese V, Chinander LL et al. Human effect of FABP4 demonstrated in this work and in previous adipocyte lipid-binding protein: purification of the protein and cloning of its publications5 results from binding to a surface receptor on complementary DNA. Biochemistry 1989; 28: 8683–8690. cardiomyocytes. We showed that the heart-specific FABP (FABP3) 8 Xu A, Wang Y, Xu JY, Stejskal D, Tam S, Zhang J et al. Adipocyte fatty acid-binding 5 protein is a plasma biomarker closely associated with obesity and metabolic leads to a similar cardiodepressive effect as FABP4. In addition, 52 – FABP3 binds to specific high-affinity-binding sites in the plasma syndrome. Clin Chem 2006; : 405 413. 47 9 Furuhashi M, Fucho R, Gorgun CZ, Tuncman G, Cao H, Hotamisligil GS. Adipocyte/ membrane of cardiomyocytes. As both FABPs have similar fatty acid-binding proteins contribute to metabolic deterioration inhibitory activity and structure, it may be anticipated that FABP4 through actions in both and adipocytes in mice. J Clin Invest 2008; also binds to a cardiac surface receptor in order to exert its 118: 2640–2650. cardiodepressive effect. The identification of such a receptor is 10 Hotamisligil GS, Johnson RS, Distel RJ, Ellis R, Papaioannou VE, Spiegelman BM. essential for understanding the detailed mechanisms of FABP4 Uncoupling of obesity from resistance through a targeted in aP2, action on cardiomyocytes. the adipocyte fatty acid binding protein. Science 1996; 274: 1377–1379. In the present study, we observed that simultaneous treatment 11 Tuncman G, Erbay E, Hom X, De Vivo I, Campos H, Rimm EB et al. A genetic variant of isolated cardiomyocytes with 5,6-EET and FABP4 strongly at the fatty acid-binding protein aP2 locus reduces the risk for hypertriglycer- idemia, , and cardiovascular disease. Proc Natl Acad Sci USA 2006; potentiated the negative inotropic effect of FABP4. EETs are very 103 – lipophilic substances, binding with high affinity to different types :6970 6975. 12 Uysal KT, Scheja L, Wiesbrock SM, Bonner-Weir S, Hotamisligil GS. Improved of proteins and are present more abundantly within tissues than and lipid metabolism in genetically obese mice lacking aP2. Endocrinology fl 16 in body uids. Therefore they are normally considered to act in 2000; 141:3388–3396. an autocrine or paracrine manner. Binding of 5,6-EET to FABP4 has 13 Engeli S, Utz W, Haufe S, Lamounier-Zepter V, Pofahl M, Traber J et al. Fatty acid not been demonstrated yet, but previous studies have also binding protein 4 predicts left ventricular mass and longitudinal function in reported a high affinity of EETs to FABP3, a homologous protein to overweight and obese women. Heart 2013; 99:944–948. FABP4.48 As FABP4 exhibits 460% sequence identity with FABP3 14 Baessler A, Lamounier-Zepter V, Fenk S, Strack C, Lahmann C, Loew T et al. (NCBI Blast; accession nos. NP_001433.1 and NP_004093.1, Adipocyte fatty acid-binding protein levels are associated with left ventricular respectively) and their crystal structures are similar,49 an interac- diastolic dysfunction in morbidly obese subjects. Nutr Diabetes 2014; 4:e106. tion between FABP4 and 5,6-EET is not unlikely. The binding of 15 Arnold C, Markovic M, Blossey K, Wallukat G, Fischer R, Dechend R et al. Arachidonic acid-metabolizing cytochrome P450 enzymes are targets of {omega}- EETs to FABPs limits their conversion to DHETs, thus regulating 285 – 50 3 fatty acids. J Biol Chem 2010; : 32720 32733. their metabolism and cellular actions. The coordinate action of 16 Roman RJ. P-450 metabolites of arachidonic acid in the control of cardiovascular EETs with FABP4 shown in this study further supports a function. Physiol Rev 2002; 82:131–185. bioavailability of EETs in heart cells. 17 Spector AA. Arachidonic acid cytochrome P450 epoxygenase pathway. J Lipid Res Obesity is known to be an independent risk factor for heart 2009; 50:S52–S56. failure.51,52 Several mechanisms have been proposed to explain 18 Node K, Huo Y, Ruan X, Yang B, Spiecker M, Ley K et al. Anti-inflammatory this direct association, including hemodynamic changes, altered properties of cytochrome P450 epoxygenase-derived eicosanoids. Science 1999; myocardial fatty acid metabolism, lipoapoptosis of myocardial 285:1276–1279. 53–56 19 Sudhahar V, Shaw S, Imig JD. analogs and vascular cells as well as neurohormonal activation. We have previously 17 – proposed a new pathophysiological mechanism through a direct function. Curr Med Chem 2010; : 1181 1190. 4 20 Harder DR, Campbell WB, Roman RJ. Role of cytochrome P-450 enzymes and effect of adipocyte factors on cardiac contractility parameters. In metabolites of arachidonic acid in the control of vascular tone. JVascRes1995; the present study, we demonstrated that human adipocytes 32:79–92. release high amounts of 5,6-EET, which, in association with FABP4, 21 Medhora M, Daniels J, Mundey K, Fisslthaler B, Busse R, Jacobs ER et al. strongly suppresses cardiomyocyte contraction in vitro. This Epoxygenase-driven angiogenesis in human lung microvascular endothelial cells. synergistic action may further explain the link between obesity Am J Physiol Heart Circ Physiol 2003; 284: H215–H224. and heart failure. 22 Michaelis UR, Fisslthaler B, Barbosa-Sicard E, Falck JR, Fleming I, Busse R. Cytochrome P450 2C8 and 2C9 are implicated in hypoxia-induced endothelial cell migration and angiogenesis. J Cell Sci 2005; 118: 5489–5498. CONFLICT OF INTEREST 23 Chen JK, Capdevila J, Harris RC. Cytochrome p450 epoxygenase metabolism of arachidonic acid inhibits apoptosis. Mol Cell Biol 2001; 21: 6322–6331. The authors declare no conflict of interest. 24 Dhanasekaran A, Al-Saghir R, Lopez B, Zhu D, Gutterman DD, Jacobs ER et al. Protective effects of epoxyeicosatrienoic acids on human endothelial cells from ACKNOWLEDGEMENTS the pulmonary and coronary vasculature. Am J Physiol Heart Circ Physiol 2006; 291: H517–H531. We thank the expert technical assistance of Uta Lehnert. This study was supported by 25 Campbell WB, Fleming I. Epoxyeicosatrienoic acids and -dependent the ‘Kompetenznetz Adipositas’ (Competence Network for Obesity) funded by the responses. Pflugers Arch 2010; 459: 881–895. German Federal Ministry of Education and Research (grant number 01GI1122G). 26 Campbell WB, Gebremedhin D, Pratt PF, Harder DR. Identification of epox- yeicosatrienoic acids as endothelium-derived hyperpolarizing factors. Circ Res 1996; 78: 415–423. REFERENCES 27 Larsen BT, Miura H, Hatoum OA, Campbell WB, Hammock BD, Zeldin DC et al. 1 Kershaw EE, Flier JS. Adipose tissue as an endocrine organ. J Clin Endocrinol Metab Epoxyeicosatrienoic and dihydroxyeicosatrienoic acids dilate human coronary 2004; 89: 2548–2556. arterioles via BK(Ca) channels: implications for soluble hydrolase 2 Hauner H. The new concept of adipose tissue function. Physiol Behav 2004; 83: inhibition. Am J Physiol Heart Circ Physiol 2006; 290: H491–H499. 653–658. 28 Oltman CL, Weintraub NL, VanRollins M, Dellsperger KC. Epoxyeicosatrienoic acids 3 Look C, Morano I, Ehrhart-Bornstein M, Bornstein SR, Lamounier-Zepter V. and dihydroxyeicosatrienoic acids are potent vasodilators in the canine coronary Adipocyte-derived factors suppress heart contraction. Int J Obes (Lond) 2010; 35: microcirculation. Circ Res 1998; 83: 932–939. 84–90. 29 Carroll MA, Schwartzman M, Capdevila J, Falck JR, McGiff JC. Vasoactivity of ara- 4 Lamounier-Zepter V, Ehrhart-Bornstein M, Karczewski P, Haase H, Bornstein SR, chidonic acid . Eur J Pharmacol 1987; 138:281–283. Morano I. Human adipocytes attenuate cardiomyocyte contraction: characterization 30 Imig JD, Navar LG, Roman RJ, Reddy KK, Falck JR. Actions of epoxygenase of an adipocyte-derived negative inotropic activity. FASEB J 2006; 20:1653–1659. metabolites on the preglomerular vasculature. J Am Soc Nephrol 1996; 7: 5 Lamounier-Zepter V, Look C, Alvarez J, Christ T, Ravens U, Schunck WH et al. 2364–2370. Adipocyte fatty acid-binding protein suppresses cardiomyocyte contraction: 31 Katoh T, Takahashi K, Capdevila J, Karara A, Falck JR, Jacobson HR et al. Glomerular a new link between obesity and heart disease. Circ Res 2009; 105:326–334. stereospecific synthesis and hemodynamic actions of 8,9-epoxyeicosatrienoic 6 Hertzel AV, Bernlohr DA. The mammalian fatty acid-binding protein multigene acid in rat kidney. Am J Physiol 1991; 261: F578–F586. family: molecular and genetic insights into function. Trends Endocrinol Metab 32 Imig JD. Epoxides and soluble in cardiovascular physiology. 2000; 11: 175–180. Physiol Rev 2012; 92: 101–130.

International Journal of Obesity (2015) 755 – 761 © 2015 Macmillan Publishers Limited Eicosanoids and FABP4 in heart function V Lamounier-Zepter et al 761 33 Jiang H, McGiff JC, Fava C, Amen G, Nesta E, Zanconato G et al. Maternal and fetal 44 Xiao YF, Huang L, Morgan JP. Cytochrome P450: a novel system modulating Ca2+ epoxyeicosatrienoic acids in normotensive and preeclamptic pregnancies. channels and contraction in mammalian heart cells. J Physiol 1998; 508:777–792. Am J Hypertens 2013; 26:271–278. 45 Chen J, Capdevila JH, Zeldin DC, Rosenberg RL. Inhibition of cardiac L-type cal- 34 Minuz P, Jiang H, Fava C, Turolo L, Tacconelli S, Ricci M et al. Altered release of cium channels by epoxyeicosatrienoic acids. Mol Pharmacol 1999; 55:288–295. cytochrome p450 metabolites of arachidonic acid in renovascular disease. 46 Cao H, Sekiya M, Ertunc ME, Burak MF, Mayers JR, White A et al. Adipocyte lipid 2008; 51: 1379–1385. chaperone AP2 is a secreted adipokine regulating hepatic glucose production. 35 Theken KN, Schuck RN, Edin ML, Tran B, Ellis K, Bass A et al. Evaluation of Cell Metab 2013; 17: 768–778. cytochrome P450-derived eicosanoids in humans with stable atherosclerotic 47 Burton PB, Hogben CE, Joannou CL, Clark AG, Hsuan JJ, Totty NF et al. Heart fatty 222 – cardiovascular disease. 2012; :530 536. acid binding protein is a novel regulator of cardiac myocyte hypertrophy. Biochem 36 Ward NC, Croft KD, Blacker D, Hankey GJ, Barden A, Mori TA et al. Cytochrome Biophys Res Commun 1994; 205: 1822–1828. P450 metabolites of arachidonic acid are elevated in patients compared 48 Widstrom RL, Norris AW, Spector AA. Binding of cytochrome P450 121 – with healthy controls. Clin Sci 2011; : 501 507. and lipoxygenase pathway products by heart fatty acid-binding 37 Ramirez CE, Shuey MM, Milne GL, Gilbert K, Hui N, Yu C et al. Arg287Gln variant of protein. Biochemistry 2001; 40: 1070–1076. EPHX2 and epoxyeicosatrienoic acids are associated with insulin sensitivity in 49 Hirose M, Sugiyama S, Ishida H, Niiyama M, Matsuoka D, Hara T et al. Structure of humans. Prostaglandins Other Lipid Mediat 2014; e-pub ahead of print 28 August the human-heart fatty-acid-binding protein 3 in complex with the fluorescent 2014; doi:10.1016/j.prostaglandins.2014.08.001. probe 1-anilinonaphthalene-8-sulphonic acid. J Synchr Radiat 2013; 20:923–928. 38 Moffat MP, Ward CA, Bend JR, Mock T, Farhangkhoee P, Karmazyn M. Effects of 50 Widstrom RL, Norris AW, Van Der Veer J, Spector AA. Fatty acid-binding proteins epoxyeicosatrienoic acids on isolated hearts and ventricular myocytes. inhibit hydration of epoxyeicosatrienoic acids by soluble epoxide hydrolase. Am J Physiol 1993; 264: H1154–H1160. Biochemistry 2003; 42: 11762–11767. 39 Rastaldo R, Paolocci N, Chiribiri A, Penna C, Gattullo D, Pagliaro P. 51 Kenchaiah S, Evans JC, Levy D, Wilson PW, Benjamin EJ, Larson MG et al. Obesity Cytochrome P-450 metabolite of arachidonic acid mediates -induced and the risk of heart failure. N Engl J Med 2002; 347: 305–313. negative inotropic effect. Am J Physiol Heart Circ Physiol 2001; 280: H2823–H2832. 52 Kenchaiah S, Sesso HD, Gaziano JM. Body mass index and vigorous physical 119 – 40 Gross GJ, Hsu A, Falck JR, Nithipatikom K. Mechanisms by which epoxyeicosa- activity and the risk of heart failure among men. Circulation 2009; :44 52. trienoic acids (EETs) elicit cardioprotection in rat hearts. J Mol Cell Cardiol 2007; 42: 53 Alpert MA. Obesity cardiomyopathy: pathophysiology and evolution of the clin- 321 – 687–691. ical syndrome. Am J Med Sci 2001; :225 236. 41 Nithipatikom K, Moore JM, Isbell MA, Falck JR, Gross GJ. Epoxyeicosatrienoic acids 54 Peterson LR, Herrero P, Schechtman KB, Racette SB, Waggoner AD, Kisrieva-Ware in cardioprotection: ischemic versus . Am J Physiol Heart Circ Z et al. Effect of obesity and insulin resistance on myocardial substrate Physiol 2006; 291: H537–H542. metabolism and efficiency in young women. Circulation 2004; 109: 2191–2196. 42 Lee HC, Lu T, Weintraub NL, VanRollins M, Spector AA, Shibata EF. Effects of 55 Poirier P, Giles TD, Bray GA, Hong Y, Stern JS, Pi-Sunyer FX et al. Obesity and epoxyeicosatrienoic acids on the cardiac sodium channels in isolated rat cardiovascular disease: pathophysiology, evaluation, and effect of weight loss. ventricular myocytes. J Physiol 1999; 519:153–168. Arterioscler Thromb Vasc Biol 2006; 26: 968–976. 43 Lu T, Hoshi T, Weintraub NL, Spector AA, Lee HC. Activation of ATP-sensitive K(+) 56 Zhou YT, Grayburn P, Karim A, Shimabukuro M, Higa M, Baetens D et al. Lipotoxic channels by epoxyeicosatrienoic acids in rat cardiac ventricular myocytes. heart disease in obese rats: implications for human obesity. Proc Natl Acad Sci USA J Physiol 2001; 537:811–827. 2000; 97:1784–1789.

© 2015 Macmillan Publishers Limited International Journal of Obesity (2015) 755 – 761